Discursive Pragmatics Handbook of Pragmatics - Highlights - 2011

Discursive Pragmatics
Volume 8
Edited by Jan Zienkowski, Jan-Ola stman and Jef Verschueren
Handbook of Pragmatics Highlights (HoPH)
Editors
Jef Verschueren
University of Antwerp
Jan-Ola stman
University of Helsinki
The ten volumes of Handbook of Pragmatics Highlights focus on the
most salient topics in the feld of pragmatics, thus dividing its wide
interdisciplinary spectrum in a transparent and manageable way.
Eachvolume starts with an up-to-date overview of its feld of interest
and brings together some 1220 entries on its most pertinent aspects.
Since 1995 the Handbook of Pragmatics (HoP) and the HoP Online (in
conjunction with the Bibliography of Pragmatics Online) have provided
continuously updated state-of-the-art information for students and
researchers interested in the science of language in use. Their value as a
basic reference tool is now enhanced with the publication of a topically
organized series of paperbacks presenting HoP Highlights.
Whether your interests are predominantly philosophical, cognitive,
grammatical, social, cultural, variational, interactional, or discursive, the
HoP Highlights volumes make sure you always have the most relevant
encyclopedic articles at your fngertips.
For an overview of all books published in this series, please see
http://benjamins.com/catalog/hoph
Edited by
Jan Zienkowski
Jan-Ola stman
University of Helsinki
Jef Verschueren
John Benjamins Publishing Company
Amsterdam / Philadelphia
Library of Congress Cataloging-in-Publication Data
Discursive pragmatics / edited by Jan Zienkowski, Jan-Ola stman, Jef Verschueren.
p. cm. (Handbook of Pragmatics Highlights, issn 1877-654X ; v. 8)
Includes bibliographical references and index.
1. Discourse analysis. 2. Pragmatics. 3. Semantics. I. Zienkowski, Jan. II. stman, Jan-Ola. III.
Verschueren, Jef.
P302.D569 2011
401.45--dc22 2011012728
isbn 978 90 272 0785 2 (Pb ; alk. paper)
isbn 978 90 272 8915 5 (Eb)
:o:: John Benjamins B.V.
No paii of ihis book may be iepioduced in any foim, by piini, phoiopiini, miciolm, oi any oihei
means, wiihoui wiiiien peimission fiom ihe publishei.
John Benjamins Publishing Co. P.O. Box 6::( :o:o mv Amsieidam Te Neiheilands
John Benjamins Noiih Ameiica P.O. Box :,,: Philadelphia v. :::8-o,: Us.
The paper used in this publication meets the minimum requirements
of American National Standard for Information Sciences Permanence
of Paper for Printed Library Materials, ANSI z39.48-1984.
8
TM
Table of contents
Preface to the series x:::
Acknowledgements xv
Discursive pragmatics: A platform for the pragmatic study of discourse :
Jan Zienkowski
Appraisal :|
Peter R.R. White
:. Introduction :|
i. Overview :o
i.: Attitude the activation of positive or negative positioning :o
i.:.: Afect :o
i.:.i Judgement :o
i.:. Appreciation :
i.:.| Modes of activation direct and implied :
i.:., Typological criteria :8
i.:.o Te interplay between the attitudinal modes :
i.i Intersubjective stance io
. Attitudinal assessment a brief outline i:
.: Afect i:
.i Judgement ii
. Appreciation i,
|. Engagement: An overview i
|.: Dialogic contraction and expansion i8
|.i Further resources of dialogic expansion i
|.i.: Acknowledge o
|.i.i Entertain o
|. Further resources of dialogic contraction o
|..: Pronounce o
|..i Concur :
|.. Disclaim (Deny and Counter) :
VI Discursive Pragmatics
|..| Disclaim: Deny (negation) :
|.., Disclaim: Counter i
|.| Engagement resources summary
,. Conclusion
Cohesion and coherence
Wolfram Bublitz
:. Introduction
i. Focus on form: Cohesion 8
. Cohesion as a condition for coherence |o
|. Focus on meaning: Connectivity |i
,. Semantic connectivity as a condition for coherence |
o. Coherence: A general view ||
. A hermeneutic, context and interpretation based view of coherence |,
8. Coherence as a default assumption |
. Perspectives |
Critical Linguistics and Critical Discourse Analysis ,o
Ruth Wodak
:. Defnitions ,o
i. Historical note ,
. Principles of CL ,
|. Trends ,,
|.: Social Semiotics ,,
|.i Orders of discourse and Foucauldian poststructuralism ,8
|. Te socio-cognitive model oo
|.| Discourse-Historical approach o:
|., Lexicometry oi
|.o Lesarten approach o|
,. Conclusion o,
nonciation: French pragmatic approach(es) :
Marjut Johansson & Eija Suomela-Salmi
:. Introduction :
i. Historical overview from the pre-theoretical to the present phase i
i.: Origins and the pre-theoretical phase i
i.i First phase: Forerunners |
i.i.: Charles Bally (18651947) |
i.i.i Gustave Guillaume (18831960) 8
i. Second phase: Main theoretical foundation 8o
i..: Emile Benveniste (19021976) 8:
Table of contents VII
i.| Tird phase: Modern developments 8,
i.|.: Antoine Culioli (born in 1924) 8,
i.|.i Oswald Ducrot (born in 1930) 88
i.|. Jacqueline Authier-Revuz
(born in 1940) o
. Some basic notions i
.: Enunciation and enunciator i
.i Situation/Context
. Subjectivity and deixis
.| Reported speech |
., Modality and modalization ,
.o Modalities of enunciation
(modalits dnonciation) o
. Utterance modalities (modalits dnonc)
Figures of Speech :oi
Manfred Kienpointner
:. Introduction :oi
i. Ancient rhetoric :oi
. Contemporary treatments of FSP :o|
.: Defnition of FSP :o|
.i Classifcation of FSP :o8
|. Across the lines of discipline: Te cognitive and communicative
role of FSP :::
Genre ::
Anna Solin
:. Introduction ::
i. Historical precedents :io
. Genre research in language studies :i:
.: Sydney School :i:
.i New Rhetoric :i
. English for Specifc Purposes :i,
|. Issues and debates :i
|.: Genre as class :i
|.i Stability of genres :i
Humor :,
Salvatore Attardo
:. Introduction and defnition :,
i. Referential and verbal humor :,
VIII Discursive Pragmatics
. Semantics :o
.: Te isotopy-disjunction model :o
.i Te script-based semantic theory of humor :
. Longer texts :8
|. Te cooperative principle and humor :8
|.: Grice and Gricean analyses :8
|.i Humor as non-bona-fde communication :8
|. Relevance-theoretic approaches to humor :
|.| Informativeness approach to jokes :|:
|., Two-stage processing of humor :|i
,. Conversation analysis :|
,.: Canned jokes in conversation :|
,.:.: Preface :|
,.:.i Telling :|
,.:. Response :||
,.i Conversational humor :||
,.i.: Functional conversational analyses :||
,.i.i Quantitative conversational analyses :|,
o. Sociolinguistics of humor :|
o.: Gender diferences :|
o.i Ethnicity and humor :|8
. Computational humor :|8
8. Cognitive linguistics and humor :|
. Conclusion :|
Intertextuality :,o
Stef Slembrouck
:. From literature to text as a productivity which inserts
itself into history :,o
i. Text linguistics on textuality :,
. Dialogism and heteroglossia in a social-diachronic
theory of discourse :,8
|. Voloinov, pragmatics and conversation analysis:
Sequential implicativeness and the translation
of the others perspective :oi
,. Synoptic and participatory views of human activity: Bakhtin, Bourdieu,
sociolinguistic legitimacy (and the body) :o,
o. Natural histories of discourse: Recontextualization/entextualization
and textual ideologies :o
Table of contents IX
Manipulation :o
Paul Chilton
:. Te ancient technique of rhetoric :o
i. Te twentieth-century nightmare of
thought control :
. Manipulation is not inherent in language structure :
|. So lets look at thought and social action :8:
|.: Drumming it in :8:
|.i Ideas that spread :8i
,. What might override the cheat-checker? :8|
o. Conclusion: Manipulation and counter-manipulation :8o
Narrative :o
Alexandra Georgakopoulou
:. Narrative as a mode of communication :o
i. Referential properties :i
. Textual properties :
.: Narrative organization :
.i Narrative evaluation :o
|. Contextual properties ioo
Polyphony io8
Eddy Roulet
:. Preliminaries io8
i. Polyphony in Bakhtins work io8
. Polyphony in Ducrots work i:i
|. Te description of the polyphonic organization
of discourse i:,
,. Te interrelations between polyphony and other
dimensions of discourse structures i:8
o. Conclusion ii:
Pragmatic markers ii
Karin Aijmer & Anne-Marie Simon-Vandenbergen
:. Te tradition and the present state of research on
pragmatic markers ii
i. Defning the feld ii|
. Te terminology: Pragmatic marker or discourse marker? iio
X Discursive Pragmatics
|. Classifcation ii
,. Pragmatic markers and multifunctionality ii8
o. Teoretical approaches to the study of pragmatic markers ii
. Methodology i:
8. Pragmatic markers in the languages of the world ii
. Te diachronic study of pragmatic markers i|
:o. Te contrastive study of pragmatic markers i|
::. Pragmatic markers in translation studies io
:i. Pragmatic markers in native versus non-native
speaker communication io
:. Pragmatic markers and sociolinguistic aspects i
:|. Pragmatic markers and the future i8
Public discourse i|8
Srikant Sarangi
:. Introduction i|8
:.: Multiple readings of publicness i|
i. Te situation-talk dialectic: public as a feature of setting vs. public
as a feature of talk i,o
i.: (Socio)linguistic markers of public discourse i,o
i.i Interaction-based approach i,:
. Gofman and the public order i,i
|. Habermas and the public sphere i,
,. Transformation of the public sphere: Public discourse
as mediated communication i,o
,.: Te states role in the confation of public and private
discourses in contemporary societies i,8
,.i Surveillance and control: Information exchange
as a site of struggle i,
o. Pragmatic theories of information exchange and the public sphere:
Towards a social pragmatics ioo
Text and discourse linguistics ioo
Jan-Ola stman & Tuija Virtanen
:. On terminology ioo
i. Historical overview io
. Important felds of study io
.: Information structure io
.i Cohesion io
. Coherence i:
Table of contents XI
.| Grounding i
., Discourse types and genres i|
|. Other trends io
,. Applications i
,.: Practical applications i8o
,.i Acquisitional and diachronic studies i8o
o. Final remarks i8:
Text linguistics i8o
Robert de Beaugrande
:. Te rise of text linguistics i8o
i. Some central issues i|
Index i
Preface to the series
In 1995, the frst installments of the Handbook of Pragmatics (HoP) were
published. Te HoP was to be one of the major tools of the International Pragmatics
Association (IPrA) to achieve its goals (i) of disseminating knowledge about prag-
matic aspects of language, (ii) of stimulating various felds of application by making
this knowledge accessible to an interdisciplinary community of scholars approaching
the same general subject area from diferent points of view and with diferent methodolo-
gies, and (iii) of fnding, in the process, a signifcant degree of theoretical coherence.
Te HoP approaches pragmatics as the cognitive, social, and cultural science of
language and communication. Its ambition is to provide a practical and theoretical
tool for achieving coherence in the discipline, for achieving cross-disciplinary intel-
ligibility in a necessarily diversifed feld of scholarship. It was therefore designed
to provide easy access for scholars with widely divergent backgrounds but with
converging interests in the use and functioning of language, in the topics, traditions,
and methods which, together, make up the broadly conceived feld of pragmatics. As
it was also meant to provide a state-of-the-art report, a fexible publishing format was
needed. Tis is why the print version took the form of a background manual followed
by annual loose-leaf installments, enabling the creation of a continuously updatable
and expandable reference work. Te fexibility of this format vastly increased with
the introduction of an online version, the Handbook of Pragmatics Online (see
www.benjamins.com/online).
While the HoP and the HoP-online continue to provide state-of-the-art informa-
tion for students and researchers interested in the science of language use, this new
series of Handbook of Pragmatics Highlights focuses on the most salient topics in the
feld of pragmatics, thus dividing its wide interdisciplinary spectrum in a transparent
and manageable way. Te series contains a total of ten volumes around the following
themes:
Key notions for pragmatics
Pragmatics and philosophy
Grammar, meaning and pragmatics
Cognition and pragmatics
Society and language use
Culture and language use
Te pragmatics of variation and change
XIV Discursive Pragmatics
Te pragmatics of interaction
Discursive pragmatics
Pragmatics in practice
Tis topically organized series of paperbacks, each starting with an up-to-date over-
view of its feld of interest, each brings together some 12-20 of the most pertinent HoP
entries. Tey are intended to make sure that students and researchers alike, whether
their interests are predominantly philosophical, cognitive, grammatical, social, cul-
tural, variational, interactional, or discursive, can always have the most relevant ency-
clopedic articles at their fngertips. Afordability, topical organization and selectivity
also turn these books into practical teaching tools which can be used as reading mate-
rials for a wide range of pragmatics-related linguistics courses.
With this endeavor, we hope to make a further contribution to the goals underly-
ing the HoP project when it was frst conceived in the early 1990s.
Jan-Ola stman (University of Helsinki) &
Jef Verschueren (University of Antwerp)
Acknowledgements
A project of the HoP type cannot be successfully started, let alone completed, without
the help of dozens, even hundreds of scholars. First of all, there are the authors
themselves, who sometimes had to work under extreme conditions of time pressure.
Further, most members of the IPrA Consultation Board have occasionally, and
some repeatedly, been called upon to review contribu tions. Innumerable additional
scholars were thanked in the initial versions of handbook entries. All this makes
the Handbook of Pragmatics a truly joint endeavor by the pragmatics community
world-wide. We are greatly indebted to you all.
We do want to specifcally mention the important contributions over the years
of three scholars: the co-editors of the Manual and the frst eight annual install-
ments, Jan Blommaert and Chris Bulcaen were central to the realization of the
project, and so was our editorial collaborator over the last four years, Eline Ver-
sluys. Our sincerest thanks to all of them.
Te Handbook of Pragmatics project is being carried out in the framework of the
research program of the IPrA Research Center at the University of Antwerp. We are
indebted to the university for providing an environment that facilitates and nurtures
our work.
Jan-Ola stman (University of Helsinki) &
Jef Verschueren (University of Antwerp)
Discursive pragmatics
A platform for the pragmatic study of discourse
Jan Zienkowski
Most introductions to the interdisciplinary felds of discourse analysis and linguistic
pragmatics begin by emphasizing that there is no consensus on the meaning of these
terms. Neither discourse nor pragmatics are straightforward concepts and it may
be correct to suggest that only the most vague and the most general defnitions are
able to unite the variety of disciplines, researchers, research programmes, heuristics,
methodologies, objects of investigation, and terminologies lumped together under
these headings. A book title such as discursive pragmatics might therefore be inter-
preted as an oxymoron consisting of two signifers whose referents are diverse in both
quantitative and qualitative senses of the term. In this sense, the usefulness i.e. the
pragmatic value of the label discursive pragmatics may be called into question. Yet,
I will argue that this notion holds great potential as an instrument for establishing a
platform for inter-disciplinary and inter-theoretical cross-fertilization. In this sense,
the usefulness of the term resides in its communicative potential for researchers who
tend to view their activities as either pragmatic or discursive ways of approaching
empirical language use.
Many of the topics, themes, theories and perspectives discussed in this volume
would also ft in publications related to rhetoric, cognitive linguistics, semantics,
studies on language acquisition, sociolinguistics, communication studies, philosophy,
literature, and many other (inter)disciplinary felds. But whatever their focus, all of
the contributions collected in this volume contribute in one way or another to the
linguistic or discursive turns that have taken place throughout the humanities and
social sciences since the second half of the twentieth century. Moreover, they are all
products of an increasing pre-occupation with pragmatic concerns related to func-
tional and communicative language use conceived in terms of interactional processes
and context generation. Tis point of convergence has generated an ever-growing
number of eclectic studies of real-life discourse(s) described in terms of empirically
observable language use. Such discourses may be defned in terms of topic (such as
racism, ecology, doctor/patient interaction), genre (such as literature, scientifc publi-
cations, newspaper articles), or function (such as narrative, manipulative, rhetorical),
depending on the researchers main interests.
2 Jan Zienkowski
In this introduction, I will trace the pragmatic features of the perspectives and
approaches to discourse as presented by the various contributors to this book. How-
ever, this collective volume is far from exhaustive. I will argue that approaches to
discourse such as those presented under the headings of post-structuralist discourse
analysis, discursive psychology, or conversation analysis would also belong here. In
some cases, there were good reasons to deal with them in one of the other volumes
of the Handbook of Pragmatics Highlights series In other cases, they represent gaps to
be flled in future updates of the Handbook. Tis introduction does not explore the
various potential areas of application of discursive pragmatics.
In their article on Text and discourse linguistics, Jan-Ola stman and Tuija Virtanen
point out that diferentiating between discourse analysis, text linguistics, pragmatics,
semiotics, and even linguistics has become a feld of inquiry in itself. Neither their article
nor this volume is intended to present an overview of this meta-feld. Recognizing that
the term discourse analysis is used in a rather restricted sense by some scholars, they
advocate a perspective in which the notion functions as an umbrella term for all issues
that have been dealt with in the linguistic study of text and discourse. With respect to
the relationship between pragmatics and discourse studies, they note that a discussion
has emerged around the question as to which notion is more general. Even though
they prefer to think of pragmatics as the more general term, they emphasize that it
is more productive to consider the debate as a matter of perspectives: Discourse and
pragmatics have the same felds of interest, but diferent aspects in focus. Tus, discourse
will typically require larger stretches of text or conversation, whereas for pragmatics this
is not necessarily the case.
Such a perspective view is also relevant with respect to other approaches con-
tained in this volume. For instance, Critical linguistics and critical discourse studies as
discussed by Ruth Wodak can easily be viewed as constituting a perspective as well.
Wodak confates CL and CDA on pragmatic grounds and goes on to characterize
CDA basically as a research program instead of a theory or a methodology. For her,
CDA is frst and foremost an approach the specifcity of which can be defned in terms
of a concern with power as a central condition in social life that may be indexed in
language use. But ultimately, CDA is considered to be an approach characterised by
interdisciplinary and problem-oriented principles. In the same sense, it is open to a
multiplicity of theories, methodologies, and objects of investigation.
Interestingly, in his article on Text linguistics, Robert de Beaugrande describes the
development of this approach in terms of three historical stages. He points out how
text linguistics was initially an attempt to extend grammatical principles to texts and
how from the seventies on this approach became increasingly preoccupied with tex-
tuality as a structure with both linguistic and social aspects. Te most recent stage
witnessed an increasing focus on textualization processes in the actual production and
reception of texts. He goes on to suggest that these developments allow for a prag-
matics which includes a critical view of communication as an ongoing interaction
Discursive pragmatics 3
whereby the signifcance of a situation (real or hypothetical) is being negotiated, speaking
turns are assigned and relations of power or solidarity are enacted. Te convergences
with (critical) discourse analysis and pragmatics become even more obvious when he
points out that today, the text is widely defned as an empirical communicative event
given through human communication rather than specifed by a formal theory.
At frst sight, the contributions that go under the headers Text and discourse lin-
guistics, Critical discourse analysis and critical linguistics, and Text linguistics seem to
situate themselves at a higher level of abstraction than the contributions that focus on
particular aspects of discourse such as the French notion of nonciation, the principles
of Polyphony, Intertextuality, Cohesion and coherence, the concepts of Appraisal, Public
discourse, Pragmatic markers, Genre, Figures of speech, or notions such as Narrative,
Manipulation, or Humor. However, if one takes a closer look, it becomes clear that
these labels do not merely designate an object of investigation for discursive pragmatic
approaches to language. Rather, researchers trained in a variety of disciplines deploy
them as principles structuring our language use at various levels of abstraction. As
such, they also provide perspectives or points of view from which phenomena related
to language use can be approached.
Te linguistic or discursive turn in the social sciences and in the humanities in
general is not merely a turn towards new objects of investigation that had been pre-
viously neglected. Rather, it is a turn which implies a change in perspective fore-
grounding and backgrounding both new and old phenomena for empirical research
with a linguistically inspired toolbox at hand. Tis process was ofen realized by
means of a double movement whereby concepts were decontextualized from their
normal use and metaphorically applied to new areas of investigation. Tis is def-
nitely the case with notions such as Narrative, Enonciation, Polyphony, Intertextuality,
Cohesion and Coherence.
Marjut Johansson and Eija Suomela-Salmi explicitly describe French enunciation
theory or enunciative linguistics as a set of several diferent diversifed approaches to
pragmatic questions. Tese questions are not merely linguistic but are relevant to the
issue of subject formation as well. In spite of its diversity, it is possible to identify three
stages in the development of enunciative pragmatics which culminate in Benvenistes
statement that nothing is in language that has not been before in discourse. Tis quote
bears witness to the struggle of pragmatically oriented authors to emancipate them-
selves with respect to dominant perspectives on language which tended to focus on
the formal rules and structures that supposedly work at a deeper level of semiotic
organization than individualized utterances, parl es, or noncs. Ultimately, French
pragmaticians do not merely study individual noncs and their indexical functions
in relation to the I or to other aspects of contextual reality. Rather, they arrived at a
view of the human subject as a subject split by the discourse it produces. Tis view was
most clearly expressed in Ducrots work by means of his integration of Bakhtinian dia-
logism and French enunciation theory. Johansson and Eija point out that Jacqueline
4 Jan Zienkowski
Authier-Revuz voiced similar points of view underscored by an integration of Lacanian
psycho-analysis into the theoretical mixture. Interestingly but not surprisingly, some
of the concepts central to French linguistic pragmatics proved to be important to the
Anglo-American strands as well: this goes for notions such as deixis, context, reported
speech and modality.
Enunciation moved from being strictly an empirical object of investigation towards
a perspective in which even the subject should be seen as an efect of discourse under-
stood as a set of enunciations. A similar movement may be observed in the intercon-
nected theories of Polyphony discussed by Eddy Roulet and Intertextuality discussed by
Stef Slembrouck. To Bakhtin, the notion of polyphony is just one type of dialogism an
instance of multi-voiced discourse. It does not refer to an actual dialogue between two
speakers but rather to the fact that even monological texts incorporate and articulate a
multiplicity of voices through various linguistic forms and processes. Whereas Bakhtin
illustrates his point frst and foremost with forms of reported speech, Ducrot extends
the dialogic and polyphonic principles to French linguistics in general. He investigates
how polyphony is realized by means of a variety of linguistic forms and pragmatic pro-
cesses including irony, negation, negative polarity, discourse markers, and presuppo-
sitions. Polyphony is therefore one of the metaphors that allow for a perspective on
discourse in which the unity of the speaking subject is challenged without falling into
the trap of removing all of its agency in the process.
Slembrouck points out that it is possible to consider intertextuality as a matter of
those factors which make the utilizaton of one text dependent upon knowledge of one
or more previously encountered texts. In this sense, intertextuality is a phenomenon or
a function of discourse which fnds itself at the same level of abstraction as cohesion,
coherence, intentionality, acceptability, informativity and situationality. Intertextuality
is then an object of investigation rather than a principle constitutive of a perspective
on (discursive) reality. Such is the view of most text linguistic approaches. A broader
notion of intertextuality is to be found in the writings of Kristeva and Bakhtin whose
works have been rediscovered by one of the main voices in the feld of critical discourse
analysis Norman Fairclough. Here intertextuality gains the status of a perspective on
language use. As Slembrouck points out, the Bakhtinian/Kristevian notion of intertex-
tuality is broader than the text linguistic one. In addition to the manifest type of inter-
textuality which is a shared point of interest with text linguisistics, Faircloughs notion
of intertextuality is a constitutive one. In a Foucaultian argument, Fairclough claims
that intertextual analysis draws our attention to: the orders of discourse that consist
of confgurations of conventionalised practices at the disposal of producers and con-
sumers of texts; to the way texts depend upon society and history; and to the way
texts transform social and historical resources. Put diferently, intertextuality becomes
a metaphor constitutive of a perspective on social relations and historical develop-
ment in general. As Slembrouck points out, this is the main reason why the notion of
intertextuality can be linked to the concepts of hegemony, polyphony and heteroglossia
in critical discourse analysis. However, one may ask in what sense we are dealing with
a pragmatic mode of investigation when the notion of intertextuality is taken to such
abstract heights. Slembrouck argues that teasing out the various dimensions of inter-
textuality both its manifest linguistic forms and its constitutive senses allows us
to take the dialogic beyond the boundedness of recognized domains of language study.
Slembrouck goes on to draw out the implications of this view on intertextuality and
concludes by shedding a rather pragmatic light on the issue. Arguing that recent work
on (re)contextualization and (en)contextualization processes in language essentially
deals with the same phenomena as the issues discussed in Kristevas original article on
intertextuality, he claims that (inter)text is a contextually grounded process whereby
publicly recognized textualities count as by-products of situated practice and cannot be
divorced from language ideologies. Moreover, intertextuality is about how history is
inscribed into situated practice. Te pragmatic dimension of CDA thus resides in its
intersubjective stance, in its emphasis on the importance of socio-political and textual
context, and in the emphasis on the process-based nature of discourse.
Like CDA, the study of discourse as performed within the framework of Appraisal
theory is informed by the principles of SFL (Systemic Functional Linguistics) as pro-
posed by Halliday. It is a decidedly functional perspective on language use in the sense
that it holds that linguistic phenomena can best be explained by reference to the social
functions performed by language and by reference to the functional demands placed
upon language by its users. In contrast to the approaches to discourse we have encoun-
tered thus far, the appraisal framework is rather typological. It categorizes the various
functions that a variety of linguistic forms such as adjectives, nouns, adverbs or verbs
may perform. Also, it has not engaged with social and political theory to the same
extent as critical discourse analysis, even though both perspectives are informed by
Bakhtinian insights and SFL principles. On the other hand, appraisal does deal with
the social function of texts in the sense that it is concerned with the systematic analy-
sis of evaluation and stance as they operate in whole texts and in groupings of texts. It is
concerned with the social function of these resources, not simply as the means by which
individual speakers/writers express their feelings and take stands, but as the means by
which they engage with socially-determined value positions and thereby align and dis-
align themselves with the social subjects who hold these positions. White argues that the
appraisal perspective is pragmatic in the sense that it recognizes that some evaluations
are not marked explicitly in a text. In such cases, one has to take recourse to aspects of
context in order to understand the evaluative dimension of a text. Tis view of context
is defnitely not the most radical one in the landscape of linguistic pragmatics. On the
other hand, there is a clear pre-occupation with pragmatic concerns: the functions of
actual language use, the role of context and form with respect to the realization of these
functions; and an emphasis on evaluation as an ongoing and intersubjective process.
6 Jan Zienkowski
Te article on Public discourse by Srikant Sarangi is not merely about public texts
in the sense of a genre or a set of discursive genres in a reifed public sphere. Rather,
it is about publicness defned in terms of social processes of talk and text in the public
domain which have institutionally ratifed consequences. Moreover, discourse is taken
to mean a stretch of talk or text (including semiotic icons) as well as a form of knowl-
edge and the social processes of production and consumption of such knowledge in the
Foucauldian sense. Afer presenting three dichotomies (private/public; expert/lay
knowledge; microcosmic/macrocosmic) that structure dominant understandings of
the notion of publicness, he moves on to consider various sociolinguistic and interac-
tion-based approaches to public discourse. In addition to work in conversation analy-
sis and rhetorics, he explores Gofmans interactional concept of the public order. On
the other hand, Sarangi contrasts these approaches with Habermas philosophical and
normative perspective on the public sphere. He describes pragmatics as a concern with
meaning-making against a set of parameters such as power, status, rights and obliga-
tions and claims that as such, various pragmatic theories of human communication
can be reassessed to account for public discourse. To Sarangi, both pragmatic theories
and theories of the public sphere are ultimately concerned with information exchange.
More specifcally, he points out that information exchange lies at the heart of the three
most dominant models of pragmatic inquiry: the cooperative principle and conventional/
conversational implicatures, face wants and positive/negative politeness strategies, and
principles of relevance and message processing. Sarangi argues that these theories can
be extended to theories of the public domain as a domain in which concerns related to
information exchange are intertwinted with issues related to role-relationships, iden-
tity construction, strategic goal imposition and so on. He claims that this would allow
for a research programme that will blend well with the Habermasian characterisation
of public discourse as systematically distorted communication, without subscribing to his
universal pragmatics model.
Public discourse is ofen described as a particular genre of discourse. However,
discursive pragmatic approaches to discourse have problematized the notion of genre
itself. Anna Solin points out that the notion is traditionally associated with the idea
that texts (and other cultural products) can be grouped into types or classes according to
formal and/or functional similarities. However, Solin points out that it is also possible
to consider genre not merely as forms, but as social and symbolic action. In her over-
view of genre studies, she discusses three contemporary approaches to genre research.
Te frst one is the Sydney school which is based on SFL principles. It inherits the
SFL preference for functionalist typologies and defnes genres as stage goal-oriented
social processes. Te specifc identity of a genre is defned by recourse to text-internal
linguistic criteria. Tere are strong afnities with text linguistics. At the same tyme
the Sydney approach is both pedagogical and political in the sense that its researchers
assumed that it was possible to teach a number of key genres to social groups with less
estensive discursive resources in particular societies (e.g. migrants and working class
students). A more thorough familiarity with key genres was supposed to help them in
their emancipation processes. Te Sydney approach was therefore decidedly pragmatic
in its functional emphasis and in its goals with respect to increasing peoples language
awareness with respect to genre conventions. Te New Rhetoric approach is even more
pragmatic. Following Bakhtin, genres are perceived as intertextually related to other
genres. Moreover, they are defned as particular types of social action: a rhetorically
sound defnition of genre must be centred not on the substance or the form of discourse
but on the action it is used to accomplish. At frst sight, this does not go against the
idea of an SFL goal-oriented attitude to genre. However, within New Rhetoric, social
actions always take place within particular communities. Consequently, genres can
only be investigated and taught if one ethnographically understands the communal
and institutional contexts in which they are deployed. Within this perspective, genre
education on the basis of formal linguistic principles only is an absurdity: if genres
are responses to contexts, can they be learned at all out of context?. Drawing on the eth-
nographic insights of New Rhetoric and on the text linguistics of the Sydney school,
English for Specifc Purposes is characterized by its pedagogical agenda: to assist non-
native English speakers in mastering academic and professional genres. It defnes genres
as sets of intersubjective and goal-oriented functions or as communicative actions. As
such, we are dealing with a pragmatic perspective.
Troughout her discussion of Narrative, Alexandra Georgakopoulou describes
how the reifcation of narrative has nonetheless increasingly been shied away from, as
attention has shifed from narrative as text to narrative as social practice. She describes
how studies move towards viewing narrative as a process of the participants joint con-
struction, negotiation and contestation of alternative interpretative viewpoints on the
events related as well as on the local relevance of the storytelling activity itself. Interest-
ingly, narrative as a perspective is also engaging with the principle of subjectivity.
Here, it is not so much French pragmatics which serves as a source of inspiration for
the exploration of the metaphorical split subject, but Gofmans notion of footing and
his corresponding view on the self.
Te shif towards a pragmatic and functional consideration of discourse in terms
of (inter)subjective social action and processes, is one of the factors contributing to
the success of the linguistic turn across the disciplines. In this sense, I do not think
of pragmatics as the study of a limited set of phenomena such as speech acts or the
establishment of cohesion or coherence. Rather, linguistic pragmatics is a perspective
on and a going concern with language use. Pragmatics-oriented researchers investi-
gate empirical data of language-related actions and processes without losing sight of
the various contextual layers that play a role in these actions and processes. Pragmatic
interpretations of abstract concepts such as narrative, dialogue or genre thus allow
for the empirical analysis of large-scale processes which leave material traces at the
8 Jan Zienkowski
micro-level of language. Put diferently, they allow for a de-reifcation of the things we
consider discourse to be.
In his article on Cohesion and coherence, Wolfram Bublitz points out that both
notions refer to meaning and bear on relations of connectedness which may or may
not be linguistically encoded. Whereas cohesion refers to inter-sentential semantic
relations which link current items with preceding or following onesby lexical and struc-
tural means, coherence is a cognitive category depending on the language users
interpretation. Bublitz provides an overview of research on the relationship between
cohesion and coherence and arrives at a pragmatics-oriented perspective. He consid-
ers the establishment of coherence as an interactive achievement resulting from nego-
tiations within specifc socio-cultural settings. As such, coherence becomes a context
dependent, user oriented and comprehension based notion. Moreover, it is not texts
that cohere but rather people who make texts cohere. On the other hand, speakers and
writers can use linguistic cohesive tools to help create coherence and the associated
gestalts. Once again, we are faced with a pragmatic perspective on discourse.
Interestingly, the article of Karin Aijmer and Ann-Marie Simon- Vandenbergen
on Pragmatic markers illustrates how pragmatic perspectives do not necessarily
originate from macro-discursive concepts such as dialogue, narrative or intertex-
tuality. On the contrary, the notion of pragmatic markers is connected to a whole
range of associated concepts among which the notion of indexicality. As Aijmer and
Simon-Vandenbergen point out, the main function of pragmatic markers is indexical.
Tey point us to features of context we need to take into account in order to con-
nect discourse with relevant aspects of context. Context involves ideational relations
between ideas and propositions, relations with respect to how speech acts relate to pre-
ceding, following or intended actions, linguistic exchanges or turns, information, and
social relations between participants in discourse. Pragmatic markers index elements
which are less concrete than more prototypical deictic elements such as pronouns,
temporal or spatial expressions. But the pragmatic potential of discourse markers or
particles also resides in their refexivity: they comment on the utterance and thus assist
in the interpretation of that utterance. Aijmer and Simon-Vandenbergen point out that
pragmatic markers perform a meta-linguistic or meta-pragmatic role. Put diferently,
they assist in the (inter)subjective establishment of coherence. A theorization of meta-
pragmatic markers is therefore not restricted to the study of a particular linguistic
phenomenon but constitutes a perspective on the overall functioning of discourse.
Aijmer and Simon-Vandenbergen emphasize that pragmatic markers are diverse.
Connectives, modal particles, adverbs, interjections, routines, feedback signals, voca-
tives, disjuncts, approximators, conjunctions, and reformulations may all perform a
multiplicity of (meta)pragmatic functions. Moreover, they may be deployed in linguis-
tic theories and discursive perspectives ranging from speech act theory, conversation
analysis, discourse analysis, semantics, socio-linguistics and SFL. Te types of texts in
which they are studied also vary greatly. However, regardless of the specifc pragmatic
perspective in which pragmatic markers are studied, they force the researcher to face
questions with respect to pragmatic processes constitutive of (inter)subjectivity and
functional context-generation.
Of course, there are many other linguistic phenomena that are investigated in the
pragmatic study of discourse. Manfred Kienpointners article on Figures of speech draws
our attention to argumentative patterns and linguistic forms deployed in both ancient
rhetoric and in the contemporary study of discourse. He argues for a perspective on
fgures of speech that difers from the ancient deviation view in which the so-called
ornamental features of language e.g. metaphor and metonymy are considered to
be deviations of a supposedly pure form of expression. Kienpointner states that FSP
are not merely ornamental or aesthetic devices but inevitably shape our cognition and
culture-specifc views of reality. Moreover, they are the output of discourse strategies
which we use to select units from linguistic paradigms of diferent levels (phonetics/pho-
nology, morphology, syntax, semantics) to create texts (in the sense of both written and
spoken genres of discourse) which are adequate as far as their communicative purpose in
some context is concerned. He goes on to present a typical pragmatic argument against
structural typologies by stating that fgures of speech should be considered as linguistic
elements having certain communicative functions like clarifcation, stimulation of inter-
est, aesthetic and cognitive pleasure, modifcation of the cognitive perspective, intensifca-
tion or mitigation of emotions etc..
Paul Chilton touches upon the issue of rhetoric as well. He points out that the
theoretical question underlying the topic of manipulation amounts to does or can
language determine the thoughts you have?. For Chilton, linguistic structures are not
inherently deceptive or manipulative. He states that linguistic structures are just lin-
guistic structures which hold varying degrees of potential for manipulation by human
users with goal-directed intentions. Troughout his article, Chilton explores some
conditions which would allow for a manipulation of the ideas of language users: expo-
sure of a population to a repeated message; the wide-spread dispersal of a linguistically
encoded notion; and/or the malfunctioning or potential over-riding of the human
consistency-checking device. Troughout his argumenation, Chilton argues for a per-
spective on manipulation that we may call pragmatic. He writes that producers and
consumers of discourse can only be partly aware of what they are doing when they
engage in discourse. In this sense, all communication is manipulative to some degree
and not just some particular kind of communication that we call manipulative. Con-
sequently the issue of responsibility appears. In this context, Chilton points out that
the human mind has the capacity or the potential to inspect or introspect. Moreover,
this capacity to refect on their unconscious cognitions, communicative behavior and
physical action can be trained and encouraged by a variety of factors. One of these is
education and it is here that Chilton attributes a role to the enlightening potential of
10 Jan Zienkowski
critical approaches to discourse such as CDA. Increasing the awareness of speakers
and writers with respect to the various functions of discourse is probably not specifc
to CDA. Rather it is characteristic of discursive pragmatics in general.
Te last article in this reader deals with Humor and is written by Salvatore Attardo.
He explores various approaches that exhibit this phenomenon. He points out that lin-
guistics did not contribute much to research on humor until the eighties. Attardo starts
out with some semantic theories such as the isotopy-disjunction model, the script-
based semantic theory of humor, and the genertal theory of verbal humor. He then
moves towards a discussion of approaches who exhibit a more pragmatic point of view.
Tese include Gricean approaches to humor as it relates to conversational maxims and
principles of relevance, but also humor as it relates to informativeness and to the way it
is processed in psycholinguistic terms. Conversation analysis and sociolinguistics are
also incorporated in this overview. As such, Attordo concludes that the feld of humor
research is defnitely taking a pragmatic perspective.
At the beginning of this overview, I emphasised that the articles collected in
this volume on discursive pragmatics do not constitute an exhaustive collection of
the themes, issues, terminologies and perspectives which take pragmatic insights as a
point of departure and/or as a topic of investigation. For instance, this is the case with
respect to an approach such as discursive psychology where discourse is considered to
consist of text and talk as parts of social practices and where the relationship between
cognition and the use of discourse as conceptualised in cognitive psychology is called
into question. From a discursive psychological point of view both knowledge, reality
and cognition are discursively constructed through situated language use. In this sense,
we are dealing with a discursive pragmatic perspective within the domain of qualita-
tive psychology (Potter 1996). Also, a whole set of socio-linguistic, anthropological
and ethnographical approaches that are implicitly and explicitly pragmatics-oriented
have not been discussed within the context of this book (but see other volumes of this
series, in particular: Jaspers et al. 2010; Senf et al. 2009).
Tere are also strong pragmatic tendencies in approaches that are not usu-
ally considered to be pragmatic at all. Tis is especially the case with respect to Fou-
caultian and other post-structuralist approaches to discourse. One might argue that
these approaches are frst and foremost philosophical or macro-sociological at best.
One might also claim that they are not so much dealing with a linguistics-based dis-
course analysis as with a perspective on self, subjectivity and social processes which
deploy linguistic categories in a metaphorical way in order to account for the phenom-
ena under investigation. In this sense, one might even argue that we are basically deal-
ing with non-empirical and non-pragmatic modes of investigation. Nevertheless, there
are strong parallels to be drawn. For instance, Foucaults archaeological approach to
discourse explicitly focuses on the basic units of discourse in terms of enunciative func-
tions (Foucault 1969: 11920). Te unity of discourse and its delineation in terms of
genre, narrative, subject, and so on is therefore always a functional enterprise that needs
to be performed in relation to contextual correlates. Moreover, it is precisely because of
this concern with function that Foucault refuses to reduce his archaeologies defned
as descriptions of statements to the linguistic features that may characterize specifc
statements (Foucault 1969: 9798). In this sense, his proposal to add a level of discur-
sive description to those levels of analysis provided by linguistics constitutes a decidedly
pragmatic agenda. Moreover, there is an inter-subjective dimension to his work as well.
Foucault explicitly stated that he was interested in how it was possible for individuals
to produce so many contradictory statements and to occupy such varying positions at a
particular point in time (Foucault 1969: 221). His theory of discourse is therefore one of
a historically over-determined process. Linguistic pragmatic perspectives which focus
on the inter-subjective and processual aspects of establishing cohesion, coherence, nar-
rative, dialogue, enunciation, manipulation, genres, public discourse, and so on might
be able to help answer these questions.
Today, the approach to discourse that is best known in socio-political theory and
philosophy was formulated by Ernesto Laclau and Chantal Moufe. Teir discourse
theory is a theory of articulations in the sense of a set of connections between semiotic
elements that can be identifed at various levels of discursive organization (Critchley &
Marchart 2004; Howarth 2005; Laclau & Moufe 1985). Tese elements include state-
ments, subject positions (Laclau & Moufe 1985: 115), subjectivities (Laclau & Zac
1994), empty signifers (Laclau 1994), nodal points (Laclau & Moufe 1985: 11013),
and other notions that can be traced back to French linguistic sources of inspiration
(Laclau & Moufe 1985: 106, 12), Foucaultian theory and Lacanian psycho-analysis
(Glynos & Stavrakakis 2004; Stavrakakis 1999). At frst sight, their terminology is
inspired by structuralist tendencies. However, their general line of argumentation is
typically pragmatic. Even though they hold on to a structuralist notion of the sign
based on de Saussures distinction between the signifer and the signifed, their general
line of argumentation is decidedly pragmatic. Tey are interested in the ways in which
political meanings are partially and inter-subjectively fxed. Tey claim that meanings
can only be temporarily fxed to a certain degree within specifc contexts because of
the sliding values of signifers and signifeds. Moreover, these contexts are themselves
results of a series of articulations of not necessarily linguistic elements. To name a
context is to alter the signifcance of a context (Butler & Laclau 2004). Linguistic prag-
maticians will not contest such statement. However, there is a problem with respect to
the linguistic aspect of their theory.
It is important to realise that the model of the sign as set out in the discourse
theories of Laclau, Moufe and successors such as Howarth, Glynos and Stavrakakis,
is not always compatible with the model of the sign as it is found in pragmatic
linguistic approaches to discourse. Most importantly, the Essex variant of post-
structuralist discourse theory does not deal with the notions of indexicality and
12 Jan Zienkowski
deixis (Levinson 1998; Sidnell 2009), i.e. with those signifers whose functional
interpretation is dependent on the non-linguistic aspects of extra-linguistic real-
ity. Tis neglect can be explained by the fact that according to Laclau and Moufe
nothing can be thought outside of discourse. Since these authors do not distin-
guish between linguistic and non-linguistic discourse in their texts, this has led
to a methodological stalemate. Torfng points out that the post-structuralist anti-
positivist stance has led many discourse theorists to throw the methodological baby
out with the epistemological bath water (Torfng 2005: 27). As a result, there is an
urgent need for critical, explicit and context-bound discussion of what we do in dis-
course analysis, why we do it, and what the consequences are (Torfng 2005: 28). Te
type of discourse analysis he refers to is the type of discourse analysis as performed
in the post-structuralist line of thought.
Today, a response to the methodological defcit of post-structuralist discourse
theory has started to emerge. Especially authors dealing with CDA (Fairclough &
Chouliaraki 1999) have started to engage with Foucaultian and Essex-based
approaches to discourse (Glynos & Howarth 2007; Howarth 2005; Howarth & Torfng
2005). Nevertheless, I would like to argue that an explicit recognition of pragmatics-
oriented perspectives would add even more depth to this discussion. As the articles
in this volume demonstrate, the pragmatic focus on language use does not need to
lead to a neglect of the non-linguistic semiotic elements that allow us to articulate
ourselves to others and to the world in general. In this sense, the increasing interest in
studies on the multimodal aspects of discourse as considered from the point of view
of interactional (Kress & van Leeuwen 2001), systemic functionalist (OHalloran
2004) and pragmatically oriented approaches to language use holds great promise.
As Scollon and Levine argue, language in use, whether this is in the form of spoken
language or text, is always and inevitably constructed across multiple modes of com-
munication, including speech and gesture not just in spoken language but through such
contextual phenomena as the use of the physical spaces in which we carry out our
discursive actions or the design, papers, and typography of the documents within which
our texts are presented (Scollon & Levine 2004: 12).
All of the articles collected in this volume present some recent tendencies. Some
even suggest future trends. I would like to conclude by suggesting a direction for future
refection that starts from a platform that we may choose to call discursive pragmatics.
It is a route that allows the linguistic approaches presented in this volume to engage
with the theories and approaches to discourse located at the macro-level of analysis.
It is a road across the methodological defcit characterizing some of the dominant
approaches to discourse outside of the feld of linguistics. Te various perspectives that
may be grouped together under the heading of discursive pragmatics are characterized
by an internal evolution towards an investigation of language use in terms of contex-
tualized performances and in terms of the interactional functional processes involved.
Te specifc terminologies deployed within each of the perspectives discussed within
the context of this book vary greatly. Te continuing success of this discursive prag-
matic turn will depend on its ability to demonstrate its own pragmatic value in a vari-
ety of empirical studies across disciplines. Success is guaranteed if one considers the
varying areas of application that are touched upon in the individual contributions to
this volume.
References
Butler, J. & E. Laclau (2004). Appendix I: the uses of equality. In S. Critchley & O. Marchart (eds.):
32944.
Critchley, S. & O. Marchart (eds.) (2004). Laclau: a critical reader. Routledge.
Fairclough, N. & L. Chouliaraki (1999). Discourse in late modernity. Edinburgh University Press.
Foucault, M. (1969). Te archaeology of knowledge. (2007 edition) Routledge.
Glynos, J. & Y. Stavrakakis (2004). Encounters of the real kind: sussing out the limits of Laclaus
embrace of Lacan. In S. Critchley & O. Marchart (eds.): 20116.
Glynos, J. & D. Howarth (2007). Logics of critical explanation in social and political theory. Routledge.
Howarth, D. (2005). Applying discourse theory: the method of articulation. In D. Howarth &
J. Torfng (eds.): 31649.
Howarth, D. & J. Torfng (eds.) (2005). Discourse theory in European politics: identity, policy and
governance. Palgrave/Macmillan.
Jaspers, J., J.-O, stman & J. Verschueren (eds.) (2010). Society and language use (Handbook of
Pragmatics Highlights, 7). John Benjamins Publishing Company.
Kress, G. & T. van Leeuwen (2001). Multimodal discourse: the modes and media of contemporary
communication. Edward Arnold.
Laclau, E. (1994). Why do empty signifers matter to politics? In E. Laclau (ed.), Emancipation(s):
3646. Verso.
Laclau, E. & Ch. Moufe (1985). Hegemony and socialist strategy: towards a radical democratic politics.
Verso.
Laclau, E. & L. Zac (1994). Minding the gap: the subject of politics, in Ernesto Laclau (ed.), Te mak-
ing of political identities: 1139. Verso.
Levinson, S. (1998). Deixis. In Jacob Mey (ed.), Concise encyclopedia of pragmatics: 200205. Elsevier.
OHalloran, K.L. (2004). Multimodal discourse analysis: systemic functional perspectives. Continuum.
Potter, J. (1996). Representing reality. Sage Publications.
Scollon, R. & Ph. Levine (2004). Multimodal discourse analysis as the confuence of discourse and
technology. In Ph. Levine & R. Scollon (eds.), Discourse and technology: multimodal discourse
analysis: 16. Georgetown University Press.
Senf, G, J.-O. stman & J. Verschueren (eds.) (2009). Culture and language use (Handbook of prag-
matics highlights, 2). John Benjamins Publishing Company.
Sidnell, J. (2009). Deixis. In J. Verschueren & J.-O. stman (eds.), Key notions for pragmatics: 11438.
John Benjamins Publishing Company.
Stavrakakis, Y. (1999). Lacan and the political. Routledge.
Torfng, J. (2005). Discourse theory: achievements, arguments, and challenges. In D. Howarth &
J. Torfng (eds.): 132.
Appraisal
Peter R.R. White
University of Birmingham
1. Introduction
Appraisal is a framework for analyzing the language of evaluation. It has emerged
from within systemic functional linguistics (see, for example, Halliday 1994; Martin
1992; Matthiessen 1995) and was driven in its early days by work in the feld of edu-
cational linguistics and the development of Australias genre-based literacy programs
(see, for example, Iedema, Feez & White 1994; Christie & Martin 1997; Martin 2000).
It provides techniques for the systematic analysis of evaluation and stance as they
operate in whole texts and in groupings of texts. It is concerned with the social func-
tion of these resources, not simply as the means by which individual speakers/writers
express their feelings and take stands, but as the means by which they engage with
socially-determined value positions and thereby align and dis-align themselves with
the social subjects who hold to these positions.
Te systemic functional linguistics out of which the framework has emerged holds
that linguistic phenomena can best be explained by reference to the social functions
performed by language, by reference to the functional demands placed upon language
by its users (see, for example, Halliday 1971: 33068). Additionally, it holds that these
social functions fall into three broad types: those by which language represents the
world of experience (the ideational), those by which social roles and relationships
are constructed (the interpersonal), and those by which texts are made coherent,
both internally and with respect to the context in which they operate (the textual)
(see Halliday 1994). Within this context, the appraisal framework is directed towards
developing the account of interpersonal functionality, with extending descriptions
and understanding of those aspects of language by which speakers/writers construct
for themselves particular identities or personae and by which they position themselves
and those they address.
An array of text analytical interests and issues have shaped its development over
the past decade or so. However, some three or four of these have had the greatest
infuence.
In the late 1980s, a group of functional linguists in Australia were exploring
modes of narrative and were interested in criteria for articulating a taxonomy of story
telling sub-types. Tey noted, for example, that what they termed the anecdote (Plum
1988; Martin & Plum 1997) had a distinctive evaluative orientation in acting to evoke
Appraisal 15
a shared emotional reaction between narrator and audience. Tis contrasted with what
they termed the exemplum, a sub-type concerned with evaluations of human actors in
terms of morality, social esteem and social acceptability. At the same time, the group
was interested in an observed disjunction between the approach to English literature
essay writing adopted by many secondary-level students in New South Wales schools
and what was looked for in these essays by teachers. Te students devoted themselves
primarily to describing how they personally felt about the characters or the plots or the
texts as whole, whereas the teachers were looking for analyses of the texts in terms of
the insights they provided into the moral order and the human condition (Rothery &
Stenglin 1997, 2000). In the early 1990s, other members of the group turned their atten-
tion to observed variation in the style of journalistic discourse according to whether
the author performed the role of general reporter, correspondent or commentator.
Tey noted that these diferent styles or voices were associated with certain combi-
nations of diferent types of appraisal, certain syndromes of choices from the resources
of evaluation and stance (Iedema et al. 1994; Martin 2002). Tis then led to an interest
in the role of these syndromes more generally in the discursive construction of autho-
rial/speaker personae and the modelling by texts of ideal or intended readerships/
audiences (for example, Fuller 1998; White 2000; Krner 2001; White 2003).
Two central issues ran through these various projects. Te frst is concerned with
the question of the nature of attitude, with how texts activate positive and negative
assessments. Te second is concerned with how texts adopt a stance towards these
assessments and related evaluative meanings, with how these assessments and related
meanings are negotiated intersubjectively. It is the answers which the group has pro-
posed for these questions which have given the appraisal framework its current shape.
Accordingly, the discussion which follows will be organized around explorations of
these two issues.
It needs to be noted that, in concerning itself with questions of attitudinal posi-
tioning and the discursive construction of communities of shared values, the appraisal
framework addresses an area of linguistic enquiry which has only relatively recently
come into the linguistic mainstream. It is only in the last decade or so, for example,
that work within corpus linguistic into semantic prosodies has revealed just how
thoroughly sufused the language is with attitudinal associations and implications.
(See, for example, Sinclair 1991 or Louw 1993). (It is noteworthy that Malrieu in his
Evaluative Semantics declares that, Despite this variety of approaches [to dealing with
meaning in linguistics], very little attention has been paid to evaluation in language,
1999: 114.) Accordingly, the development of the appraisal framework has required the
exploration of new approaches to linguistic taxonomizing and new modes of linguistic
argumentation in support of those taxonomies. Inevitably, then, the project remains a
work in progress with some of its analytical typologies still having the status of propos-
als or hypotheses requiring further investigation and testing.
16 Peter R.R. White
2. Overview
2.1 Attitude the activation of positive or negative positioning
In their early work into evaluation in narrative and student writing, the Australian
group drew on the well-established tradition of research into the language of afect, a
tradition which was exemplifed by a special issue of Text concerned with the poten-
tial of language to express diferent emotions and degrees of emotional intensity
(Ochs 1989: 1). Te group shared with this tradition the view that emotion is crucially
implicated in attitudinal assessment, in the activation by texts of positive and nega-
tive viewpoints. Te group, however, has departed from the tradition in its view that
afect in its broadest sense should not be too closely tied to emotion and that, in
order to address the text analytical issues with which the appraisal research has been
concerned, it is useful to identify additional modes of afectual or attitudinal meaning.
(See, for example, Martin 1997, 2000). Specifcally, the appraisal framework proposes
that attitudinal meanings (positive and negative assessments) can be grouped into
three broad semantic domains.
2.1.1 Afect
Firstly there are these fundamentally attitudinal meanings associated with emotion
texts indicate positive or negative views via either reports of the speaker/writers
emotional responses or reports of the emotional responses of third parties. For
example (values of Afect are underlined):
(1) I am disappointed and ashamed that two of our most admired and respected
sportsmen could behave in such a manner. To play for your country is an hon-
our and a privilege, not a right. (Te West Australian 11/12/98: 12, letter to
the editor, Jennifer Black, Riverdale)
Te traditional term Afect has been taken up as a label for such meaning.
2.1.2 Judgement
Secondly, there are meanings by which a view is indicated of the social acceptability
of the behaviour of human actors, assessment by reference to some system of social
norms or morality.
(2) Tose who are chosen to represent Australia should not only be talented but
they should be above reproach. Sport is supposed to teach honour, fair play,
teamwork, leadership and social skills. It is not supposed to create or support
greed and egos. (Te West Australian 11/12/98: 12, letter to the editor,
Jennifer Black, Riverdale)
Te term Judgement has been adopted to reference such meanings.
Appraisal 17
2.1.3 Appreciation
Tirdly, there are meanings by which assessments are made of semiotic and natural
phenomena by reference to their value in a given feld, perhaps most typically by refer-
ence to their aesthetic qualities.
(3) It [the E-type Jaguar] is a masterpiece of styling whose proportions are
dramatic; its crisp details are in complete harmony with the broader
outlines of the gorgeous general arrangement (Te Independent, Weekend
Review: p.1 27/01/2001)
Te term Appreciation has been adopted to reference these meanings.
2.1.4 Modes of activation direct and implied
Te framework makes a distinction in terms of the way in which such meanings are
activated in text. Least problematically they can be activated by explicitly attitudinal
terms, terms which generally carry a negative or positive meaning. For example,
(4) Without the intervention of a partisan, right-wing Supreme Court to ensure the
election of a Republican, Mr Bush would now be a forgotten loser. Te Observer
considers his election an afront to the democratic principle with incalculable
consequences for America and the world. (Te Observer,
Jan 21, 2001 leader page)
More problematic are activations which rely on implication, inference and associa-
tion, which rely on the reader/listener to interpret the depicted happening or state
of afairs as positive or negative according to the value system they bring to the text.
For example,
(5) George W. Bush delivered his inaugural speech as the United States
President who collected 537,000 fewer votes than his opponent.
(Te Observer, Jan 21, 2001 leader page)
Here an essentially factual depiction points the reader towards a negative assessment
of Bush and/or the US electoral process and clearly has the potential to activate such
an assessment, depending, of course, on the viewpoint the individual reader brings to
the text. Te sentence, however, contains no explicitly attitudinal (positive or negative)
lexis, no meanings which of themselves convey negative assessments.
Clearly such implied appraisal poses major theoretical and analytical problems.
In moving from direct to indirect activation, we step here from what certain Anglo-
American traditions would see as semantics into what would be seen as pragmat-
ics, from meanings seen to be inscribed in the text to meanings seen to be operating
only in the context. Accordingly, those operating with analytical philosophical notions
of semantics might want to exclude such formulations from treatments of evaluative
18 Peter R.R. White
language, arguing that there is nothing about the actual language here which is
attitudinal. While this might be attractive in terms of avoiding complication and pro-
viding for more easily replicable analyses of texts, it would mean that much of the evalu-
ative work being done by texts would simply be missed out. Analyses would not only be
unable to attend to the role of implied or indirect evaluation generally, but they would
also be unable to take account of the ofen rhetorically crucial interaction between
direct and indirect assessment. (For a demonstration of how appraisal analysis can be
used to explore this interaction, see Cofn 1997.) Under the appraisal framework, then,
such formulations are seen as falling very much within the scope of linguistic analyses
of evaluative efects.
In this, the framework stands with writers such as Gruber (1993) and Malrieu
(1999). Gruber, for example, includes in his taxonomy of Evaluative Units the use
of directly quoted material which is likely to be seen by the reader as evidence of
that quoted sources ethical shortcomings. Although the quoted source need not in
any way be overtly evaluated, the use of the sources own words still clearly serves an
evaluative function. Malrieu provides what is, perhaps, even stronger support, argu-
ing that when expressions are considered in their actual textual context, it is difcult
to conceive of any phrase which would be evaluation free. In context, even adverbs
and complement such as always and with a knife have an evaluation (Malrieu
1999: 134). In taking this view of evaluation, the appraisal framework, of course,
also takes a lead from corpus linguistic research into semantic prosodies, an area of
research already mentioned above.
Accordingly, appraisal proposes a distinction between what it terms inscribed
evaluation, where the positive/negative assessment is directly inscribed in the dis-
course through the use of attitudinal lexis, and what it terms invoked evaluation,
where it is not possible to isolate such explicitly attitudinal vocabulary. Te extract in
(5) would thus provide an example of invoked attitude. Considerably more research
is required in order to provide a systematic account of the mechanisms by which this
process of invocation operates.
2.1.5 Typological criteria
Te exact criteria upon which this three-way taxonomy are based remains a central
issue for the ongoing appraisal analysis project. Te distinction between Afect and
the other categories (Judgement and Appreciation) is a relatively unproblematic one.
Te semantics of Afect is one by which meanings are most typically realized through
a verbal process undergone or experienced by a conscious human participant
the reactional Mental Processes of systemic functional linguistics (Halliday 1994)
Your decision saddens me, I fear ghosts. But no such canonical realizational mode
applies for Judgement and Appreciation, with the picture complicated by the fact that
all three categories have diversifed realizations as, for example, adjectives (adored
Appraisal 19
leader [afect], corrupt leader [judgement], handsome man [appreciation]); adverbs
(adoringly, corruptly, beautifully); nouns (his love of, tyrant, masterpiece); and verbs
(I love bagpipe music, he corrupted the political process).
At this stage there is some evidence suggesting that the distinctions are refected
in patternings in collocational frames. (For more general work on evaluation and
collocational frames, see Hunston & Sinclair 2000). We fnd, for example, that
Judgement values are available for the frame, It was X-Judgement value of Y to
(It was dishonest of him to. .. , It was lazy of her to , It was wise of her to)
while Appreciation values are not. Tus, It was thoughtless of you to leave the cat
out in the rain is possible but not It was elegant of you to wear that outft. Similarly
when terms such as beautiful operate as Appreciation, the It was X of Y to frame
is not available (It was beautiful of you to wear your hair like that.), but when they
operate as Judgement it is available (It was beautiful of you help out those street
kids the way you did.). More work, however, is required in this area. At this stage
the three-way taxonomy is proposed as an hypothesis about the organisation of the
relevant meanings, being ofered as a point of comparison for those with alternative
classifcations, as a resource for those who need something to manage the analysis
of evaluation in discourse, and as a challenge to those concerned with developing
appropriate reasoning.
2.1.6 The interplay between the attitudinal modes
It must be stressed, however, that while the framework extends established notions
of the afectual in this way, it still sees the three categories as fundamentally inter-
connected in that they are all to do with the expression of feelings. It is just that the
grounding of that feeling varies across the three modes. Under Afect, the action of
emotion is directly indicated feelings are presented as the contingent, personalized
mental reactions of human subjects to some stimulus. But under both Judgement and
Appreciation, these feelings are institutionalized in some way and are recast as quali-
ties which inhere in the evaluated phenomenon itself. Tus I like that picture grounds
the evaluation in the contingent, individual reactions of the speaker while that is a
beautiful picture grounds the evaluation in the objective properties of the evaluated
phenomenon itself. Under Judgement, feelings are reconstrued as proposals about
correct behaviour how we should or shouldnt behave. Tus, in He cruelly lef the cat
out in the rain the negative feeling towards the perpetrator of this act is here reworked
as a proposal about what it right and wrong behaviour towards cats. Under Apprecia-
tion, feelings are reconstrued as propositions about the value of things. Tus in Tats
a beautiful picture, the positive feeling towards the picture is reworked as a proposal
about the pictures aesthetic worth. (For a more extended discussion of Judgement and
Appreciation as institutionalizations of feeling, see Martin 2000). Te role of Afect at
the heart of institutionalized feelings is depicted diagrammatically below.
20 Peter R.R. White
Feeling institutionalized as ethics/morality (rules & regulations)
Feeling institutionalized as
aesthetics/value (criteria & assesment)
Judgement: she is naughty
Appreciation: she is an unattractive child
Afect: she disappoints me
Figure 1. Judgement & Appreciation as institutionalized afect (from Martin 2000)
2.2 Intersubjective stance
In its modelling of the resources of intersubjective stance, the framework is concerned
with formulations which have traditionally been analyzed under such headings as
modality (see for example Palmer 1986), polarity (see for example Pagano 1994), evi-
dentiality (Chafe & Nichols 1986), hedging/boosting (Markkanen & Schrder 1997;
Hyland 1996; Myers 1989; Meyer 1997), vague language (Channell 1994), intensi-
fcation (Labov 1984), and meta-discourse (Crismore 1989). Under the appraisal
framework, these lexico-grammatically diverse wordings are brought together on the
grounds that they are all resources which vary the terms of the speakers engagement
with propositions and proposals, which vary what is at stake interpersonally both in
individual utterances and as the texts unfolds cumulatively. Tese resources of inter-
subjective stance are divided into two further broad categories (a) resources by which
the textual voice positions the current proposition with respect to actual or potential
alternatives to that proposition (given the label Engagement) and (b) resources which
provide grading or scaling (given the label Graduation), either in terms of the degree
of the textual voices personal investment in the proposition (intensifers/down-toners)
or in terms of choices the textual voice makes with respect to the preciseness of focus
of its formulations. For reasons of space I will confne myself here to considering only
Engagement. (For a full account of both Engagement and Graduation, see Martin &
White 2003).
Te approach taken to accounting for the intersubjective functionality of these
values of Engagement is informed by Bakhtins now widely infuential notion of dia-
logism and heteroglossia under which all verbal communication, whether written or
spoken, is dialogic in that to speak or write is always to refer to, or to take up in some
Appraisal 21
way, what has been said/written before, and simultaneously to anticipate the responses
of actual, potential or imagined readers/listeners. As Voloinov states,
Te actual reality of language-speech is not the abstract system of linguistic forms, not
the isolated monologic utterance, and not the psychological act of its implementation,
but the social event of verbal interaction implemented in an utterance or utterances.
Tus, verbal interaction is the basic reality of language.
Dialogue [] can also be understood in a broader sense, meaning not only direct,
face-to-face, vocalised verbal communication between persons, but also verbal
communication of any type whatsoever. A book, i.e. a verbal performance in print,
is also an element of verbal communication. [] [it] inevitably orients itself with
respect to previous performances in the same sphere [] Tus the printed verbal
performance engages, as it were, in ideological colloquy of a large scale: it responds
to something, afrms something, anticipates possible responses and objections, seeks
support, and so on. (Voloinov 1995: 139)
Te approach adopted by the appraisal framework holds that the functionality of these
resources can only be adequately explained when such dialogistic efects are taken
into account. Tat is to say, it holds that by the use of wordings such as possibly, It
is my contention that, naturally, admittedly, I believe, the textual voice acts
frst-and-foremost to acknowledge, to engage with and to align itself with respect to
positions which are in some way alternatives to that being advanced by the text.
In this, the appraisal framework represents a departure from much of the modal-
ity and evidentiality literature (see for example, Lyons 1977; Palmer 1986; Chafe &
Nichols 1986) and at least some of the hedging literature (see Markkanen & Schrder
1997) where accounts of epistemic modals and similar resources, for example, ofen
assume that the sole function of these wordings is to reveal the writer/speakers state of
mind or knowledge, to indicate that the speaker/writer is uncertain or tentative and is
not committed to the truth value of the proposition.
3. Attitudinal assessment a brief outline
3.1 Afect
Te appraisal framework is oriented towards mapping semantic domains as they
operate in discourse. Accordingly, as indicated above, the categorizations frequently
bring together diverse grammatical structures under single discourse semantic head-
ings. Afect is typical in this regard its values are sometimes construed as qualities
(adjectives I am happy about that), sometimes as processes (verbs Tis pleases
me) and sometimes as comment adjuncts (Happily). Tey may also be realised as
virtual entities (nouns) via nominalization happiness.
22 Peter R.R. White
Following Martin (1997) and Martin (2000), the appraisal framework classifes
diferent instances of afect according to the following 6 factors:
i. Are the feelings popularly construed by the culture as positive (enjoyable) or neg-
ative ones (unenjoyable)?
ii. Are the feelings represented as a surge of emotion involving some kind of para-
linguistic or extralinguistic manifestation (for example, weeping or trembling), or
more internally experienced as an emotive state or ongoing mental process?
behavioural surge She broke down crying.
mental process/state She was distraught.
iii. Are the feelings represented as targeting or responding to some specifc emotional
stimulus or are they represented as a general ongoing mood?
reaction to stimulus Her mothers absence is upsetting her.
undirected mood She is sad.
iv. Where do the feelings lie on a scale from low to high intensity?
low I dislike bagpipe music.
median I hate bagpipe music.
high I detest bagpipe music.
v. Do the feelings involve intention (rather than reaction), with respect to a stimulus
that is not yet actualised (irrealis) as opposed to an actual stimulus (realis)?
realis Im upset by what she said.
irrealis I fear what she might say.
vi. Finally, emotions can be grouped into three major sets having to do with
un/happiness, in/security and dis/satisfaction. Te un/happiness variable
covers emotions concerned with afairs of the heart sadness, anger, happi-
ness and love; the in/security variable covers emotions concerned with eco social
well-being anxiety, fear, confdence and trust; the dis/satisfaction variable
covers emotions concerned with telos (the pursuit of goals) ennui, displeasure,
curiosity, respect.
un/happiness I am sad.
in/security I am anxious.
dis/satisfaction I am bored.
3.2 Judgement
Judgement is the domain of meanings by which attitudes are construed with respect to
human behaviour approval/disapproval of human behaviour by reference to social
Appraisal 23
acceptability/social norms; assessments of a persons character or how they measure
up to social requirements of expectations. Te framework divides these Judgements
into those dealing with social esteem and those oriented to social sanction. Judge-
ments of social sanction involve an assertion that some set of rules or regulations,
more or less explicitly codifed by the culture, are at issue. Tose rules may be legal,
moral or religious. Judgements of social esteem involve evaluations under which the
person judged will be lowered or raised in the esteem of their community, but which
do not have legal or moral implications. Tus negative values of social esteem will
be seen as dysfunctional or inappropriate or to be discouraged but they will not be
assessed as sins or crimes.
Judgements of social esteem can be to do with normality (how usual someones
behaviour is), capacity (how capable they are) and tenacity (how resolute they are).
Judgements of social sanction have to do with veracity (how truthful someone is) and
propriety (how ethical someone is).
Table1. Judgement (afer Iedema et al. 1994)
Social esteem positive [admire] negative [criticize]
normality (custom)
is the persons
behaviour unusual,
special, customary?
standard, everyday,
average; lucky,
charmed; fashionable,
avant garde
eccentric, odd, maverick;
unlucky, unfortunate;
dated, unfashionable
capacity
is the person
competent, capable?
skilled, clever,
insightful; athletic,
strong, powerful;
sane, together
stupid, slow, simple-
minded; clumsy, weak,
uncoordinated; insane,
neurotic
tenacity (resolve)
is the person
dependable, well
disposed?
plucky, brave, heroic;
reliable, dependable;
indefatigable, resolute,
persevering
cowardly, rash,
despondent; unreliable,
undependable;
distracted, lazy,
unfocussed
Social sanction positive [praise] negative [condemn]
veracity (truth)
is the person
honest?
honest, truthful,
credible; authentic,
genuine; frank,
direct;
deceitful, dishonest;
bogus, fake; deceptive,
obfuscatory
propriety (ethics)
is the person
ethical, beyond
reproach?
good, moral, virtuous;
law abiding, fair, just;
caring, sensitive,
considerate
bad, immoral, lascivious;
corrupt, unjust, unfair;
cruel, mean, brutal,
oppressive
(Please note that the lists of terms in the right-hand columns are intended only
as a guide to the types of meanings which are involved here and not as some sort of
24 Peter R.R. White
dictionary of Judgement sub-types. In actual analysis, it is always meaning in context
which is addressed. Tus the evaluative meaning of a particular word form may vary
from text to text under the infuence of co-textual factors.)
Under the appraisal framework, this fve-way taxonomy (normality, capacity,
tenacity, veracity, propriety) is grounded in the semantics of modalization as articu-
lated by Halliday (1994). Tat is say, each of the sub-categories of Judgement can be
understood as a lexicalization of one of the grammatical categories of modality. Tis
relationship operates in the following proportions: normality is to usuality, as capacity
is to ability, as tenacity is to inclination, as veracity is to probability, as propriety is to
obligation. In the early work on Judgement (Iedema et al. 1994), the labels for the fve
sub-types were closer to these modal oppositions, as refected in Figure2 below (fate
for normality, resolve for tenacity, truth for veracity, ethics for propriety).
Type
Modalization
Probability Probably
Truth
Sanction
Esteem
Fate
Ethics
Resolve
Capacity
Usually
Supposed to
Keen to
Able to
Usuality
Obligation
Inclination
Ability
Readiness
Modulation
Figure 2. (from Iedema et al. 1994)
Te bridge between the underlying grammatical modal options and the lexical
categories of Judgement is provided by Hallidays notion of interpersonal metaphor
(Halliday 1994). Under Hallidays framework, modal values can be realized congru-
ently (non-metaphorically) by modal auxiliaries (may, must, might, could etc.) and
adjuncts (perhaps, possibly, certainly) and metaphorically through more lexical for-
mulations such as It is possible that, It is required that, I think that , etc. It is pos-
sible to construct a series of realizations for the epistemic modal values of probability,
usuality and capacity which begins with congruent realizations (via grammatical for-
mulations) and extends through metaphorical ones (more lexicalized) towards lexis
which is clearly appraising in nature. In this way modalizations of probability can be
related to lexicalized Judgements of veracity:
He took the money. (unmodalized)
He must have taken the money.
Certainly he took the money.
Its certain he took the money.
Its true that he took the money.
She was honest in stating that he took the money. [Judgement: veracity]
Appraisal 25
Similarly, modalities of usuality can be related to Judgements of normality:
My students dress entirely in black. (unmodalized)
My students ofen dress entirely in black.
Its usual for my students to dress entirely in black.
Its normal for my students to dress entirely in black.
Its fashionable for my students to dress entirely in black. [Judgement: normality]
Likewise for ability and capacity:
He can go.
Hes able to go.
Hes capable of going.
Hes strong enough to go.
Hes healthy enough, mature enough to go [Judgement: capacity]
A similar relationship of proportionality holds for the deontic modal values (obliga-
tion, inclination) and the Judgement values of propriety and tenacity. Tus inclination
can be related to lexicalized tenacity:
Ill go.
Im determined to go.
Im intent on going.
Im resolved to go
Im resolute, steadfast, unyielding, unfinching etc. [Judgement: tenacity]
And obligation can be related to lexicalized Judgements of propriety:
Dont do that.
You shouldnt do that.
Youre not supposed to do that.
Its expected that you wont do that.
Itd be unfair for you to do that.
Its corrupt, insensitive, arrogant, selfsh, etc. of you to do that. [Judgement:
propriety]
(For a full account see Martin & White 2003)
3.3 Appreciation
As indicated above, Appreciation is the domain of meanings for construing evalua-
tions of the products of human endeavour such as artefacts, buildings, texts and works
of art, and also of natural phenomena and states of afairs. Te semantic is one by
which such objects are assigned a value (negative or positive) in a given discourse or
feld of activity. One of the most salient systems for the assignment of such value is that
of aesthetics. Human subjects may be appreciated rather than judged, but only when
26 Peter R.R. White
it is, for example, their aesthetic qualities which are being addressed rather than the
social acceptability of their behaviour.
Te appraisal framework sub-divides Appreciations into those assessments which
turn on our reactions to things (do they catch our attention; do they please us?), their
composition (balance and complexity), and their value (how innovative, authentic,
timely, efective, healthy, relevant, salient, signifcant etc.). Some illustrative examples
are supplied below in Table2.
Table2. Values of appreciation
Positive Negative
reaction: impact
did it grab me?
arresting, captivating,
engaging; fascinating,
exciting, moving; lively,
dramatic, intense;
remarkable, notable,
sensational
dull, boring, tedious; dry,
ascetic, uninviting; fat,
predictable, monotonous;
unremarkable, pedestrian
reaction: quality
did I like it?
lovely, beautiful, splendid;
appealing, enchanting,
welcome
plain, ugly, grotesque;
repulsive, revolting,
of-putting
composition: balance
did it hang together?
balanced, harmonious,
unifed, symmetrical,
proportioned; consistent,
considered, logical; shapely,
curvaceous, willowly
unbalanced, discordant,
irregular, uneven,
fawed; contradictory,
disorganised; shapeless,
amorphous, distorted
composition: complexity
was it hard to follow?
simple, pure, elegant; lucid,
clear, precise; intricate, rich,
detailed, precise
ornate, extravagant,
byzantine; arcane, unclear,
woolly; plain, monolithic,
simplistic
valuation
was it worthwhile?
penetrating, profound,
deep; innovative, original,
creative; timely, long
awaited, landmark;
inimitable, exceptional,
unique; authentic, real,
genuine; valuable, priceless,
worthwhile
shallow, reductive,
insignifcant; derivative,
conventional, prosaic;
dated, overdue, untimely;
dime-a-dozen, everyday,
common; fake, bogus,
glitzy; worthless, shoddy,
pricey
Te fact that afectual values underpin all three sub-categories of Attitude (Afect,
Judgement, Appreciation) is perhaps most obviously demonstrated in reaction values
of Appreciation such as a fascinating book, a boring piece of music. Such instances, of
course, involve lexis which could otherwise be used to realize Afect the book fas-
cinates me, that music bores me. Under the framework, such instances are kept apart
on the grounds that there is something rhetorically signifcant at stake in choosing
between explicitly grounding the evaluation in the emotional reactions of an individual
Appraisal 27
human subject (Tat book bores me) and externalizing that feeling by representing it
as a characteristic which inheres in the evaluated entity itself. Tat is to say, it is seen
as important to be able to distinguish between construing the emotions someone feels
(Afect) and ascribing the power to trigger such feelings to things (Appreciation).
4. Engagement: An overview
As indicated previously, the treatment of the resources of intersubjective positioning
developed within the appraisal framework is informed by the view that all verbal utter-
ances are ultimately dialogic. To illustrate this style of treatment I will consider briefy
the functionality of the phrase there is an argument, is there in the following short
extract from a radio interview. Te interviewer quizzes the then conservative Aus-
tralian Prime Minister, John Howard, about the behaviour of the Australian banks in
raising interest rates at a time when they have been making record profts.
(6) Tere is an argument, though, is there, the banks have been a bit greedy I mean,
the profts are high and good on them, theyre entitled to have high profts, but
at the same time the fees are bordering on the unreasonable now.
Tere is, of course, a backwards looking dialogistic aspect to the use of this phrase.
Te interviewer presents himself as simply taking up the words of some other, non-
specifed prior group of speakers. He represents himself as conveying community
concerns rather than his own, individual views. But there is rather more going on here
in terms of the way the text recognizes and hence engages with potential alternatives
to the current proposition. By such a device the interviewer indicates that this is a con-
tested, debated assessment of the banks behaviour he acknowledges that this it is but
one of a number of views currently in play in the community. He thereby indicates that
he anticipates that at least some elements in society will object to, and challenge such
a suggestion. By representing the proposition as arguable in this way, he represents
himself as not personally committed to this position and hence signals a preparedness
to enter into debate on the issue. In this sense, then, the formulation can be seen as an
instance of dialogistic anticipation.
Under the framework, the following options (which may be multiply present in
a given utterance) are seen as enabling the textual voice to vary the terms by which it
engages with alternative voices and alternative positions.
Disclaim the textual voice positions itself as at odds with, or rejecting, some
contrary position:
(deny) negation
(counter) concession/counter expectation
28 Peter R.R. White
Proclaim by representing the proposition as highly warrantable (compelling, valid,
plausible, well-founded, generally agreed, reliable, etc.), the textual voice sets itself
against, suppresses or rules out alternative positions:
(concur) naturally, of course, obviously, admittedlyetc.; some types of
rhetorical question
(pronounce) I contend, the truth of the matter is, there can be no doubt
thatetc.
(endorse) X has demonstrated that; X as compellingly arguedetc.
Entertain by representing the proposition as grounded in a contingent, individual
subjecthood, the textual voice represents the proposition as but one of a range of pos-
sible positions it thereby entertains or invokes these dialogic alternatives:
it seems, the evidence suggests, apparently, I hear
perhaps, probably, maybe, its possible, may/will/must; some types of rhetorical
question
Attribute by representing proposition as grounded in the subjecthood of an external
voice, the textual voice represents the proposition as but one of a range of possible
positions it thereby entertains or invokes these dialogic alternatives:
(acknowledge) X said.., X believes, according to X, in Xs view
(distance) X claims that, the myth that., its rumoured that
4.1 Dialogic contraction and expansion
Tese various options, then, are seen as enabling variations in stance they all provide
for a diferent orientation to the heteroglossic diversity in which the text operates.
Tey are further seen as falling into two wider categories according to a broader-scale
axis of variation in rhetorical functionality. Tey are characterized as either dialogi-
cally expansive or dialogically contractive, with the distinction turning on the degree
to which an utterance, by dint of one or more of these wordings, entertains dialogically
alternative positions and voices (dialogic expansion), or alternatively, acts to challenge,
fend of or restrict the scope of such (dialogic contraction).
Consider the following by way of illustration of this distinction.
(7) (Endorsement) Follain punctures the romantic myth that the mafa started as
Robin Hood-style groups of men protecting the poor. He shows that the mafa
began in the 19th century as armed bands protecting the interests of the absen-
tee landlords who owned most of Sicily. He also demonstrates how the mafa has
forged links with Italys ruling Christian Democrat party since the war
(Cobuild Bank of English)
Appraisal 29
(8) (Distance) Tickner said regardless of the result, the royal commission was a
waste of money and he would proceed with a separate inquiry into the issue
headed by Justice Jane Matthews. His attack came as the Aboriginal women
involved in the demanded a female minister examine the religious beliefs they
claim are inherent in their fght against a bridge to the island near Goolwa in
South Australia. (Cobuild Bank of English)
Both extracts are obviously dialogistic in that they explicitly reference the utterances
and viewpoints of external voices. But there is more at stake here than simple attribu-
tion, than a simple multiplying of voices. Extract (7) is an example of a formulation in
which a special type of reporting verb has been used (show, demonstrate) one which
presupposes the warrantability of the attributed proposition, which, for example, holds
it to be true, reliable, plausible or fair. (Reporting verbs of this type have, of course,
been widely discussed in the literature on attribution and direct and indirect speech.
See, for example, Hunston 2000 or Caldas-Coulthard 1994). By such endorsing for-
mulations, the textual voice aligns itself with some external voice which is represented
as correct, authoritative or otherwise argumentatively compelling, at least in the con-
text of this particular proposition. By indicating in this way a heightened personal
investment by the author, by co-opting some authoritative second party to the current
rhetorical cause, such formulations set themselves against, or at least fend of, actual or
potential contrary positions. Tat is to say, they increase the interpersonal cost to any
who would advance such an alternative. Tus in the above instance, show and dem-
onstrate are employed as the textual voice sets itself against the discredited alternative
view of the Mafa as Robin Hood types. Such endorsements, then, can be construed as
dialogically contractive the close down the space for dialogic alternatives.
Extract (8) has the opposite efect. Here, of course, the textual voice distances itself
from the proposition framed by claim, representing it as, if not doubtful, then at least
as doubtable, as potentially open to question. Te efect is to invite or at least enter-
tain dialogic alternatives, to lower the interpersonal cost for any who would advance
such an alternative. Accordingly, such distancing formulations can be seen as dialogi-
cally expansive, as opening up the dialogic space for alternative positions. (It must be
stressed that it is not proposed that a verb such as to claim necessary has this function
in all cases. Te rhetorical potential of such formulations, for example, may vary sys-
tematically across registers, genres and discourse domains.)
In this distinction, then, between Endorsing and Distancing we see the funda-
mental contrast between dialogic contraction and expansion.
4.2 Further resources of dialogic expansion
Te values of Acknowledge and Entertain align with such Distancing formulations
in acting to open up the dialogic space to alternative positions.
30 Peter R.R. White
4.2.1 Acknowledge
Te category of Acknowledge involves attribution where neutral frames are employed
to simply report the words and viewpoints of external voices by the use, for example,
of reporting verbs such as say, report, state and wordings such as according to, in
her view. Just as in the case of the other attributions (Distance, Endorse), such formu-
lation are obviously dialogic in that they explicitly introduce an alternative voice into
the text. And once again they are dialogistic in an additional sense in that, by this
explicit grounding of the proposition in an individual subjecthood, they represent the
proposition as individual and contingent, as but one of a range of possible proposi-
tions. Tose alternatives to the current proposition are, in this sense, recognized and
the heteroglossic context in which the text operates is thereby revealed.
4.2.2 Entertain
Formulations which actively entertain dialogic alternatives include
deductive formulations such as it seems, apparently, the evidence suggests;
forms which represent the proposition/proposal as more or less likely, including
modals of probability, as well as certain rhetorical uses of questions.
In such contexts, the proposition is shown to be grounded in the subjecthood of the
textual voice since it provides assessments of the propositions likelihood or evidential
basis. Grounded in this way, the proposition is shown to be contingent and associated
with an individualized point of view. As such, it is revealed to be but one of a num-
ber of possible alternative position. In this way, these alternatives are entertained or
acknowledged and the dialogic space for those alternatives is thereby expanded.
4.3 Further resources of dialogic contraction
4.3.1 Pronounce
Under the appraisal framework, Pronouncements are formulations which involve cer-
tain types of intensifcation, authorial emphases or explicit authorial interventions or
interpolations. For example: I contend, Te facts of the matter are that, Te truth
of the matter is that, We can only conclude that, You must agree that, intensifers
with clausal scope such as really, indeed etc. and, in speech, appropriately placed stress
(e.g. Te level of tolerance is the result of government intervention).
For example, the phrase, It is absolutely clear to me in the following performs
this function
(9) It is absolutely clear to me that what Charlotte was arguing was that Crouching
Tiger was a bad flm to which liberal audiences imputed a signifcance shaped
by their own prejudices about Chinese cinema and the Chinese in general.
(Letter to the www.dimsum.co.uk website from
Ian Katz, features editor, the Guardian.)
Appraisal 31
Such intensifcations and interpolations are dialogically motivated. Te textual voice
doesnt indicate this heightened personal investment in the proposition in a commu-
nicative vacuum. Rather it does so against some opposed dialogic alternative here
against a contrary view of what Charlotte was arguing. Tus such formulations are
dialogic in that they acknowledge an alternative while simultaneously acting to chal-
lenge or fend it of. Tey are dialogically contractive by dint of this action of confront-
ing and fending of the contrary position.
4.3.2 Concur
Te Concur option involves wordings such as of course and naturally. Tese formu-
lations are like Pronouncements in that they also provide for the textual voice to
explicitly convey its investment in the viewpoint being advanced and thereby to con-
front or rule out possible alternatives. Tey difer, however, in that they represent the
proposition/proposal as uncontentious within the current communicative context, as
a given, as being in accord with what is generally known or expected. Te textual voice
is represented as taking up a viewpoint held by people generally, and hence the reader/
listener. Consider by way of example the use of of course in the following.
(10) When, belatedly, their selectors chose Paul Adams, who would assuredly have
won them the second Test in Johannesburg, their attack became very good in
the opinion of Trevor Bailey, who has seen a few in his time. Bailey, of course,
was that rarity, a cricketer who at his best was world-class with both bat and ball.
(From the Bank of English OzNews corpus)
Here the writer represents himself/herself as simply agreeing with the reader, as
recounting a view (that Bailey was a cricketing rarity) which is already held by the dia-
logic partner and by people in general. Te location of the current proposition within a
dialogistic exchange is thus employed to increase the cost of any subsequent challeng-
ing or rejecting of the proposition.
4.3.3 Disclaim (Deny and Counter)
Te fnal dialogistically contractive option is supplied by meanings by which some
prior utterance or some alternative position is invoked so as to be directly rejected,
replaced or held to be unsustainable. Obviously to deny or reject a position is maxi-
mally contractive in that, while the alternative position has been recognized, it is
held not to apply the alternative position is thus directly confronted. Tis is the
domain of negation and concession/counter-expectation. Te term Disclaim is
used to reference such formulations which operate in this way, with two further
sub-types identifed.
4.3.4 Disclaim: Deny (negation)
From the appraisal frameworks dialogistic perspective, negation as a resource for
introducing the alternative positive position into the dialog, and hence acknowledging
32 Peter R.R. White
it and engaging with it, and then rejecting it. Tus in these interpersonal/dialogistic
terms, the negative is not the simple logical opposite of the positive, since the negative
carries with it the positive, while the positive does not reciprocally carry the negative.
Tis aspect of the negative, though perhaps at odds with common-sense understand-
ings, has been quite widely noted in the literature see for example, Leech (1983: 101);
Pagano (1994); Fairclough (1992: 121). Consider, for example, the following extract
from an advertisement placed in magazines by the British Heart Foundation.
(11) We all like something to grab hold of. But sometimes you can have too much
of a good thing. And a man whose table diet consists of double cheeseburgers
and chips can end up looking like a tub of lard. Teres nothing wrong with meat,
bread and potatoes. But how about some lean meat, wholemeal bread and jacket
potatoes?
Here the denial, Tere is nothing wrong with meat, bread and potatoes, is clearly
dialogic in the sense that it invokes, and presents itself as responding to, claims/beliefs
that Tere is something wrong with meat, bread and potatoes. A prior and alternative
position is thus clearly engaged with dialogistically.
4.3.5 Disclaim: Counter
Related to such negating formulations are those which represent the current proposi-
tion as replacing and supplanting a proposition which would have been expected in its
place. Consider, for example
(12) Tey [Kevin and Ian Maxwell, sons of Robert Maxwell] have a lot to prove in
the coming years. Now they will not only seek to make their own fortunes but
to clear their fathers besmirched name. Tey grew up to see him as the eternal
outsider, the man who had fought Establishment prejudice and pettifogging
bureaucracy to get where he was. Sure, he broke rules. Yes, he ducked and dived.
Admittedly, he was badly behaved. But look at what he had achieved. From
nothing, he had become a multinational businessman with an empire stretching
across the world, the confdant of statesmen and just as famous himself.
(From the Bank of English UKMags corpus)
Te extract (from Te Times) is concerned with the notorious British businessman,
newspaper magnate and former Labour MP, Robert Maxwell (now deceased) and his
two sons, Kevin and Ian. In the extract, the writer seeks to explain, even justify, why the
two sons might have continued to regard their father favourably, despite the negativity
with which Maxwell had come to be viewed generally. (Maxwell had been found afer
his death to have secretly diverted millions of dollars from two of his companies and
from employee pension funds in an efort to keep his business empire solvent.) For our
current purposes we are concerned with the latter part of this sequence, the utterances
which follow But But look at what he had achieved. From nothing he had become
Appraisal 33
a multinational businessman Te textual voice is here setting itself against what is
represented as a generally applying negative view of Maxwell. By the formulation, the
negative view is held not to apply, at least for Maxwells sons. Tus through a dialogic
interaction, a certain view is referenced and then rejected.
4.4 Engagement resources summary
Te following table provides an overview of the resources of Engagement.
Table3. Engagement resources
Dialogic contraction
Disclaim:
Deny: e.g. It is a review which doesnt consider the feelings of the Chinese community.
Counter: e.g. What is surprising is to fnd such ofensive opinions in the Guardian..
Proclaim:
Concur: e.g. Te Premier, of course, wants us to think what a fne anti-racist fellow he is.
Pronounce: e.g. It is absolutely clear to me that what Charlotte was arguing was that Crouching
Tiger was a bad flm.
Endorse: Dr Rufmans work has shown that parents or carers who talk to their children about
mental states -thoughts, beliefs, desires and feelings end up with children who know much earlier in
life what another person is thinking.
Dialogic expansion
Entertain: e.g. Perhaps the most telling thing about Charlotte Ravens review of Crouching Tiger isnt in
the review itself but in the one line preamble on the Guardian website.
Attribute
Attribute/Acknowledge: e.g. It states: Crashing bore, wooden drama: Charlotte Raven dares to
difer from the unanimous acclaim for Ang Lees Crouching Tiger, Hidden Dragon.
Attribute/Distance: e.g. and someone went as far as to suggest that by using the phrase it seemed
to contain multitudes to describe the performance of the cast, Charlotte was alluding to Western
images of Chinese masses.
5. Conclusion
Tis, then, in outline is the model provided by the appraisal framework of some of he
key resources of evaluation and stance. In its taxonomy of values of Attitude it provides
an account of the options available for construing diferent types of positive or nega-
tive assessment. In its notion of direct versus invoked attitude, it provides an account
of the options available for activating these assessments. By its account of the resources
of Engagement, it ofers a framework for exploring how the textual voice positions
itself with respect to such assessments, a framework for characterizing the diferent
intersubjective stances available to the textual voice.
34 Peter R.R. White
It is not possible or appropriate in the current context to go beyond this descrip-
tive outline and to set about demonstrating applications of the framework to text ana-
lytical issues. For such demonstrations, see, for example, Iedema et al. 1994; Eggins &
Slade 1997; Christie & Martin 1997; White 2002; Martin & White 2003; Macken &
Martin 2003). By way of conclusion, however, it may be useful to note that the appli-
cations which have been developed to this point have attended to the following types
of issues:
diferences in attitudinal profles (diferent patterns of occurrence of attitudi-
nal subtypes) by which individual texts or groupings of texts (for example those
representing a given register or genre) can be contrasted;
patternings within a text of the occurrence of attitudinal values by which functional
stages can be identifed;
associations between given social actors and particular types of attitudinal
assessment;
the role of implicit or invoked attitude in providing for the strategic impersonal-
ization of texts;
the association of particular patternings of dialogistic resources with rhetorical
efects such as the construction of authorial personas or the modelling of an
intended audience;
patterns of interplay between Attitude and Engagement which reveal the ideologi-
cal assumptions operating in the text.
Te ongoing research project out of which the appraisal framework has emerged is one
which seeks to provide a systematic account of the social semiotic principles by which
attitudinal assessments are activated and negotiated in texts and by which those texts
construct for themselves communities of shared feelings, values, tastes and beliefs. In
this it provide an account in which the lexico-grammatical, semantic and the social
and contextual are integrated, and by which, therefore, it becomes possible to provide
linguistically based explanations of such social efects as attitudinal positioning, the
construction of authorial personae and negotiations of solidarity.
References
Caldas-Coulthard, C.R. (1994). On reporting reporting: Te representation of speech in factual and
factional narratives. In M. Coulthard (ed.): 295309.
Chafe, W.L. & J. Nichols (eds.) (1986). Evidentiality: Te linguistic coding of epistemology. Ablex.
Channell, J. (1994). Vague language. Oxford University Press.
Christie, F. & J.R. Martin (eds.) (1997). Genres and institutions: Social processes in the workplace and
school. Cassell.
Appraisal 35
Cofn, C. (1997). Constructing and giving value to the past: an investigation into second school
history. In F. Christie & J.R. Martin (eds.): 196230.
Coulthard, M. (ed.) (1994). Advances in written text analysis. Routledge.
Crismore, A. (1989). Talking with readers: Metadiscourse as rhetorical act. Peter Lang.
Eggins, S. & D. Slade (1997). Analysing casual conversation. Cassell.
Fairclough, N. (1992). Discourse and social change. Polity Press.
Fuller, G. (1998). Cultivating science: Negotiating discourse in the popular texts of Stephen Jay
Gould. In J.R. Martin & R. Veel (eds.) Reading science Critical and functional perspectives on
discourses of science: 3562. Routledge.
Gruber, H. (1993). Evaluation devices in newspaper reports. Journal of Pragmatics 19(5): 469486.
Halliday, M.A.K. (1971). Linguistic function and literary style: An inquiry into the language of
William Goldings Te Inheritors. In S. Chatman (ed.) Literary style: 330368. Oxford Univer-
sity Press.
(1994). An introduction to functional grammar. Edward Arnold.
Hunston, S. (2000). Evaluation and the planes of discourse: Status and value in persuasive texts. In
S. Hunston & G. Tompson (eds.): 176206.
Hunston, S. & J. Sinclair (2000). A local grammar of evaluation. In S. Hunston & G. Tompson (eds.):
75100.
Hunston, S. & G. Tompson (eds.) (2000). Evaluation in text: Authorial stance and the construction of
discourse: 176206. Oxford University Press.
Hyland, K. (1996). Writing without conviction: Hedging in science research articles. Applied Linguis-
tics 17(4): 43354.
Iedema, R., S. Feez & P.R.R. White (1994). Media literacy. NSW Department of School Education,
Sydney.
Krner, H. (2001). Negotiating authority: Te logogenesis of dialogue in common law judgements.
Ph.D. Diss., University of Sydney.
Labov, W. (1984). Intensity. In D. Schifrin (ed.) Meaning, form and use in context: 4370. Georgetown
University Press.
Leech, G. (1983). Te principles of pragmatics. Longman.
Louw, B. (1993). Irony in the text of insincerity in the writer? Te diagnostic potential of semantic
prosody. In M. Baker, G. Francis & E. Tognini-Bognelle (eds.) Text and technology: 157176.
John Benjamins.
Lyons, J. (1977). Semantics. Cambridge University Press.
Macken, M. & J.R. Martin (eds.) (2003). Special issue on Appraisal of Text 23(2).
Malrieu, J.P. (1999). Evaluative semantics Cognition, language and ideology. Routledge.
Markkanen, R. & H. Schrder (eds.) (1997). Hedging and discourse: Approaches to the analysis of a
pragmatic phenomenon in academic texts. De Gruyter.
Martin, J.R. (1997). Analysing genre: Functional parameters. In F. Christie & J. Martin (eds.): 339.
(2000). Beyond exchange: APPRAISAL systems in English. In S. Hunston & G. Tompson
(eds.): 14275.
Martin, J.R. & G.A. Plum (1997). Construing experience: Some story genres. Journal of Narrative and
Life History 7: 299308.
Martin, J.R. & P.R.R. White (2003). Appraisal: the language of attitude and intersubjective stance.
Palgrave.
Matthiessen, C.M. (1995). Lexicogrammatical cartography: English systems. International Language
Sciences, Tokyo.
36 Peter R.R. White
Meyer, P.G. (1997). Hedging strategies in written academic discourse: Strengthening the argument
by weakening the claim. In R. Markkanen & H. Schrder (eds.): 2142.
Myers, G. (1989). Te pragmatics of politeness in scientifc articles. Applied Linguistics 10: 135.
Ochs, E. (ed.) (1989). Te pragmatics of afect. Special issue of Text 9(3).
Pagano, A. (1994). Negatives in written text. In M. Coulthard (ed.): 25065.
Palmer, F.R. (1986). Mood and modality. Cambridge University Press.
Plum, G. (1988). Textual and contextual conditioning in spoken English: a genre-based approach. Ph.D.
Diss., University of Sydney.
Rothery, J. & M. Stenglin (1997). Entertaining and instructing: Exploring experience through story.
In F. Christie & J.R. Martin (eds.): 231263.
(2000). Interpreting literature: Te role of Appraisal. in L. Unsworth (ed.) Researching lan-
guage in schools and functional linguistic perspectives: 222244. Cassell.
Rothery, J., M. Stenglin & J. Sinclair (1991). Corpus concordance collocation. Oxford University Press.
Voloinov, V.N. (1995). Marxism and the philosophy of language. Routledge.
White, P.R.R. (2000). Dialogue and inter-subjectivity: Reinterpreting the semantics of modality and
hedging. In M. Coulthard, J. Cotterill & F. Rock (eds.) Working with dialog: 6780. Niemeyer.
(2002). Appraisal the language of evaluation and intersubjective stance. http://www.
grammatics.com/appraisal.
(2003). Beyond modality and hedging: a dialogic view of the language of intersubjective
stance. In M. Macken & J. Martin (eds.).
Cohesion and coherence
Wolfram Bublitz
University of Augsburg
1. Introduction
Linguists use the two notions of cohesion and coherence to refer to the (linguistically
encoded or just assumed) connectedness of spoken as well as written discourse or text.
Of course, connecting relations also hold among elements of structure within grammat-
ical units such as word, phrase, clause or sentence. But these intra-sentential relations
are diferent in kind because they are determined by phonological and grammatical
rules and described, inter alia, as syntactic-semantic relations of valency, dependency,
constituency, modifcation. Cohesion, operating inter-sententially, and coherence are
key notions in text and discourse analysis, as well as in pragmatics because they also
relate to the complex interrelationship between form, meaning and use of linguistic
expressions in specifc (social) contexts.
Native speakers have intuitions about which sequences of utterances do or do not
constitute discourse or text. If, by way of an experiment, we deliberately distort a per-
fectly comprehensible and acceptable text by, e.g. changing the order of its utterances
or its linguistic, situational or socio-cultural context, the efect will be one of confusion
on the part of our hearers or readers. Tey may still understand each individual utter-
ance but not the resulting string of utterances as a whole, i.e. as one unit with a defnite
function in its environment. In the eye of the language user who is trying to interpret
them, they do not hang together in a reasonable way. I.e., they do not display order
and do not form a meaningful gestalt that fts both into the linguistic environment as
well as the social situation, serves the accepted communicative goal and contributes
to the topic at hand; in other words, they are not coherent. Accordingly, the defning
characteristic of such instances of discourse or text is coherence, which itself rests on
text-forming resources such as cohesion and general structural properties determined
by register or genre.
Tough both cohesion and coherence refer to meaning resting on relations of
connectedness (between individual propositions and sets of propositions), which may
or may not be linguistically encoded, they are descriptive categories which difer in
kind. Cohesion refers to inter-sentential semantic relations which link current items
with preceding or following ones by lexical and structural means (cf. below). Cohe-
sion is a kind of textual prosody. Since J.R. Firth, who perceived prosodic efects as
phonological colouring, we use prosody to refer to the property of a feature to extend
38 Wolfram Bublitz
its domain, stretch over and afect not just one but several units. Analogously, textual
prosody refers to cohesive colouring involving more than one element in discourse or
text. As cohesion is anchored in its forms, we can argue that it is an invariant, user and
context independent property of a piece of discourse or text. Coherence, on the other
hand, is a cognitive category that depends on the language users interpretation and is
not an invariant property of discourse or text.
Tough both cohesion and coherence have found their place as key terms in text
and discourse analysis, they still mean diferent things to diferent people. Simplifying
matters drastically, we can say that form and structure oriented linguists, who regard a
text as a kind of long sentence, i.e. as a unit beyond the sentence, focus on cohesion as
an essential feature of textuality. Function oriented linguists, on the other hand, who
equate text with any linguistic expression of any length which is used to perform a
specifc function, focus on coherence as the defning feature of textuality.
2. Focus on form: Cohesion
Te most infuential account of cohesion was developed by Halliday and Hasan in
their book Cohesion in English. Teir concept of cohesion is a semantic one because
it refers to relations of meaning that exist within the text (1976: 4) and enable one
part of the text to function as the context for another (Halliday & Hasan 1989: 489).
Cohesion manifests itself in linguistic means that appear at the surface level of lan-
guage. An example of such a cohesive means is them in Wash and core six cooking
apples. Put them into a freproof dish (1976: 2), which presupposes for its interpreta-
tion something other than itself (1976: 4), here the expression six cooking apples in
the preceding sentence. Cohesion, they argue, occurs where the interpretation of
some element in the discourse is dependent on that of another. Te one presupposes
the other, in the sense that it cannot be efectively decoded except by recourse to it.
When this happens, a relation of cohesion is set up (1976: 4).
Halliday and Hasan focus on fve kinds of cohesive ties between utterances: (1)
reference, (2) substitution, (3) ellipsis, (4) conjunction, and (5) lexical cohesion. (1)
Reference is described as situational or exophoric, resting on means that refer to the
situation at hand (as in Would you like to join me for a cup of tea this afernoon?), and as
textual or endophoric reference, resting on anaphora or cataphora that refer to partici-
pant roles (personal reference as in Doris likes him very much), to (degrees of) proxim-
ity (demonstrative reference as in Joe wont get tenure. Tis is what I cant understand),
or to (degrees of) similarity (comparative reference as in He always eats meat balls with
his fngers. I detest such manners). (2) Substitution involves noun phrases (My beer is
warm. Please get me a chilled one), verb phrases (I have never talked in Jena before. But
Cohesion and coherence 39
I did in Halle a couple of years ago) and clauses (I trust you know how to open a bottle
of wine? I believe so). (3) Ellipsis is described as substitution by zero (1976: 142) and
likewise involves noun phrases, verb phrases and clauses. (4) Tere are four types of
conjunction: additive (expressed by and, or, furthermore etc.), adversative (but, how-
ever etc.), causal (so, therefore etc.), temporal (then, next, fnally etc.). (5) Lexical cohe-
sion, which rests on identity of reference between two items, is realized by forms
of reiteration such as word-for-word repetitions, synonyms, superordinates, general
words, so-called collocations, i.e. members of the same lexical feld (euro, cent; moun-
tain, climb, peak) and items which, though they are not hyponyms or synonyms in
the language system, are nonetheless used as such in discourse or text, acquiring their
coreferential status only instantially (1976: 289).
In the wake of Halliday and Hasans proposal, much work has been devoted to

re-classify and extend their inventory of cohesive means (cf. Martin 1992). Essentially,
all means can be categorized into three types because cohesion stretches three ways:
there is reference to what has been said, to what will be said and also to what could have
been said instead. Tis three-way distinction corresponds to Roman Jakobsons well-
known two modes of arrangement of verbal means, selection (paradigmatic choice,
relatedness in the system) and combination (syntagmatic relatedness), to which, for
a higher degree of delicacy, we can add the two vectors of retrospection and prospec-
tion (cf. Hasan 1984). Te following overview categorizes cohesive means into classes
according to their prospective or retrospective orientation.
Retrospective cohesive means are proforms (as anaphora), substitutions, syn-
onyms, hyponyms, word-for-word repetitions, general words or labels and second
pair parts of adjacency pairs. Except for the latter, they are based on similarity of form,
structure, content and function and can therefore be labelled forms of repetition (or
parallelism). Te three essential properties of repetition are quality, quantity and dis-
tribution. Quality is best described as a scale which represents a continuum of fxity
vs. looseness of form and indicates the degree of formal as well as semantic corre-
spondence between the parallel items involved; there is a cline from total equivalence
to paraphrastic substitution. Quantity refers to the length of the repeating item and
is likewise best represented on a scale with a single word (or even phone) at one end
and a string of utterances at the other. Distribution refers to the distance between the
repeated and the repeating item, which can be anywhere between closely adjacent and
considerably removed.
Prospective devices are used to create a slot (to be flled), thereby reducing the
number of options and, generally, setting up expectations. Tey include proforms (as
cataphora), text structuring discourse markers (incidentally, actually, cf. Lenk 1998)
and gambits (frst of all, before I forget, and another thing), and also general nouns or
labels (Francis 1994) like aspect, dilemma, episode, pattern, problem. Labels do not just
40 Wolfram Bublitz
refer to sections but categorize them; in (1) there is both a prospective (stuf) and a
retrospective (things) label:
(1) I always meant going down to the shops on Saturday to get fresh all the stuf
to get meat and vegetables beefurgers sausages all these things the children
eat for tea. (London Lund Corpus 4.3.90 f)
To these two classes of cohesive means we can add forms of ordered arrangements that
function both prospectively and retrospectively at the same time. Tey rest on prin-
ciples such as information assessment and iconicity. Te former has been described
from a variety of diferent points of view, prominent among them the Prague Schools
principle of functional sentence perspective, according to which each sentence part is
evaluated for its relative importance for the information carried by the sentence as a
whole and thus for its communicative function within its contextual environment.
Known or given information (i.e. theme or topic status) has a low degree of com-
municative dynamism (because it contributes little to informational progress) while
new information (i.e. rheme or comment status) has a high degree of communicative
dynamism. Te assignment of theme and rheme to parts of an utterance has a clear
cohesive function because it sets up links with preceding and following utterances
(cf. Firbas 1992). Te second principle of arrangement, iconicity, is easiest illustrated
by Harvey Sacks much-quoted example Te baby cried. Te mommy picked it up,
where the temporal sequence of events is refected in the linear sequence of the two
adjacent utterances. Adjacency based on iconicity is thus another powerful means of
creating cohesive linkage.
3. Cohesion as a condition for coherence
Cohesion should be kept strictly apart from coherence. It is neither a sufcient nor a
necessary condition for coherence. Referring to van Dijks example
(2) We will have guests for lunch. Caldern was a great Spanish writer
(van Dijk 1972: 40)
Edmondson shows that, though the two utterances are not cohesively connected, they
are nonetheless coherent in a context like (3):
(3) Do you know Caldern died exactly 100 years ago today? Good heavens!
Id forgotten. Te occasion shall not pass unnoticed. We will have guests for
lunch. Caldern was a great Spanish writer. I shall invite Professor Wilson and
Senor Castellano right away. (Edmondson 1981: 13)
Furthermore, in authentic discourse or text, we can encounter long stretches of utter-
ances which, though they are not cohesively connected, are nonetheless accepted as
coherent texts because they are, e.g., parts of enumerations or cases of dissociated
interior monologue.
However, while cohesion is not a necessary condition of coherence, studies
have shown that discourse and text tend to be cohesive to a greater or lesser extent,
depending on genre. Following a general principle of cooperation, speakers/writers
are anxious to generate cohesion as a means of guiding their recipients interpretation
of coherence and thus, ultimately, of securing comprehension. Cohesive means are
cues which signal or indicate the preferred line of coherence interpretation. A lack of
cohesive means may disturb the hearers/readers interpretation of coherence. Easy and
unimpeded interpretation of both discourse and text as coherent depends consider-
ably on the presence of communicative and meta-communicative textual links.
Of course, the equation more cohesion = more coherence is not valid per se. And
yet, there are genres, among them everyday face-to-face conversation, in which the
wealth and abundance of cohesive means is quite extraordinary, as in (4):
(4) A well I must admit I feel I mean Edwards mother and his great and his
grandfather will come up on Christmas Day but I feel somehow the sheer
fact of not having to have to have this really sort of its for one thing it does
nark me that its so bloody expensive that he wont eat anything except the
largest most splendid pieces of meat you know
B how annoying
A and it upsets me you know if he needed it I wouldnt mind
D come to think of it hes also hes also an extremely greedy individual who
A yeah so that if you buy enough [f]
D he isnt satisfed with a normal portion
A for cheese for for three days if he sees its there hell eat it you know
(London Lund Corpus 4.3)
Cohesive means come on all levels. In (4) we fnd on the level of phonetics and pros-
ody alliteration, assonance, rhythm (cf. Couper-Kuhlen 1983) and sound-rows or
-sequences as in cheese three he sees eat (Sacks 1992). On the lexico- syntactic
level we notice articles, proforms, deictics, conjunctions, complementizers, labels
(somehow, do, anything, individual) and members of a common script or semantic
feld (mother, grandfather; eat, meat, greedy, portion; nark, upset, annoy). And from
a pragmatic and text-analytic viewpoint we recognize speech act pairs (statement
supportive statement accompanied by supportive elements like yes, how annoying,
yeah), hedges, gambits, discourse markers, collocations and tags (well, I mean, sort of,
you know, come to think of it, I must admit, I feel somehow, the sheer fact, I wouldnt
mind, for one thing). Tere is an abundance of cohesive means to the point where they
add up to a rather high degree of redundancy. Hearers, however, who have to process
42 Wolfram Bublitz
online, to interpret ongoing talk in actu (without the possibility to look ahead or the
time to recall what was said before in any detail) depend on a certain degree of redun-
dancy. Redundancy helps relieve their memory and gives them time to understand.
In everyday face-to-face discourse, at least, securing comprehension as a principle
ranks higher than avoiding redundancy. Occasionally, speakers may come up with
too many cohesive cues which add up to a degree of redundancy that is no longer
tolerable and can irritate or even annoy the hearer. Such cases of cohesive overkill
are triggered by the speakers misjudgement of the hearers interpretive competence
and range of knowledge.
While there is no direct correlation between (lack of) cohesion and (lack of) coher-
ence, to claim that cohesion is not a sufcient condition for coherence is much more
arguable especially if we accept that the speakers/writers primary motive for using
cohesive means is to help secure coherence. And indeed, the examples which are given
to prove that cohesion alone does not generate coherence, are not very convincing:
(5) Te heads of the citys uniformed services polished their contingency plans for
a strike. Queen Wilhelmina fnalized her own plans for the evening. In a nearby
Danish town, two fshmongers exchanged blows. Anders, by far the stronger,
had a cousin in prison. Many criminals are in prison. (Samet & Schank 1984: 63)
Te authors claim that (5) is cohesive but not coherent because the utterances
exhibit [] connections but make no sense at all (Samet & Schank 1984: 63).
However, as we saw when we looked at Edmondsons reading of the non-cohesive
sequence of utterances in example (2), judgements of coherence, or, for that mat-
ter, incoherence are not invariantly triggered by the text at hand for any hearer or
reader alike. If the text in (5) was authentic, i.e. an instance of naturally occurring
data such as a narrative, its hearers or readers would easily succeed in reaching a
plausible interpretation of coherence by resorting to the larger context, to the situa-
tion at hand, to the overall communicative goal, to their encyclopedic knowledge and
to other sources of supporting data. But even as a piece of constructed, context and
situation free text, (5) allows for a coherent reading. To ask people whether or not an
isolated sequence of utterances has coherence is tantamount to asking whether or
not they have enough imagination to come up with a context in which the sequence is
indeed coherent. To reach coherence, they will then quite naturally rely on the cohe-
sive means given in the text. Hence, cohesion is normally a sufcient condition for
coherence because it serves as a powerful and suggestive guideline for the hearers/
readers interpretation.
4. Focus on meaning: Connectivity
It has been argued that text which is not cohesive can, nonetheless, be coherent provided
the propositions underlying its utterances are semantically related to each other. In
this view, coherence is a semantic notion resting on (a net of) semantic relations. Te
variety of semantic or connectivity relations that have been described in the literature
so far can be categorized into a few core classes. Among them we fnd causality, refer-
ence, coordination (parallel, contrastive), elaboration (example, generalization, para-
phrase), overlap and contiguity (temporal, spatial, aspectual, referential) (cf. Samet &
Schank 1984) as well as so-called scripts, schemata or frames which refer to socially
defned activities and events and help participants apply their knowledge of the world
to the interpretation of what goes on in an encounter (Gumperz 1982: 154). Viewing
coherence as a semantic notion usually leads to the assumption that it is a feature of, or
rather in the text. It is there, in the text for people to fnd it. Returning to example (2)
with its interpretation in (3), we can now argue that its coherence rests on the semantic
relations of causality and coordination, which do not only link the propositions of the
two utterances but also two additional though latent, i.e. not realized propositions; here
is a possible paraphrase: We will have guests for lunch because we want to celebrate
because it is Calderns birthday today and because Caldern was a great Spanish
writer. Te connectivity of (2) is only partly refected on the surface level of cohesion
by the linear ordering of the two utterances. To make a clear distinction between con-
nectivity and cohesion (with coherence resting on either or on both of these), is justifed
on theoretical grounds. In practical discourse analysis, however, the two concepts are
not always easy to keep apart.
5. Semantic connectivity as a condition for coherence
From the fact that coherence is frequently based on semantic connectivity we may
conclude that the latter is both a sufcient and a necessary condition for coherence.
It has been argued, however, that this is too strong a claim. Basically, the argument is
supported by fve observations. Firstly, adjacent utterances can be semantically related
without being coherent. In (5) there is connectivity (resting on the semantic relations
of contiguity, time and coordination) but not coherence. Samet & Schank point out
that our ability to render certain elements of a text understandable by using these
[semantic] connections cannot itself make a text coherent to the point of making
sense (1984: 64). Secondly, semantic relations can involve utterances which are not
adjacent as, e.g., in enumerations where they are related to a superordinate topic but
not to each other. Tirdly, utterances can be semantically related in ways that are not
clearly identifable or which allow for alternative identifcation. Tus, in authentic dis-
course or text it may be difcult or even impossible to fgure out exactly which seman-
tic relation is involved, as in (6):
(6) John bought a raincoat. He went shopping yesterday on Queen Street
and it rained. (Morris & Hirst 1991: 25)
44 Wolfram Bublitz
Each of several connectivity relations is a likely candidate for coherence: Te coher-
ence relation here could be elaboration (on the buying), or explanation (of when, how,
or why), or cause (he bought the raincoat because it was raining out) (Morris & Hirst
1991: 25). Fourthly, two adjacent utterances can be semantically related (and thus pos-
sibly coherent) for one hearer/reader but not for another, as in (7):
(7) S Well unless youre not a member, if yer a member of TM people do, ah
simply because its such a fucking high price to get in there (1.0 sec) its like
thirty fve dollars
C its like water polo (2.0 sec)
S Why, is it expensive (Vuchinich 1977: 246)
Tis exchange is part of an experiment. In order to test the reaction of his/her inter-
locutor the organiser C makes a deliberately incoherent contribution, i.e. one that is
not semantically related to the preceding utterances. However, as is clear from his/her
reaction, S does see a semantic relation (of comparison) between its like water polo and
the preceding text and understands it as coherent (though not straight away, he/she
needs more processing time than usual). Fifhly, two or more utterances can be con-
nected by a semantic relation which, however, cannot be inferred from the linguistic
or the non-linguistic context but only from previously acquired knowledge or from
experience (as is ofen the case with allusions).
Of course, not one of these fve lines of reasoning precludes connectivity from
being at least a major condition for coherence. Afer all, connectivity rests on seman-
tic relations of the kinds mentioned above rather than on transparence, adjacency or
absoluteness.
6. Coherence: A general view
Coherence is a concept which in its complexity is still not fully understood and a
matter of continuing debate. Tough it has doubtless found its place as a key term
in text and discourse analysis, its usage continues to vary to the extent that to give
a comprehensive overview of even the major views advocated goes well beyond the
scope of a handbook article. While afer the publication of Halliday & Hasan (1976)
the notion of cohesion was widely welcomed and accepted as a well-defned and
useful category, coherence was ofen regarded or even occasionally dismissed as a
vague, fuzzy and rather mystical notion (Sinclair 1991: 102) with little practical
value for the text or discourse analyst. Tis view was held by parts of the linguistic
community with, however, some notable exceptions, prominent among them her-
meneutic, context- and interpretation-based dynamic concepts of coherence (cf.
below). Since the late eighties, there has been renewed interest in the intriguing
notion of coherence. Te remarkable number of almost fve hundred titles listed
in a recent bibliography (Bublitz 2010) bears witness to this development and to a
rapidly changing scene in coherence research which is moving away from reducing
coherence to a mere product of (formally represented) cohesion and/or (semantically
established) connectivity.
In her overview, Hellman (1995) distinguishes, inter alia, between approaches
which see coherence (a) as a formal property of texts (p.191f), (b) as a discourse
processing concept (p.194f), referring primarily to work by Sanders & Spooren(who,
in a 1999 paper, juxtapose the linguists view of coherence as a relational concept and
the cognitivists view of coherence as a realisation of participant intentions manifest
in each section of a discourse), (c) as a result of computing referential, causal or
other relations, i.e. of a complex problem-solving process in which the reader infers
relations among the ideas, events and states that are described in the text (p.195),
and (d) as a result of computing, i.e. recognising the intention(s) of a discourse
producer (p.196).
7. A hermeneutic, context and interpretation based
view of coherence
Much recent work describes coherence as a mental notion which is interactively nego-
tiated within a given socio-cultural setting and less dependent on the language of
discourse or text itself (cf. the readers by Gernsbacher & Givn 1995; Bublitz et al.
1999). Such a hermeneutic approach describes coherence as a context dependent, user
oriented and comprehension based notion. Tis view, which dominates work by Fritz
(1982), Brown & Yule (1983) and many others, is in accordance with a fairly long
interpretive tradition in Europe (represented by Ludwig Wittgenstein, Alfred Schtz,
Harold Garfnkel, Hans-Georg Gadamer, Anthony Giddens and others) and the USA.
Within a sociological and ethnographic framework, Dell Hymes, John Gumperz and
others (cf. the reader edited by Bauman & Sherzer 1974) argue for analysing language
within its sociocultural settings. Speakers and hearers alike come to an understand-
ing of the ongoing communicative interaction by linking linguistic and non-linguistic
cues with their background knowledge. Tey thus continually and jointly negotiate
meaning by constructing a shared context. From such a contextualizing, interpretive
viewpoint, speakers/writers are said to intend, anticipate and (overtly and/or covertly)
suggest coherence while hearers/readers ascribe coherence to utterances within their
linguistic, situational and socio-cultural context.
Because much recent research into coherence follows a context-, negotiaton-,
interpretation-dependent view of coherence, a snapshot account of it seems indicated.
According to this view, coherence is not a discourse or text inherent property, i.e. it is
46 Wolfram Bublitz
not given in discourse or text independently of interpretation. Consequently, one can-
not say a text has coherence in the same way as one can say a text has a beginning
or an end, or indeed, a text has cohesion (the latter being a text inherent property).
We can only say someone understands a text as coherent. Of course, coherence is
based on the language of the text in the same way as it is based on other information
provided by the linguistic context, the socio-cultural environment, the valid commu-
nicative principles and maxims, the interpreters encyclopedic knowledge etc.
Since it is not texts that cohere but rather people who make texts cohere, we can
say that for one and the same text there exist a speakers/writers, a hearers/readers and
an analysts coherence, which may or may not match. Typically, diferent interpreta-
tions of the coherence of a text depend on its linguistic complexity, the temporal, local
and social setting, the interpreters familiarity with genre and content as well as his/
her knowledge of the speakers/writers background (motives, preferences, interests).
Being dependent on interpretation means that speakers/writers can never produce
coherence which is binding for hearers/readers. It is the latter who have to arrive at
their own understanding of coherence. Normally speakers/writers are set to help cre-
ate coherence by (more or less subtly) guiding their hearers/readers to a suggested line
of understanding. Conversely, hearers/readers use these guiding signals as instruc-
tions to align their interpretations with what they take to be the speakers/writers
intentions. Hearers/readers assemble and subsequently test a view of coherence which
they assume comes closest to that of the speaker/writer. Hence, coherence is rarely
static but frequently dynamic, i.e. a process rather than a state. It can be tentative and
temporary because it is continually checked against any new information which may
make adaption and updating necessary. Of course, eventually, coherence (especially of
written texts) can lose some of its provisional and temporary character and acquire a
higher degree of permanence.
Coherence is the outcome of the language users gestalt creating power. People
are driven by a strong desire to identify forms, relations, connections which they
can maximize in order to turn fragments into whole gestalts, i.e. to see coherence in
strings of utterances. Coherence is also a cooperative achievement (in ongoing dis-
course more than in petrifed text) because it depends on both the speakers/writers
and the hearers/readers willingness to negotiate coherence. Mutual understanding
not only rests on the participants sharing the same socio-cultural background, the
same range of knowledge and communicative assumptions, but also on their ability
to fgure out unshared experience, i.e. to adjust their own world-view to that of their
interlocutors. Hearers/readers are constantly engaged in trying to re-create coherence
as an equivalent of the speakers/writers coherence, but despite their eforts they can
never succeed in coming up with an exact copy. Coherence is only approximate and a
matter of degree and best described as a scalar notion. Any interpretation of coherence
is restricted and, accordingly, partial to diferent degrees.
8. Coherence as a default assumption
As a rule, speakers/writers and hearers/readers alike operate on a standard or default
assumption of coherence. Tis explains why despite the fact that ascribing coherence
frequently involves making difcult and complex acts of inference, cases of disturbed
coherence are not abundant, and why people try to understand coherence even though
the data to go by may look sadly insufcient. As long as there is no evidence to the
contrary, they proceed from the assumption that what they hear and read can be made
coherent, even if this involves making extremely remote and unlikely connections.
Speakers/writers likewise rely on the default principle and expect hearers/readers to
assume that what they say or write is coherent. Te default principle of coherence is
accepted for two reasons. Firstly, it relates to a more general strategy according to
which it is normally assumed by the interactants that hearers/readers do as little pro-
cessing as possible (Brown & Yule 1983: 60), opting for the obvious and likely reading.
Secondly, it follows from the general principle of cooperation (in the Gricean sense)
and is therefore an essential normative basis of communication, which leads to ratio-
nal behaviour as the only path to efcient communication. (Cf. also Lindes observa-
tion that the process of creating coherence is a social obligation 1993: 16).
9. Perspectives
Tis overview can only touch upon a limited number of methodological and theoreti-
cal approaches to the description of cohesion and coherence, on the one hand, and of
ways and means of (re-)creating them, on the other. For some time now there have been
several common tendencies, among them the tendency to describe a far larger inven-
tory of cohesive means than originally proposed by Halliday and Hasan, the tendency
to refrain from accepting only one canonic defnition of coherence, and the tendency
to observe a basic stock of fundamental theoretical and methodological assumptions.
Much recent work demands such descriptive principles as the need to rely on authentic
data, to proceed in an interdisciplinary way, to assume a cultural and common sense
basis for coherence (cf. Linde 1993: 192f), to relate micro-linguistic to macro-linguistic
(e.g. socio-cultural) issues, and to focus on the powerful coherence-securing role of
such (sometimes long neglected) means as gestures (cf. McNeill & Levy 1993), dis-
course topics, collocational orientation.
Discourse topic and the various procedures of handling it (introducing, changing,
shifing, digressing from it) are prevalent and strong means of (re-)creating coherence
(cf. Geluykens 1999). In many societies, speakers/writers are expected to stick to a
topic. Taking up social space is only justifed if an utterance is evidently a contribution
to the topic at hand, i.e. an acceptable answer to the generic question [] why that
48 Wolfram Bublitz
now (Scheglof 1990: 55). Te relationship between discourse topic and overarch-
ing communicative goals is one of the stubborn problems that future research has to
address. Collocations based on shared polarity, following from the nodes negative or
positive semantic prosody, contribute to coherence because they are likely to recur
in the local as well as global environment. Te semantic orientation of the individual
collocation is then not regarded as a single occurrence with no bearing on context but
rather as indicating the overall tenor of the discourse, i.e. its general semantic orienta-
tion. On account of such predictive force, they are regularly instrumental in dispersing
meaning in discourse (cf. Bublitz 1996).
Future research will also have to deal with unsolved problems such as how means
of securing coherence (and ways of describing them) vary from spoken to written
language, from genre to genre, from text type to text type (cf., e.g., Fritz 1999 on
coherence in hypertext), from one society to another, from earlier to later stages of a
language (the diachronic perspective), and from earlier to later stages of acquiring a
language (both as a native speaker and a foreign language learner).
References
Bauman, R. & J. Sherzer (eds.) (1974). Explorations in the Ethnography of Speaking. Cambridge
University Press.
Brown, G. & G. Yule (1983). Discourse analysis. Cambridge University Press.
Bublitz, W. (1996). Semantic prosody and cohesive company. Leuvense Bijdragen 85: 132.
Bublitz, W. (2010). A bibliography of coherence and cohesion. http://www.philhist.uniaugsburg.de/de/
lehrstuehle/anglistik/sprachwissenschaf/downloads_intern/Bibliography_of_Coherence.pdf
Bublitz, W., U. Lenk & E. Ventola (eds.) (1999). Coherence in spoken and written discourse: how to
create it and how to describe it. John Benjamins.
Couper-Kuhlen, E. (1983). Intonatorische Kohsion. Zeitschrif fr Literaturwissenschaf und Lin-
guistik 49: 74100.
Edmondson, W. (1981). Spoken discourse. Longman.
Francis, G. (1994). Labelling discourse: an aspect of nominal-group lexical cohesion. In M. Coulthard
(ed.) Advances in written text analysis: 83101. Routledge.
Firbas, J. (1992). Functional sentence perspective in written and spoken communication. Cambridge
University Press.
Fritz, G. (1982). Kohrenz. Narr.
(1999). Coherence in hypertext. In W. Bublitz et al. (eds.): 221232.
Geluykens, R. (1999). It takes two to cohere: the collaborative dimension of topical coherence in
conversation. In W. Bublitz et al. (eds.): 3353.
Gernsbacher, M.A. & T. Givn (eds.) (1995). Coherence in spontaneous text. John Benjamins.
Gumperz, J.J. (1982). Discourse strategies. Cambridge University Press.
Halliday, M.A.K. & R. Hasan (1976). Cohesion in English. Longman.
(1989). Language, context, and text: aspects of language in a social-semiotic perspective. Oxford
University Press.
Hasan, R. (1984). Coherence and cohesive harmony. In J. Flood (ed.) Understanding reading
comprehension: 181219. International Reading Association.
Hellman, C. (1995). Te notion of coherence in discourse. In G. Rickheit & C. Habel (eds.) Focus and
coherence in discourse processing: 190202. de Gruyter.
Lenk, U. (1998). Marking discourse coherence. Narr.
Linde, C. (1993). Life stories. Te creation of coherence. Oxford University Press.
Martin, J.R. (1992). English text. System and structure. John Benjamins.
Mcneill, D. & E.T. Levy (1993). Cohesion and gesture. Discourse Processes 16: 363386.
Morris, J. & G. Hirst (1991). Lexical cohesion computed by thesaural relations as an indicator of the
structure of text. Computational Linguistics 17: 2148.
Sacks, H. (1992). Lectures on conversation. Blackwell.
Samet, J. & R. Schank (1984). Coherence and connectivity. Linguistics and Philosophy 7: 5782.
Sanders, T. & W. Spooren (1999). Communicative intentions and coherence relations. In W. Bublitz
et al. (eds.): 235250.
Scheglof, E. (1990). On the organization of sequences as a source of coherence in talk-in-interac-
tion. In B. Dorval (ed.) Conversational organization and its development: 5178. Ablex.
Sinclair, J. (1991). Corpus, concordance, collocation. Oxford University Press.
Van Dijk, T. (1972). Some aspects of text grammars. Mouton.
Vuchinich, S. (1977). Elements of cohesion between turns in ordinary conversation. Semiotics 20:
229257.
Critical Linguistics
and Critical Discourse Analysis
Ruth Wodak
University of Lancaster
1. Defnitions
Te terms Critical Linguistics (CL) and Critical Discourse Analysis (CDA) have been
frequently used interchangeably. Recently, however, the term CDA seems to have
been preferred and is being used to denote the theory formerly identifed as CL.
Tus, I will continue to use CDA exclusively in this paper (see Anthonissen 2001 for
an extensive discussion of these terms). Te roots of CDA lie in classical Rhetoric,
Text linguistics and Sociolinguistics, as well as in Applied Linguistics and Pragmatics
(see also Wodak & Meyer 2009; Fairclough 2003; Wodak 2004, 2007; Renkema 2004;
Blommaert 2005).
Deconstructing the label of this research program I view CDA basically as a
research program, the reasons for which I will explain below entails that we have to
defne what CDA means when employing the terms critical and discourse. Michael
Billig (2003) has clearly pointed to the fact that CDA has become an established aca-
demic discipline with the same rituals and institutional practices as all other academic
disciplines. Ironically, he asks the question whether this might mean that CDA has
become uncritical or if the use of acronyms such as CDA might serve the same
purposes as in other traditional, non-critical disciplines; namely to exclude outsiders
and to mystify the functions and intentions of the research. I cannot answer Billigs
questions extensively in this chapter. But I do believe that he points to potentially very
fruitful and necessary debates for CDA.
It is necessary to stress that CDA has never been and has never attempted to be or
to provide one single or specifc theory. Neither is one specifc methodology character-
istic of research in CDA. Quite the contrary; studies in CDA are multifarious, derived
from quite diferent theoretical backgrounds, oriented towards diferent data and
methodologies. Researchers in CDA also rely on a variety of grammatical approaches.
Te defnitions of the terms discourse, critical, ideology, power and so on are
also manifold. Tus, any criticism of CDA should always specify which research or
researcher they relate to. I myself would suggest using the notion of a school for CDA,
or of a program, which many researchers fnd useful and to which they can relate. Tis
program or set of principles has changed over the years (see Fairclough & Wodak 1997).
Critical Linguistics and Critical Discourse Analysis 51
Such a heterogeneous school might be confusing for some; on the other hand, it
allows for open discussions and debates, for changes in the defnition of aims and goals,
and for innovation. Tis heterogeneity of methodological and theoretical approaches
that can be found in this feld would tend to confrm Van Dijks point that CDA and CL
are at most a shared perspective on doing linguistic, semiotic or discourse analysis
(Van Dijk 1993b: 131). Below, I summarize some of these principles, which are shared
by most researchers.
Most importantly, CDA sees language as social practice (Fairclough & Wodak
1997), and considers the context of language use to be crucial (Wodak 2000, 2008;
Benke 2000):
CDA sees discourse language use in speech and writing as a form of social
practice. Describing discourse as social practice implies a dialectical relationship
between a particular discursive event and the situation(s), institution(s) and social
structure(s), which frame it: Te discursive event is shaped by them, but it also shapes
them. Tat is, discourse is socially constitutive as well as socially conditioned
it constitutes situations, objects of knowledge, and the social identities of and
relationships between people and groups of people. It is constitutive both in the sense
that it helps to sustain and reproduce the social status quo, and in the sense that it
contributes to transforming it. Since discourse is so socially consequential, it gives
rise to important issues of power. Discursive practices may have major ideological
efects that is, they can help produce and reproduce unequal power relations
between (for instance) social classes, women and men, and ethnic/cultural majorities
and minorities through the ways in which they represent things and position people.
(Fairclough & Wodak 1997: 258)
Of course, the term discourse is used very diferently by diferent researchers and also
in diferent academic cultures. In the German and Central European context, a distinc-
tion is made between text and discourse, relating to the tradition in text linguistics
as well as to rhetoric (see Brnner & Graefen 1994; Wodak 1996 for summaries). In the
English speaking world, discourse is ofen used both for written and oral texts (see
Schifrin 1994). Other researchers distinguish between diferent levels of abstractness:
Lemke (1995) defnes text as the concrete realization of abstract forms of knowledge
(discourse), thus adhering to a more Foucauldian approach (see also Jger et al. 2001).
Te shared perspective and program of CDA relate to the term critical, which in
the work of some critical linguists could be traced to the infuence of the Frankfurt
School and Jrgen Habermas (Tompson 1988: 71f; Fay 1987: 203; Anthonissen
2001). Nowadays this concept is conventionally used in a broader sense, denoting, as
Krings argues, the practical linking of social and political engagement with a socio-
logically informed construction of society, (Krings et al. 1973: 808). Hence, critique
is essentially making visible the interconnectedness of things (Fairclough 1995: 747;
see also Connerton 1976: 1139). Te reference to the contribution of Critical Teory
52 Ruth Wodak
to the understanding of CDA and the notions of critical and ideology are of par-
ticular importance. (See Anthonissen 2001 and Chilton et al. 2010 for an extensive
discussion of this issue)
1
Critical theories, thus also CDA, are aforded special standing as guides for
human action. Tey are aimed at producing enlightenment and emancipation.
Such theories seek not only to describe and explain, but also to root out a particular
kind of delusion. Even with difering concepts of ideology, critical theory seeks to
create awareness in agents of their own needs and interests. Tis was, of course, also
taken up by Pierre Bourdieus concepts of violence symbolique and mconnais-
sance (Bourdieu 1989). One of the aims of CDA is to demystify discourses by
deciphering ideologies.
In agreement with its Critical Teory predecessors, CDA emphasizes the need for
interdisciplinary work in order to gain a proper understanding of how language func-
tions in constituting and transmitting knowledge, in organizing social institutions or
in exercising power (see van Dijk 2008; Graham 2002; Lemke 2003; Martin 2002; Gee
2004; Blommaert 2005; Wodak 2009).
An important perspective in CDA related to the notion of power is that it is
very rare that a text is the work of any one person. In texts discursive diferences are
negotiated; they are governed by diferences in power which is in part encoded in
and determined by discourse and by genre. Terefore texts are ofen sites of struggle
in that they show traces of difering discourses and ideologies contending and strug-
gling for dominance.
Tus, defning features of CDA are its concern with power as a central condi-
tion in social life, and its eforts to develop a theory of language, which incorporates
this as a major premise. Not only the notion of struggles for power and control, but
also the intertextuality and recontextualization of competing discourses in various
public spaces and genres are closely attended to (Iedema 1999, 2003; Muntigl et al.
2000; Wodak & Koller 2008). Power is about relations of diference, and particularly
about the efects of diferences in social structures. Te constant unity of language and
other social matters ensures that language is entwined in social power in a number of
ways: language indexes power and expresses power; language is involved where there
is contention over and a challenge to power. Power does not derive from language,
but language can be used to challenge power, to subvert it, to alter distributions of
:. In the 1960s, many scholars adopted a more critical perspective in language studies. Among
the rst was the French scholar Pcheux (1982 [1975]), whose approach traced its roots to the
work of Russian theorists Bakhtin (1981) and Voloinov (1973), both of whom had postulated an
integration of language and social processes in the 1930s. Te term itself was apparently coined
by Jacob Mey (1974).
power in the short and the long term. Language provides a fnely articulated vehicle
for diferences in power in hierarchical social structures.
CDA might thus be defned as being fundamentally interested in analyzing opaque
as well as transparent structural relationships of dominance, discrimination, power and
control when these are manifested in language. In other words, CDA aims to investi-
gate critically social inequality as it is expressed, constituted, legitimized, and so on,
by language use (or in discourse). Most critical discourse analysts would thus endorse
Habermas claim that language is also a medium of domination and social force. It
serves to legitimize relations of organized power. Insofar as the legitimizations of power
relations, , are not articulated, ., language is also ideological (Habermas 1967: 259).
2. Historical note
In the 1960s and 1970s, many scholars adopted a more critical perspective in language
studies. Among the frst was the French scholar Pcheux (1982), whose approach
traced its roots to the work of Russian theorists Bakhtin and Voloinov, who had pos-
tulated an integration of language and social processes in the 1930s. In the late 1970s, a
group of Hallidayan linguists at the University of East Anglia began applying the term
critical linguistics (CL) in their research on language use in diferent institutions (see
e.g. Fowler et al. 1979; Kress & Hodge 1979). Kress and Hodge assumed strong and
pervasive connections between linguistic structure and social structure, claiming that
discourse cannot exist without social meanings. Te authors reacted strongly against
contemporary trends in pragmatics (e.g. speech act theory) and Labovian quantita-
tive sociolinguistics. Since 1979, this general approach has been refned, broadened,
changed and re-applied by other linguists coming from very diferent traditions, many
of whom believe that the relationship between language and the social, because of
its complex and multifaceted character, requires interdisciplinary research. Scholars
from backgrounds including sociolinguistics, formal linguistics, social psychology
and literary studies have contributed to the growth of the tradition, and have directed
CL research into subject domains such as racism, ethnicity, gender studies, political
oratory, etc.
3. Principles of CL
Some of the basic questions informing critical linguistic research include: How does
the naturalization of ideology come about? Which discursive strategies legitimate
control or naturalize the social order? How power is linguistically expressed? How
are consensus, acceptance and legitimacy of dominance manufactured? Who has
access to which instruments of power and control? Who is discriminated against in
54 Ruth Wodak
what way? Who understands a certain discourse in what way with what results? In
order to be able to answer these very complex and broadly formulated questions, we
might suggest the following general characteristics, aims and principles as being con-
stitutive of CDA, in general.
1. Te approach is interdisciplinary. Problems in our societies are too complex to be
studied from a single perspective. Tis entails diferent dimensions of interdisci-
plinarity: the theories draw on neighbouring disciplines and try to integrate these
theories. Teamwork consists of diferent researchers from diferent traditionally
defned disciplines working together. Lastly, the methodologies are adapted to the
data under investigation.
2. Te approach is problem-oriented, rather than focused on specifc linguistic items.
Complex social problems are the items of research, such as racism, identity,
social change, which, of course, are and could be studied from manifold perspec-
tives. Te CDA dimension, i.e. discourse and text analysis, is one of many possible
approaches.
3. Te theories as well as the methodologies are eclectic; i.e. theories and methods are
integrated which are adequate for an understanding and explanation of the object
under investigation.
4. Te study usually incorporates feldwork and ethnography to explore the object under
investigation (study from the inside) as a precondition for any further analysis and
theorizing. Tis approach makes it possible to avoid ftting the data to illustrate
a theory and cherry-picking. Rather, we deal with bottom-up and top-down
approaches at the same time.
5. Te approach is abductive: a constant movement back and forth between theory and
empirical data is necessary. Tis is a prerequisite for principle 4.
6. Multiple genres and multiple public spaces are studied; intertextual and interdiscur-
sive relationships are investigated. Recontextualization is one of the most important
processes in connecting genres as well as topics and argumentation patterns.
7. Te historical and broader socio-political contexts should be analyzed and inte-
grated into the interpretation of discourses and texts. Te notion of change (see
principle 6) has become constitutive for the study of text and discourse.
8. Te categories and tools for the analysis are defned in accordance with all these steps
and procedures and also with the specifc problem under investigation. Tis entails
some eclecticism, as well as pragmatism. Diferent approaches in CDA use difer-
ent grammatical theories.
9. Grand Teories might serve as a foundation; in the specifc analysis, Middle-Range
Teories serve the aims better. Te problem-oriented approach entails the use and
testing of middle-range theories. Grand Teories result in large gaps between
structure/context and linguistic realizations (although some gaps must necessar-
ily remain unbridgeable).
10. Practice and application are aimed at. Te results should be made available to
experts in diferent felds and, as in a second stage, be applied, with the goal of
possibly changing certain discursive and social practices.
4. Trends
4.1 Social Semiotics
As early as 1970, M.A.K. Halliday had stressed the relationship between the grammati-
cal system and the social and personal needs that language is required to serve. Halliday
distinguished three interconnected meta-functions of language: (1) the ideational func-
tion through which language lends structure to experience. Te ideational structure
has a dialectical relationship with social structure, both refecting and infuencing it.
(2) Te interpersonal function which accounts for relationships between the partici-
pants, and (3) the textual function which accounts for coherence and cohesion in texts.
Gunter Kress was heavily infuenced by the Hallidayan school of thought. His
work serves as an example of the original tradition which frst labelled itself CL. He
has developed his model and methodology ever since, shifing over the years to a
more social-semiotic mode of working. Kress is concerned with the central notion of
the sign as an indissoluble conjunct of meaning and form. In his own words, he wants
to connect the specifcities of semiotic forms, in any medium, with the specifcities
of social organizations and social histories (Kress 1993: 176f). Tis theory in turn has
the efect that language is seen as a semiotic system in which meaning is made directly,
rather than as a linguistic system and in which meaning is indirectly associated with
linguistic form. Kress work has displayed a growing interest in the description, analy-
sis and theorizing of other semiotic media, in particular visual media (e.g. Kress &
van Leeuwen 1990). Kress concentrated on the political economy of representational
media: that is, an attempt to understand how various societies value diferent modes
of representation, and how they use these diferent modes of representation. A central
aspect of this work is the attempt to understand the formation of the individual human
being as a social individual in response to available representational resources. His
most recent position as part of an institute on educational research has oriented much
of Kress eforts into thinking about the content of educational curricula in terms of
representational resources and their use by individuals in their constant transforma-
tion of their subjectivities, the process usually called learning. One by-product of
this research interest has been his increasing involvement in overtly political issues,
including the politics of culture.
56 Ruth Wodak
Te theory on multimodality put forward by Kress & van Leeuwen (1996) should
be mentioned here, as this provides a useful framework for considering the com-
municative potential of visual devices in the media. Van Leeuwen studied flm and
television production as well as Hallidayan linguistics. His principal publications are
concerned with topics such as the intonation of disc jockeys and newsreaders, the lan-
guage of television interviews and newspaper reporting, and more recently, the semi-
otics of visual communication and music. Van Leeuwen developed a most infuential
methodological tool: the Actors Analysis (1993, 2009). Tis taxonomy allows for the
analysis of (both written and oral) data, related to agency in a much diferentiated way.
Te taxonomy has been widely applied in data analysis.
Van Leeuwen also focused on other areas of visual communication, especially the
semiotics of handwriting and typography and the question of colour. He is increas-
ingly moving away from using a systemic-functional approach as the single model
and feels that it is important for social semiotics to realise that semiotic discourses and
methods are linked to semiotic practices, and that grammars are one type of semiotic
discourse which is linked to a specifc kind of control over specifc kinds of semiotic
practices. To give an example of a very diferent type of discourse, histories of art
and design focus on the semiotic innovations of specifc individuals in their histori-
cal contexts, rather than on a synchronous approach to semiotic systems. However,
they, too, are linked to the ways in which production and consumption is regulated
in that area. It is important for social semiotics to provide models of semiotic practice
that are appropriate to the practices they model, and as diferent semiotic practices
are very diferently organised, it is not possible to apply a single model to all. All of
this is closely related to the role and status of semiotic practices in society, and this is
currently undergoing change as a result of the fact that it is increasingly global corpo-
rations and semiotic technologies, rather than national institutions, which regulate
semiotic production and consumption.
Tis emphasis on regulatory practices has led to a research approach in three
stages, starting with the analysis of a particular category of texts, cultural artefacts or
communicative events, then moving to a second set of texts (and/or cultural artefacts
and/or communicative events), namely those that seek to regulate the production and
consumption of the frst set, and fnally moving to a third set of texts, namely actual
instances of producing or consuming texts (etc) belonging to the frst set. For instance,
in a study of baby toys, van Leeuwen and his team analysed the toys and their semiotic
potential, as objects-for-use and as cultural icons, then studied discourses seeking to
infuence how they are used, e.g. relevant sections of parenting books and magazines,
toy advertisements, texts on toy packaging etc., and fnally transcribed and analysed
videos of mothers and babies using these same toys together (Caldas-Coulthard and
van Leeuwen 2002). Tis type of work leads to a particular relation between discourse
analysis, ethnography, history and theory in which these disciplines are no longer
contributing to the whole through some kind of indefnable synergy or triangulation,
but are complementary in quite specifc ways.
Jay Lemke and Ron and Suzie Scollons research also have to be mentioned in
this context. In the last few years Lemkes work has emphasized multimedia semiot-
ics, multiple timescales, and hypertexts/traversals. He extended his earlier work on
embedded ideologies in social communication from analysis of verbal text to integra-
tion of verbal text with visual images and other presentational media, with a particular
focus on evaluative meanings. Tis work emphasizes the implicit value systems and
their connections to institutional and personal identity.
Te work on multiple timescales is an extension of earlier work on ecological-social
systems as complex dynamical systems with semiotic cultures. It is very important in
considering all aspects of social dynamics to look across multiple timescales, i.e. how
processes and practices which take place at relatively faster rates are organized within
the framework of more slowly changing features of social institutions and cultures. Tis
is a promising practical approach to the so-called micro/macro problem, both theo-
retically and methodologically (Lemke 2000, 2001). His more recent work has com-
bined both these themes to develop the idea that although we tell our lives as narratives,
we experience them as hypertexts. Building on research on the semantic resources of
hypertext as a medium, he proposed that post-modern lifestyles are increasingly liber-
ated from particular institutional roles and that we tend to move, on multiple times-
cales, from involvement in one institution to another, creating new kinds of meaning,
less bound to fxed genres and registers, as we surf across channels, websites, and
lived experiences. Tis is seen as a new historical development, not supplanting institu-
tions, but building up new socio-cultural possibilities on and over them.
In all this work, Lemke uses critical social semiotics as an extension of critical
discourse analysis, combined with models of the material base of emergent social phe-
nomena. His concern is with social and cultural change: how it happens, how it is
constrained, and the ways in which is it expectably unpredictable.
Te problem Ron and Suzie Scollon have addressed in their work (which Suzie
Scollon is continuing afer Rons death in December 2008) is to build a formal theo-
retical and a practical link between discourse and action. It is an activist position
that uses tools and strategies of engaged discourse analysis and thus requires a for-
mal analysis of how its own actions can be accomplished through discourse and its
analysis. Te problems in developing this framework are that action is always mul-
tiple both in the sense that there are always simultaneous parallel and interacting
actions at any moment we chose to analyze as well as in the sense that these multiple
actions operate across difering timescales so that it is not at all clear that we can see
higher level actions as simple composites of lower level actions. Te linkages are
more complex. Jay Lemkes work is, of course, also an important resource in studying
this problem.
58 Ruth Wodak
Ron Scollons work elaborates the idea developed in Mediated Discourse: Te Nexus
of Practice (2001). Tis suggests that practice in general is most usefully understood as
many separate practices which are linked in nexus of practice. Te relations between
discourse and a nexus of practice are many and complex and rarely direct. He tried
to open up and explicate these linkages through what could be called nexus analysis.
Tis work was integrated in two projects. In the frst which Ron and Suzie Scollon
have written about in Discourses in Place: Language in the Material World (2002) is a
kind of geosemiotics which is the integration of social interactionist theory (includ-
ing, of course, all forms of spoken discourse), visual semiotics, and place semiotics,
especially the built environment. Teir interest in this work has been to theorize the
link between indexicality in language (and discourse and semiotics more generally)
and the indexable in the world. Tis could also be understood as theorizing the link
between producers of communications and the material world in which those com-
munications are placed as a necessary element of their semiosis.
4.2 Orders of discourse and Foucauldian poststructuralism
Norman Fairclough (1985, 1989, 1992, 2003) sees the value of CDA as a method to be
used alongside others in research on social and cultural change, and as a resource in
struggles against exploitation and domination (1993: 133134). Fairclough is mainly
concerned with the study of power and institutional discourse, stressing the intertextu-
ality of diferent forms of social practice, like for example the relationships between the
social practices education systems and advertising. In his view, language use is always
simultaneously constitutive of social identities, social relations and systems of knowl-
edge and belief. Like Kress and van Leeuwen, Fairclough also relies on Hallidayan
linguistics for his analysis of discursive events.
Conventions underlying discursive events are termed orders of discourse (inter-
discourse) in this framework. Te order of discourse of some social domain is the
totality of its discursive practices, and the relationships (of complementarity, inclu-
sion/exclusion, opposition) between them. Te order of discourse of a society is thus
the set of these more local orders of discourse, and the relationships between them
(e.g. the relationship between the orders of discourse of the school and of home). Te
boundaries and insulations between and within orders of discourse may be points of
confict and contestation, open to being weakened or strengthened, as a part of wider
social conficts and struggles (1993: 135f). In his investigation of several examples of
university discourses on the background of an extensive analysis of the contemporary
post-traditional society, Fairclough claims that a new discourse, a consumer culture
discourse, infuences many other domains, thus also the universities. Traditional
genres, like the curricula vitae are suddenly created in the mode of advertisements
(the marketization of public discourse). Tese changes naturally have an impact on
the institutions, on hierarchy structures and on the identities of scholars.
Later, Chouliaraki & Fairclough (1999) explain and elaborate some advances in
CDA, showing not only how the analytical framework for researching language in
relation to power and ideology developed, but also how CDA is useful in disclosing
the discursive nature of much contemporary social and cultural change. Particularly
the language of the mass media is scrutinized as a site of power, of struggle and also as
a site where language is ofen apparently transparent. Media institutions ofen purport
to be neutral, in that they provide space for public discourse, refect states of afairs dis-
interestedly, and give the perceptions and arguments of the newsmakers. Fairclough
shows the fallacy of such assumptions, and illustrates the mediating and constructing
role of the media with a variety of examples.
Fairclough has also been concerned with the Language of New Labour (2000).
His work has centered around the theme of Language in New Capitalism focusing
on language/discourse aspects of the contemporary restructuring and re-scaling
(shif in relations between global, regional, national and local) of capitalism. He has
also worked with sociological theorists Bob Jessop and Andrew Sayer in theorizing
language (semiosis) within a critical realist philosophy of (social) science (Fairclough,
Jessop & Sayer 2002).
Phil Graham elaborates the research on the problems of New Capitalism
(Graham 2002, 2003a, b). Te historical investigation of hortatory genres compares
the emergence and struggles between Church, Divine Right Royalties, and secular
forces over legitimate uses of the sermon form in Western Europe between the tenth
and fourteenth centuries with contemporary struggles over genres that are used to
motivate people on a mass scale. Te main focus of the study is to explore and explain
the relationships between new media, new genres, institutions, and social change at
a macro level. Te perspective is primarily historical, political-economic, relational,
and dynamic. Genres are produced, textured, and transformed within institutional
contexts over long periods of time. In turn, institutions invest years in some cases,
millennia developing, maintaining, and adapting generic forms to changing social
conditions in order to maintain or to gain power. Graham believes that at certain
times in history, specifc genres become very efective for motivating or manipulating
large sections of society. Because genres are developed within institutions, and thus
within the realms of vested interests, they display inherent axiological biases (since
about 15 years, quite a few critical discourse analysts have become more and more
concerned with the link between history, historical sources and discourse theory; see
also Heer et al. 2008; Martin & Wodak 2003; Blommaert 2005; Ensink & Sauer 2003;
Tiesmeyer 2004; Flowerdew 2002; Reisigl & Wodak 2001, 2009; Wodak et al. 1990,
1994, 1999, 2001 and below).
60 Ruth Wodak
4.3 The socio-cognitive model
Te main representative of this approach is Teun van Dijk, most of whose critical work
focuses on the (re)production of ethnic prejudices and racism in discourse and com-
munication. Tus, earlier studies examined the ways white Dutch and Californians
talk about minorities (van Dijk 1984, 1987). Besides identifying various structures of
such talk, the study also aimed at reconstructing ethnic attitudes and ideologies from
everyday conversation. An analysis of frequent topics, for instance, suggests what
speakers have on their minds, or as he explains it, what the hierarchies are of their
personal mental models of ethnic events as well as the structures of ethnic attitudes.
Te overall strategy of talk about Others, according to van Dijk, combines strategies
of positive self-presentation with negative other-presentation.
In another study, van Dijk examined the role of the news media in the reproduc-
tion of racism (van Dijk 1991). Combining both quantitative and qualitative analyses
of thousands of news reports in the British and Dutch press, van Dijk concluded
that the most frequent topics on racism in the press corresponded to prevailing eth-
nic prejudices expressed in everyday talk: immigration as invasion, immigrants and
refugees as spongers, crime, violence and problematic cultural diferences. Tese per-
vasive topics were also refected in the style, rhetoric, and the local semantic moves
of news reports, Op Ed articles and editorials, especially in the conservative and
tabloid press.
In another book on discourse and racism, van Dijk investigated a hypothesis
which increasingly suggested itself in the previous studies, namely, that the elites play
a crucial role in the reproduction of racism (van Dijk 1993a). In an analysis of parlia-
mentary debates, corporate discourse, textbooks and media, van Dijk claims that in
many ways the elites preformulate and thus instigate popular racism. Among the many
strategies of what he describes as elite racism are the consistent denial of racist beliefs
and the attribution of racism to others: people in other countries, people from earlier
times, or poor whites in the older cities.
van Dijk has also turned to more general questions of abuse of power and the
reproduction of inequality through ideologies. In his view, which integrates elements
from his earlier studies on cognition, those who control most dimensions of discourse
(preparation, setting, participants, topics, style, rhetoric, interaction, etc.) have the
most power. He argues that no direct relation can or should be constructed between
discourse structures and social structures, but that they are always mediated by the
interface of personal and social cognition. Cognition, according to van Dijk, is the
missing link of many studies in CDA, which fail to show how societal structures infu-
ence discourse structures and precisely how societal structures are in turn enacted,
instituted, legitimated, confrmed or challenged by text and talk.
More recently, Teun van Dijk has taken up a more detailed study of the role of
knowledge in discourse. Another topic in his research is a new approach to the study
of context. One of the main arguments of this research is that there is much inter-
est in context and contextualization, but hardly any in theory of context. Van Dijk
proposes to defne context in terms of context models in episodic memory, that is, in
terms of subjective, dynamic representations of the ongoing communicative event and
situation. It is these context models that, in van Dijks view, control all discourse and
communication, and especially all dimensions of discourse that adapt it to the current
situation as it is understood by the participants such as style and rhetoric. (Van
Dijk 2001, 2008, 2009).
4.4 Discourse-Historical approach
Ruth Wodak (now Lancaster) and her group in Vienna base their model on sociolin-
guistics in the Bernsteinian tradition, and on ideas of the Frankfurt school, especially
those of Jrgen Habermas. Wodak conducted studies on institutional communication
and speech barriers in court, in schools and in hospital clinics (Wodak 1996), and
later, she has focused on sexism, and contemporary antisemitism and racism in set-
tings of various degrees of formality as well as on national and trans-national identity
politics (Krzyanowski & Wodak 2008; Wodak 2009). One of the groups major aims is
the practical application of critical research, e.g. in guidelines for non- discriminatory
language use towards women, in guidelines for doctors on how to communicate more
efectively with their patients, and in providing expert opinions for courts on anti-
Semitic and racist language use by journalists in newspapers.
In an interdisciplinary study of post-war anti-Semitism in Austria completed in
1990, Wodak and her colleagues devised what they have termed the discourse histori-
cal approach. Te distinctive feature of this approach is its attempt to integrate sys-
tematically all available background information in the analysis and interpretation of
the many layers of a written or spoken text. Te study in which and for which this
approach was developed attempted to trace in detail the constitution of an anti-Semitic
stereotyped image as it emerged in public discourse in the 1986 Austrian presidential
campaign of Kurt Waldheim. Te study addressed the problem of anti-Semitic lan-
guage behavior in contemporary Austria, in other words, linguistic manifestations
of prejudices towards Jews. Wodak et al. (1990) were able to show that the context of
the discourse had a signifcant impact on the structure, function, and content of the
anti-Semitic utterances (see also Reisigl & Wodak 2001; Wodak & Reisigl 2002; Wodak
2004b; Pelinka & Wodak 2002; Heer et al. 2008).
Several other studies on prejudice and racism have led the group in Vienna to more
general and theoretical considerations on the form and content of racist discourse
62 Ruth Wodak
about foreigners, indigenous minorities, immigrant workers etc. Although the forms
of racist and prejudiced discourse may be similar, the contents vary according to the
stigmatized groups as well as to the settings in which certain linguistic realizations
become possible. In comparing anti-Semitic with racist discourse, for example, Wodak
& Matouschek (1993) suggest that the norms and taboos controlling the utterances
about foreigners and Jews difer, depending on the targeted discriminated group and
the specifc historical traditions and socio-political contexts of the speakers and dis-
courses. In the anonymous contexts of conversations tape-recorded on the street,
sexist, racist and anti-Semitic prejudice stories were interwoven and expressed in
the same discursive event, whereas in ofcial discourse (print and electronic media,
speeches of politicians) explicit anti-Semitic utterances are taboo, yet explicit racist
remarks towards foreigners are not. Te discourse-historical approach is designed to
enable the analysis of indirect prejudiced utterances, as well as to identify and expose
the codes and allusions contained in prejudiced discourse. Wodak & Van Dijk (2000)
were able to study and compare parliamentary discourses in six European countries
on issues of immigration (see also Blommaert & Verschueren 1998; Krzyanowski &
Wodak 2009). Tis study illustrates overriding discourses in the European Union on
the one hand, and context-dependent, historically rooted arguments on the other,
when immigration and Human Rights issues are discussed.
More recently, Wodak has been concerned with investigating identity politics
and patterns of decision-making in EU organizations (Wodak 2009). Fieldwork
in European Union organizations and at the European Convention have allowed
insight in doing politics behind closed doors (Muntigl et al. 2000; Wodak &
Weiss 2005; Krzyanowski 2005; Oberhuber 2005; Oberhuber et al. 2005; Wodak
et al. 1999). In this research, together with sociologists and political scientists,
models were proposed to explain the context-dependent tensions and contradic-
tions which necessarily arise in such a historically complex entity as Europe. EU
ofcials were interviewed, policy documents analyzed, political speeches on a
vision of Europe collected, and media reporting on EU events confronted with
the in-side view of EU organizations. Te analysis of various genres illustrates the
recontextualization of salient arguments, metaphors, and slogans in transnational
and national contexts.
4.5 Lexicometry
Te combination of political science and political philosophy (predominantly under
a strong Marxist infuence) on the one hand and French linguistics on the other hand
is typical of French critical discourse analysis. Basically, two diferent approaches may
be distinguished.
Te frst is political lexicometry, a computer-aided statistical approach to political
lexicon, developed at the cole Normale Suprieure at Saint-Cloud. A text corpus (e.g.
texts of the French Communist Party) is prepared. Texts are then compared on the basis
of relative frequency (cf: Bonnafous & Tournier 1995). One study shows, for example,
how the relative frequency of the words travailleur and salari varies signifcantly
between French trade unions, refecting diferent political ideologies, and how the fre-
quency changes over time. (Groupe de Saint-Cloud 1982; Bonnafous & Tournier 1995).
Althussers theory on ideology and Foucaults theory were major points of refer-
ence for the second tendency in French discourse analysis, notably the work of Michel
Pcheux (1982) (see above). Discourse is the place where language and ideology meet,
and discourse analysis is the analysis of ideological dimensions of language use, and
of the materialization in language of ideology. Both the words and the meanings
of words vary according to the class struggle position from which they are used
according to the discursive formation they are located within. For instance, the word
struggle itself is particularly associated with a working class political voice, and its
meaning in that discursive formation is diferent from its meanings when used from
other positions. Pcheuxs main focus was political discourse in France, especially the
relationship between social-democratic and communist discourse within lef political
discourse. Pcheux stresses the ideological efects of discursive formations in position-
ing people as social subjects. Echoing Althusser, he suggests that people are placed in
the imaginary position of sources of their discourse, whereas actually their discourse
and indeed they themselves are efects of their ideological positioning. Te sources
and processes of their own positioning are hidden from people. Tey are typically not
aware of speaking/writing from within a particular discursive formation. Moreover,
the discursive formations within which people are positioned are themselves shaped
by the complex whole in dominance of discursive formations, which Pcheux calls
interdiscourse but people are not aware of that shaping. Radical change in the way
people are positioned in discourse can only come from political revolution.
Pcheux and his colleagues changed their views on this and other issues in the late
1970s and early 1980s (Pcheux 1988; Maingueneau 1987). Te infuence of Foucault
increased, as did that of Bakhtin. Studies began to emphasize the complex mixing of
discursive formations in texts, and the heterogeneity and ambivalence of texts (see, for
example, Courtine 1981). Some other French researchers investigate detailed rhetorical
patterns, for example in the presidential campaigns of 1988 and 1995 (Groupe de Saint
Cloud 1995). Also the infuence of Anglo-Saxon pragmatics is prominent, and that of
the French linguist Benveniste (1974), whose work on nonciation focused on deic-
tic phenomena. In this framework, Pierre Achard produced detailed accounts of the
political functioning of a very wide range of text types (Achard 1995). (See Wodak &
de Cillia 2006 for more details).
64 Ruth Wodak
4.6 Lesarten approach
National Socialist language frst became the object of critical philological observations
by Viktor Klemperer (Klemperer 1975
3
). Utz Maas, however, was the frst linguist and
discourse-analyst to subject the every-day linguistic practice of National Socialism to
an in-depth analysis: he used NS texts to exemplify his approach of Lesweisenanalyse
(Maas 1984, 1989a, 1989b). His historical argumentation analysis, based on the the-
ories of Michel Foucault, demonstrates how NS discourses were determined by the
German society and NS ideology, i.e. in what may be termed a social practice. In his
analysis of language practices during the National Socialist regime between 1932 and
1938 he illustrates how the discursive practices of society in Germany were impacted
by the NS discourse characterized by social-revolutionist undertones. Nazi discourse
had superseded almost all forms of language (practices), a fact that made it difcult for
an individual who did not want to cherish the tradition of an unworldly Romanticism
to use language in a critical-refective way.
Discourse is basically understood as the result of collusion: the conditions of the
political, social and linguistic practice impose themselves practically behind the back
of the subjects while the actors do not guess the rules of the game (cf. also Bourdieus
violence symbolique). Discourse analysis identifes the rules, which make a text, for
example, a fascist text. In the same way as grammar characterizes the structure of
sentences, discourse rules characterize utterances/texts that are acceptable within a
certain practice. Te focus is not on National Socialist language per se, but the aim
is to record and analyze the spectrum of linguistic relations based on a number of
texts dealing with various spheres of life. Tese texts represent a complicated network
of similarities, which overlap and intersect. Terefore it is also important to do jus-
tice to the polyphony of texts resulting from the fact that societal contradictions are
inscribed into texts. Texts from diverse social and political contexts (cooking recipes,
local municipal provisions on agriculture, texts by NS politicians, but also by critics of
this ideology, who are ultimately involved in the dominant discourse) are analyzed in
a sample representative of NS discourse.
Te method of reading analysis proposed by Maas may be described as a con-
centric hermeneutic approach to the corpus in fve systematic steps.
(A) Statement of the self-declared content of the text, (B) description of the
staging (Inszenierung) of the content, (C) analysis of the sense and meaning of the
staging, (D) provisional summery of the analysis, and (E) discussion of competing
readings. (Maas 1984: 18) In this context it should be stressed that competing readings
of texts may result from disclosing the diference between self-declared and latent con-
tent. Applications of this method (Titscher et al. 1998: 232) can be found in Januscheks
analysis of Jrg Haiders allusions to NS discourse (Januschek 1992) and in Sauers
analysis of texts of the Nazi occupation of the Netherlands (Sauer 1989, 1995).
Te Duisburg School of CDA (Jger 1993, 1999, 2001) draws on Foucaults
notion of discourse. According to (Jger 1999: 116) discourse is materiality sui gene-
ris and discourse theory is a combination of the materialistic cultural theory, on
the one hand, and Alexej N. Leontjevs speech activity theory (Leontjev 1984) and
Jrgen Links collective symbolism (Link 1988), on the other hand. As institution-
alized and conventionalized speech modes, discourses express societal power rela-
tions, which in turn are infuenced by discourses. Tis overall discourse of society,
which could be visualised as a diskursives Gewimmel (literally: discursive swarm-
ing), becomes comprehensible in diferent discourse strands (composed of discourse
fragments on the same subject) at diferent discourse levels (science, politics, media,
and so on). Every discourse is historically embedded, and has repercussions on cur-
rent and future discourse. In addition to the above levels, the structure of discourse
may be dissected into: discursive events and discursive context, discourse position,
overall societal discourse and interwoven discourses; themes, bundles of discourse
strands, history, present and future of discourse strands. Discourse Analysis makes
a contribution to (media) impact research, as it analyzes the impact of discourse on
individual and collective consciousness. Individual discourse fragments are selected
from the archived material for concrete analysis. Tese fragments are analyzed in fve
steps (institutional framework, text surface, linguistic-rhetorical means, program-
matic-ideological messages, and interpretation), for which a range of concrete ques-
tions regarding the text is formulated (Jger 1999: 175187). Te uniformity of the
hegemonic discourse implies that analysis requires only a relatively small number of
discourse fragments.
5. Conclusion
Te feld of CDA is developing fast, and the critical perspective is penetrating into
many felds of investigation of language usage. Te critical dimension of linguistic
pragmatics was forcefully advocated by Jacob Mey (1985), and the lack of critical
awareness in traditional sociolinguistics exposed by Williams (1992). In the mean-
time critical approaches to specifc topics in language studies have emerged (see e.g.
Meeuwis ed. 1994 on intercultural communication) and more and more researchers
are arguing that the study of language should be based on a sound socio-political
intellectual basis allowing for better analyses of power in language and language
usage (see Chilton et al. 2010; Heller 1988; Woolard 1985; Rickford 1986; Meeuwis &
Blommaert 1994). CDA is also developing into interdisciplinary research domains
par excellence, and thus ofers interesting perspectives for integrated research on lan-
guage in society.
66 Ruth Wodak
References
Achard, P. (1995). Formation discursive, dialogisme et sociologie. Langages 117: 8295.
Anthonissen, C. (2001). On the Efectivity of Media Censorship: An Analysis of Linguistic, Paralinguis-
tic and other Communicative Devices used to defy Media Restrictions. Ph.D. thesis, University of
Vienna.
Bakhtin, M. (1981). Te Dialogic Imagination. University of Texas Press.
Beneviste, E. (1974). [1966]. Problmes de linguistique gnrale (2 vols.). Gallimard.
Benke, G. (2000). Diskursanalyse als sozialwissenschafliche Untersuchungsmethode. SWS
Rundschau(2): 140162.
Billig, M. (2003). Critical Discourse Analysis and the rhetoric of critique. In G. Weiss & R. Wodak
(eds.) Critical Discourse Analysis: theory and interdisciplinarity: 3546. Palgrave Macmillan.
Blommaert, J. & J. Verschueren (1998). Te Diversity Debate. Routledge.
Blommaert, J. (2005). Discourse. Cambridge University Press.
Bonnafous, S. & M. Tournier (1995). Analyse du discours, lexicomtrie, communication et politique.
Mots 117: 6781.
Bourdieu, P. (ed.) (1989). La noblesse detat. Editions de Minuit.
Brnner, G. & G. Graefen (1994). Texte und Diskurse. Westdeutscher Verlag.
Caldas-Coulthard, C. & T. van Leeuwen (2002). Stunning, Shimmering, Iridescent: Toys as the
Representation of Gendered Social Actors. In L. Litosseleti & J. Sunderland (eds.) Gender
Identity and Discourse Analysis: 91108. John Benjamins.
Chilton, P., Tian, H. & Wodak, R. (2010). Refections on Discourse and Critique in China and the
West, Journal of Language and Politics 9/4, 489510.
Chouliaraki, L. & N. Fairclough (1999). Discourse in Late Modernity: Rethinking Critical Discourse
Analysis. Edinburgh University Press.
Connerton, P. (1996). [1976]. How Societies Remember. Cambridge University Press.
Courtine, J. -J. (1981). Analyse du discours politique. Langages 62.
van Dijk, T. (1984). Prejudice in discourse. John Benjamins.
(1987). Communicating racism. Sage.
(1991). Racism and the press. Routledge.
(1993a). Elite discourse and racism. Sage.
(1993b). Principles of critical discourse analysis. Discourse and Society 4(2): 249283.
(2001). Critical Discourse Analysis. In D. Tannen, D. Schifrin & H. Hamilton (eds.): Handbook
of Discourse Analysis: 352371. Blackwell.
(2008). Society in Discourse. How Context Controls Text and Talk. CUP.
(2009). Critical Discourse Studies: a sociocognitive approach, in: Wodak, R. & Meyer, M.
(eds.) Methods of CDA: 6286. Sage.
Ensink, T. & Ch. Sauer (eds.) (2003). Te Art of Commemoration. Benjamins.
Fairclough, N. (1985). Critical and descriptive goals in discourse analysis. Journal of Pragmatics 9:
73963.
(1989). Language and power. Longman.
(1992). Discourse and Social Change. Polity Press.
(1993). Critical discourse analysis and the marketization of public discourse. Discourse and
Society 4(2): 133168.
(1995). Critical Discourse Analysis: Te Critical Study of Language. London.
(2000). New Labour, New Language? Routledge.
(2003). Analysing Discourse: Text Analysis for Social Research, Routledge.
Fairclough, N. & R. Wodak (1997). Critical Discourse Analysis. In T.A. van Dijk (ed.) Introduction to
Discourse Studies: 258284. Sage.
Fairclough, N., B. Jessop & A. Sayer (2002). Critical realism and semiosis. Journal of Critical
Realism 5(1).
Fay, B. (1987). Critical Social Science. Polity Press.
Flowerdew, J. (2002). Rhetorical Strategies and Identity Politics in the Discourse of Colonial With-
drawal. Journal of Language and Politics 1(1): 149180.
Fowler, R., B. Hodge, G. Kress & T. Trew (1979). Language and control. Routledge & Kegan Paul.
Gee, J. (2004). Discourse Analysis. Teory and Method. Routledge.
Graham, P. (2002). Space and cyberspace: On the enclosure of consciousness. In J. Armitage &
J. Roberts (eds.) Living With Cyberspace: Technology & Society in the 21st Century: 156164.
Te Athlone Press.
(2003a). Time, space, and political economy: A history of hype and hypercapitalism.
Discurso y Sociedad.
(2003b). Te Digital Dark Ages: New media, new literacies, and socio-political change. Peter
Lang.
Groupe De Saint-Cloud (1982). La Parole syndicale: tude du vocabulaire confdral des centrales
ouvrires franaises (19711976). PUF.
(1995). Prsidentielle: regards sur les discours tlviss. INA-Nathan.
Habermas, J. (1967). Erkenntnis und Interesse. Suhrkamp.
Halliday, M.A.K. (1970). Language as social semiotic. Edward Arnold.
(1985). Introduction to Functional Grammar. Edward Arnold.
Heer et al. (eds.) (2008). Te discursive construction of history. Remembering the German Wehrmachts
war of extermination. Palgrave.
Heller, M. (1988). Where do we go from here? In M. Heller (ed.) Codeswitching: 265272. Mouton
de Gruyter.
Iedema, R. (2003). Post-bureaucratic Organizations. John Benjamins.
Wodak, R. (1999). Introduction: Organizational discourse and practices. Discourse & Society
10: 1520.
Jger, S. (1993). Kritische Diskursanalyse. Eine Einfhrung. DISS.
(1999). Kritische Diskursanalyse. Eine Einfhrung. DISS.
(2001). Discourse and knowledge: Teoretical and methodological aspects of critical dis-
course analysis. In R. Wodak & M. Meyer (eds.): 3262.
Januschek, F. (1992). Rechts-populismus und NS-Anspielungen am Beispiel des sterreichischen
Politikers Jrg Haider. DISS-Texte 15.
Klemperer, V. (1975
3
) [1947]. LTI. Lingua Tertii Imperii. Die Sprache des Dritten Reiches. Reclam.
Kress, G. & T. van Leeuwen (1990). Reading images. Deakin University Press.
(1996). Reading Images. Routledge.
Kress, G. & R. Hodge (1979). Language as ideology. Routledge.
Kress, G. (1993). Against arbitrariness. Discourse and Society 4(2): 169191.
Krings, H. et al. (eds.) (1973). Handbuch philosophischer Grundbegrife. Ksel-Verlag.
Krzyanowski, M. (2005). European Identity Wanted! On Discursive Dimensions of the European
Convention. In R. Wodak & P. Chilton (eds.) New Research Agenda in CDA: Teory and Multi-
disciplinarity: 137164. John Benjamins.
Krzyanowski, M. & Wodak, R. (2008). Te Politics of Exclusion. Te Austrian Case. Transaction Press.
68 Ruth Wodak
Krzyanowski, M. & Wodak, R. (2009). Teorising and Analysing Social Change in Central and East-
ern Europe: Te Contribution of Critical Discourse Analysis, In: Galasinska, A. & Krzyanowski,
M. (eds.). Discourse and Transformation in Central and Eastern Europe: 1740. Palgrave.
van Leeuwen, T. (1993). Genre and feld in critical discourse analysis. Discourse and Society 4(2):
193223.
(2009). Discourse as the recontextualisation of social practice: a guide, in Wodak, R. & Meyer,
M. (eds.) Methods of CDA: 144161. Sage.
Lemke, J. (1995). Textual Politics: Discourse and Social Dynamics. Taylor & Francis.
(2000). Multiplying Modalities: Presentational, Orientational and Organizational Meaning.
Unpubl. Paper.
(2001). Discursive Technologies and the Social Organization of Meaning. Folia Linguistica
35(1/2): 7997.
(2003). Texts and Discourses in the Technologies of Social Organization. In G. Weiss &
R. Wodak (eds.) Critical Discourse Analysis. Teory and Interdisciplinarity: 130149. Palgrave
Macmillan.
Leontjev, A.N. (1984). Der allgemeine Ttigkeitsbegrif. In D. Viehweger (ed.) Grundfragen einer
Teorie der sprachlichen Ttigkeit: 1330. Akademie-Verlag.
Link, J. (1988). ber Kollektivsymbolik im politischen Diskurs und ihren Anteil an totalitren
Tendenzen. kultuRRevolution 17/18: 4753.
Maas, U. (1984). Als der Geist der Gemeinschaf eine Sprache fand. Sprache im Nationalsozialismus.
Westdeutscher Verlag.
(1989a). Sprachpolitik und politische Sprachwissenschaf. Suhrkamp.
(1989b). Sprache im Nationalsozialismus. Analyse einer Rede eines Studentenfunktionrs. In
K. Ehlich (ed.) Sprache im Faschismus: 162197. Suhrkamp Verlag.
Maingueneau, D. (1987). Nouvelles tendances en analyse du discours. Hachette.
Martin, J.R. (2002). Voicing the Other: Reading and Writing Indigenous Australians. In G. Weiss &
R. Wodak (eds.) Critical Discourse Analysis. Teory and Interdisciplinarity: 199219. Palgrave
Macmillan.
Martin, J.R. & R. Wodak (eds.) (2003). Re/reading the past. Critical and functional perspectives on
time and value. John Benjamins.
Meeuwis, M. (ed.) (1994). Critical perspectives on intercultural communication. Pragmatics 4(3)
(Special Issue).
Meeuwis, M. & J. Blommaert (1994). Te markedness model and the absence of society. Multilingua
13(4): 387422.
Mey, J. (1994). [1974]. Pragmatics. Blackwell.
(1985). Whose language? John Benjamins.
Muntigl, P., G. Weiss & R. Wodak (2000). European Union Discourses on Unemployment. An inter-
disciplinary approach to employment policy-making and organizational change. John Benjamins.
Oberhuber, F. (2005). Deliberation or Mainstreaming: Empirically Researching the European Con-
vention. In R. Wodak & P. Chilton (eds.) New Research Agenda in CDA: Teory and Multidisci-
plinarity: 165188. John Benjamins.
Oberhuber, F., C. Brenreuter, M. Krzyanowski, H. Schnbauer & R. Wodak (2005). Debating the
European Constitution: On representations of Europe/the EU in the press. Journal of Language
and Politics 4(2): 227 271.
Pcheux, M. (1982). Language, semantics and ideology. MacMillan.
(1988). Discourse structure or event. In C. Nelson & L. Grossberg (eds.) Marxism and the
Interpretation of Culture: 633650. MacMillan.
Pelinka, A. & R. Wodak (2002). Dreck am Stecken. Politik der Ausgrenzung. Czernin Verlag.
Reisigl, M. & R. Wodak (2001). Discourse and Discrimination. Routledge.
(2009). Te Discourse-Historical Approach. In: Wodak, R. & Meyer, M. (eds.) Methods of
CDA: 87121. Sage.
Renkema, J. (2004). Introduction to Discourse Studies. John Benjamins.
Rickford, J. (1986). Te need for new approaches to social class analysis in sociolinguistics. Language
and Communication 6: 215221.
Richardson, J. (2004). (Mis)representing Islam. John Benjamins.
Richardson, J. & Wodak, R. (2009). Visual Argumentation in Racist Discourse, Controversia 2,
4577.
Sauer, C. (1989). Nazi-Deutsch fr Niederlnder. Das Konzept der NS-Sprachpolitik in der
Deutschen Zeitung in den Niederlanden 19401945. In K. Ehlich (ed.) Sprache im Faschismus:
237288. Suhrkamp.
(1995). Sprachwissenschaf und NS-Faschismus. Lehren aus der sprachwissenschaflichen
Erforschung des Sprachgebrauchs deutscher Nationalsozialisten und Propagandisten fr den
mittel- und osteuropischen Umbruch? In K. Steinke (ed.) Die Sprache der Diktaturen und Dik-
tatoren: 996. Carl Winter.
Schifrin, D. (1994). Approaches to Discourse. Blackwell.
Scollon, R. (2001). Mediated Discourse. Te Nexus of Practice. Routledge.
Scollon, R. & S. Scollon (2002). Nexus analyis: Discourse and the Emerging Internet. Routledge.
Tompson, J.B. (1988). Critical Hermeneutics (4th edition). Cambridge University Press.
Tiesmeyer, L. (ed.) (2004). Discourse and Silencing. John Benjamins.
Titscher, S., R. Wodak, M. Meyer & E. Vetter (1998). Methoden der Textanalyse. Westdeutscher Ver-
lag (English translation 2000, Sage).
Voloinov, V.I. (1973). [1928]. Marxism and the Philosophy of Language. Seminar Press.
Williams, G. (1992). Sociolinguistics. Routledge.
Wodak, R. (1996). Disorders in Discourse. Longman.
(2000). Recontextualisation and the transformation of meaning: a critical discourse analysis
of decision making in EU-meetings about employment policies. In S. Sarangi & M. Coulthard
(eds.) Discourse and Social Life: 185206. Pearson Education.
(2004a). Discourses of Silence: Anti-Semitic Discourse in Post-war Austria. In L. Tiesmeyer
(ed.) Discourse and Silencing: 124 156. John Benjamins.
(2004b). Critical discourse analysis. In C. Seale, G. Gobo, J.F. Gubrium & D. Silverman (eds.)
Qualitative Research Practice: 197213. Sage.
(2007). Pragmatics and Critical Discourse Analysis, Pragmatics and Cognition 15(1): 203225.
(2008). Introduction: Terms and Concepts, in R. Wodak & M. Krzyanowski (Eds.),Qualitative
Discourse Analysis in the Social Sciences. Basingstoke: Palgrave, pp. 142.
(2009). Te Discourse of Politics in Action: Politics as Usual. Palgrave.
Wodak, R., J. Pelikan, P. Nowak, H. Gruber, R. de Cillia & R. Mitten (1990). Wir sind alle unschuldige
Tter. Suhrkamp.
Wodak, R. & B. Matouschek (1993). We are dealing with people whose origins one can clearly tell
just by looking. Discourse and Society 4(2): 225248.
Wodak, R., F. Menz, R. Mitten & F. Stern (1994). Die Sprachen der Vergangenheiten: fentliches
Gedenken in sterreichischen und deutschen Medien. Suhrkamp.
Wodak, R., R. de Cillia, M. Reisigl & K. Liebhart (1999). Te Discursive Construction of National
Identity. Edinburgh University Press (2nd edition 2009).
Wodak, R. & T. van Dijk (eds.) (2000). Racism at the Top. Drava.
70 Ruth Wodak
Wodak, R. & M. Meyer (eds.) (2001). Methods of Critical Discourse Analysis. Sage (2nd edition 2009).
Wodak, R. & M. Reisigl (2002). wenn einer Ariel heist Ein linguistisches Gutachten zur
politischen Funktionalisierung antisemitischer Ressentiments in sterreich. In A. Pelinka &
R. Wodak (eds.) Dreck am Stecken: 134172. Czernin Verlag.
Wodak, R. & R. de Cillia (2006). Language and/in Politics. In K. Brown (ed.) Encyclopaedia for Lan-
guage and Linguistics: 707719. Elsevier.
Wodak, R. & G. Weiss (2005). Analyzing European Union Discourses. Teories and Applications. In
R. Wodak & P. Chilton (eds.) A New Agenda in (Critical) Discourse Analysis: 121135. Benjamins.
Wodak, R. & Koller, V. (eds.) (2008). Handbook of Communication in the Public Sphere. De Gruyter.
Wodak, R. & Meyer, M. (2009). Critical discourse analysis: history, agenda, theory, and methodo-
logy, in: Wodak, Ruth & Meyer, Michael (Eds.) Methods of CDA: 133. Sage.
Woolard, K. (1985). Language variation and cultural hegemony. American Ethnologists 12: 738748.
nonciation
French pragmatic approach(es)
Marjut Johansson & Eija Suomela-Salmi
University of Turku
1. Introduction
French enunciation
1
theory is not a unifed theory, but represents several diferent
diversifed approaches to pragmatic questions. It developed out of French structural-
ism, from the heritage of Saussure and grammatical traditions, whereas pragmatics
in the Anglo-American tradition grew out of analytical philosophy and logic (Fuchs
1981). It bears resemblances to other pragmatically oriented theories (see, for instance,
Banks 2004), but presents an original view on language use. First of all, nonciation
is not a modern discovery, but rather a rediscovery of old problematics of language
concentrating mainly on issues that concern language-internal systems, such as deic-
tics and modality (Fuchs 1981: 35). Basically, enunciation deals with utterance-level
meaning from the perspective of diferent linguistic elements. In other words, the
activity of the speaker is the focus: on the one hand there are traces or indices lef by
the speaker in the utterance; on the other hand there is the relationship the speaker
maintains with her/his interlocutor (Dubois 1969). Enunciation is understood as an
act of producing an utterance as an individual instance of language use in a specifc
situation (Benveniste 1974: 80).
In this article, the terms enunciation theory and enunciation linguistics are
occasionally used. Tey refect the French tradition and are literal equivalencies of
French terms. But since our perspective is a pragmatic one, we consider it appropriate
to also use the terms enunciative pragmatics and French pragmatics.
Te term nonciation itself derives from Latin enuntio, which means disclosing,
verbal expression or more concretely a phrase. In logic it refers to assertion and in
:. Most of the translations of the French linguistic terminology used in this article are based on
the recent literature or articles on the subject. However, in some cases this has not been possible
and the authors have decided on a dierent type of translation or created a new equivalency in
their use of translated terms. All the examples have been translated by the authors.
72 Marjut Johansson & Eija Suomela-Salmi
grammar to pronunciation.
2
Tere are three basic concepts: enunciation, the act of
producing an utterance, or for short, the act of uttering; nonc, the product of an act
of uttering, i.e. an utterance; and enunciator, speaker, locutor or utterer depending on
the theoretical approach to enunciation. In other words: an act, its product and its pro-
ducer. In this theory, the basic view of language is that it is communicative: its object
of study is language in use from the point of view of the speaker how the speaker,
or rather, the speaking subject expresses her/his subjectivity in spatio-temporal situa-
tions and how the speaking activity leaves traces in the utterance. Enunciation theory
is also a theory about subjectivity, the speaking subject being in the centre of interro-
gation, not as a unifed subject, but a heterogeneous one. Tis speaking subject is not
autonomous, but depends on the situation of communication. In enunciation theory,
diferent trends in the study of language use and its explanation depend on whether
description is based on actual language use as documented in corpora, or whether
theoretically oriented linguistics is the starting point. Moreover, depending on the the-
oretical premises, approaches or schools in French pragmatics can vary considerable:
from cognitively oriented to sociolinguistic ones.
In this article, we frst trace the context in which French enunciation theory has
developed, and we survey its traditions and the formulation of its main concepts
from a historical point of view up to the present day. Tis is done in order to examine
the basic theoretical claims and assumptions the theory makes more generally, and
to see how it is situated vis--vis neighbouring disciplines. Secondly, we present its
main concepts from the point of view of their basic features and as they are used in
research today.
2. Historical overview from the pre-theoretical to the present phase
2.1 Origins and the pre-theoretical phase
In the development of the theory of enunciation, chronologically, a pre-theoretical
phase and three diferent rather fragmented phases can be distinguished. During the
pre-theoretical phase the term enunciation itself is not introduced, but the underlying
notion shows up in ways of pragmatic thinking that led to the theoretical concept of
enunciation (Nerlich & Clark 1996). Te origins of enunciation theory can be found
in various directions and its roots go back to rhetoric, grammar and logic (Fuchs
1981; Groupe Relpred 1989). In the development of its pre-theoretical phase, French
i. We thank Jyri Vaahtera, Professor of Ancient Languages and Culture at the University of
Turku for this information.
nonciation 73
pragmatic thinking was infuenced by Cartesian philosophy and the Port-Royal gram-
mar as well as by Lockes philosophy (Nerlich & Clark 1996: 62).
In the 19th century, several linguists in Europe became occupied with problems
that were essential to the development of French pragmatics, i.e. problems linked to
the presence of the speaker, as in Bhlers work. Also awareness of the interrelatedness
of linguistic forms and their functions grew stronger in the protopragmatic phase in
Europe (Nerlich & Clark 1996). In France, it was Ernest Renan who noted the relation
between the meaning of words and their sounds (Groupe Relpred 1989), but this prag-
matic thinking emerges most clearly in the work of Michel Bral who tried to establish
a theory of signs:
Te fact is that most linguists have directed their attention to the form of words: the
laws which govern changes in meaning, the choice of new expressions, the birth and
death of phrases, have been lef behind or have been noticed only in passing. Since
this subject deserves a name as much as does phonetics or morphology, I shall call it
semantics () that is, the science of meanings. (Bral 1991: 137)
Te new science of semantics was developed in the Essai de la smantique published in
1897 in which Brals pragmatic and historical perspective on language can be traced
(Desmet & Swiggers 1995: 292327). Bral was interested in the meaning that could
be called, from our contemporary perspective, contextual: he considered the spatio-
temporal changes as well as the sociolinguistic factors such as professional context in
the meaning of words (Bral 1991). He also considered the subjectivity that can be
seen in words or phrases, in grammatical forms and in language in general (Desmet &
Swiggers 1995: 313). In fact, he shows that the language a speaker uses refects his or
her impression of the world (Bral 1991: 174); subjectivity is also coded in language
expressing the speakers evaluation (Desmet & Swiggers 1995: 313319). Te examples
Bral uses are interesting: in order to show how the speaker expresses diferent degrees
of certainty, she or he can use adverbs like certainly, maybe, possibly (ibid.) a type of
analysis reproduced later by several linguists in enunciation theory.
In the following section, we discuss the main phases in the development of
enunciation theory. Te two frst phases are easily distinguishable. According to
Joly (1987), there are two generations: the frst enunciation linguistics (premire lin-
guistique de lnonciation) is the generation of linguists like Charles Bally, Gustave
Guillaume, grammarians Jacques Damourette and Edouard Pichon, and Ferdinand
Brunot; the second generation (seconde linguistique de lnonciation) is that of mod-
ern enunciation theory. In other words, the frst phase in enunciation pragmatics
(linguistics) is that of forerunners of the theory such as Charles Bally and Gustave
Guillaume. Te second phase is the stage of the main theoretical foundation that was
laid by Emile Benveniste over a long period of time, from the end of 1930s up to the
1970s. Te main contribution to the development of enunciation theory is to be found
in the two volumes Problmes de linguistique gnrale III (1966,1970), a collection of
articles by Benveniste. In a third phase, there have been several modern developments
by linguists like Antoine Culioli, Oswald Ducrot and Jacqueline Authier-Revuz who
have either developed their own approach to enunciation or redefned some concepts
that are essential to this theory. Teoretical thinking has developed ever since in sev-
eral, ofen diferent directions.
2.2 First phase: Forerunners
2.2.1 Charles Bally (18651947)
Charles Bally, a disciple of Saussure and one of the editors of Saussures Cours de
linguistique gnrale, is today considered one of the early predecessors of enuncia-
tive pragmatics even though his infuence was not really recognized until the 1980s
(Chiss 1985; Lala 2006), probably because he was closely identifed with Saussurian
structuralism and its emphasis on langue (system) at the expense of parole (discourse)
which led to the study of linguistic forms and to the elimination of the speaking sub-
ject (sujet parlant) from serious linguistic research.
3
However, Ducrot himself, for
example (cf. infra), acknowledges his debt to Bally for the development of his theory of
polyphony (Ducrot [1986]1991). Tere are four aspects which relate him with enun-
ciative pragmatics: (1) the emphasis on the synchronic study of ordinary language, (2)
the inherent subjectivity in all language use and hence the inscription of the subject/s
in discourse, (3) the role of interlocutors in the construction of discourse (the concept
of co-nonciation developed later by Culioli, cf. infra), and (4) the relationship between
language and discourse, i.e. the question of how entities belonging to the level of lan-
guage (virtual entities) are actualized in discourse, or questions related to the reference
of linguistic signs. In the following we will only discuss the frst three, omitting the
question of reference which will shortly be taken up when we discuss the insights of
Gustave Guillaume (cf.infra).
Bally may be best known for his early work on stylistics (Trait de la stylistique
franaise 1909). It must immediately be pointed out, however, that his conception
of stylistics is quite original since he deliberately excluded literary stylistics from his
study. In fact, he makes a distinction between literary and linguistic stylistics. Accord-
ing to Bally, linguistic stylistics observes and asserts whereas the aim of literary sty-
listics is to learn to use language for aesthetic reasons. In fact, Bally considers literary
language to be a result of several individual styles used for aesthetic reasons. He also
compares literary language to special languages:
. More recent research has shown that this is not true. Te interpretation must be attributed to
F. Bopp (Valette 2004)
nonciation 75
Literary language has frst of all a social value, it is a symbol of distinction, of high
intellect and high education; stylistics cannot but consider it (literary language)
as a special language [.]. As such it has its place a place of honour beside
administrative and scientifc language as well as that of sports. (Bally 1952: 28)
Instead of literary stylistics, Bally is interested in the language ordinary people use
when communicating orally in everyday situations. He stresses the importance of
studying the stylistic usages of a speech community instead of individual styles.
Te language of ordinary people refects, not pure ideas, but emotions, feelings,
desires, impulsions, in short, it is a means of expression and action. (Bally 1952: 133)
From the above citation it becomes evident that Bally considers language as a prag-
matic phenomenon it is a means of expression and action but as we will see below, it
is also a means of expressing thoughts.
He replaces the term stylistics by theory of enunciation in the frst edition of
Linguistique gnrale et linguistique franaise (1932) and by general theory of enun-
ciation in the second edition (1944). According to Chiss (1986:165), Ballys theory of
enunciation makes it possible to relate the principles of general linguistics with the
theory of the linguistic specifcity of a given language. As Chiss and Puech point out
(1997:162), enunciation in the sense used by Bally includes all the elements that make
a language a language diferent from any other (its specifc syntactic, lexical phonetic
and gestual dimensions) i.e. all the means by which a language gives form to com-
municated thought. Bally works with the linguistic resources of language (the French
language), not with the actual discourses produced. Chiss and Puech propose to inter-
pret Ballys enunciation as semantics, which broadens the feld of general linguistics
integrating into it the logical, psychological and linguistic conditions of all communi-
cation (nonciation) of thoughts by means of language (ibid).
It is not always clear whether Ballys nonc refers to utterance as a pragmatic
phenomenon or as a phenomenon belonging to the level of language structure (cf.
Ducrot [1986]1991: 3). For example, when defning the process of actualization he
says that in order to become an element of a sentence a concept has to be actualized
(Bally 1944: 77) or when discussing the relation of utterance and thought he afrms
that the sentence is the simplest possible form of communication of thought (Bally
1944: 35).
In accordance with a long and well-established philosophical tradition (cf. Ducrot
[1986]1991: 4, Chiss & Puech 1997: 159), Bally is concerned with the relation of lan-
guage and thought. To him, language is put into operation in order to express thoughts
or rather representations.
To think is to react to a representation by asserting it, by appreciating it or by desiring
it. Tought is thus not just pure and simple representation without any active
participation by the thinking subject. (Bally 1944: 35)
For Bally there are two kinds of mental representations, that is, ideas and feelings (senti-
ments). According to Bally, ideas refect the intellectual component of representations,
but they are nevertheless not the primary aspect in language use. Te main function of
language would, according to Bally, be the expression of feelings (Gouvard 2005). For
Bally, language is a two-level system of communication composed of (1) expressions of
ideas and (2) expressions of feelings, the latter being divided into individual (sentiments
individuels) and collective (sentiments sociaux). Te aim of stylistics is, in Ballys view, to
describe the emotive, subjective, dynamic and fugitive nature of language.
Nearly all representations of reality are subjectively coloured and deformed as they
penetrate language; furthermore language deforms itself constantly. What do you
expect to remain of reality in an image that has gone through such a double prism?
(Bally 1952: 155)
Since the function of language is to express thought (representation of reality) it
necessarily implies that the speaking subject is engaged in the activity either explicitly
or implicitly (Durrer 1998: 109). Bally emphasizes the relation of the speaking subject
to his utterance (nonc) since an utterance is not a true representation of the extra-
linguistic reality but a reaction of the speaking subject to a representation. Te reaction
may be intellectual, afective or volitional. One does not necessarily exclude the other,
and it is not a question of dominance but of degrees (Durrer 1998: 115). It follows from
this, according to Ducrot [1986]1991:4) that all thought breaks down into an active or
subjective element the reaction and a passive or objective element the representa-
tion. Tis brings us to the dichotomy of modus and dictum. According to Bally, every
utterance contains a content (dictum) and the attitude of the speaking subject towards
this content (modus).
An explicit sentence consists of two parts: the frst one corresponds to what constitutes
the representation (for example the rain) and this we call in accordance with the
logicians, the dictum. Te other is the master piece of the sentence, without which there
would be no sentence, in other words the expression of modality, corresponding to the
operation of the thinking subject. Te modality has as its logical analytic expression a
modal verb and its subject, the modal subject. Te two together constitute the modus.
(Bally1944:36)
According to Ducrot (1991:5), Ballys meaning of a sentence can be expressed in the
form,: X has this particular reaction to this particular representation where X has
this particular reaction is the modus and the type of reaction is expressed by the
modal verb. Te dictum is the representation of the object of the reaction. Te modal
subject is the subject to whom the communicated thought is attributed, but it is not
necessarily identical with the speaking subject even if the speaking and the modal
subjects coincide in most cases like in the following examples:
nonciation 77
(1) I want you to leave (Je veux que vous sortiez)
(2) I think that the accused is innocent (Je crois que laccus est innocent)
But the modal subject can also be diferent from the speaking subject, it can encompass
several subjects, or remain vague:
(3) We dont think it is going to rain (Nous ne croyons pas quil pleuvra)
(4) Galileo and astronomers think that the earth rotates (Galile, les astronomes
pensent que la terre tourne)
(5) Tey think that the king is dead (On croit que le roi est mort) (Bally1944:37)
Te distinction between modal and speaking subject makes it also possible to describe
negation as a dialogic phenomenon, even if Bally does not develop this line of thought
further. According to Bally, negation is a refusal to accept an assertion. Te meaning
of a negative utterance such as Te sun does not rotate around the earth is, accord-
ing to Bally, equivalent to I deny, it is false (to say) that the sun rotates around the
earth (1944:219). Such an assertion comes close to the concept of polemic negation
developed by Ducrot much later (cf. infra and Ducrot 1984).
But even in cases where the thinking subject is identical with the speaking
subject, care must be taken not to confuse personal thought (pense personnelle) and
communicated thought (pense communique). Tis distinction is of utmost impor-
tance and it can be explained by the nature and function of the linguistic sign itself.
In fact, the subject can communicate a thought that he pretends to be his own even
if it is not (bien quelle lui soit trangre (1944: 37)). Bally calls this split personality
(ddoublement de la personnalit) a concept that bears some resemblance to Authier-
Revuzs ddoublement nonciatif (cf. infra). As examples of such instances, Bally men-
tions deliberate lies and irony.
Since utterances are bound to express subjectivity inherent in all language use,
it is obvious that what Bally considers most worthy of analysis is the modus. Not all
utterances, however, are explicitly modalized, but there is a continuum from explicit to
implicit marking of modality. Even the implicit marking of modality is, however, not a
problem since the spirit (lesprit) can easily complete the insufciencies of expression
and thus nothing essential is lef out of the utterance (1944: 41). Such a line of thought
seems to anticipate speech act theories, especially those of indirect speech acts, their
illocutionary force and the interferences made by interlocutors.
In Ballys view, an utterance is, in the frst place, the result of the intentions of the
speaking subject but at the same time it is the result of interaction between the speaker
and his interlocutors, be they real or imagined.
One can hardly speak without speaking to someone. [.] Language should be seen
as expression of a thought communicated to someone else or having in mind the
representation of someone else. It is of no importance whatsoever whether the other
be an individual, a crowd or everyone.
(Bally, Trait de la stylistique franaise, vol.2 1909: 8)
Tis means that an utterance, or rather the meaning of an utterance, is co-constructed
in interaction, but at the same time it is a means of acting upon the interlocutors
(Durrer 1998: 123), that is to say that utterances contain always an argumentative
(implicit or explicit) dimension too.
When we say that it is hot or that it is raining, it is very rarely just an afrmation of
the state of afairs but rather an afective impression or a judgement which is likely to
determine an action. (Bally 1952: 17)
Here again the afnities with modern pragmatics are striking: there is a general agree-
ment in pragmatics that the informative function of the language is never the only one
operating in any discourse.
2.2.2 Gustave Guillaume (18831960)
Gustave Guillaume, a contemporary of Bally and a disciple of Antoine Meillet,
4
can be
regarded a representative of a psycho-cognitive orientation of the frst phase of enun-
ciative pragmatics, especially in his early scholarly production, whereas his later work
can rather be associated with cognitive linguistics la franaise (Joly & Roulland 1981;
Valette 2003). Being a representative of his time, Guillaume is obsessed with the rela-
tionship of language and thought. He tries to give thought a linguistic status which
would establish the thought/language relation permanently. According to Guillaume,
thinking takes place in time. In other words, the conception of ideas is an intellectual
process which requires a certain, even if at times a minimal lapse of time. To think
about a notion means to construct it and the time needed for it is called operative time
(temps opratif). Tis process is characterized by a twofold activity which moves from
the maximal to the minimal extension of a concept and then again from the mini-
mal to the maximal extension. Te maximal extension corresponds to universality and
the minimal extension to a singular case. Te movements in operative time serving
to construct a notion are generalization and particularization. (Ducrot & Schaefer
1995: 5960). From the enunciative point of view the following elements are essential in
Guillaumes refexion: the use of language is an act, language is dynamic, and the rela-
tion of language and thought is intrinsic.
|. Antoine Meillet (18661936) was one of the leading linguists of his time. He was a specialist in
the history and structure of Indo-European languages. He seems also to have been one of the rst
to use the concept of grammaticalization.
nonciation 79
According to Guillaume (1912), language is on the one hand always an act, moti-
vated by the speakers will to infuence the other, and, on the other hand, it is a means
by which the human thought changes and evolves in time (Valette 2006: 61).
[] language (parole) is a force, a means to act upon others. Tis action upon
other human beings who are like us, foreseen before the act of speaking, guides our
thought, gives it a form and at certain moments becomes so imperious that it can
break ancient moulds and create new ones which are more suitable for our purposes.
(Guillaume 1912: 4)
Guillaume rejects Saussures opposition between synchronic and diachronic. Accord-
ing to Guillaume, language can only be understood as a system that is dynamic and
in constant evolution. Te system of language is also dynamic because of the relation
of language to discourse (a term analogous to Saussures parole). Te linguistic phe-
nomena to be explained (diverse uses of a grammatical form, polysemy, ambiguity,
etc., in short, everything that has to do with actualized meaning (efet de sens)) mani-
fest themselves in discourse, whereas the operations which lead to such usages are
psychomechanic in nature and thus belong to the level of the language system. Tey
cannot be directly apprehended but have to be discovered via methodological refec-
tion (cf. Culiolis model of enunciative operations infra). In other words language pre-
exists discourse, but the linguistic analysis is based on an ascending description, from
discourse to language (Valette 2006: 24).
At the beginning of the 20th century the concept of actualization (cf. Bally supra)
is one of the key theoretical notions in the frst phase of enunciative pragmatics
(linguistique de la parole). It seems as though it was invented by Bally and Guillaume
almost simultaneously. Bally uses it for the frst time in 1922 and Guillaume in 1929
(Valette 2004). As we have seen, Bally defned actualization as a process that converts
a virtual linguistic entity belonging to language to an actual entity of discourse. For
Guillaume, actualization is necessary in order to produce discourse. In both cases
actualization necessarily implies an activity instantiated by the speaking subject. Tere
is still a marked diference between Bally and Guillaume. Guillaumes psychomechanic
approach is confned to the level of the word; actualization is consequently a process
that concerns the word which, by virtue of the article, is activated in discourse; its
status changes from puissance in posse (potential meaning) to efet in esse (meaning
in discourse) (Valette 2006: 40). He had developed a similar line of thought already
in 1919, in Le problme de larticle et sa solution en langue franaise, but instead of
mentioning actualization, there he proposed the hypothesis of fundamental kinesis
(cintisme fondamental), the way in which thought creates the specifc/universal rela-
tion by moving from the maximal extension of a notion to the minimal, and then
from the minimal to the maximal extension again. Such a movement (kinesis) explains
the diferent referential meanings (efets de sens) actualized by French indefnite and
defnite articles un/le ranging from universal to particular. Un/le soldat franais (a/the
French soldier) in its universal, generalizing meaning encompasses all French soldiers
(un/le soldat franais ne connait pas la fatique a/the French soldier is never tired),
whereas their particularizing function refers to a specifc individual (un/le soldat fran-
ais que je connais a/the French soldier I know).
As for Bally, the minimal unit is not the word but the sentence and hence for him
the concept is actualized in order to become a term of a sentence.
As pointed out by Valette (2003), the notion of transcendental subject, i.e. the
speaking and thinking subject, gradually loses weight in Guillaumes theoretical
refection, it no longer is a concept with the help of which the linguistic operations
might be explained. On the contrary, it becomes a problematic concept. Language is
seen as invading the subject, it interferes with the subject who has to speak and takes
control over him. Slips of the tongue, reformulations, self-repairs, diferent versions
of a text or of a defnition are examples of the interference of language in the com-
municating subject. Te power of language tends to expand inevitably at the expense
of subjective control, but, paradoxically, to the beneft of the subject. Tus, the focus
in Guillaumes work changes from the speaking and thinking subject to the study of
language systems (psychosystmatique) or as Valette (2006:72) puts it, he abandons
the speaking subject for language. Tis is to say that Guillaumes psychomechanic
model would not really include a theory of enunciation as proposed by Joly et Roul-
land (1981) but rather that it deals with some of the problematics of enunciation
(Valette 2006: 75).
2.3 Second phase: Main theoretical foundation
Tere are several linguists and other theorists who are closely linked to enunciation
theory or who have infuenced it a great deal. Tree names come to mind in par-
ticular: Roman Jakobson, Mikhail Bakhtin and Michel Foucault. As pointed out by
Normand (1985a), the notion of shifers developed by Jakobson as early as 1957 had a
remarkable infuence on the linguistic scene in France. In fact, the work of Jakobson
was much more frequently cited in the sixties than that of Emile Benveniste (cf.
infra), considered the founder of the theory of enunciation (Normand 1985a: 810).
Likewise, the ideas put forward by Bakhtin had a great impact on the theory on
enunciation, especially his conception of the social and dialogical nature of all lan-
guage use even if these notions did not spread among French linguists until the late
seventies with the frst translation of Bakhtins work (Le marxisme et la philosophie
du language 1977).
From an epistemological point of view, the philosophy of Michel Foucault is
an entirely separate theory of the pragmatics of enunciation. However, Foucault has
been so infuential for quite a few French linguists, especially in the feld of discourse
nonciation 81
analysis, that some of his theoretical conceptions need to be mentioned. His focuses on
instance dvnement nonciatif the instance of uttering event. His conception is not
structural, grammatical or pragmatic, but rather social. Te social subject produces
an utterance that is not independent but contextual, existing only in the discourse
(Foucault 1969).
In the following, we will concentrate on some of the main theorists of the second
phase of enunciative pragmatics.
2.3.1 Emile Benveniste (19021976)
Te main theorist of enunciation theory is Emile Benveniste even if the expression
theory of enunciation does not appear anywhere in his writings. Benveniste grew
up in the structuralist tradition and his work overall was remarkably infuenced by
Saussurian thinking (cf. van Hecke 2002) and apparently he, too, considered himself
an heir of this line of thought by positioning himself as linguiste smioticien (Fontaine
1986: 213). However, his elaboration of the theory of enunciation difers from the
structuralist emphasis on language excluding the level of parole and consequently the
speaking subject as a serious object of linguistic study (cf. however, Normand 1985b:
5 an object of study not scientifc yet, and Brunet 2004 suggesting that the problem
of subject was never really totally excluded from linguistics). It is only in the 1960s
that aspects related to enunciation are systematically taken into account in linguistic
analysis in France. Tis is largely due to the succes of Benvenistes refection presented
in his Problmes de linguistique gnrale (1966). As Ducrot points out (Encyclopedia
Universalis), the ffh section of the book has an emblematic title: Lhomme dans la
langue (Man in language). In the sixties, the ideas of Benveniste spread more rapidly
among French philosophers who saw in his refexion a way out of rigid structural-
ism whereas in linguistic circles Benveniste was largely ignored or, if mentioned, he
was always closely related to structuralism (Normand 1985a). In what follows we will
only deal with aspects of Benvenistes work related to enunciative pragmatics omit-
ting other aspects of his ambitious scholarly work already well documented by van
Hecke (2002).
Even though an heir of Saussurian thinking, Benveniste does not approve of the
dichotomy langue/parole but argues quite the opposite: Nothing is in language that
has not been before in discourse (1974: 80). As we have seen, the term enunciation
had already appeared in the writings of Bally in the early 1930s but the defnition of
the term must be attributed to Benveniste. He writes: Enunciation means activating a
language in discourse by an individual usage act. [] Tis is the act of a speaker who
mobilizes a language for his own use (L nonciation est cette mise en fonctionnement
de la langue par un acte individuel dutilisation (1974:80). And he continues by saying
that before enunciation, language is but the possibility of language. Afer enunciation
language is realised in an instance of discourse which has its origins in the speaker
[1970]1974:81). By enunciation is meant the actual act and circumstances of producing
an utterance.
Two of the most important articles in which Benveniste develops his theoretical
thinking on enunciation are La nature des pronoms [1956]1966), where he already
sketches the milestones of the concept of enunciation (the notion of shifers, third
person as a non-person, diferent elements of deixis) without calling it enunciation,
and secondly, L appareil formel de l nonciation [1970]1974), where he explicitly lays
the foundations of the theory of enunciation.
One of the questions Benveniste problematizes is: what does I signify ? And he
answers: I refers to the individual speech act (acte de discours) where it is pronounced,
and it designates the locutor (1966:261). For Benveniste there are two kinds of lan-
guage usages: cognitive and enunciative. Whereas the utterance Socrates is a man
has a universally valid meaning, the meaning of the utterance: I am a woman is
dependent on the person saying so precisely because it contains the frst person singu-
lar I. Tus, in language there are elements whose meaning is more or less stable but in
contrast there are others, like I, the meaning of which is dependent of the situation of
enunciation in which they are uttered.
Utterances produced by individual speakers bear traces of how language is
mobilized in discourse. Tis means that the use of certain grammatical constructions,
words and expressions entails or contributes to creating specifc relations between the
interlocutors. Enunciation can be seen from the individuals point of view as a process
of appropriation. Te speaking subject appropriates the formal apparatus of language
and utters his position as a speaking subject by resorting to specifc linguistic signs.
Tese are traces (indicateurs in 1956) that translate the subjectivity of the speaker,
for example when designating himself I, the speaker appropriates, makes his own the
formal apparatus of language.
Te individual act of appropriating a language inserts the speaker into his speech
act. [] Tis situation is evidenced by specifc forms whose function is to
place the speaker in a constant and necessary relationship with his enunciation.
(Benveniste 1974: 82)
Te forms are anchored in an utterance situation of which the constitutive parameters
are: speaker, interlocutor, place and time of interaction, i.e. person deixis and spatio-
temporal deixis. But as the subject positions himself as the speaker/locutor he neces-
sarily implants the other in his discourse since all enunciation is explicitly or implicitly
interactional: it postulates an interlocutor. Furthermore, reference is an integral part
of enunciation since enunciation is used to express a certain relation to the world.
According to Benveniste (1974:82), the prerequisite for mobilizing and appropriating
language is the need for the locutor and his interlocutor to refer to extra-linguistic
reality by means of discourse in an identical way.
nonciation 83
Te interpretation of frst and second person pronouns is entirely dependent on the
situation of enunciation.
In the case of the frst and second person pronouns there is always a person implied
and a discourse about him. I refers to the person speaking and it implies at the same
time an utterance on his account. When I am saying I, I cannot but speak of myself.
Te 2nd person (tu) is necessarily designated by I and it cannot be conceived of
outside a situation created by I (je). (Benveniste1966: 228)
What characterizes the frst and second person is their uniqueness (unicit), whereas
the reference of third

person is vague.
Te I who speaks and the you (tu) to whom I addresses himself are always unique.
But he (il) can be anyone or nobody. He as such does not refer to anything or anybody
in particular. You (tu) could be defned as a non-subjective person opposed to the
subjective person I and these two together are in opposition to the form of non-
person designated by he (il). (Benveniste 1966: 232)
Tis leads to the conclusion that frst and second person singular pronouns do not
have any fxed semantic meaning, they are but discursive entities anchored in the
discursive reality of enunciation. In addition to frst and second person pronouns,
there are other pronouns (demonstratives this), adverbs (today, now, here) and
adverbial expressions (the year before) of time and place which organize spatial and
temporal relations in relation to the speaking subject, taken as the point of reference,
and which can only be identifed through the instance of discourse i.e. the utterance
situation where they have been uttered, in other words under the dependence of I
who expresses himself (Benveniste 1966: 262).
Benveniste also applies the idea of the subjectivity and circularity of the meaning
of sui-referential linguistic entities to the verbal system. Te present tense coincides
with the moment of uttering (instance du discours) in relation to which other tem-
poral determinations can be situated. A certain organisation of the notion of time is,
according to Benveniste, a universal feature of languages irrespective of their formal
structural or typological properties. And in all cases the demarcation line is the pres-
ent tense coinciding with the instance of discourse which it describes.
[] there are no other criteria nor any other expressions to describe the moment
where one is but to consider it as the moment when one speaks. Te present tense is
eternally present but it never relates to the same event of objective chronology because
it is defned in relation to every instance of discourse and every locutor. Linguistic
time is sui-referential. In the last resort, the human conception of temporality with
its linguistic apparatus reveals the inherent subjectivity operating in language.
In his article Relations de temps dans le verbe franais in Problmes de linguistique
gnrale I (1966), Benveniste proposes two levels of discourse (plans dnonciation)
according to whether the verbal system (among other indicators of subjectivity)
expresses the presence of the speaking subject or not. When the engagement of the
locutor is heavily marked in his discourse, it is called discourse (discours), whereas
utterances without the presence of the speaking subject belong to the level history
(histoire). As far as the verbal forms are concerned, discourse is characterized in par-
ticular by the use of the present, future and perfect tenses, whereas history is charac-
terized by preterit and aoristic tenses and the present tense expressing general truths.
Discourse belongs to speech and is also characterized by the presence of all pronomi-
nal forms but especially of the frst and second person and deictic markers. History is
the domain of writing, characterized by the third person singular or absence of pro-
nominal reference and it is devoid of deictic markers. Since the term history was not
transparent, it was later replaced by rcit (see for example Mainguenau 1994). Te
dichotomy histoire or rcit vs. discours has in recent years met with much criticism as
the title of an article by Adam et al., Pour en fnir avec le couple rcit/discours (To put
an end to the couple rcit/discours, 1998), indicates. Maingueneau (1994) proposes
instead a dichotomy of deictic (embray) and non-deictic (non-embray) discourse.
Since enunciation is characterized by subjectivity but also by inter-subjectivity, the
speaking subject uses language not only to convey the propositional content of utter-
ances but also in order to act upon his interlocutor. Tis leads Benveniste to account
for what he calls grandes fonctions syntaxiques (1974: 84), the kind of relation the
speaker wants to establish with his interlocutor. We would today call them speech acts
and their illocutory force such as the function of interrogation, assertion, imperative
and vocative. Furthermore, the speakers attitude as to dictum (the content of his utter-
ance) realized through the use of a large variety of modalities (cf. infra), some of which
are related to the category of verb and others to lexical entities belonging to phraseol-
ogy (ibid:85), can be accounted for in relation to the speaking subject I indicating
the subjectivity inherent in language use. But this manifestation of subjectivity is only
efective in the case of the frst person singular.
When I say I suppose that, I presume that, it indicates an attitude taken, not a
description of an operation []. I imply that I have a certain attitude towards the
proposition to follow. (Benveniste 1966: 264)
Benveniste has been considered as the pre-fondateur of the theory of enunciation. But
as already mentioned, neither the theory of enunciation nor the subject of enunciation
(sujet dnonciation) are explicitly mentioned in his texts. Tis has animated two kinds
of critical re-evaluation.
First, even if Benveniste repeatedly refers to the subjectivity inherent in discourse,
he does not specify what he means by subject. Either he uses traditional grammatical or
nonciation 85
psychological terminology such as subject of a verb, ego, subject as the home of sentiments,
of intentions, or a-theoretical concepts taken from everyday language to describe sub-
jects such as individual, locutor, participant, oneself (Normand 1986: 201). Tis con-
tradiction, absence of a well-defned subject of enunciation and recurrent references
to subject and subjectivity in discourse, has given rise to the question as to whether
Benveniste really intended to create a theory of the subject. Or was the attribution of
the concept sujet de l nonciation to Benveniste rather an over-interpretation of those
who in the 1960s were engaged in the debate concerning the subject investing the
notion with psychoanalytic, marxist and linguistic assumptions? A second epistemo-
logical problem is the fact that Benveniste was in essence interested in the meaning of
linguistic signs (the domain of semiotics), but also in meaning in discourse (the domain
of semantics) hence the necessity of the concept of subject. Tere can be no meaning,
no predication, without a subject which he, however, did not succeed in defning except
in an unsatisfactory and vague manner (Normand 1986: 202).
Te second re-evaluation is related to the absence of the term theory of enunciation
in the writings of Benveniste. Ever since the 1990s it has been suggested that
Benveniste did not really propose a unifed theory of enunciation but rather a ques-
tioning around enunciation, a beginning of a theory to be completed by his successors
(Valette 2006: 43).
2.4 Third phase: Modern developments
Te third phase starts at the end of 1960s when scholars such as Antoine Culioli and
Oswald Ducrot take up the structuralist heritage in their work (Ducrot 1968) and
develop it further, in rather diferent directions based on the theoretical thinking of
that time. In this section, we discuss the work of Antoine Culioli, Oswald Ducrot, and
Jacqueline Authier-Revuz and their connection to enunciation theory.
2.4.1 Antoine Culioli (born in 1924)
Te starting point for all linguistic investigation is according to Culioli symbolic
activity, which is by nature cognitive and afective, language activity apprehended
through the diversity of languages, texts and situations. Texts in Culiolis sense include
both written and spoken texts. Consequently, linguistic research should not limit itself
to one language or even to a language family in order to avoid hasty and simplistic con-
clusions. Culioli is looking for invariants through a theoretical framework that would
take into account both the diversity and resemblances between languages. Secondly,
all linguistic research should be based on an analysis of authentic empirical material
validated by native informants. Tirdly, Culiolis conception of language is holistic:
there is no need for a separate analysis of production and reception. Furthermore,
all observable facts should be accounted for (slips of the tongue, ambiguity), and
fnally, Culioli does not separate syntax and semantics but claims that we can provide
a unifed theory which will integrate phenomena which at present are treated sepa-
rately in diferent sectors (Culioli [1978] 1990:73).
Culiolis main preoccupation is the mechanism underlying the production of
enunciation, i.e. the uncovering of operations through which the meaning of utter-
ances is constructed. Enunciation as understood by Culioli is the metalinguistic res-
titution, based on utterance (nonc) of the meaning construction (construction du
sens) (Valette 2004). For Culioli, enunciation is not a question of appropriation of the
apparatus of language as it is for Benveniste, but a work of reconstruction performed
by the linguist. For Culioli there is no subject a priori in enunciation, it has to be recon-
structed through the analysis performed by the linguist (Valette 2006: 42).
Culioli is probably best known for his theory of enunciative operations (Torie
des oprations nonciatives, TOE) which he started to elaborate as early as 1968 and
which he has been developing until recent years. It is a cognitively oriented seman-
tic theory of enunciation in which some of his commentators see resemblances with
Guillaumes psychomechanic model (cf. supra), an infuence he himself does not really
acknowledge (Valette 2006: 258260). Te main hypothesis underlying the theory is
the following: enunciative activity is based on the interpretative activity which takes
place when producing and recognizing abstract forms (cf. Ducard 2006: 13).
[] we hope to construct an ensemble of formal operations in order to understand,
through the diversity of languages and discursive phenomena, these fundamentally
generalizable operations underlying enunciative activity, production on the one hand
and comprehension on the other. (Culioli [1978] (1990:304)
In the following we will present the outlines of Culiolis theory of enuncative operations
very briefy. For the reader wishing to get a deeper insight into the theory we refer to a
collection of Culiolis texts in English, selected, edited and introduced by Michel Liddle
(1995). Te theory has three levels:
I. Te epilinguistic level: which is unconscious and never directly accessible: the level
of afective, cognitive and senso-motoric activities giving rise to mental repre-
sentations which Culioli calls either physico-cultural representations or primitive
properties. Tese are based on our experience of the world. Examples of primitive
properties are animate/inanimate, male/female, discrete/continuous (Groussier
2000: 169).
II. Te linguistic level: the level where we can observe textual traces (utterances as
a system of markers that can be lexical and grammatical) of cognitive processes
creating representations. Tere is no one-to-one correspondence between the
terms of levels I and II. As a consequence, a marker can correspond to several
notional values or several markers can correspond to one and the same notional
value on level I.
nonciation 87
III. Te metalinguistic level: the level of theoretical schematization based on observa-
tions of phenomena belonging to level II. It follows from this that the metalin-
guistic representations of level III are representations of representations and not
representations of level I (Normand 2005; Valette 2006: 265266).
Hence, the theory of enunciative operations is not a theory of cognitive operations of
linguistic activity. Culiolis enunciative operations are strictly theoretical, metalinguis-
tic in nature. Nevertheless, Culioli hopes level III to be in such an adequate relation to
level II that a simulation of the correspondence between the levels I and II would be
possible (que le niveau III sera dans une relation dadquation (de correspondance) au
niveau II, telle que, par le biais de cette relation explicite entre II et III, nous puissions
simuler la correspondance entre I et II (1990:23)).
Te term enunciative in the name of the theory refers to the uttering situation
(situation dnonciation) composed of the utterer (nonciateur) and the circumstances
of enunciation, i.e. time and place of uttering. Enunciator (nonciateur) should in
Culiolis model not be confused with the speaker/locutor. Te locutor is the one who
actually pronounces the utterance whereas enunciator is an abstract entity. Te enun-
ciator is responsible for all the discursive choices made in the text and he also assumes
responsibility for the modal choices in the utterance. As we have seen (cf. Bally, Ducrot
supra), the speaking subject may coincide with the enunciator, but this is not neces-
sarily the case. Co-enunciator is equally an abstract entity, not to be mixed up with
the concrete addressee. His presence and his relationship to the enunciator afects the
construction of the utterance. In fact, this notion in Culiolis work is slightly ambigu-
ous. It can either refer to the interlocutor/s, to the enunciator as his own co-enunciator,
or it can just be the theoretical counterpart of enunciator. As for operations (oprations
de rperage), they relate to formalized processes by means of which utterances are con-
structed from notional representations that belong to level I (cf. Groussier 2000). Te
concept of notion is slightly problematic in Culiolis theory.
We call notion a complex system of representations which structure physico-
cultural properties of cognitive order (lexical, grammatical) and in general every
relation between notions. A notion precedes categorization into nouns, verbs etc. []
(Culioli [1982] 1999: 100)
According to Grossier (2000:164), notions are not given phenomena but constructs.
Tey are at the same time clusters of representations of primitive properties (cognitive
representations) and the frst phase of metalinguistic representation which can only be
apprehended through specifc realizations that are called occurrences.
One only encounters notions through occurrences. We do not study activity as it
appears through non-verbal behaviour, but only through verbalized behaviour [].
(Culioli [1995] 1999: 41)
As we have seen, language (langage), in Culiolis interpretation, is an activity which
presupposes a continuous epilinguistic activity (unconscious activity) and a relation
between a model and its actualization, of which we have phonic or graphic traces, that
is texts (Culioli [1968]1999:19).
Language use (activit langagire) is signifcative (signifante) i.e. it bears a meaning
and furthermore, this meaning is co-constructed. But at the same time language is also
refexive:
[..] because in communication there are two-ended operations by virtue of which
utterances receive their meanings (the operations are complex because the sender is
at the same time receiver and vice versa). (Culioli [1968]1990: 19)
Tis means that when speaking we are simultaneously monitoring what we are saying
and anticipating what we have not yet uttered. Te enunciator is thus also his own
co-enunciator, a mirror of himself as enunciator (Culioli 2002: 67).
When we talk about subjectivity, we should always talk about inter-subjectivity,
of which the relation to oneself is one special case, a subject in relation to himself.
(Culioli 2002: 208)
Culiolis theory of enunciative operations is very formal, combining linguistic concepts
and concepts from other disciplines such as mathematics, logic and topology. Te
emphasis is on the formalization of his model of mechanisms underlying enuncia-
tion. Nevertheless, the following citation reveals the importance of enunciation as a
pragmatic concept for Culioli:
[] to utter (noncer) is to construct a space, to orient, to determine, to establish a
network of referential values, in short a system of location (systme de reprage). Every
utterance is located in relation to the uttering situation (situation dnonciation), which
in turn is defned in relation to the enunciator, or to be exact to the frst enunciator
and to the time of uttering. [1973]1999: 49)
2.4.2 Oswald Ducrot (born in 1930)
Te theoretical work of Oswald Ducrot is to be situated centrally in the feld of
semantics and partly in pragmatics in France. His approach is not clearly a prag-
matic approach, but one that combines a strongly semantic approach with a prag-
matic one, called integrated pragmatics (pragmatique intgre). Ducrot has worked
on several pragmatic issues, for instance, on presupposition (Ducrot 1972, 1984) and
argumentative particles (Ducrot 1980a), and has developed several diferent theo-
retical approaches, such as a theory of argumentation in language (Anscombre &
Ducrot 1983), the theory of argumentative scales (Ducrot 1980b) and more recently,
the theory of semantic blocks (Ducrot 2004), to name just a few (for an elaborate
nonciation 89
introduction on these issues, see Moeschler 2006). Here we take up the perspective
Ducrot ofers on enunciation theory, more specifcally on enunciative polyphony.
Te theoretical view of enunciation created by Ducrot focuses on understanding
the utterance meaning from a very specifc semantic perspective. Tis view is based on
an internal view of language and it analyzes utterance-level meaning. Even though he
sees that the situation attributes a meaning to an utterance (Ducrot 1980a: 8), Ducrots
approach is not based on the analysis of the situational context, but rather on the cotext
or the linguistic context present in an utterance. In his view, enunciation is a momen-
tary apparition of an utterance and therefore a historical event and it is in the utter-
ance that the enunciation describes its act of uttering (Ducrot 1984: 179). Terefore,
every utterance contains a commentary about how it has been uttered by the speaker.
Ducrot distinguishes three levels of description: phrase, that is a theoretical object,
an object of grammar, utterance, a particular manifestation of the spatio-temporal
nature of a phrase, a linguist object, and discourse, a linear string of utterances (Ducrot
1984: 174). He also distinguishes phrase-level meaning (signifcation) from utterance-
level meaning (sens) that is a description of the act of uttering.
Created to contest the views by structuralism and generative grammar about the
unity of the subject, the theory of enunciative polyphony (Ducrot 1984) is inspired
by the Bakhtinian notion of polyphony. However, Ducrot presents quite a diferent
interpretation of Bakhtinian multivoicedness, placing polyphony at the utterance
level. Ducrot (1980a:44) claims that at the utterance level, polyphony is created by
the simultaneous presence of voices. According to Ducrot (1980a:44), at the utter-
ance level, there is a plurality of voices, or opinions expressed that are not those of
the speaker. In order to describe this, Ducrot introduces the notion of split subject.
In an utterance, one can distinguish the speaking subject (sujet parlant), the locutor
or speaker (locuteur) and an enunciator or utterer (nonciateur): the frst one is the
empirical being in the world producing the utterance, the second is present in the
discourse uttering the act and the third is the voice that has responsibility over
the opinion, the attitude etc. of the utterance. Terefore, an utterance can have several
meanings that can be opposing, as in cases of polyphonic negation or irony shown in
the following example (Ducrot 1984: 215):
(6) Pierre is not nice (Pierre n est pas gentil)
In this utterance, there are two diferent opposing enunciators, utterers who produce
the utterance meaning:
Enunciator 1: Pierre is nice. (Pierre est gentil)
Enunciator 2: the contrary meaning to the frst enunciators
In the utterance, the meaning is a result of the combination of the meanings produced
by two enunciators. In the case of irony, an utterance is presented from the perspective
of one enunciator who does not take responsibility for the utterance, does not commit
her/himself, but instead distances her/himself from it as in the next example:
(7) What a disaster! (In quite the opposite situation)
In this view, the speaker can commit her/himself to the utterance meaning or distance
her/himself from the utterance meaning and leave the responsibility for the utterance
meaning to somebody else. In the case of frst person marking, the commitment is clear:
(8) I think that X (Je pense que X)
In the following example, the speaker shows a diferent enunciator who takes the
responsibility for the utterance meaning:
(9) It seems that the weather is going to be nice (Il parat quil va faire beau)
By using an impersonal pronoun (il) in a impersonal verbal construction (il parat
que) that introduces a subordinate clause, the speaker distances her/himself from the
responsibility that the meaning the weather is nice is produced by him.
Tis split subject has as its counterpart the split hearer: the listener (auditeur),
who is hardly mentioned by Ducrot, the hearer (allocutaire), present at the discourse,
and the addressee or receiver (destinataire) of the enunciators opinion and the act of
utterance. Speaker and hearer are the ones who can use the frst and second person
pronouns. To Ducrot, polyphony can be described with the help of a theatre metaphor:
in the enunciation, the speaker gives a certain kind of representation in the utterance
(cf. Ducrot 1984: 231).
2.4.3 Jacqueline Authier-Revuz (born in 1940)
Quite a diferent theoretical approach is developed by Jacqueline Authier-Revuz, who
also takes not only Bakhtinian polyphony as an inspiration, but also Jacques Lacan
and his conception of the division of the subject. In her earlier work, Authier-Revuz
introduces one basic conception about the nature of enunciation which she sees as
enunciative heterogeneity (htrognit nonciative). Tis means that the production of
utterances, speech, parole, is essentially heterogenous as the subject is divided. Te use
of language is fundamentally polyphonic and the production of discourse is a result of
interdiscursivity (Authier-Revuz 1984: 100101). Furthermore, Authier-Revuz (1982)
makes a distinction between two types of heterogeneity, namely constitutive and shown
heterogeneity. Te constitutive heterogeneity is that of the unconsciousness and the way
it traces itself in language: therefore the subject is an efect of language, divided, split,
decentralized; the subject cannot be a center, but what is central is the encounter of
the subject with the Other (Authier-Revuz 1984: 101). However, it is the shown het-
erogeneity that allows a linguist to see how the Other is inscribed in the discourse. In
other words, these are the phenomena found on the metadiscursive or metalinguistic
nonciation 91
level in which certain types of linguistic forms encode the presence of the otherness.
In practice, this means that linguistic forms are either non-marked forms or marked
forms (Authier-Revuz 1984: 98, 1992: 41). Te former are to be found in irony and free
indirect speech, the latter are those of reported direct speech, quotation marks, italics
or mentioned fragments or words. An example of the latter would be:
(10) according to Z (selon Z)
(11) the term X (le terme X)
In these examples, there is a mention of the fragment according to and the term Z
which creates in both cases a refexive reference. One of the linguistic terms Authier-
Revuz (1992: 41, 1996) introduces is that of textual islet (lot textuel), which is a word
placed in quotation marks that refers to itself:
(12) John said that his villa was in bad shape.
(Jean a dit que sa villa tait en mauvais tat.)
Tis is a type of discourse frequently used in journalism, but as a metalinguistic
phenomenon it is indirect discourse reporting another enunciation, or a production
of utterance with what is preserved from the original utterance in quotation marks
(1992 :41).
In her main work, Authier-Revuz examines the shown heterogeneity as a refexive
loop (boucle rfexive). By this term, Authier-Revuz means the kind of forms that show
the representation of speech in the making (Authier-Revuz 1995a, b) for instance from
what is said, to ensure a logical progression in speech, to demonstrate a phase in a con-
versation (Authier-Revuz 1995b: 23). In Authier-Revuz (1995) these cases are divided
into four categories: (1) the relationship of the co-enunciators.
(13) what you call (ce que vous appelez)
(14) if you like (si vous voulez)
(2) something that has been already said.
(15) as is said (comme on dit),
(3) a non-coincidence between words and things.
(16) the word is not appropriate (le mot ne convient pas),
(17) that is the word (cest le mot)
and (4) the meaning is afected by other meanings:
(18) in the sense of p (au sens p),
(19) in all the meanings of the word (dans tous les sens du mot).
Tese examples show diferent types of metadiscursive utterances: it is also called
enunciative dedoubling (ddoublement nonciatif) as the enunciator takes a posi-
tion which is that of an observer who comments on the enunciation. Tis means that
in the production of the utterance there is something that does not go by itself , that
is not automatic or transparent in the language; this is described by Authier-Revuz
in the following way: non-coincidence of saying, a non-unity (non-un), auto-repre-
sentation of saying in the making (auto-reprsentation du dire en train de se faire). In
sum, Authier-Revuzs enunciative linguistics is an explanation of the metasemantic
and metapragmatic use of language in a speakers (single) utterance.
3. Some basic notions
3.1 Enunciation and enunciator
Te main concepts are enunciation, act of uttering, nonc, utterance, a linguistic object
that is the product of the act of uttering, and nonciateur, enunciator, the one who
produces utterances or utterer (see above) or speaker depending on the theoretical
approach in which the terms are used. In order to consider a situation communicative,
the counterpart of the speaker is either receiver or addressee. Tese notions refect the
communication scheme developed by Jakobson and the unilateral model of commu-
nication. In more recent accounts, this explanation of the relationship between the
enunciator and the addressee has been given more broad meanings. A dialogic view of
enunciation has produced the concept of co-enunciation which takes the co-presence
of two or several speakers in the speaking situation and assigns them as co-enunciators
or co-speakers. However, in the framework of diferent enunciation theories, the last
distinction between speaker and utterer is a fuctuating one between the utterance
level (Ducrot) and the discourse level. Te adjective enunciative (nonciatif, -ive) refers
to enunciation.
One of the basic distinctions of enunciation theory is between restricted enun-
ciation (nonciation restreinte) and extended enunciation (nonciation tendue). Te
former analyzes only the speaker and the traces she or he leaves in the utterance
(Kerbrat-Orecchioni 2006: 3536), whereas the latter perspective takes into consider-
ation the participants, the situation, the spatio-temporality and the general conditions
of the production and reception of the message (Kerbrat-Orecchioni 2006: 35).
3.2 Situation/Context
Since the terms situation and context have acquired so many diferent meanings in
the linguistic literature, the following classifcation is an attempt at clarifcation (cf.
Charaudeau 2002: 535536).
nonciation 93
1. Communicative situation refers to the extra-linguistic situation where interac-
tion takes place.
2. Utterance situation (situation dnonciation) refers to the situation in which
language is actualized by an individual act of usage characterized by deictic and
illocutionary markers.
3. Discursive situation refers to inter-discursive knowledge and to institutionalized
ways of thinking and speaking. It sets the limits of what can be spoken and, more
importantly, how something may be spoken of (cf. Maingueneau 1997)
3.3 Subjectivity and deixis
Enunciation theory can be considered as the theory of the speaker, or the speaking
subject. Te notion of subjectivity is one of the most central notions that appear in the
writings of several theoreticians during the development of this approach and also in
contemporary writings (Vion 1998). It was Benveniste who noted how subjectivity is
the speakers capacity to position her/himself as subject:
It is in and through the language that the human being becomes a subject; because it
is language alone that provides a foundation for the concept of ego in reality, in his
reality which is that of existence.
Cest dans et par le langage que lhomme se constitue comme sujet ; parce que le
langage seul fonde en ralit, dans sa ralite qui est celle de ltre, le concept dego.
One of the main axes of investigation is the personal deictic system by which the
subjectivity and the relationship of the speaker becomes manifest intersubjectively.
Accounts concerning deixis consider person, place and time and what their gram-
matical forms and their meanings are (Cervoni 1987; Morel & Danon-Boileau 1992;
Maingeneau 1994; Kerbrat-Orecchioni 2006). Deictics refer to the way in which they
take the spatio-temporal environment (entourage) into consideration when describing
an occurrence: indexical meaning is given to the referent by the contextual information
(Kleiber 1986: 19).
In accounts deriving from Bhler, the speakers point of view is seen as the deictic
origo or centre; in deictic accounts, the main criteria in identifying the referent of the
person form are the time and the place of the situation. Te focus is on the varying
reference of pronouns or other type of shifers, or embrayeurs as they are also called
(Jakobson 1963). From the point of view of subjectivity in enunciation theory, the
person form je is the one who speaks; it is also the name of the speaker when she or
he leaves a trace in her/his utterance. According to Benveniste (1966: 253), the present
situation of producing an utterance is the one from which the je derives its mean-
ing. In the French tradition, the referential nature of frst and second person personal
pronouns has been studied as marking the subjectivity of speech. Tis refers to the way
in which the speaker stages her/himself as a subject (Benveniste 1966: 259). According
to Danon-Boileau (1994:162), the personal pronoun of the frst person functions in
propositions as an indicator of subjectivity or point of view, especially when associated
with the verb. Te subjectivity or objectivity of the discourse has also been described
in terms of a dichotomy: speakers may stage themselves or make themselves mani-
fest in utterances, or on the other hand may decide to distance themselves from it,
leaving no explicit signs of their presence or manifesting their attitude in utterances.
Refections on deictics also consider time and place.
3.4 Reported speech
Te linguistic phenomena of voices and of reported speech has a long history in
enunciation theory from Bally to Benveniste, from the third generation to the pres-
ent day (see for instance Rosier 1999). It is an extremely heterogeneous feld, where
the terminology concerning the phenomena varies considerably, refecting the difer-
ences in explanation among the analytical approaches. Te ways of conceptualizing
this linguistic phenomenon can be divided into the traditional approaches that take
into consideration the internal characteristics of reported speech, and those that dis-
cuss the external qualities (Johansson 2000: 67). Te former approach stems from the
lexico-grammatical tradition and its analysis of the diferent forms of direct, indirect,
free (in)direct speech and the relationship of the reporting and reported parts, the use
of diferent verbal and deictic forms. In enunciation theory there are also consider-
ations about how the speaker takes distance from or adheres to the reported utterance.
In the modern approaches there are several new forms that are linked to the analysis
of this phenomenon, such as the verb form in the conditional as well as prepositional
groups such as according to X (selon X) and their capacity to mediate information from
other contexts (Rosier 1999). Te latter approach based on the logical and philosophi-
cal tradition of oratio obliqua and oratio recta discusses how reported speech is refex-
ive, how it keeps fdelity when reproducing the speech (Johansson 2000: 67). Several
theoreticians discussed above give their own defnition and description of reported
speech. However, in the modern accounts there are several approaches in which the
nature of reproduction is problematized and in which diferent sequential dimensions
are taken into consideration, whether cognitive, textual, interactional or discursive,
in which the relationship to the enunciation varies (Waugh 1995, Vincent & Dubois
1997; Marnette 2005).
3.5 Modality and modalization
Te notion of modality in French linguistics is a complex one. Tis is due to its
origins (cf. infra) and to the many diferent usages it has in the French linguistic
nonciation 95
literature. As Vion (2001:213) points out, it is a concept closely related to enuncia-
tive pragmatics and hence very little used in the Anglo-American literature. We will
frst trace the origins of the concept and then look at some more contemporary
approaches to modality.
Te concept of modality has a long history dating back to Greek antiquity (cf.
Aristotles modalities of the sentence: assertive, interrogative, optative, predictive and
imperative). Modality is a recurrent topic whenever the problem of thought and lan-
guage is dealt with. Tus, for example, Arnauld and Lancelot state in their Grammaire
gnrale et raisonne (1660): Te verbs are the kinds of word that signify the way we
think and the form of our thoughts, of which the foremost is afrmation (Seconde
partie, chapitre XVI: Des divers Modes ou Manires de Verbes).
Traditionally, the concept of modality was used to refer to the construction of
a sentence around the verb but also to the production of a sentence as an utterance.
It is in this sense that Ferdinand Brunot treats issues related to modality inherent in
utterances in his book La pense et la langue even if the speaking subject does not yet
explicitly appear in his defnition:
An action uttered, be it a question, a positive or a negative utterance, presents itself
for our judgment, our sentiments, our will and this utterance has extremely varied
properties. It is presented as certain or possible, as desirable or doubtful, it expresses
an order or it dissuades etc. Tese are modalities of ideas (modalits de lide). []
Tere is modality in all kinds of elements of the sentence. (Brunot 1922: 507508)
Te defnition of the term modality which explicitly mentions the speaking subject is
attributed to Bally who defnes it as:
Modality is the soul of the sentence; just as thought, modality is mainly realized
through the action of the speaking subject. Tus one cannot attribute the value of a
sentence to an utterance unless one has discovered the expression of modality of the
utterance. (Bally 1932 28)
Bally includes among signs that mark modality a great variety of markers such as into-
nation, mimic, gesticulation, modus of verbs, adverbs and adjectives.
As for Benveniste, the range of explicit modal expressions is much more limited.
He includes only modal verbs expressing desire, necessity, volition, and above all,
obligation and permission. Te latter two he considers as modal verbs par excellence
whereas the others are only occasionally modalizing verbs according to Benveniste
(Benveniste 1974 :188). However, when speaking of the grandes fonctions syntaxiques
(cf. supra) Benveniste deals precisely with categories related to modality.
For Culioli [1968] 1999 :24), there are four types of modality, each of which
expresses a particular attitude of the subject to the predication: (1) positive, negative,
injunctive and fctive (hypothetic) ; (2) certainty, probability, necessity etc., i.e. epis-
temic modality; (3) judgemental (apprciative) (how sad, luckily..) and (4) pragmatic
or inter-subjective, the principal markers of which are deontic auxiliaries, voicing the
relation between the subjects, i.e. the interlocutors.
Modality can be seen as a subcategory of a wider process, that of modalization
which according to Coquet (1976 :64) means the attribution of modalities to an
utterance, by which an enunciator expresses an attitude towards his interlocutor and
towards the contents of his utterance (cf. also Culioli [1968]1999 :24 : to modalize is
to assign modalities to an utterance []). But as Vion (2001: 219220) points out,
defnitions of modality do not radically difer from those of modalization and the con-
cept of modalization is used fairly seldom in French linguistics. Vion himself defnes
modalities as the linguistic markers signalling the subjective attitude of the locutor in
discourse whereas modalization, according to Vion, refers to the process of expressing
a subjective attitude (ibid:214).
Since the seventies, the concept of modality in the French linguistic literature has
been divided into modalities of enunciation (modalits dnonciation) and utterance
modalities (modalits dnonc). Tis division is attributed to Meunier (1974). A more
recent defnition is Nlkes (1993), according to which the modalities of enunciation
express the way in which the speaker looks at his own enunciative activity (Ce sont les
regards que le locuteur jette sur son activit nonciative). With the help of the modali-
ties of enunciation the speaker can in fact make commentaries which directly concern
illocutionary acts or the act of enunciation in which he is engaged (Nlke 1993 :85). As
for utterance modalities they express the way the speaker looks at the contents of his
utterance []. He can in fact evaluate in diferent ways the argumentative and truth
values of his utterance (ibid :143).
3.6 Modalities of enunciation (modalits dnonciation)
Modalities of enunciation in their traditional sense correspond to the Aristotelean
division of modalities they deal with the relation the speaker wants to establish
between himself and his interlocutor/s and the way in which he wants to act upon
them. Meuniers classifcation includes the assertive, interrogative, injunctive and
exclamative modalities. Following Nlke (1993) one should also include what he
calls modalisateurs dnonciation, a heterogenous class of illocutionary markers which
give instructions as to how to interpret an utterance, such as illocutionary adverbs
(fnally a metatextual marker; if I may say, unless I am mistaken commenting on
the enunciative activity; or no joking (sans rire) excluding the possibility of humorous
or ironic interpretations (Nlke 1993 :114115).
3.7 Utterance modalities (modalits dnonc)
Utterance modalities express the attitude of the speaking subject (modus) towards
the contents of his utterance (dictum). Bally adopted this distinction between modus
nonciation 97
and dictum from the medieval philosophical tradition (cf. supra). Depending on the
point of view adopted, utterance modalities can include logical modalities, apprecia-
tive modalities (Kerbrat-Orecchioni 1980) and subjective vs. objective modalities
(Le Querler 1996). Te typologies resemble each other to a certain extent and are
partly overlapping. Still other categories have been suggested (cf. Vion 2001), but in
the following we shall discuss very briefy logical modalities and in more detail the
appreciative and subjective vs. objective modalities.
Te classical logical modalities can be divided into (1) alehtic modalities which
deal with necessity, possibility and contingency, the latter designating a state of afairs
that is not necessarily true but not necessarily false either, a limited case of possibility;
(2) deontic modalities which are related to either moral or social norms and express
permission and obligation ; and (3) epistemic modalities related to knowledge and
beliefs, dealing with the certainty of propositions (Cervoni 1987: 7477).
Te appreciative modalities (Kerbrat-Orecchioni 1980) express the subjective
evaluations and judgements of the speaker. According to Cervoni (1987:177178),
there is no reason why evaluative modalities, of which the appreciative modalities are
a subcategory, should not be included in the category of modal expressions since they
too refer to a certain norm, social or individual. Appreciative modalities are expressed
via axiological expressions which, according to Kerbrat-Orecchioni (1980 : 82), are
in an implicit relation with the speaker. Tis means that unlike other subjective ele-
ments, such as modal verbs and deictic expressions, axiological expressions allow the
speaker to take a stance without explicitly revealing himself as the source of the judge-
ment expressed (ibid). Appreciative modalities can be expressed via nouns, adjectives,
adverbs and verbs.
Nouns expressing the appreciation of the speaker can be pejoratively or melio-
ratively marked (Pierre is an idiot, He is a genius), be negatively marked by virtue of
sufxes (-asse, -ard in French- femmard, blondasse) or be nouns derived from inher-
ently subjective verbs and adjectives (corruption, beauty). Furthermore, nouns may be
marked because of the register they belong to (fic, cop vs. police).
As for adjectives, they are frst divided into objective and subjective adjectives,
secondly the subjective ones into afective (splendid, funny) and evaluative adjectives,
which in turn can be further subdivided into non-axiologic adjectives, which express
neither evaluation nor personal engagement of the speaker (expensive, big, hot) and
axiologic adjectives (beautiful, good) which express either positive or negative evalua-
tions of the speaker (Kerbrat-Orecchioni 1980 :8586).
Verbs expressing the appreciation of the speaker must, according to Kerbrat-
Orecchioni (1980 : 101), be considered from three diferent points of view :
1. Who expresses the evaluation: the speaker (Helene is screaming, Pierre pretends
that Paul never came) or an actor of the process (Paul wishes that).
2. What is evaluated: the process (to brawl) or the object of the process (Pierre hates
Paul)
3. Te type of evaluation: axiologic evaluation as in good/bad or modal evaluation
such as true/false (uncertain).
Furthermore, some verbs are occasionally subjective, when used in the frst person
singular (I hope, I regret), whereas others are inherently subjective, expressing an eval-
uation made by the speaker (He stinks, he admits ).
Le Querlers typology (1996) consists of a continuum ranging from subjective via
inter-subjective to objective modalities. Her subjective modalities express the speakers
attitude to the propositional content of his utterance. Tey include epistemic modali-
ties (cf. logical modalities supra) and appreciative modalities. In the case of epistemic
modalities, the locutor expresses the degree of certitude towards what he is saying
(I do not know if that is true, Tat is certainly, probably, maybe true, He might be
in his sixties ) In case of appreciative modalities, the locutor expresses his evalu-
ation or judgement (Unfortunately, it is not at all clear, I do appreciate your pres-
ence here today ) When resorting to inter-subjective modalities the speaker indicates
directly or indirectly, the illocutory force of his utterance. Te inter-subjective modali-
ties correspond to deontic modalities and have to do with order, permission, request,
advice..). (I would like you to close the window, you can close the window, close the
window). Objective modalities (called also implicative modalities) do not, according
to Le Querler, depend on the volition or the judgement of the speaker. Te speaker
subordinates the propositional content of the utterance to another proposition (Le
Querler 1996: 64). Te speaking subject only asserts that p implies q (In order to grow,
one must eat, Te door has either to be closed or to be lef open).
Especially the subcategory of objective modalities has met criticism. It is difcult
to accept the idea that the relationship between propositions would be independent
of the speaking subject (Vion 2001: 218). Vion also points out that the proposi-
tional content of an utterance does not refect any objective reality or state of afairs
since the dictum is always a result of subjective choices made by the speaker (Vion
2001: 220). Consequently, modus as well as dictum should be taken into account
when analyzing modality. Tis leads Vion to propose the term double enunciation
(double nonciation) which implies that modalization should be seen as a phenom-
enon taking place at two diferent levels, that of content and that of modal attitude,
or as Vion expresses it: the speaker stages his utterance from two enunciatively dif-
ferent positions (ibid).
Note: Eija Suomela-Salmi is responsible for the Sections 2.2.1, 2.2.2., 2.3.1, 2.3.4 and 3.5. Marjut
Johansson is responsible for the rest.
nonciation 99
References
Adam, J.-M., G. Lugrin & F. Revaz (1998). Pour en fnir avec le couple rcit/discours. Pratiques 100:
8198.
Anscombre, J.-C. & O. Ducrot (1983). Largumentation dans la langue. Pierre Mardaga.
Arnauld, A. & C. Lancelot [1660] (1998). Grammaire gnrale et raisonne. Allia.
Authier-Revuz, J. 1982. Htrognit montre et htrognit constitutive: lments pour une
approche de lautre dans le discours. DRLAV, Revue de linguistique 26: 91151.
(1984). Htrognits nonciatives. Langages 73: 98111.
(1992). Repres dans le champs du discours rapport. Information grammaticale 55: 3842.
(1995a). Ces mots qui ne vont pas de soi. Boucles rfexives et non-concidences du dire. Tomes
III. Larousse.
(1995b). Mta-nonciation et (d)fgement. Cahiers du franais contemporai 2: 1739.
(1996). Remarques sur la catgorie de llot textuel. Cahiers du franais contemporain 3: 91115.
Bakhtine, M. [1929] (1977). Le marxisme et la philosophie du langage. Essai dapplication de la mthode
sociologique en linguistique. Traduit par M. Yaguello. Publi sous le nom de V. Voloshinov.
Editions du Minuit.
Bally, C. (1909). Trait de stylistique franaise. Vol 1. et 2. Carl Winters Universittsbuchhandlung.
[1913] (1952). Le langage et la vie. Attar.
[1932] (1944). Linguistique gnrale et linguistique franaise. Francke.
Banks, D. (2004). Anglophone systemicists and French enunciativists: shall the twain ever meet?
Language sciences 26: 391410.
Benveniste, E. (1966). Problmes de linguistique gnrale I. Gallimard.
(1974). Problmes de linguistique gnrale II. Gallimard.
Bral, M. (1991). Te beginnings of Semantics. Edited and translated by G. Wolf. Duckworth.
Brunet, E. (2004). Continuit ou discontinuit en histoire des sciences du langage? La place du sujet
parlant depuis Saussure. Actes des VIIe RJ EDC268 Langage et Langues: 2529.
Brunot, F. (1922). La pense et le langage. Methode, principe et plan dune thorie nouvelle du langage
applique au franais. Masson et Cie.
Cervoni, J. (1987). L nonciation. Presses universitaires de France.
Charaudeau, P. (2002). Situation de communication. In P. Charaudeau et D. Maingueneau (eds.),
Dictionnaire danalyse du discours: 532536.
Chiss, J.-L. (1985). La stylistique de Charles Bally. De la notion de sujet parlant la thorie de
lnonciation. Langages 77: 8594.
(1986). Charles Bally: Quest-ce quune Torie de lnonciation? In: Histoire, Epistmologie,
Langue. Tome 8 fascicule II: 165176.
Chiss, J.-L. & C. Puech (1997). Fondations de la linguistique. Etudes dhistoire et pistmologie. 2
me

dition. Duculot.
Coquet, J.-C. (1976). Les modalits du discours. Langages 43: 6470.
Culioli, A. (1990). Pour une linguistique de lnonciation. Formalisation et oprations de reprage.
Tome 1. Orphys.
(1999). Pour une linguistique de lnonciation. Formalisation et oprations de reprage. Tome
2. Orphys.
(2002). Variations sur la linguistique. Entretiens avec F. Fau et notes de M. Viel. Klincksieck.
Danon-Boileau, L. (1994). La personne comme indice de modalit. Faits de langues 3: 159168.
Desmet, P. & P. Swiggers (eds.) (1995). De la grammaire compare la smantique. Textes de Michel
Bral publis entre 19641898. Introduction, commentaires et bibliographie par P. Desmet et P.
Swiggers. Peeters.
Dubois, J. (1969). Enonc et nonciation. Langages 13: 100110.
Ducard, D. (2006). Seuils, passages, sauts. In D. Ducard & C. Normand (dir), Antoine Culioli. Un
homme dans la langue. Colloque de Cerisy: 1318. Orphys.
Ducrot, O. (1968). Le structuralisme en linguistique. Seuil.
(1972). Dire et ne pas dire. Principes de smantique linguistique. Hermann.
(1980a). Les Mots du discours. Minuit.
(1980b). Les chelles argumentatives. Editions de Minuit.
(1984). Le dire et le dit. Editions de Minuit.
[1986] (1991). Charles Bally and pragmatics. Translated by C. Porter, K. Rabbit & L. Waugh.
Diacritics 21:4 319.
(2004). Argumentation rhtorique et argumentation linguistique. In M. Doury & S. Moirand
(eds.), Largumentation aujourdhui. Positions thoriques en confrontation. Presses Sorbonne
Nouvelle.
(s.a.) Enonciation. Encyclopdie Universalis. www.universalis.fr.
Ducrot, O. & J.-M. Schaefer (eds.) (1995). Nouveau dictionnaire encyclopdique des sciences du lan-
gage. Seuil.
Durrer, S. (1998). Introduction la linguistique de Charles Bally. Delacaux et Niestl.
Fontaine, J. (1986). Lnonciation de Benveniste Weinrich. Histoire, Epistmologie, Langage. Tome
8 fascicule II: 207220.
Foucault, M. (1969). Larchologie du savoir. Gallimard.
Fuchs, C. (1981). Les problmatiques nonciatives: Esquisses dune prsentation historique et cri-
tique DRLAV, Revue linguistique (Dans le champ pragmatico-nonciatif) 25: 3560.
Groupe Relpred (Bernard, G. & al.) (1989). Enonciation. In Encyclopdie Philosophique Universelle.
Vol. Notions, tome 1. Presses universitaires de France.
Groussier, M.-L. (2000). On Antoine Culiolis theory of enunciative operations. Lingua 110(3):
157187.
Gouvard, J.-M. (2005). Linguistique, stylistique et style chez Charles Bally. In J.-M. Gouvard (ed.), De
la langue au style: 520. Presses universitaires de Lille.
Guillaume, G. (1912). Etudes de grammaire franaise logique compare. Le lieu du mode dans le temps,
dans lespace. Fascicule 1: Larticle. Fichbacher.
(1919). Le problme de larticle et sa solution dans la langue franaise. Hachette.
Jakobson, R. [1957] (1963). Shifers, Verbal Categories, and the Russian Verb. In Essais de linguistique
gnrale. Les Editions de Minuit.
Johansson, M. (2000). Recontextualisation du discours dautrui. Discours reprsent dans linterview
politique mdiatique. University of Turku.
Joly, A. (1987). Essais de systmatique nonciative. Presses universitaires de Lille.
Joly, A. & D. Roulland (1981). Pour une approche psychomcanique de lnonciation. In A. Joly &
W.H. Hirtle (eds.), Langage et psychomcanique du langage. Etudes ddies Roch Valin:
537581. Presses de luniversit de Laval.
Le Querler, N. (1996). Typologie des modalits. Presses Universitaires de Caen.
Kerbrat-Orecchioni, C. (1980). L nonciation. De la subjectivit dans le langage. Armand Colin.
(2006). Lnonciation. De la subjectivit dans le langage. Forth edition. Armand Colin.
Kleiber, G. (1986). Dictiques, embrayeurs, token-refexives, symboles indexicaux etc.: comment les
dfnir? Linformation grammaticale 30: 4244.
nonciation 101
Lala, M.-C. (2006). Points de vue croiss la source dune thorie de lnonciation. Chiss, Jean-Louis
(ed.) Charles Bally (19651947) Historicit des dbats linguisitiques et didactiques. Stylistique,
nonciation, crise du franais: 94120. Peeters Publishing.
Liddle, M. (ed) (1995). Antoine Culioli. Cognition and representation in liguistic theory. John Benjamins.
Maingeneau, D. (1994). Lnonciation en linguistique franaise. Hachette.
(1997). Lanalyse du discours. Hachette.
Marnette, S. (2005). Speech and Tought Presentation in French. John Benjamins.
Meunier, A. (1974). Modalits et communication. Langue franaise 21: 825.
Moeschler J. (2006). Te French tradition in pragmatics: From structuralism to cognitivism. Intercul-
tural pragmatics 3(4): 381307.
Morel, M.-A. & L. Danon-Boileau (dir.) (1992). La deixis. Presses universitaires de France.
Nerlich, B. & D.D. Clark (1996). Language, action and context. John Benjamins.
Normand, C. (1985a). Le sujet dans la langue. Langages 77: 719.
(1985b). Le sujet est de retour. Langages 77: 5.
(1986). Les termes de lnonciation de Benveniste. In Histoire, Epistmologie, Langage. Tome
8 fascicule II: 191206.
(2005). Culioli, Antoine. Encyclopedia of Language and Linguistics: 307. Elsevier.
Nlke, H. (1993). Le regard sur le locuteur. Pour une linguistique des traces nonciatives. Volume 1.
Kim.
Rosier, L. (1999). Le discours rapport. Histoire, thories, pratiques. Duculot.
Valette, M. (2003). Enonciation et cognition: deux termes in absentia pour des notions omniprsents
dans luvre de Guillaume. Le franais moderne Tome LXXXI no 1: 625.
(2004). Actualisation et nonciation. Retour sur une gmllit problmatique. In G. Hassler &
G. Volkmann (eds.), History of linguistics in texts and concepts: 813821. Nodus Publikationen.
(2006). Linguistiques nonciatives et cognitives franaises. Gustave Guillaume, Bernard Pottier,
Maurice Toussaint, Antoine Culioli. Honor Champion.
Waugh, L. (1995). Reported speech in journalistic discourse: the relation of function and text. Text
15(1): 129173.
van Hecke, T. (2002). Emil Benveniste. Handbook of Pragmatics. www.benjamins.com /cgi-bin/
bbr_hop.cgi.
Vincent, D. & S. Dubois (1997). Le discours rapport au quotidien. Nuit blanche.
Vion, R. (ed). (1998). Les sujets et leurs discours. Enonciation et interaction. Publications de lUniversit
de Provence.
(2001). Modalits, modalisations et actes langagires. Marges linguistiques 2: 209231. www.
revue-texto.net.
Figures of Speech
Manfred Kienpointner
University of Innsbruck
1. Introduction
Figures of Speech (= FSP, also called rhetorical fgures or fgures of rhetoric; the
term FSP will be used in its broadest sense, including both tropes and FSP in the nar-
row sense, that is, fgures of diction and fgures of thought) can be briefy defned as the
output of discourse strategies for creating communicatively adequate texts (cf. below
Section 3.1. for a more detailed defnition). From antiquity onwards, they have been
studied as one of the main branches of rhetoric. In ancient rhetoric, FSP were mainly
characterized as a kind of ornament. Moreover, a detailed typology of FSP was devel-
oped. In recent times, various theories of style have been developed which present quite
diferent defnitions and characterizations of FSP. Several authors have tried to establish
standards of classifcation and demarcation of FSP which are more satisfactory than
those of ancient rhetoric. Tere also have been attempts to extend the concept of FSP to
nonverbal domains (e.g. FSP in the visual arts: painting, cinema etc.). Not very surpris-
ingly, metaphor has attracted far more interest than any other FSP: therefore, studies on
metaphor will be mentioned more frequently than work on other FSP.
In the following sections, I will frst present an outline of the description of FSP in
ancient rhetoric (Section2). Ten I wish to deal with contemporary attempts to defne
(Section3.1.) and classify (Section3.2.) FSP in disciplines like rhetoric, speech com-
munication studies, linguistics, literary criticism, philosophy and psychology. Finally,
the important cognitive and communicative roles of FSP and their treatment in difer-
ent disciplines will be discussed (Section4).
2. Ancient rhetoric
In ancient rhetoric, FSP were characterized as a kind of ornament added to plain speech
to improve its persuasive impact. Tus FSP are conceived of as a kind of clothing, an
ornament which makes ordinary speech, which is merely clear and plausible, more
attractive and efcient (cf. Quintilian Inst.Orat. 8.3.61). Further metaphorical charac-
terizations of FSP in ancient rhetoric are fores (fowers), lumina (highlights) and
colores (colours) (cf. Cicero or. 3.19, 3.96, 3.201; Knape 1996: 303). More specif-
cally, tropes (cf. Greek tropos turn, from the verb tropein to turn) are characterized
Figures of Speech 103
by Quintilian as a kind of semantic change or deviation, where the proper meaning of
a word or phrase is replaced by a new meaning which enhances the quality of speech
(cf. Quintilian Inst.Orat. 8.6.1). FSP in the narrow sense (cf. Latin fgura < Greek
schema form, shape) are defned as some artistic innovation of speech (cf. Quintilian
Inst.Orat. 9.1.14). Te Greek origin of most rhetorical terms and the following trans-
lation into Latin (and, later, into English, French and other languages) is not without
problems concerning the accuracy and adequacy of the respective translations and the
resulting terminology.
From antiquity onwards, FSP have been classifed according to the trichotomy of
(1) tropes (e.g. metaphor, metonymy, hyperbole, irony), (2) fgures of diction (e.g. ana-
phor, parallelism, climax, ellipsis) and (3) fgures of thought (e.g. rhetorical question,
exclamation, persuasive defnition, personifcation) (Quintilian Inst.Orat. 9.1.47,
9.1.17, Cic. or. 3.52.101).
Already Quintilian, however, conceded that there is no consensus as to the num-
ber of tropes and the delimitation of certain other FSP (Inst.Orat. 8.6.1). Moreover,
the traditional typology ofers no clear demarcation of linguistic levels, apart from
the problematic dichotomic distinction of fgures of diction and fgures of thought
(cf. Knape 1996: 310f.). According to this dichotomy, fgures of diction are funda-
mentally changed if the specifc wording of a fgure is altered. Figures of thought are
said to remain the same even if the formulation is (slightly?) modifed (cf. Cicero or.
3.200). Furthermore, Quintilians (Inst. Orat. 1.5.38) promising attempts to classify all
FSP according to four basic linguistic operations which are used to realize FSP were
not carried out consistently. Tese four linguistic operations are: addition, subtraction,
substitution, permutation.
As far as particular FSP are concerned, already Aristotle treated metaphor in some
detail, defning it (Poetics 1457b) as a transfer of a noun from its proper species or
genus to another one, or as an analogical transfer. Te art of creating metaphors is
greatly appreciated by Aristotle, who also recognized the cognitive dimensions of met-
aphors when he writes that they lead to a kind of intellectual pleasure in the audience:
metaphors ofer some cognitive surprise, but this surprise is soon followed by a growth
of knowledge which is easily achieved (cf. also below Section4).
Te classical treatment of FSP appears, for example, in the catalogue of 65 FSP in
the Rhetorica ad Herennium (1st century B.C.). It served as a model for many treatises
on FSP and remained more or less unchanged in medieval and early modern times,
although occasionally additions, reductions and slight modifcations were made. In
the Middle Ages (cf. Murphy 1974) FSP were a part of treatises on the art of making
poems (artes poetriae) and grammars (e.g. Alexander de Villa Deis highly infuen-
tial Doctrinale, 12th century). Petrus Ramus (16th century) was highly infuential
for the further development of rhetoric when he suggested that the art of rhetoric
should be reduced to a theory of style and presentation. In the year 1572 his friend
104 Manfred Kienpointner
Audomarus Talaeus was the frst in a series of rhetoricians who published rhetorical
treatises dealing mainly or exclusively with style and FSP. In this respect, Talaeus was
followed by French authors like B. Lamy (17th century), C. Dumarsais (18th century)
and P. Fontanier (19th century).
Te classical typology of FSP survives even into our times: despite some modifca-
tions, the classifcation of FSP in the neo-classical handbook by Lausberg (1990) still
resembles the ancient doctrine and the treatment in Ueding & Steinbrink (1986: 264f.)
is even closer to the ancient typology (on ancient and modern theories of FSP cf. also
Bradford 1982; Bacry 1992: 9f.).
However, the strong infuence of the ancient tradition did not prevent a num-
ber of attempts to replace the classical trichotomy of tropes, fgures of diction and
fgures of thought by diferent classifcations. For example, Fontanier (1968: 66f.),
called the Linn of Rhetoric, suggested a division of all FSP into seven classes
(1. fgures of signifcation (= the tropes), 2. fgures of expression, 3. fgures of diction,
4. fgures of construction, 5. fgures of elocution, 6. fgures of style, 7. fgures of thought).
Nash (1989: 112) distinguishes two classes of FSP: (1) schemes (fgures involving word
order and syntactic patterning) and (2) tropes (subclassifed into a. fgures involving
word-semantics and b. discourse semantics). Corbett & Connors (1999: 379) similarly
distinguish schemes (a deviation from the ordinary pattern or arrangement of words)
and tropes (a deviation from the ordinary and principal signifcation of a word).
3. Contemporary treatments of FSP
3.1 Defnition of FSP
Te ancient perspective concerning FSP has been taken up and refned by some mod-
ern linguistic theories of style and poetic language (cf. Freeman 1970 and Enkvist
1973: 67f. for an overview of competing concepts).
Within some of these frameworks, FSP are conceived as deviations from everyday
language. In his classical article on poetics and linguistics, Jakobson (1960: 356) has
tried to formulate a general semiotic framework for a deviational perspective where
Te set (Einstellung) toward the MESSAGE as such, the focus on the message for its
own sake, is the POETIC function of the language. Tis specifc focus on the message
for its own sake is realized by the principle of equivalence (ibid. 358): Te poetic
function projects the principle of equivalence from the axis of selection into the axis
of combination. Semantically equivalent or similar linguistic units are selected from
a paradigm in a way that phonetically equivalent or similar units are established on
the syntagmatic level. For example, the adjective horrible is selected from a lexical
paradigm also containing the synonyms disgusting, frightful, terrible etc. to produce
the alliterating noun phrase horrible Harry.
Following the basic semiotic principles outlined by Jakobson, linguists and literary
critics have developed detailed and sophisticated deviational approaches, for example
Leech (1966), Todorov (1967), the Belgian groupe m (Dubois et al. 1974) and Plett
(1975, 2000). I will return to these approaches when dealing with the classifcation of
FSP (cf. Section3.2.).
Despite its intuitive appeal, deviation theory has had to face severe criticism (e.g.
by Coseriu 1994: 159f.; Spillner 1974: 39f.; Ricur 1975: 173f.; Knape 1996: 295f.).
Tis criticism is basically justifed. It is true that modern deviation theory has devel-
oped much more sophisticated standards of explicitness and consistency than ancient
rhetoric. But still, the following weak points remain:
1. It is hardly possible to isolate a zero-variety of language which could serve as
the basis from which fgurative language is derived via the principle of equiva-
lence and more specifc linguistic operations (cf. Ricur 1975: 178: le langage
neutre nexiste pas, that is, neutral language doesnt exist). Tat a zero-vari-
ety is difcult to establish is conceded even by deviation theorists (cf. Todorov
1967: 97f.; Dubois et al. 1974: 59). FSP occur and are sometimes even extremely
frequent in many instances of everday language: in conversations, political
speeches, advertising texts, fairy tales, slogans, idioms and proverbs (cf. Kloepfer
1975). Moreover, empirical studies provide frequency counts of metaphors
within diferent types of text. Tese studies show that speakers use 1.80
creative and 4.08 dead metaphors per minute of discourse (cf. Gibbs 1994: 123f.;
Aitchison 1994: 149). Finally, recent publications on metaphor stress the
crucial role of metaphor even in languages for specifc purposes, for example,
in scientifc and philosophical language (cf. McCloskey 1983; Kittay 1987: 9;
Pielenz 1993: 76f., Gibbs 1994: 169f.; Debatin et al. 1997).
2. Quite ofen FSP are not even replaceable by a proper expression because such
a proper expression simply does not exist and we have to rely on fgurative lan-
guage (cf. Weydt 1987; Coseriu 1994: 163). Tis was already acknowledged by
ancient deviation theorists like Quintilian (cf. inst.orat. 8.6.6.; cf. also Aristotle
poet. 1457b 2526). Tis is not to deny the fact that in many cases we can
distinguish proper and fgurative language. As far as metaphor is concerned,
even dead or frozen metaphors, which have almost become literal expres-
sions, can be recognized as such at a given synchronic moment in a given
language community (cf. Kittay 1987: 22). Moreover, the psycholinguistic evi-
dence referred to by Gibbs (1994: 80f., 399f.; cf. also Aitchison 1994: 150) pro-
vides reasons to assume that conventionalized FSP are processed as naturally
and quickly as proper expressions. Tus, this evidence challenges the view
that fgurative speech requires special or additional mental processes for it to
be understood.
3. Deviation theory could imply the assimilation of FSP with mistakes, which are
indeed deviations. However, unlike mistakes, FSP are the result of intentional
operations. It would be most implausible to assume that we frst choose proper
but less adequate expressions from a zero-variety, then recognize that they are not
adequate and then substitute them with fgurative expressions. It is more plausible
to assume that we directly select the most suitable verbal tool from the available
paradigms in our language. Terefore, deviation theory should be replaced by
a selection theory of style (cf. e.g. Marouzeau 1935: Xf., Spillner 1974: 64, van
Dijk 1980: 97) or, on a more general level, by a pragmatic theory which describes
language use as a process of selective adaptation to context (cf. Verschueren 1998).
4. FSP are phenomena occurring at the textual level of language. Traditionally
prevailing views of sentence grammarians (which reappear in modern rhetorical
studies like, for example, those of Dubois et al. 1974: 260f.) claimed that the
text level does not belong to the language system proper. However, textlinguists
have amply demonstrated in the past few decades (cf. e.g. van Dijk 1980;
Coseriu 1994) that the text level, too, is at least partially organized according
to language-specifc rules. FSP are realized by verbal strategies which form an
essential part of our textual and/or communicative competence. Terefore, they
are not merely secondary phenomena of parole or linguistic performance, but par-
take in a defnition of language as a creative, communicative activity (cf. Coseriu
1956; Ricur 1975: 87f.).
From these arguments the following conclusion can be derived: FSP are not merely
ornamental or aesthetic devices. Many FSP are linguistically and cognitively basic.
Terefore, they inevitably shape our cognition and culture-specifc views of reality.
Tis has been especially stressed in recent studies on metaphor: Our ordinary concep-
tual system, in terms of which we both think and act, is fundamentally metaphorical in
nature (Lakof & Johnson 1980: 3; cf. similarly Ricur 1975: 25; Kvecses 2002: viii).
In this respect, also conventionalized tokens of FSP are particularly important because
unlike newly created instances of FSP they have ceased to attract our attention and
have thus become unnoticed cognitive background phenomena.
Against this conclusion, deviation theorists could still argue that a restricted
version of deviation theory is correct under the following conditions: (1) Its claims
are restricted to poetic literature in the narrow sense of the word; (2) the diference
between everyday language and poetic language is no longer seen as a qualitative devi-
ation but as a quantitative deviation: truly poetic texts like poems usually contain more
FSP than, say, editorials or parliamentary speeches; (3) it is restricted to creative FSP,
which have not (yet) been conventionalized (cf. Fontanier 1968: 64).
While these arguments are relevant and important, they can still be refuted.
It is true that there is a clear statistical diference between the frequency of FSP in
poetic texts and everyday language. However, this diference does not justify consid-
ering poetic language as a deviation from everyday language. As FSP partake in the
defnition of language as a creative activity and FSP are present in all kinds of language
varieties, the diference of frequency simply does not justify seeing poetic language as
a deviation from other varieties of language. Rather, everyday language or scientifc
language could be seen as deviations from or reductions of poetic language. Only
the latter fully exhausts the creative potential of language by using all available verbal
strategies (cf. Coseriu 1994: 159f.).
In this respect, conventionalized FSP are not substantially diferent from creative
FSP (cf. Kvecses 2002: 44f.). In both cases, fgurative patterns are used to produce
texts for some communicative purpose, only that in the case of creative FSP, new ways
of formulating meaningful texts are recognized and implemented. Conventionalized
FSP are verbal strategies used to repeat these original creations, to re-create tokens of
utterances according to stylistic patterns which are already accepted and widely used
in a speech community. But this is only a diference of degree, because new realizations
of FSP can gradually spread to the whole speech community and sooner or later also
become conventionalized. It is in this context that Spitzer (1961: 517f.) rightly charac-
terized syntax and grammar as frozen stylistics.
In the light of the preceding discussion, the weaknesses of ancient and recent
attempts to defne FSP as deviations from a neutral variety of language (cf. Quintilian
Inst.Orat. 9.1.11: fgura = any rational deviation, either in thought or expression,
from the ordinary and simple method of speaking, a change analogous to the diferent
positions our bodies assume when we sit down, lie down, or look back or Corbett &
Connors (1999: 379): We will use fgures of speech as the generic term for any artful
deviations from the ordinary mode of speaking or writing) should be avoided by the
following defnition of FSP (cf. also Ricur 1975: 10):
FSP are the output of discourse strategies which we use to select units from
linguistic paradigms of diferent levels (phonetics/phonology, morphology,
syntax, semantics) to create texts (in the sense of both written and spoken genres
of discourse) which are adequate as far as their communicative purpose in some
context is concerned.
From the perspective of the audience, the same process can be conceived of as an
infnite sequence of the interpretations of texts. Tis time, the FSP in a text are used
as interpretive clues or hints for the attentive hearer/reader. Again, most of the time
we repeat or re-create standard interpretations of these texts. But the list of stan-
dard interpretations can be creatively extended by detecting new FSP in the text or by
detecting non-traditional interpretations of well-known FSP. In this way, non-standard
interpretations can be found. Te use of FSP can thus be defned as an open-ended,
creative communicative activity by both speaker and hearer.
3.2 Classifcation of FSP
Typologies of FSP in ancient rhetoric sufer from a variety of weaknesses. Terefore,
it is not surprising that recent typologies (cf. Leech 1966; Todorov 1967; Dubois et al.
1974; Plett 1975, 1985; Bacry 1992) try to overcome the defciencies of the traditional
typology. Most of these typologies use two basic principles for the classifcation of
FSP: (1) a distinction of linguistic levels at which the FSP are situated; (2) a distinc-
tion of several basic operations which realize particular FSP. Tese principles of
classifcation are highly valuable even if one does not want to accept the deviational
framework underlying such typologies.
To illustrate these typologies, Plett (1975, 2000) will serve as an example. Plett
starts from a deviational perspective, but refnes it considerably, distinguishing
between rule-violating deviation and rule-strengthening deviation (cf. also Todorov
1967: 108). In this way he avoids part of the criticism against deviation theory men-
tioned above (cf. Section 3.1.): at least some FSP are rightly classifed as operations
which enforce rules of everyday language rather than violating them. Rule-violating
deviation is based on the four elementary operations of addition, subtraction, permu-
tation and substitution, which are executed at all linguistic levels, from the phonemic/
morphemic level to text level.
Some examples of FSP derived by these four operations are: prosthesis, paren-
thesis, tautology (addition); syncope, ellipsis, oxymoron (subtraction); metathesis,
inversion, hysteronproteron (permutation); fnally, substitution of sounds or syllables,
exchange of word classes, metaphor, metonymy, irony (substitution).
Rule-strengthening deviation is based on various kinds of repetition, which con-
cern either the position of linguistic elements or their size or their degree of similarity,
frequency and distribution. Examples of FSP resulting from rule-strengthening devia-
tion, again taken from various levels, are: alliteration, assonance, anaphor, parallelism,
synonymy, simile, allegory (position); rhyme (size); partial equivalence of vowels/con-
sonants, puns, parison (similarity); the same FSP can be studied as to their frequency
and distribution in a text.
From the perspective of a selection theory of style, one could see the four opera-
tions as an implementation of selection strategies, as a means to transform linguistic
units of a paradigm into other units of the same paradigm (at the same level) and vice
versa. Terefore, unlike deviation theorists, one need not assume that one of these lin-
guistic units (phonemes, morphemes, phrases, clauses, word meanings and sentence
meanings) is part of a zero-variety of language and all the others are derived from it.
Rather than assuming a unidirectional transformation of linguistic elements, the four
operations could be better conceived of as multilateral transformations which operate
in all directions.
Plett classifes all FSP according to the respective operations and levels of analysis.
Te following display shows a summary of Pletts typology (1975: 148; 2000: 22, 50f.).
Note that rule-strengthening deviation is a further operation (equivalence) which
has to be added to the 4 basic rule-violating deviations:
Linguistic levels Operations
Phonology I. Rule-violating deviation:
Morphology 1. Addition: a+b+c a+b+c+d
Syntax 2. Subtraction: a+b+c a+b
Semantics 3. Permutation: a+b+c a+c+b
Graphemics 4. Substitution: a+b+c a+z+c
II. Rule-strengthening deviation:
5. Equivalence: a+b+c a+b+c+a
Figure1.
Plett illustrates his typology with examples which are mainly taken from Ger-
man and English poetic texts (cf. Plett 1975: 252f.). Below, some of these examples
are quoted:
I. Rule-violating Deviation at the semantic level:
1. Addition: Tautology (with the help of synonymic expressions, the same semantic
content is predicated twice):
(1) Hamlet: Teres neer a villain dwelling in all Denmark
But hes an arrant knave (W. Shakespeare, Hamlet I.5: 123f.)
2. Subtraction: Oxymoron/Paradoxon (antonymic semantic properties are ascribed
to the same object simultaneously, within the same noun phrase or sentence):
(2) Romeo: Why, then, o brawling love! O loving hate! O heavy lightness! cold
fre, sick health! Still-waking sleep, that is not what it is!
(W. Shakespeare, Romeo and Juliet I.1: 182187)
3. Permutation: Hysteronproteron (a clash between temporal succession and linear
word order: a chronologically later event 1 is presented earlier in the text and fol-
lowed by the chronologically earlier event 2 later in the text):
(3) Ihr Mann ist tot und lt Sie gren. (W. v.Goethe, Faust I: 2916)
(Your husband is dead and sends you his regards)
4. Substitution: Metaphor (expressions containing semantic features like [+abstract]
or [+visual] interact with other expressions containing semantic features like
[+concrete] or [+acoustic] within the same phrase, clause or sentence):
[+abstract] [+concrete]:
(4) hands/ Tat lif and drop a question on your plate
(T.S. Eliot, Te Love Song of J. Alfred Prufrock: 29f.)
[+visual] [+acoustic]:
(5) Pyramus: I see a voice
(W. Shakespeare, A Midsummer Nights Dream V.1: 194)
Apart from its deviation-theoretical background, Pletts typology is to be praised for
the following reasons: (1) the principles of classifcation are much more transparent
than in traditional classifcations; (2) standards of explicitness and demarcation are
thus raised considerably; (3) the cross-classifcation based on basic operations and
linguistic levels is carried out much more consistently than in traditional typologies
following the classical tripartite classifcation.
Important points of criticism still remain. First, a high degree of intersubjec-
tive reliability has apparently been realized only at the phonological, morphological
and syntactic levels. Here, a comparison of the typologies by Plett and the Belgian
groupe m (Dubois et al. 1974: 80f.; Plett 1975: 150f.) shows that diferent typologies
classify in much the same way. But at the semantic level there are considerable discrep-
ancies. For example, Plett defnes oxymoron as a FSP derived via semantic subtraction
(cf. above), whereas Dubois et al. (1974: 200) derive it via semantic substitution:
(6) Cette obscure clart qui tombe des toiles
(Tis dark brightness which falls from the stars)
(P. Corneille, Le Cid IV.3: 1273)
Tere are more examples for discrepancies at the semantic level: Dubois et al.
consider only a small part of the classical tropes to be semantic fgures (Mta-
smmes, 1974: 152f.), while they treat most of them as extra-linguistic logical fgures
(Mtalogismes, 1974: 204f.). Plett, however, treats all tropes as fgures of semantic
deviation. Te view of Dubois et al. could be criticized as a problematic equation of
semantic phenomena of natural languages with semantic phenomena of formal logic.
Another example is provided by problematic attempts to reduce semantic FSP like
metaphor, metonymy and synecdoche to one basic type (e.g. metaphor to a double
synecdoche, cf. Dubois et al. 1974) or one dimension of language. Jakobson (1971)
tried to associate metaphor with selection (the paradigmatic dimension of language)
and metonymy with combination (the syntagmatic dimension of language). Moreover,
he extended this dichotomy to the description of types of aphasia, literary traditions
and nonverbal codes (painting and cinema). Tese attempts are highly interesting, but
have plausibly been criticized as instances of reductionism (cf. Ricur 1975: 222f.;
Eggs 1994: 188f.; Eco 1985: 169f.).
Second, a great part of the traditional fgures of thought has not been system-
atically integrated into recent typologies (cf. the relatively short remarks in Todorov
1967: 110; Dubois et al. 1974: 218f.; 260f.; Plett 1975: 302; Plett 2000: 31f., 236f.).
Take, for example, praeteritio (faked omission), where the speaker mentions a fact
which is harmful to his opponent while pretending at the same time not to refer to it.
In the following example, Cicero pretends to omit the fnancial problems of his enemy
Catilina while at the same time mentioning them:
(7) Praetermitto ruinas fortunarum tuarum.
(I omit the ruin of your fortune) (Cicero, In Catilinam 1.6.14)
Or see the following example of another important fgure of thought, namely, rhetori-
cal question (interrogatio):
(8) What use is a smoke alarm with dead batteries?
Dont forget it, check it.
(Te Observer, 3.3.1991, quoted afer Ilie 1994: 103)
Te neglect of fgures of thought in modern typologies may be due to the fact that
unlike other FSP they cannot easily be classifed with the help of formal and structural
features.
Finally, structural typologies have to be criticized because they tend to neglect the
fact that FSP should be considered as linguistic elements having certain communica-
tive functions like clarifcation, stimulation of interest, aesthetic and cognitive pleasure,
modifcation of the cognitive perspective, intensifcation or mitigation of emotions etc.
(cf. Knape 1996: 320f.) Ideally, each structurally identifed FSP should be assigned one
or more functions as a complementary part of its description (for a frst attempt cf. Plett
1985: 77, quoted afer Knape 1996: 340). Tis is also important because one and the
same FSP can have diferent functions according to the type of text in which it is used.
From a functional perspective, FSP have been studied and defned as verbal means
expressing the specifc way in which a speech act is performed (cf. Sandig 1986).
4. Across the lines of discipline: The cognitive
and communicative role of FSP
Some philosophers have postulated that rational discourse has to avoid all kinds of
FSP. Take the example of John Locke (1975: 508; in: An Essay Concerning Human
Understanding, 3.10.34):
But yet, if we would speak of Tings as they are, we must allow, that all the Art of
Rhetorick, besides Order and Clearness, all the artifcial and fgurative application
of Words Eloquence hath invented, are for nothing else but to insinuate wrong Ideas,
move the Passions, and thereby mislead the Judgment; and so indeed are perfect cheat:
And therefore however laudable or allowable Oratory may render them in Harangues
and popular Addresses, they are certainly, in all Discourses that pretend to inform and
to instruct, wholly to be avoided.
As the big problems of isolating a zero-variety of language have already made clear, it
is simply not possible to refrain from using FSP. Even Locke in his severe criticism uses
dead metaphors like to move the passions and to mislead judgment (cf. Kittay 1987: 5).
Nor would it be desirable to speak or write without using FSP. A postulate to present
the naked facts and nothing else note that the postulate, too, makes use of a meta-
phorical expression! is not only completely unrealistic, it also obscures the fact that
FSP can make a positive contribution to the rationality of argumentation by making
good arguments even stronger (cf. van Eemeren 2010).
It is also impossible to ban or stigmatize specifc types of FSP unconditionally. A
prominent example is metaphor. Aristotle (together with other distinguished authors)
is ofen quoted as an authority for the prohibition of metaphors in rational discourse
(Topics 139b, 34f.: metaphorical expressions are always obscure). However, Aristo-
tle supports a much more sympathetic view of metaphor in his Rhetoric and Poetics,
where he acknowledges its positive cognitive role (cf. Rhetoric 1410b 1015; Poetics
1459a 58; Ricur 1975: 49; Eco 1985: 151; Kittay 1987: 2f.):
To learn easily is naturally pleasant to all people, and words signify something: so
whatever words create knowledge in us are the pleasantestMetaphor most brings
about learning, for when he calls old age stubble, he creates understanding and
knowledge through the genus, since both old age and stubble are [species of the genus
of] things that have lost their bloom.
Even in the Topics, Aristotle recognizes the fact that metaphors can make a certain con-
tribution as to the better recognition of an object (Top. 140a 810). Modern linguists,
philosophers and psychologists (e.g. Black 1954; Ricur 1975; Lakof & Johnson 1980;
Perelman & Olbrechts-Tyteca 1983; Pielenz 1993; Gibbs 1994; Kvecses 2002) have
gone beyond this and stress the argumentative function of metaphors and their highly
important cognitive role: they shape our experience of reality, creative metaphors even
restructure our vision of the world; moreover, they are not isolated entities, but com-
plex cognitive networks.
Moreover, it has to be made clear that metaphorical thought, in itself, is neither
good nor bad; it is simply commonplace and inescapable (Lakof 1991: 73; cf. also
Lakof 2005). Te same metaphors can be used in a more or less rationally accept-
able way, for example, light and darkness as metaphorical symbols for more or less
respectable ideological concepts. Furthermore, diferent metaphorical domains, only
some of which are rationally acceptable, are applied to the same concept: love can be
portrayed metaphorically as war, madness, but also as a collaborative work of art
or a journey (Lakof & Johnson 1980: 139f.). Likewise, anger can be linguistically
portrayed in many diferent ways, for example, as boiling liquid in a container, fre,
explosion or madness (Lakof 1987: 397f.; Kvecses 2002: 184f.).
Te same holds for other FSP like metonymy, irony, personifcation, anaphor,
parallelism, climax, rhetorical question etc., which can be used for enhancing clear,
plausible and honest forms of communication, but also for producing confused and/
or aggressive and/or manipulative texts (of course, most uses of FSP in everyday com-
munication fall in between of these extremes).
Terefore, it seems to be impossible to stop using FSP; rather, their inevitable use
should be critically evaluated according to standards of cooperative communication
and argumentation, that is, they should not be used to make texts inadequate as far as
their communicative purpose is concerned. More particularly, they should not be used
to formulate (1) overly aggressive or (2) manipulative or (3) highly obscure texts (cf.
Kienpointner 1996: 185f.). Of course, these general standards have to be interpreted
in relation to specifc text types and speech situations. For example, the last criterion
only holds for everyday communication in standard situations, but certainly not for
certain literary genres (cf. Weinrich 1976: 295f.; Eco 1985: 176f.).
FSP also are highly important communicative tools as far as the interpersonal
dimension of language use is concerned. Tey are ofen used for expressing com-
municative intentions indirectly. Terefore FSP like tautology, metaphor, metonymy,
synecdoche, irony, meiosis, hyperbole etc. are compared with indirect speech acts in
speech act theory (cf. Searle 1979); or FSP are discussed as means for fouting max-
ims of conversation and as triggers for various kinds of conversational implicatures
(cf. Grice 1975). Tey are amply treated also in studies which criticize and/or elabo-
rate and modify Grices framework (cf. relevance theory (Sperber & Wilson 1986);
Kerbrat-Orecchioni 1986; Fogelin 1987; 1988). Leech (1983) deals with FSP as part
of the interpersonal rhetoric, which is used for verbalizing maxims and strategies of
politeness. Terefore, FSP like irony, understatement, metaphor etc. frequently appear
in empirical studies of politeness (cf. Brown/Levinson 1987). As far as text compre-
hension is concerned, FSP can make it easier to recall what has been said or written.
Terefore, attention to fgurative structures can enable better retrieval of the informa-
tion. Tis has also been confrmed by the results of comprehension tests (cf. van Dijk &
Kintsch 1983: 92f.; 253f.). In this context, the diferent functions of FSP in spoken and
written discourse have to be mentioned. Te distinction between orality and literacy
can also be based on the diferent functions of FSP in oral discourse, where they are
functional as mnemotechnic devises, and literary texts, where they have primarily aes-
thetic functions (cf. Ong 1982).
Goatly (1997) provides a good example for a fruitful integration of a number of
pragmatic frameworks for a comprehensive description of metaphor. Goatly posi-
tions metaphor on a continuum between literal and fgurative uses of linguistic
expressions. Using relevance theory (cf. Sperber & Wilson 1986), Goatly (1997: 137f.)
distinguishes creative (active) metaphors from sleeping (inactive) metaphors: the
former require more processing eforts, but also trigger far more contextual implica-
tions and in this sense are optimally relevant. However, he also criticizes relevance
theorys lack of social dimension and uses Hallidays (1994) functional grammar
to distinguish between ideational, interpersonal and textual functions and a corre-
sponding classifcation of a variety of communicative functions of metaphor. More-
over, he uses Hallidayan concepts like register to explain the use of metaphors within
specifc genres. Faircloughs (1989) model of text interpretation is used to combine
relevance theory and functional grammar within an integrated model of text inter-
pretation (Goatly 1997: 283f.).
As far as cross-cultural communication is concerned, on the one hand FSP can
be considered as universals, which occur as discourse strategies in all languages and
cultures; on the other hand, there exist a number of language-specifc or culture-
specifc contrasts as far as the use and the frequency of particular FSP are concerned
(cf. Kvecses 2000). In this context, studies comparing the use of FSP in Indo- European
and non-Indo-European languages are especially interesting (cf. e.g. Jakobson 1977: 89f.
on parallelism; Jakobson & Waugh 1979, ch. 4, on onomatopoeia; Yu 1998 on metaphor
in Chinese; Kienpointner 2004 on metaphor in German, English, Turkish, Chinese and
Native American languages such as Hopi, Navaho, Tohono Oodham).
Te ubiquity of FSP calls for an interdisciplinary approach: their use in literary
texts, in visual codes, their relationship with slips of the tongue and other speech
errors and language disorders, the problems of their treatment in machine translation
and studies on artifcial intelligence cannot be studied other than by a joint efort of
disciplines like linguistics, literary criticism, the graphic arts, psychology, computer
science etc. Te following remarks can only provide a few illustrative examples of
research in these areas.
Of course, FSP are highly relevant for the production of literary texts. Tis is
why they have been amply discussed in literary criticism. In the last few decades, FSP
have been used as descriptive tools for the linguistic interpretation of literary texts by
authors using a variety of difering frameworks, among them quantitative approaches,
linguistic structuralism, generative grammar or text linguistics (cf. e.g. Kibdi Varga
1970; Freeman 1970; Spillner 1974; Plett 1975; Jakobson & Lvi-Strauss 1977; Coseriu
1994). More recently, postmodernist approaches have studied Nietzsches rhetorical
deconstruction of classical metaphysics (cf. especially Nietzsche 1969: 314) and prob-
lematized the interpretation of (literary) texts with the help of classical typologies of FSP
(cf. e.g. De Man 1979, quoted afer Kopperschmidt (ed.) 1991: 170f.; Cahn 1989; Sexl
2004: 178f.). Research on oral narratives in languages and cultures all over the world
suggests that these instances of ordinary talk resemble poetry in many respects: they
are organized according to Jakobsons principle of equivalence (cf. above Section3.1),
contain many FSP and have structural patterns similar to the verses and stanzas of
poems (cf. Hymes 1998 on Native American and African-American narratives).
Many researchers have followed Jakobsons pioneering work in extending the
analysis of FSP from verbal to visual codes. Tey use traditional and modern treat-
ments of FSP for studying rhetorical fgures in music, painting, cinema, advertising
and architecture (e.g. Spencer 1957; Eco 1972: 184f.; Dubois et al. 1974; Aldrich 1983;
Schmitz 1990; Becker/Kibdi Varga 1993; Kennedy/Kennedy 1993). Tere are also
recent attempts to extend concepts of argumentation theory like argument or fallacy
for explaining and evaluating visual arguments (cf. Groarke 1996, 2002, 2009).
Furthermore, the connection of elements of the doctrine of FSP (e.g. FSP like met-
aphor, irony or puns; the distinction of four basic rhetorical operations like addition,
deletion, substitution, permutation) with psychological and psycholinguistic studies
should be mentioned. In this context, research on speech errors (slips of the tongue),
schizophrenic discourse, the mental lexicon, and (Freudian) psychoanalysis provides
examples for the fruitful application of classical rhetorical concepts to still further
disciplines (cf. e.g. Hopper 1992; Reinsdorf 1993; Aitchison 1994; Gibbs 1994; Freud
1975; Todorov 1977: 285f.; Jafe 1980; Lacan 1983; Springfellow 1994; Levelt 1999).
Finally, FSP are a challenge for computer science. Tey pose specifc problems
for their treatment in studies on computer-aided lexicography, machine translation
and artifcial intelligence. Tere are several suggestions how these difculties could be
overcome and a description of FSP like metaphors or metonymies could be integrated
into computer programs designed for simulating the human capacity of understanding
and creating fgurative utterances (cf. e.g. Carbonell 1982; Fass 1988; Lewandowska-
Tomaszczyk 1993; Behrens 1993).
References
Aitchison, J. (1994). Words in the Mind. Blackwell.
Aldrich, V.C. (1983). Visuelle Metapher. In A. Haverkamp (ed.): 142159.
Aristotle (1991). On Rhetoric. Oxford University Press.
(1960). Posterior Analytics. Topica. Heinemann.
(1965). Poetics. Clarendon.
Bacry, P. (1992). Les fgures de style. Belin.
Becker, J. & A. Kibdi Varga (eds.) (1993). Te Appeal of Images. Argumentation 7(1) (Special issue).
Behrens, L. (1993). Lexikalische Ambiguitt und Disambiguierung im Kontext der maschinellen
bersetzung. In P.R. Lutzeier (ed.): 251268.
Black, M. (1954). Metaphor. Proceedings of the Aristotelian Society 55: 273294.
Bradford, A.N. (1982). Classical and Modern Views of the Figures of Speech: Ancient Teory and
Modern Manifestations. Ph.D. New York.
Brown, P. & S. Levinson (1987). Politeness. Some Universals of Language Usage. Cambridge University
Press.
Cahn, M. (1989). Paralipse und Homopathie. In H. Schanze & J. Kopperschmidt (eds.) Rhetorik
und Philosophie: 275295. Fink.
Carbonell, J.G. (1982). Metaphor: An Inescapable Phenomenon in Natural-Language Compre-
hension. In W.G. Lehnert & M.H. Ringle (eds.) Strategies for Natural Language Processing:
415434. Erlbaum.
Cicero (1951). De oratore. Clarendon.
Corbett, E.P.J. & R.J. Connors (1999). Classical Rhetoric for the Modern Student. Oxford University
Press.
Coseriu, E. (1956). La creacin metafrica en el lenguaje. University de la Repblica.
(1994). Textlinguistik. Francke.
Debatin, B. et al. (eds.) (1997). Metaphor and Rational Discourse. Niemeyer.
De Man, P. (1979). Allegories of Reading: Figural Language in Rousseau, Nietzsche, Rilke and Proust.
New Haven.
van Dijk, T.A. (1980). Textwissenschaf. dtv.
van Dijk, T.A. & W. Kintsch (1983). Strategies of Discourse Comprehension. Academic Press.
Dubois, J. et al. (1974). Allgemeine Rhetorik. Fink.
Dumarsais, C. & P. Fontanier (1984). [1818] Les Tropes. Slatkine.
Eco, U. (1972). Einfhrung in die Semiotik. Fink.
(1985). Semiotik und Philosophie der Sprache. Fink.
van Eemeren, F.H. (2010): Strategic Maneuvering in Argumentative Discourse. Benjamins.
van Eemeren, F.H. & R. Grootendorst (1992). Argumentation, Communication and Fallacies. Erlbaum.
Eggs, E. (1994). Grammaire du discours argumentatif. Kim.
Enkvist, N.E. (1973). Linguistic Stylistics. Mouton.
Fairclough, N. (1989). Language and Power. Longman.
Fass, D. (1988). An Account of Coherence, Semantic Relation, Metonymy, and Lexical Ambiguity
Resolution. In S.L. Small, G.W. Cottrell & M.K. Tanenhaus (eds.) Lexical Ambiguity Resolution:
151177. Kaufmann.
Fogelin, R.J. (1987). Some fgures of Speech. In F.H. van Eemeren et al. (eds.) Argumentation: Across
the Lines of Discipline: 263272. Foris.
(1988). Figuratively Speaking. Yale University Press.
Fontanier, P. (1968). Les fgures du discours. Flammarion.
Freeman, D.C. (ed.) (1970). Linguistics and Literary Style. Holt, Rinehart & Winston.
Freud, S. (1975). Te Psychopathology of Everyday Life. Penguin.
Gibbs, R. (1994). Te Poetics of Mind. Cambridge University Press.
Goatly, A. (1997). Te Language of Metaphors. Routledge.
Grice, H.P. (1975). Logic and Conversation. In P. Cole & J.L. Morgan (eds.) Syntax and Semantics,
vol. 3: 4158. Academic Press.
Groarke, L. (1996). Logic, Art and Argument. Informal Logic 18(2): 105129.
(2002). Toward a pragma-dialectics of visual argument. In F.H. van Eemeren (ed.) Advances
in Pragma-Dialectics: 137151. SicSat/Vale Press.
(2009). Five Teses on Toulmin and Visual Argument. In F.H. van Eemeren & B. Garssen
(eds.) Pondering on Problems of Argumentation: 229239. Springer.
Halliday, M.A.K. (1994). An Introduction to Functional Grammar. Arnold.
Haverkamp, A. (ed.) (1983). Teorie der Metapher. Wiss. Buchgesellschaf.
Hopper, R. (1992). Speech Errors and the Poetics of Conversation. Text and Performance Quarterly
12(2): 113124.
Hymes, D. (1998). When is oral narrative poetry? Generative Form and its Pragmatic Conditions.
Pragmatics 8(4): 475500.
Ilie, C. (1994). What Else Can I Tell You? A Pragmatic Study of English Rhetorical Questions as Discur-
sive and Argumentative Acts. Almqvist & Wiksell.
Jafe, S. (1980). Freud as Rhetorician: Elocutio and Dreamwork. Rhetorik 1: 4269.
Jakobson, R. (1960). Closing Statement: Linguistics and Poetics. In T. Sebeok (ed.) Style in Language:
350377. MIT-Press.
(1971a). Two Aspects of Language and Two Types of Aphasic Disturbances. In R. Jakobson
(1971b): 239259.
(1971b). Selected Writings, vol. 2. Mouton.
(1977). Huit questions de potique. Seuil.
Jakobson, R. & C. Lvi-Strauss (1977). Les Chats de Charles Baudelaire. In R. Jakobson: 163188.
Jakobson, R. & L.R. Waugh (1979). Te Sound Shape of Language. Harvester Press.
Kennedy, V. & J. Kennedy (eds.) (1993). Metaphor and Visual Rhetoric. Metaphor and Symbolic Activ-
ity 8(3) (Special issue).
Kerbrat-Orecchioni, C. (1986). Limplicite. Colin.
Kibdi-Varga, A. (1970). Rhtorique et littrature. Didier.
Kienpointner, M. (1996). Vernnfig argumentieren. Rowohlt.
(2004). Metaphern fr Emotionen: Universalien oder Kulturspezifka? In L. Zybatow (ed.)
Translation in der globalen Welt und neue Wege in der Sprach- und bersetzerausbildung: 6191.
Lang.
Kittay, E.F. (1987). Metaphor. Clarendon Press.
Kloepfer, R. (1975). Poetik und Linguistik. Fink.
Knape, J. (1996). Figurenlehre. In G. Ueding (ed.) Historisches Wrterbuch der Rhetorik, vol. 3:
289342. Niemeyer.
Kopperschmidt, J. (ed.) (1990/1991): Rhetorik, 2 vols. Wiss. Buchgesellschaf.
Kvecses, Z. (2000). Metaphor and emotion. Cambridge Univ. Press.
(2002). Metaphor. Oxford Univ. Press.
Lacan, J. (1983). Das Drngen des Buchstabens im Unbewuten oder die Vernunf seit Freud. In A.
Haverkamp (ed.): 175215.
Lakof, G. (1987). Women, Fire and Dangerous Tings. Chicago University Press.
(1991). War and Metaphor. In P. Grzybek (ed.) Cultural semiotics: Facts and Facets: 7392.
Brockmeyer.
(2005). Dont Tink of an Elephant! Know Your Values and Frame the Debate. Chelsea Green.
Lakof, G. & M. Johnson (1980). Metaphors we Live by. Chicago University Press.
Lamy, B. (1969). [1670] La rhtorique ou Lart de parler. Brighton.
Lanham, R.A. (1991). A Handlist of Rhetorical Terms. University of California Press.
Lausberg, H. (1990). Handbuch der literarischen Rhetorik. Hueber.
Leech, G.N. (1966). A Linguistic Guide to English Poetry. Longman.
(1983). Principles of Pragmatics. Longman.
Levelt, W.J.M. (1999). Producing spoken language: A blueprint of the speaker. In C.M. Brown &
P. Hagoort (eds.) Te Neurocognition of Language: 83122. Oxford Univ.Press.
Lewandowska-Tomaszczyk (1993). Bilingual Tesaurus (BIT), Field Teory, and Cognitive Linguis-
tics. In P.R. Lutzeier (ed.): 229238.
Locke, J. (1975). An Essay Concerning Human Understanding. Clarendon Press.
Lutzeier, P.R. (ed.) (1993). Studien zur Wortfeldtheorie/Studies in Lexical Field Teory. Niemeyer.
Marouzeau, J. (1935). Trait de stylistique. Les Belles Lettres.
Mccloskey, D.N. (1983). Te Rhetoric of Economics. Journal of Economic Literature 21: 481517.
Murphy, J.J. (1974). Rhetoric in the Middle Ages. University of California Press.
Nash, W. (1989). Rhetoric. Te Wit of Persuasion. Blackwell.
Nietzsche, F. (1969). ber Wahrheit und Lge im auermoralischen Sinn. In K. Schlechta (Hg.)
Friedrich Nietzsche. Werke III: 309322. Ullstein.
Ong, W.J. (1982). Orality and Literacy. Te Technologizing of the Word. Methuen.
Perelman, C. & L. Olbrechts-Tyteca (1983). [1958]. Trait de largumentation. La nouvelle rhtorique.
Ed. Universit de Bruxelles.
Pielenz, M. (1993). Argumentation und Metapher. Narr.
Plett, H.F. (1975). Textwissenschaf und Textanalyse. Quelle & Meyer.
(1985). Rhetoric. In T.A. van Dijk (ed.) Discourse and Literature: 5984. Benjamins.
(2000). Systematische Rhetorik. Fink.
Quintilianus (1970). Institutio oratoria. Oxford University Press.
Ricur, P. (1975). La mtaphore vive. Seuil.
Reinsdorf, W. (1993). Schizophrenia, Poetic Imagery and Metaphor. Imagination, Cognition and
Personality 13(4): 335345.
Sandig, B. (1986). Stilistik der deutschen Sprache. de Gruyter.
Schmitz, A. (1990). Die oratorische Kunst Bachs. In J. Kopperschmidt (ed.): 290312.
Searle, J. (1979). Expression and Meaning. Studies in the Teory of Speech Acts. Cambridge University
Press.
Sexl, M. (ed.) (2004). Einfhrung in die Literaturtheorie. Wien.
Spencer, J.R. (1957). Ut Rhetorica Pictura: A Study in Quattrocento Teory of Painting. Journal of the
Warburg and Courtauld Institute 20: 2644.
Sperber, D. & D. Wilson (1986). Relevance. Communication and Cognition. Blackwell.
Spillner, B. (1974). Linguistik und Literaturwissenschaf. Kohlhammer.
Spitzer, L. (1961). Stilstudien. Hueber.
Springfellow, F. (1994). Te Meaning of Irony: A Psychoanalytic Investigation. State University of New
York Press.
Todorov, T. (1967). Littrature et signifcation. Larousse.
(1977). Tories du symbole. Seuil.
Ueding, G. & B. steinbrink (1986). Grundri der Rhetorik. Metzler.
Verschueren, J. (1998). Understanding Pragmatics. Arnold.
Weinrich, H. (1976). Sprache in Texten. Klett.
Weydt, H. (1987). Metaphern und Kognition. In J. Ldtke (ed.) Energeia und Ergon: 303311. Narr.
Yu, N. (1998). Te Contemporary Teory of Metaphor: A Perspective from Chinese. Benjamins.
Genre
Anna Solin
University of Jyvskyl
1. Introduction
Genre is a concept used in many disciplines, ranging from linguistics, literary studies,
cultural and flm studies to rhetoric, folklore and educational research. Its roots are in
Greek antiquity, but in language studies the concept has gained salience only in the
past few decades.
Te notion of genre is closely associated with the idea that texts (and other cultural
products) can be grouped into types or classes according to formal and/or functional
similarities. However, classifcation is not all there is to genres. Genre interests scholars
across disciplines because it allows looking at products of culture not merely as forms,
but as social and symbolic action. Genres are increasingly perceived not as normative
formulas, but as sets of socially situated and variable conventions and expectations
regarding textual form. Tus, in contemporary genre research, descriptive and taxo-
nomic approaches coexist with studies on genre use in diferent settings and change
and variability in genre repertoires.
Te concept of genre is complex and even controversial, in part due to diferences
in how genres are conceptualised, modelled and analysed in diferent disciplines
and schools of genre research. Te metaphors through which genres are viewed are
illustrative: while some talk about genres as species or classes, drawing on biological
models, others prefer to think of genres as families or prototypes, and yet others as
blueprints or labels (Frow 2006: 52). What most approaches have in common, how-
ever, is the assumption that genres as forms of semiotic practice are socially based.
1

Te concept encourages analysts to look for shared conventions, typifcation, recur-
rence and standardisation. Following Devitt, genres require multiplicity, multiple
actions by multiple people (2004: 33).
In the following, I will review diferent approaches to genre research within
language studies. I will begin, however, with a brief outline of the historical roots of
the concept and particularly the work of Mikhail Bakhtin, which has been instrumen-
tal in the emergence of the concept in language studies.
:. It is notable that cognitive approaches to genre have been far less prominent than sociocultural
ones in recent years (but see Paltridge 1997).
120 Anna Solin
2. Historical precedents
Te term genre derives from the Latin genus, which refers to kind or class. Applied to
literary texts, classifcation goes as far back as Aristotles Poetics, where distinctions
are drawn between epic, tragedy and comedy. Subsequently, renaissance scholars were
infuenced by classical models to produce listings of genres. Tese lists did not pro-
vide any comprehensive coverage of existing literary genres, but rather a prescriptive
taxonomy of valued forms (Frow 2006: 26).
2
Classifcatory approaches to genre were
common also in rhetoric and the study of folklore.
Te concept of genre began to be widely used in linguistics only in the 1970s.
Infuences were drawn from the study of registers, begun in the 1960s (e.g. Halliday
et al. 1964; see also Dittmar 1995), sociolinguistics and linguistic anthropology (e.g.
Hymes 1974), as well as from folklore and literary studies. Perhaps the most important
theoretical infuence for genre research in linguistics has been the Russian philosopher
and literary scholar Mikhail Bakhtin (see Bjrklund 2005). In his now classic essay Te
problem of speech genres (1986, originally published in 1953), Bakhtin makes the claim
that all language use is necessarily framed within generic types: genre is a precondition
for communication, for the creation and interpretation of texts. It is therefore impos-
sible to communicate outside genres: We speak only in defnite speech genres, that is,
all our utterances have defnite and relatively stable typical forms of construction of the
whole (1986: 78, original italics).
3
Bakhtin defnes genres as relatively stable thematic, compositional and stylistic
types of utterances (1986: 64). Due to their stability, they act as a normative infu-
ence, in the same way as grammatical systems. However, this stability is variable:
some generic forms are rigid and standardised to the extent that their use amounts
to mechanical reproduction while others are fexible, plastic and creative (1986: 78).
Tat generic stabilities are only relative has become a central theme in
contemporary genre research: texts are perceived as being shaped by both centripetal
(unifying and centralising) and centrifugal (stratifying, decentralising) forces (Bakhtin
1981: 27072). Every genre is infuenced at the same time by normative forces, seeking
to standardise generic expression, and forces which bring out a multiplicity of social
voices (Bakhtin 1981: 263).
Another key element of Bakhtins discussion of genre is that texts are established
as dialogical in nature: they form links in chains of speech communion (Bakhtin
1986: 76). Tat is, texts are not self-sufcient, but are always to be regarded as responses
i. An example is Sir Philip Sidneys Te Defence of Poesie (1595), which lists Heroick, Lyrick,
Tragick, Comick, Satyrick, Iambick, Elegiack, Pastorall, and certaine others (1595: 10).
. Te term speech genres refers to both spoken and written genres.
Genre 121
to previous texts in the same sphere. Te notion of addressivity, the quality of being
directed to someone, emphasises that texts are constructed with such responses in
mind (1986: 95). Tus, genre as a concept directs attention not only to the stability
and organisation of individual types, but to interaction between types and interaction
between producers and consumers of texts.
3. Genre research in language studies
Contemporary genre research in language studies focuses in particular on professional
and disciplinary communication (e.g. research writing, business and technical writ-
ing) and genre literacy. In the past two decades, genre has become an infuential con-
cept particularly in the teaching of writing. Genre research has also gained increasing
visibility on the conference scene: the frst conference devoted to genre research in
language studies was held in 1992 in Ottawa, with new editions in Vancouver (1998)
and Oslo (2001). Currently, a series of conferences is being organised in Brazil under
the label SIGET.
In the following, I will review three established approaches to genre research
(cf. Hyon 1996): the Australian-based Sydney School, the North American New
Rhetoric paradigm and genre research within English for Specifc Purposes. Tese
three approaches form relatively homogeneous schools with specifc and identifable
research agendas and interests.
4
In addition, the concept of genre is drawn on in a
number of other traditions of linguistics; for example, linguistic anthropology (e.g.
Bauman & Briggs 1992; Bauman 2001), conversation analysis (e.g. Levinson 1992;
Linell 1998) and critical discourse analysis (Fairclough 2003; Wodak 2006). Tese will,
however, not be discussed further here.
3.1 Sydney School
Tis approach to genre builds on systemic-functional linguistics (SFL), pioneered by
Michael Halliday (see e.g. Halliday 1994). It has developed a coherent research pro-
gramme as well as pedagogical applications in the feld of genre literacy, both of which
have been explicitly articulated and exemplifed in an extensive literature. Te school
derives its name from its main institutional base, the Department of Linguistics at
the University of Sydney in Australia. Studies of genre in SFL developed from early
research on register, which examined how language varies according to situational
|. Genre research is also a vibrant eld in Brazil (see SIGET 2007) and in the Nordic countries
(see e.g. Berge & Ledin 2001; Berge & Maager 2005). For a review of genre research in German
speaking countries, see Muntigl and Gruber (2005).
122 Anna Solin
context. It was noted that linguistic features (such as particular types of lexis) are not
spread evenly across texts; language users make diferent linguistic choices in difer-
ent phases of a text or interaction. Tis gave rise to an examination of generic staging
(Ventola 2005).
Te key scholars associated with the Sydney School are Jim Martin, Frances
Christie and Joan Rothery (see e.g. Martin et al. 1987; Rothery 1996; Christie &

Martin 1997). In broad terms, genres are defned by their function or purpose, in
relation to a particular sociocultural context: diferent genres are diferent ways of
using language to achieve diferent culturally established tasks (Eggins & Martin
1997: 236).
5
Sydney scholars have developed, in particular, analyses of the schematic
structure of genres, as staging is assumed to be the major linguistic refex of difer-
ences in purpose (Eggins & Martin 1997: 236). Tus, the classic Sydney defnition of
genres is that they are staged goal-oriented social processes (Martin 1997: 13). Te
argument is that texts unfold through stages, since social action usually takes more
than one step to accomplish. Te stages are goal-oriented in the sense that when they
do not unfold as expected, there is likely to be a sense of incompletion.
Another important characteristic of Sydney genre research is close analysis of the
linguistic realisation of generic stages: diferent stages are assumed to draw on diferent
lexico-grammatical resources. Eggins (2004) gives the example of a recipe: while lists of
ingredients tend to be realised as noun phrases including numbers and measurements,
methods sections ofen contain clauses in the imperative mood, linked by temporal
conjunctions (then, next) and including action-oriented verbs (such as slice, fry, heat).
Tus, the identifcation of stages is not an interpretive process, but a descriptive one.
Here SFL work on genre difers signifcantly from many other approaches to genre
research: genre identity is defned by recourse to text-internal, linguistic criteria.
6
Tis emphasis on lexico-grammatical description is related to one of the principal
motivations of genre research in the Sydney School: contributing to genre literacy
education. Te modelling of genres in terms of stages was developed in the 1980s
as a response to debates around the teaching of writing in schools and to support
new literacy programmes (see Reid 1987; Cope & Kalantzis 1993; Kress 2003).
7
Tus,
,. In Sydney genre theory, genres are modelled on the level of context of culture, while register
variation is related to the context of situation (see e.g. Martin & Rose 2008: 10, 16; the two terms
are derived from Bronislaw Malinowskis work).
o. A similar argument has been proposed by Ruqaiya Hasan, who has developed a model where
genres are analysed in terms of obligatory and optional elements, constituting a generic structure
potential (GSP; Hasan 1989; see also Hasan 1996). Te obligatory elements are seen as constitutive
of a genre: they condition whether a genre text is complete or not.
. Te genre literacy programmes were a response particularly to process writing curricula then
popular in schools.
Genre 123
besides academic analyses, Sydney genre scholars have produced textbooks and other
teaching materials for literacy teaching.
Kress (1993: 28) notes that Australian models of genre pedagogy formed both
a pedagogical and a political project. Te key notion was access: it was assumed that
explicit teaching of a number of key genres to migrant and working class students
would help them gain access to educational and career opportunities, and indeed to
full citizenship (Cope & Kalantzis 1993).
8
Such pedagogical agendas have signifcantly infuenced the selection of genres
for analysis. Te early focus was on primary and secondary school genres, and later
migrant English education and workplace training programmes. Six key example
genres were report, explanation, procedure, discussion, recount and narrative. Te
recount genre, for example, contains the stages orientation and record of events
with an optional stage reorientation, while narratives consist minimally of orienta-
tion and resolution, with the possible addition of complication and evaluation
between these two stages (Macken-Horarik 2002).
Tese examples suggest that genres are ofen described on a diferent level of
abstraction compared to many other schools of genre research. Genre labels such as
exposition and narrative in fact resemble labels given to text types in typologically
oriented text linguistics (e.g. Werlich 1983; Biber 1989; see also Pilegaard & Frandsen
1996). In such models, text types may be seen as constituent elements of genres. Simi-
larly, Martin (1997, 2002) suggests that reports, narratives, expositions etc. may be
treated as elemental genres which combine into macro-genres.
3.2 New Rhetoric
Tis approach to genre research, sometimes also referred to as Rhetorical Genre
Studies, developed in North America mainly within the felds of rhetoric and compo-
sition studies in the 1980s and 1990s. To date, most of the work has focused on written
genres and L1 writers. Te approach difers from the Sydney School in many respects.
Instead of treating genres as broadly available cultural resources, genres are analysed
as socially situated, ofen within particular professional or disciplinary communities.
Besides looking at textual products, researchers examine the settings in which genres
are used, focusing on institutional, cultural and historical factors infuencing writing
practice (see e.g. Freedman & Medway 1994; Coe et al. 2002). Analysts have looked at
the writing of, for example, tax accountants (Devitt 1991), biologists (Myers 1990) and
workers at a central bank (Smart 2006). Unlike in Sydney genre research, there tends to
8. Models of SFL genre literacy have also been widely criticised (e.g. Luke 1996; Freedman 1994;
see Kress 2003 for a review).
124 Anna Solin
be little close analysis of textual structures or lexicogrammatical choices. Nonetheless,
the approach has been very infuential in language studies.
A key text by Carolyn Miller elaborates on the conceptualisation of genres as
social action: a rhetorically sound defnition of genre must be centred not on the
substance or the form of discourse but on the action it is used to accomplish (Miller
1984: 151). Miller argues that typifcation of rhetorical forms such as genres is linked
to typifcation in rhetorical situations. When situations recur they prompt similar
responses, and these similarities come to be understood as types.
Following Bakhtin, genres are perceived as dynamic forms. Despite appearing
stable, they are likely to change when rhetorical situations change (Berkenkotter &
Huckin 1995). In an of-cited formulation, genres are not fully fxed but
stabilised-for-now (Schryer 1994: 108). Tere is therefore a concern to theorise
genres as socially grounded but yet not deterministic of language use: the notion of
genre is not meant to imply that writers have little agency or creative freedom. Te
themes of genre evolution and change have been examined in a number of mono-
graphs, for example Bazerman (1988) on the experimental article and Yates (1989) on
the emergence of organisational genres.
Again following Bakhtin, genres are perceived as intertextually related to other
genres, both diachronically and synchronically. New Rhetoric scholars have developed
a number of concepts to account for such intertextual relations, notably genre rep-
ertoire and genre system (see Devitt 2004 for a review). Genre repertoires are sets of
genres that exist within a particular sphere of activity or community and which are
routinely used to conduct interaction (see e.g. Orlikowski & Yates 1994). Te term
genre system refers to interlinked genres, ofen organised in sequence, which work
together to fulfl a common purpose; examples include the genre system of job search
or organising a wedding (for analyses, see Bazerman 1994; Berkenkotter 2001; Tardy
2003). Importantly, since genres are dynamic and variable, so are genre repertoires
and systems.
Te conceptualisation of genres as dynamic forms of social action has important
methodological consequences: it demands close contextualisation of genre texts
through the use of ethnographically oriented methodologies. Analysts ofen rely
on some form of feldwork, including interviews and observation, in order to gain
insights into the community being studied. For example, Tardys (2003) study of grant
applications and Schryers (1994) research on genres in veterinary medicine draw on
writer interviews and observation of local discourse practices.
A fnal issue where New Rhetoric scholars diverge from their Australian col-
leagues regards the feasibility of the explicit teaching of genre forms. Few researchers
in this feld advocate explicit teaching. For example, Berkenkotter and Huckin (1995)
argue that genre learning is a process of situated acquisition through apprenticeship,
Genre 125
analogous to second language acquisition through immersion (see also Beaufort 2000).
Te most forceful statement against explicit teaching is Freedman (1994). She argues
that in most cases explicit teaching will not result in genre acquisition, owing to the
complexity and context-dependence of genres. Freedman argues that if genres are per-
ceived as rhetorical action, rather than as aggregations of formal features, it is doubtful
whether an explication of textual features is helpful in supporting genre acquisition: if
genres are responses to contexts, can they be learned at all out of context? (Freedman
1994: 194; see also Freedman & Richardson 1997).
3.3 English for Specifc Purposes
Genre analysis has been a very prominent approach also in applied linguistics, and
particularly in the subfeld of English for Specifc Purposes (ESP), mainly through
the infuence of John Swales. However, the agenda of ESP genre research is not as
distinctive or theoretically elaborate as those of the Sydney School and New Rhetoric.
Researchers have been infuenced both by systemic-functionalist models of analysis
and by the ethnographic orientation of New Rhetoric genre research.
A key feature of ESP genre research is its pedagogical agenda: to assist non-native
English speakers in mastering academic and professional genres. Published genre
studies ofen have implications for teaching practice, and many scholars in the feld
have taught genre-based writing courses for L2 students and produced teaching mate-
rials (see e.g. Swales & Feak 2004). Indeed, Swales states in his infuential monograph
Genre Analysis that the aim of the book is to ofer an approach to the teaching of
academic and research English (1990: 1). However, Swales work is frequently drawn
on also outside applied linguistics. Scholars interested in genre have taken on, in par-
ticular, the emphasis on defning genres through their communicative purposes rather
than their forms, as well as the contextualisation of genres as the property of discourse
communities.
In contrast to Sydney genre theory, Swales establishes communicative purpose as
a privileged criterion of genre identifcation:
A genre comprises a class of communicative events, the members of which share some
set of communicative purposes. Tese purposes are recognised by the expert members
of the parent discourse community and thereby constitute the rationale for the genre.
Tis rationale shapes the schematic structure of the discourse and infuences and
constrains choice of content and style. (Swales 1990: 58)
Here, a relation is established between communicative purpose and language choice,
via the notion of rationale: the rationale of the genre constrains topic-related, struc-
tural and linguistic choices.
126 Anna Solin
Swales (1990) further argues that genres belong to discourse communities.
Such communities are defned by common goals rather than the shared social back-
ground of community members. Established members of discourse communities are
familiar with the key genres used in the community, while new members are seen as
apprentices, in the process of becoming socialised into the relevant discoursal expec-
tations (Swales 1990: 2132). Te communities analysed are typically professional or
disciplinary ones.
As in Sydney genre research, most ESP analysts focus on the overall structure
of genre texts, modelled by Swales in terms of moves and steps (cf. SFL stages).
Swales own analyses have mainly concerned academic genres, but move analysis has
also been applied to a host of other professional genres (typically business and legal
genres). Individual genres that have been analysed for rhetorical moves include legal
case reports (Bhatia 1993), business faxes (Louhiala-Salminen 1997), fundraising
letters (Bhatia 2004) and personal statements (Samraj & Monk 2008).
Moves and steps are not linguistically but rhetorically defned. Perhaps the most
famous move-step model is that proposed by Swales on the rhetorical organisation
of introductions in research articles (1990: 140161). Te CARS model (Create a
Research Space) suggests that successful introductions consist of three moves with
a number of obligatory and optional steps in each. For example, the frst move, estab-
lishing a territory consists of three optional steps: claiming centrality, making topic
generalisation and reviewing items of previous research.
While moves are functionally defned and labelled and fexible in terms of
linguistic realisation (Swales 2004: 229), it is ofen assumed that they are associated
with specifc realisations. Unlike in SFL stage analysis, such realisations are usually
not analysed systematically, nor do ESP studies draw on any one grammar model.
Typically those grammatical and lexical features are commented on which are deemed
frequent or important by the analyst.
9
Te analysis of linguistic realisations is ofen motivated by pedagogical concerns,
but as in New Rhetoric thinking, ESP scholars tend to emphasise variation and advise
against putting forward prescriptive models of generic structure (see e.g. Flowerdew
2002). Te key issue here is that of rhetorical success. A common aim of ESP studies
is to establish what kinds of structural and linguistic choices lead to rhetorical success
(e.g. the publication vs. rejection of a research article), in order for teachers to be able
to pass on such information to their students.
. For instance, Swales (1990) outlines a number of linguistic features typically found in move 2
of the CARS model, i.e. establishing a niche. Writers in his data ofen use adversative connectors
such as however and yet as well as lexical and grammatical negation to introduce a discussion
of gaps in previous research. However, Swales does not comment at all on possible linguistic reali-
sations of the step claiming centrality.
Genre 127
Genre research in ESP is not solely based on text analysis, however. In line with
work in New Rhetoric, there is increasing interest in analysing writers and commu-
nities, sometimes using ethnographically oriented methodologies (Swales 1998 is a
prominent example). Many studies use specialist informants in order to gain access
to normative views and expert understandings of the genre analysed, for example
through text-based interviews or an analysis of manuals relevant to genre production
(see e.g. Paltridge 2002; Samraj & Monk 2008).
Te above review has illustrated a great deal of variation in the ways in which
scholars conceptualise and analyse genres. In the following, I will address two
issues which arise from the review and which particularly trouble the feld of genre
research.
4. Issues and debates
Te frst debate regards classifcation and its role in the study of genres, an issue which
the three schools approach diferently. Te second and related issue is that of stability:
again, there are signifcant diferences between those who analyse genres as essentially
stable and those who focus attention on variability and change in the forms and uses
of genres.
4.1 Genre as class
Te notion of genre is ofen associated with the classifcation of texts into types.
Tis involves asking questions such as how many genres there are in a given feld,
whether a text belongs to one genre class or another and what the identifying fea-
tures of a genre are. Criteria used in assigning individual texts into genre classes are
ofen formal ones (similarities in lexico-grammatical choices and schematic struc-
ture), but as noted above, also the communicative function of a text has been used in
defning genre identity (as in Swales work). Te Sydney School, in particular, tends
to favour a taxonomising approach: analysts take as their objects not only individual
genres, but how genres relate to each other and form systems. Tus, Martin and Rose
(2008: 44) state that Our ultimate goal has always been to map cultures as systems
of genre.
However, classifcation as an aim has been widely problematised in recent genre
research. Commentators note, for instance, that it is doubtful whether individual texts
can be seen as members of genre classes or as belonging to classes, if genres are
conceptualised as dynamic and variable and their boundaries as indeterminate. Frow
(2006: 2) argues that texts even the most simple and formulaic do not belong
to genres but are, rather, uses of them. For example, individual texts may draw on a
128 Anna Solin
generic template only partially, or they may mix genres. Frow adds a further perspective
in suggesting a focus on texts as performances of genre, rather than analysing texts
as reproductions of a class:
Texts are acts or performances which work upon a set of generic raw materials.
Te relationship is one of productive elaboration rather than of derivation or
determination. (Frow 2006: 2324)
Te issue of classifcation is closely related to the question of whether generic identity,
and thereby genre classes, can be established by recourse to shared formal features.
Te key question is whether the relation between genre identity and form is a rela-
tion of association (genre A is typically realised using forms x, y, z, and therefore such
forms act as traces of the genre) or defnition (members of genre A are defned by
the presence of forms x, y, z, without which they must be defned as non-members or
incomplete).
Devitt (2004) expresses a widely held view in rhetorical genre studies when she
claims that formal features cannot be the basis of genre identifcation or classifca-
tion. While the identifcation of recurrent forms may be useful, formal features
do not defne or constitute [genres] (2004: 11; original italics). In contrast, Sydney
genre theorising subscribes to a formalist assumption in the sense that generic stages
are understood as constitutive of genres. Particularly in the genre models devised
for literacy programmes, generic stages are identifable through the presence of
particular sets of lexicogrammatical features (cf. also Hasans generic structure
potential; Hasan 1989).
A number of commentators have criticised what they perceive as an overprivileging
of form in Sydney genre research. For example, Bex (1996: 141) notes that only consid-
ering formal features makes it impossible to distinguish parodic texts from sincere
ones, since parodies work precisely because they closely imitate (some of) the formal
elements of the texts they parody, while subverting their communicative function. A
second criticism is that a formal model fails to account for genre evolution and change
over time; the formal features of genres might change while the genre name and the
practice it designates might not (see e.g. Devitt 2004).
Classifcation on the basis of communicative purpose has also been questioned.
Fairclough (2003: 71) argues that not all genres are purpose-driven: focusing on
purposes emphasises stable genres with a determinate structure, which are part of
instrumental social systems. He notes that many analysts associate a contemporary
tendency towards purpose-driven genres with genre in general, and suggests giving
the notion of purpose a less central place in genre defnitions.
Swales himself has also later problematised the status of communicative purposes
as genre defning. Askehave and Swales (2001) in fact suggest abandoning communica-
tive purposes as a method for grouping texts into genre categories. Tis is because the
Genre 129
purposes, goals, or public outcomes [of genres] are more evasive, multiple, layered,
and complex than originally envisaged (2001: 197). However, they suggest retaining
the notion of communicative purpose, but as something that can only be established
afer considerable research (a long-term outcome of the research), not as an initial
stage of genre identifcation.
4.2 Stability of genres
A related issue which is intensely debated in genre research is the question of whether
genres are essentially stable and classifable forms or entities which are variable and
dynamic to the point of escaping defnition. Te issue revolves around the tension iden-
tifed by Bakhtin (1981) regarding the unifying and stratifying forces at work in the
production of texts. While genres provide constraints for the production of texts, they
are never fully fxed. For Bakhtin, stability is thus always relative: with each new instan-
tiation or performance, a genre enters into interaction with new contexts and demands.
While the dynamicity of genres is a central assumption in New Rhetoric
theorisations, the situation is less clear in the ESP and Sydney models. Te bulk of
both ESP and systemic-functionalist genre analyses have been concerned with stable,
institutionalised genres, and have had relatively little to say about genre emergence,
instability and hybridity. Kress (1993: 35) notes that the Martin model has tended
towards the reifcation of types, with an emphasis on the linguistic system as an
inventory of types. Te pedagogical programmes ofered by the Sydney School have
also been criticised for suggesting a static, deterministic interpretation of genres.
It should be noted, though, that Sydney School theorising increasingly addresses
questions of variability and change. For example, Martin and Rose (2008) emphasise
that genre evolution should be understood in terms of metastability:
As system, genre functions as a kind of inertia As process, genre allows for gradual
change, as texts unfold in relation to both recurrent and divergent material and social
conditions; as divergence recurs, innovative confgurations of meaning stabilise, and
new texts become familiar genres. (Martin & Rose 2008: 259)
Some disparity is still in evidence, however, between theoretical models (which tend
to agree that genres are dynamic entities) on the one hand and models of genre peda-
gogy on the other. Te ambivalence about stability seems to be a genuine source of
tension. Many writers in the feld of genre literacy espouse dynamicity in principle and
are happy to cite Bakhtins writings as informing their work, while at the same time
describing pedagogical applications in terms which suggest that genres can be pre-
sented as static formulas; that, for teaching purposes, it is possible to establish what
works in specifc settings. For example, Hyland writes about the need for clear and
explicit genre descriptions to support L2 writing (2004: 11) and the need for clear
guidelines for how to construct the diferent kinds of texts [students] have to write
130 Anna Solin
(2004: 21). In the same volume, he reminds the reader that Te dangers of a static,
decontextualised pedagogy are very real (2004: 19).
Divergent views of stability also come to the fore in the ways in which the diferent
schools approach the analysis of hybridity, or diferent forms of mixing genres. While
all three schools subscribe to the assumption that texts may mix genres or form hybrids
in other ways, they difer in terms of whether hybrids are seen as being constituted of
relatively stable elements, or whether the mixtures defy description.
Martin (2002: 264) sees hybrids as constituted of established generic resources:
In order to mix genres, we need something canonical to mix. Tis suggests that
there exists a stable and identifable set of genres from which mixtures are con-
structed. Martin (2002) and Martin and Rose (2008) distinguish between diferent
types of mixing, including shifing gears (shifing to a new genre in the middle of a
text), contextual metaphor (one genre being used to stand for another, as in parody)
and genre embedding (one genre functioning as a stage in another). An additional
form of hybridity is multimodality, the way in which texts combine verbal and visual
resources (see e.g. Kress 2003; Kress & van Leeuwen 2006).
In ESP genre research, the scholar who has written most actively about genre
mixing is Bhatia. His analyses concern in particular the mixing of communicative pur-
poses in texts (e.g. academic and promotional, or legislative and diplomatic; see Bhatia
1997, 2000). He refers to the mixing of two or more established forms to create new
hybrid forms (2000: 148). Tus, the focus here too is on the mixing of existing generic
resources, whose traces can be descriptively mapped in individual texts.
Medway (2002), in a collection of articles on rhetorical genre studies, takes a
contrasting approach. He examines a fuzzy set of texts, architecture students sketch-
books. Tese texts form an unusual object of analysis since there are hardly any textual
regularities evident in the data. Te sketchbooks Medway analyses are all widely difer-
ent structurally, linguistically and visually. Tere is also great variation in the functions
which the texts seem to serve; these include recording and preserving, thinking and
learning and preparing action (Medway 2002: 131).
Te hybridity of the texts is not describable in any simple way. Te texts combine
verbal and visual elements; writers seem to switch from sketching to writing and vice
versa in a casual manner. Medway also notes that the texts form intertextual relations
to a wide array of generic resources, such as lecture notes, calendars and confessional
journals. However, such models are not constitutive elements of the genre, nor does
their description exhaust the analysis of the texts hybridity.
Medways analysis is illuminating because it forces a consideration of whether
stability, in the form of textual regularities, is a necessary precondition of genre sta-
tus. Medway notes that if similarities in form or function are the principal criterion
of genre status, his data do not count as manifestations of a single genre. However,
he also argues that the sketchbooks are a recognised and signifcant cultural item
Genre 131
within the community of architecture students, a format students learn from their
peers when entering the community (2002: 136). Te critical question is then whether
the recognisability of a genre as a form of action overrides the need for typifcation on
the level of texts.
Te above review suggests that there is increasing agreement in contemporary
genre research that genres are not simply formal classes and that theoretical models
must take account of issues like hybridity and genre change. It is notable, in particu-
lar, that representatives of the Sydney School, which has tended to take a taxonomic
approach to mapping genres, also provide tools for the analysis of mixed texts.
However, Medways questions regarding what determines genrehood also indicate that
important diferences remain, especially regarding the status of textual regularities as
a marker of genre identity.
10
References
Askehave, I. & J. Swales (2001). Genre identifcation and communicative purpose: a problem and a
possible solution. Applied Linguistics 22: 195212.
Bakhtin, M.M. (1981). [1931]. Te dialogic imagination. Four essays by M.M. Bakhtin. (transl. by
C. Emerson & M. Holquist; ed. by M. Holquist.) University of Texas Press.
(1986). [1953]. Te problem of speech genres. In Speech genres and other late essays: 60102.
(transl. by V. McGee; ed. by C. Emerson & M. Holquist.) University of Texas Press.
Bauman, R. (2001). Te ethnography of genre in a Mexican market: form, function, variation. In
P. Eckert & J.R. Rickford (eds.) Style and sociolinguistic variation: 5777. Cambridge University
Press.
Bauman, R. & C.L. Briggs (1992). Genre, intertextuality, and social power. Journal of Linguistic
Anthropology 2: 131172.
Bazerman, C. (1988). Shaping written knowledge. Te genre and activity of the experimental article in
science. University of Wisconsin Press.
(1994). Systems of genre and the enactment of social intentions. In A. Freedman & P. Medway
(eds.): 79101.
Beaufort, A. (2000). Learning the trade. A social apprenticeship model for gaining writing expertise.
Written Communication 17: 185223.
Berge, K-L. & P. Ledin (2001). Perspektiv p genre. Rhetorica Scandinavica 18: 415.
Berge, K-L. & E. Maager (eds.) (2005). Semiotics from the North. Nordic approaches to systemic
functional linguistics. Novus.
Berkenkotter, C. (2001). Genre systems at work. DSM-IV and rhetorical recontextualisation in
psychotherapy paperwork. Written Communication 18: 326349.
Berkenkotter, C. & T. Huckin (1995). Genre knowledge in disciplinary communication. Lawrence
Erlbaum.
:o. Te article has been written with the support of research funding from the Academy of
Finland.
132 Anna Solin
Bex, T. (1996). Variety in written English. Routledge.
Bhatia, V.K. (1993). Analysing genre: language use in professional settings. Longman.
(1997). Genre-mixing in academic introductions. English for Specifc Purposes 16: 181195.
(2000). Genres in confict. In A. Trosborg (ed.) Analysing professional genres: 146161.
Benjamins.
(2004). Worlds of written discourse. Continuum.
Biber, D. (1989). A typology of English texts. Linguistics 27: 343.
Bjrklund, M. (2005). Mikhail Bakhtin. In J. Verschueren & J.O. stman (eds.) Handbook of Prag-
matics. Benjamins.
Christie, F. & J.R. Martin (eds.) (1997). Genre and institutions. Social processes in the workplace and
school. Cassell.
Coe, R., L. Lingard & T. Teslenko (eds.) (2002). Te rhetoric and ideology of genre. Hampton Press.
Cope, B. & M. Kalantzis (eds.) (1993). Te powers of literacy. A genre approach to teaching writing.
Falmer Press.
Devitt, A. (1991). Intertextuality in tax accounting: generic, referential and functional. In
C. Bazerman & J. Paradis (eds.) Textual dynamics of the professions historical and contem-
porary studies of writing in professional communities: 336357. University of Wisconsin Press.
(2004). Writing genres. Southern Illinois University Press.
Dittmar, N. (1995). Register. In J. Verschueren, J.-O. stman, J. Blommaert & C. Bulcaen (eds.)
Handbook of Pragmatics. Benjamins.
Eggins, S. (2004). An introduction to systemic functional linguistics, 2nd ed. Continuum.
Eggins, S. & J.R. Martin (1997). Genres and registers of discourse. In T.A. van Dijk (ed.) Discourse as
structure and process: 230256. Sage.
Fairclough, N. (2003). Analysing discourse. Routledge.
Flowerdew, J. (2002). Genre in the classroom: a linguistic approach. In A.M. Johns (ed.) Genre in the
classroom: multiple perspectives: 91102. Lawrence Erlbaum.
Freedman, A. (1994). Do as I say: the relationship between teaching and learning new genres. In
A. Freedman & P. Medway (eds.): 191210.
Freedman, A. & P. Medway (eds.) (1994). Genre and the New Rhetoric. Taylor & Francis.
Freedman, A. & P. Richardson (1997). Literacy and genre. In L. van Lier & D. Corson (eds.) Encyclo-
pedia of language and education, vol. 6: 139149. Kluwer Academic Publishers.
Frow, J. (2006). Genre. Routledge.
Halliday, M.A.K. (1994). Introduction to functional grammar. Arnold.
Halliday, M.A.K., A. McIntosh & P. Strevens (1964). Te linguistic sciences and language teaching.
Longman.
Hasan, R. (1989). Te identity of a text. In M.A.K. Halliday & R. Hasan (eds.) Language, context and
text: aspects of language in a social-semiotic perspective: 5269. Oxford University Press.
(1996). [1984]. Te nursery tale as a genre. In C. Cloran, D. Butt & G. Williams (eds.) Ways of
saying: ways of meaning. selected papers of Ruqaiya Hasan: 5172. Cassell.
Hyland, K. (2004). Genre and second language writing. University of Michigan Press.
Hymes, D. (1974). Foundations in sociolinguistics an ethnographic approach. University of Penn-
sylvania Press.
Hyon, S. (1996). Genre in three traditions: Implications for ESL. Tesol Quarterly 30: 693722.
Kress, G. (1993). Genre as social process. In B. Cope & M. Kalantzis (eds.): 2237.
(2003). Literacy in the new media age. Routledge.
Kress, G. & T. Van Leeuwen (2006). Reading images: a grammar of visual design. Routledge.
Genre 133
Levinson, S. (1992). Activity types and language. In P. Drew & J. Heritage (eds.) Talk at work. Interac-
tion in institutional settings: 66100. Cambridge University Press.
Linell, P. (1998). Approaching dialogue. Talk, interaction and contexts in dialogical perspectives. John
Benjamins.
Louhiala-Salminen, L. (1997). Investigating the genre of a business fax: A Finnish case study. Journal
of Business Communication 34: 316333.
Luke, A. (1996). Genres of power? Literacy education and the production of capital. In R. Hasan &
G. Williams (eds.) Literacy in society: 308338. Longman.
Macken-Horarik, M. (2002). Something to shoot for: A systemic-functional approach to teaching
genre in secondary school science. In A.M. Johns (ed.) Genre in the classroom: multiple perspec-
tives: 1742. Lawrence Erlbaum.
Martin, J.R. (1997). Analysing genre: functional parameters. In F. Christie & J.R. Martin (eds.):
339.
(2002). From little things big things grow: ecogenesis in school geography. In R. Coe,
L. Lingard & T. Teslenko (eds.): 243269.
Martin, J.R., F. Christie & J. Rothery (1987). Social processes in education: a reply to Sawyer and
Watson (and others). In I. Reid (ed.): 5882.
Martin, J.R. & D. Rose (2008). Genre relations. Mapping culture. Equinox.
Medway, P. (2002). Fuzzy genres and community identities: the case of architecture students sketch-
books. In R. Coe, L. Lingard & T. Teslenko (eds.): 123153.
Miller, C.R. (1984). Genre as social action. Quarterly Journal of Speech 70: 151167.
Muntigl, P. & H. Gruber (2005). Introduction: approaches to genre. Folia Linguistica 39: 118.
Myers, G. (1990). Writing biology. Texts in the social construction of scientifc knowledge. University
of Wisconsin Press.
Orlikowski, W.J. & J.A. Yates (1994). Genre repertoire: the structuring of communicative practices in
organizations. Administrative Science Quarterly 39: 541574.
Paltridge, B. (1997). Genre, frames and writing in research settings. John Benjamins.
(2002). Tesis and dissertation writing: an examination of published advice and actual prac-
tice. English for Specifc Purposes 21: 125143.
Pilegaard, M. & F. Frandsen (1996). Text type. In J. Verschueren, J.-O. stman, J. Blommaert &
C. Bulcaen (eds.) Handbook of Pragmatics. John Benjamins.
Reid, I. (ed.) (1987). Te place of genre in learning: current debates. Deakin University.
Rothery, J. (1996). Making changes: developing an educational linguistics. In R. Hasan & G. Williams
(eds.) Literacy in society: 86123. Longman.
Samraj, B. & L. Monk (2008). Te statement of purpose in graduate program applications: genre
structure and disciplinary variation. English for Specifc Purposes 27: 193211.
Schryer, C. (1994). Te Lab vs. the Clinic: sites of competing genres. In A. Freedman & P. Medway
(eds.): 105124.
Sidney, P. (1962). [1595]. Te prose works of Sir Philip Sidney. Volume III. (ed. by A. Feuillerat)
Cambridge University Press.
Smart, G. (2006). Writing the economy. activity, genre and technology in the world of banking. Equinox.
Swales, J. (1990). Genre analysis: English in academic and research settings. Cambridge University
Press.
(1998). Other foors, other voices: a textography of a small university building. Lawrence
Erlbaum.
(2004). Research genres. Explorations and applications. Cambridge University Press.
134 Anna Solin
Swales, J. & C. Feak (2004). Academic writing for graduate students. Essential tasks and skills.
University of Michigan Press.
Tardy, C.M. (2003). A genre system view of the funding of academic research. Written Communication
20: 736.
Ventola, E. (2005). Revisiting service encounter genre some refections. Folia Linguistica 39: 1942.
Werlich, E. (1983). A text grammar of English. Quelle & Meyer.
Wodak, R. (2006). Critical linguistics and critical discourse analysis. In J. Verschueren, J.O. stman
(eds.) & E. Versluys (coll.) Handbook of Pragmatics. John Benjamins.
Yates, J.A. (1989). Control through communication. Te rise of system in American management. Te
Johns Hopkins University Press.
Internet sources
SIGET 2007. Proceedings of the 4th international symposium on genre studies, SIGET. (University
of Southern Santa Catarina, Brazil.). Available at http://www3.unisul.br/paginas/ensino/pos/
linguagem/cd/index1.htm. (Accessed 4 December 2008.)
Humor
Salvatore Attardo
Texas A&M University-Commerce
1. Introduction and defnition
Te conceptual feld of humor is broad and only few areas within it are well
determined. Attempts at defning and subcategorizing areas within the feld such as
humor vs. comedy or ridicule have by and large failed. Lexicographic studies have
only highlighted the diferences and fuidity of the classifcations used by various lan-
guages (Attardo 1994: 27). Te term humor has emerged as technical term to be
intended as covering anything that is (or may) be perceived as funny, amusing, or
laughable. Tis does not preclude the possibility of establishing subcategorizations in
certain specifc areas, e.g. tendentious humor, or genres such as puns, jokes, etc. It
should be noted, however, that terms such as pun and joke are not technical terms
and are ultimately fuzzy. On these grounds, some have challenged the possibility of
providing a unitary account of humor (e.g. Ferro-Luzzi 1990). A case for an essen-
tialist account of humor, and a refutation of the arguments against it, is presented in
Attardo (1994). By and large, linguists (as well as scholars from most disciplines) have
operated on the assumption that humor is universal (cf. Apte 1985).
Te history of humor research has a long and prestigious gallery of scholars trying
to describe and explain the phenomena surrounding humor. Going back to Plato,
Aristotle, Cicero and Quintilian, and including Kant, Schopenhauer, Freud, Bergson,
and Pirandello, the history of refection on humor looks surprisingly like the history
of western culture itself. Certainly humor has been an open question for two millen-
nia. Reviews of the history of humor research can be found in Keith- Spiegel (1972),
Morreall (1983;1987), Raskin (1985), and Attardo (1994). Te latter has a particular
focus on linguistic issues.
2. Referential and verbal humor
Linguistics and before it the refection on language does not contribute much to
research on humor until the 80s. Before then, with few exceptions, puns and other
wordplay were the only subjects deemed interesting to or treatable by linguistics. Te
distinction between referential and verbal humor (introduced by Cicero and rediscov-
ered by more or less every scholar since then under diferent terminologies) captures
136 Salvatore Attardo
the diference between puns, which involve a reference to the surface structure of the
linguistic expression, and other forms of humor which do not. Te reference to the
signifer can take many forms: the most common is similarity (paronymy) and its limit
case identity (homonymy), other rarer forms include alliteration and syntactic ambi-
guity. What seems to have been lost on most researchers is that beyond this reference
to the surface form of (part of) the text, verbal and referential humor share the same
deep semantic and pragmatic mechanisms. Overviews of linguistic research about
puns and related phenomena can be found in Sobkowiak (1991) and Attardo (1994).
3. Semantics
Psychologists and other cognitively oriented scholars elaborated a model of the cog-
nitive processing of humor known as the incongruity-resolution (IR) model (see
Forabosco 1992 for a review of the research). Te concept of incongruity has a long
history in the psychology of humor, where it is taken as basic. A typical defnition
(McGhee 1979: 67) for example reads:
Te notions of congruity and incongruity refer to the relationship between components
of an object, event, idea, social expectation, and so forth. When the arrangement of
the constituent elements of an event is incompatible with the normal or expected
pattern, the event is perceived as incongruous. Te incongruity disappears only when
the pattern is seen to be meaningful or compatible in a previously overlooked way.
Te resolution phase of the humorous perception has been called justifcation
(Aubouin 1948), local logic (Ziv 1984), and sense in non-sense (Freud 1905). All
the works on the resolution phase of the IR model emphasize that the resolution is
not complete (i.e. the incongruity persists afer the resolution) and ofen stress the
playfulness or even the fallacious nature of the resolution. A common conceptualiza-
tion is to view IR models as two stage (Suls 1972) models. Te two stages involve
giving the text a frst interpretation which is then rejected in favor of a second inter-
pretation. Te two interpretations must coexist, at least to the extent that they are to
be judged incongruous.
3.1 The isotopy-disjunction model
An approach to humor which is entirely compatible with the IR model emerged from
European structural semantics and narratology in the work of Greimas (1966) and
Morin (1966). Widely infuential in Europe (see a survey of its infuence in Attardo
1988, 1994) it has been by and large ignored in the US (with few exceptions). Te
model, which Attardo et al. (1994) dub the Isotopy-Disjunction Model (IDM), is based
on the concepts of isotopy and of narrative function. Te text of a joke is seen as
Humor 137
developing one isotopy and then switching to a second isotopy. From a narrative view-
point, the text of the joke is divided in three functions, which set up, complicate and
resolve the narration, respectively. Within the text of the joke, the IDM distinguishes
two signifcant elements: a connector, the (optional) ambiguous or paronymic element
which allows the passage from the frst to the second isotopy, and the disjunctor, which
causes the passage from one isotopy to the other. Te connector is present only in ver-
bal humor. Te IDM, in its original formulation, runs into several problems, due to the
loose defnitions of the central concepts of isotopy and (narrative) function.
3.2 The script-based semantic theory of humor
Quite independently from the IDM, in the late 70s, a new approach to the semantics of
humor was presented by Raskin (1979, 1985), based on the concept of script. Scripts
are intended as a cover term for such AI originated concepts as frames, schemata,
daemons, and scripts; on script-based semantics see Raskin (1985, 1986), Fillmore
(1975, 1985) and in general the papers in the roundtable published in Quaderni di
Semantica (Raskin 1985b). Te Semantic Script Teory of Humor (SSTH) advocated
a new approach to humor theory and presented a falsifable theory of humor based on
two conditions: (1) the text be (at least in part) compatible with two scripts; (2) the two
scripts are opposed, in a technical sense, i.e. they are local antonyms. A list of typical
oppositions is provided, which include actual vs. non-actual and normal vs. abnor-
mal and such subtypes as good vs. bad sex vs. no-sex, etc. Te SSTH also introduces
terminology to account for the linguistic element causing the switch from one script
to the other (script-switch trigger). Te SSTHs rigorous presentation and theoretical
baggage ensured its success and it has been widely applied to various forms of humor
(although it was conceived as applying to jokes). A review of the SSTHs applications
can be found in Attardo (1994: 195229).
Te limitations of the SSTH were immediately pointed out by Raskin himself in a
series of papers and conference presentations (see Attardo 1994: 214219 for a review)
and eventually led to its revision in Attardo and Raskin (1991). Te General Teory of
Verbal Humor (GTVH) is presented as an expansion of the SSTH which it accepts as a
signifcant part of its theoretical apparatus. Essentially, the GTVH addresses three basic
issues which the SSTH does not highlight: (1) the fact that the SSTH is explicitly limited
to handling jokes; (2) the fact that some jokes are perceived as being more or less similar
to others; and (3) the fact that verbal and referential humor difer (although not seman-
tically). To account for these issues the GTVH introduces other Knowledge Resources
(KR) beyond the basic script opposition/overlap (symbolized SO). Tese include,
the LAnguage of the joke (i.e. its surface structure) signifcant for puns, but also for the
placement of the punch line), the SItuation, which includes the subject matter of the
narrative of the text, the TArget, (roughly, the butt of the joke), the Narrative Strategy,
corresponding to the genre of the text (e.g. question and answer, knock-knock joke,
etc.), and the Logical Mechanism, which is the mechanism whereby the SO is intro-
duced and may correspond to the resolution phase of the processing (see a suggestion
to that efect in Forabosco 1992: 59). Most of Attardo and Raskin (1991) is dedicated to
ranking the various KRs to determine which ones afect more directly the perception
of similarity among jokes. Te fnal ranking is (from most to least signifcant) SO, LM,
SI, TA, NS, LA. Empirical support for the GTVH is presented in Ruch et al. (1993) and
Forabosco (1994). Extensive debate and critical assessment of the SSTH/GTVH can
also be found in the papers collected in Attardo (ed. 2004).
3.3 Longer texts
Te SSTH and GTVH have been so far primarily confned to jokes (with some excep-
tion, e.g. (1987). Attardo (1994: 254270; 1996) presented some analyses and a call
for further research, respectively. Tis area of research appears particularly difcult
because of the intersection of felds of research as disparate as narratology, natural
language processing, literary criticism and text linguistics, not to mention humor
research. Recent work has seen the development of an extension of the GTVH directly
tailored to the analysis of long humorous texts (Attardo 2001, 2002a; see also Tsakona
2003; an alternative approach is presented in Chlopicki 1997, 2001).
4. The cooperative principle and humor
4.1 Grice and Gricean analyses
All jokes involve violations of one or more of Grices maxims. Te claim that jokes
could be viewed in terms of fouting of conversational maxims dates back to Grice
himself, who considers irony (as an example of implicature), and a (complex) pun. A
review of the applications of Grices insight to humor analysis can be found in Attardo
(1994: 272 f.). An analysis of irony in Gricean terms would be out of place in this
context (but see Attardo 2000 for a literature review), but puns fall squarely under the
realm of humor.
4.2 Humor as non-bona-fde communication
Raskin (1985: 100104) brought a new concept to the analysis of the pragmatics of
humor: by defning humor as a non-bona-fde (NBF) mode of communication he
excluded the possibility of accounting for it straightforwardly within the realm of
serious non-humorous language. Essentially, Raskin distinguishes, as does Grice,
between a bona-fde type of communication, in which the speaker is committed to
communicating in the most efective way, as clearly as possible, etc., in short follows
Humor 139
the CP. Humor, just as lying, involves a diferent mode of communication which does
not abide by the CP (i.e. is NBF). In the case of jokes, the perlocutionary goal of the
speaker is not to convey information but rather to elicit a humorous reaction in the
hearer. It can be overt or consist merely of the hearers recognition of the speakers
intention. Note that the use of hearer and speaker does not imply that these remarks
are limited to spoken language(s).
Raskin, to better illustrate the diference between NBF modes (such as humor)
and other modes governed by Grices CP, provided a set of maxims for joke-telling
directly paraphrased from Grice (Give as much information as is necessary for the
joke). Raskin immediately proceeded to deny any theoretical interest in these max-
ims: they do not really provide an explicit account of the semantic mechanisms of
humor (1985: 103104). Despite this explicit warning, the four joking-maxims have
been widely quoted at face value in the literature.
Te fundamental insight of Raskins discussion is that jokes do not merely fout
but violate a maxim. When the joke teller introduces a frst script he/she deliberately
misleads the hearer into believing that that script is central to the processing of the
text, only to reveal again deliberately at the end of the text that the script was in fact
incompatible with the one introduced by the script-switch trigger/disjunctor.
Attardo (1994: 271286) has systematized this approach, for example by showing
that violation of any maxim can produce a joke, and has argued against two possible
strategies that deny that any real violation of the CP takes place in jokes by argu-
ing that the violation is only mentioned, and not actually performed in the text (cf.
Yamaguchi, below).
Tis approach to humor as CP violation runs into an obvious problem: jokes are
ofen perceived as not being totally devoid of communicative efect, while the violation
theory would seem to predict that no communicative import could follow a violation
of the CP (Attardo 1994: 274275). Te apparent paradox is solved when one consid-
ers that jokes communicate on the basis of the presuppositions that the text may have
independently of its humorous nature, on the basis of metamessages (of the kind, I
think that it is an appropriate situation to be facetious), or on the basis of the suppres-
sion of the incongruity (i.e. the hearer takes the joke at face value, refusing to interpret
it as non-cooperative; this is common in teasing, cf. Drew 1987).
4.3 Relevance-theoretic approaches to humor
Sperber and Wilsons (1986) Relevance Teory (RT) has been one of the most success-
ful post-Gricean approaches to pragmatics. Relevance theoretic accounts of humor
necessarily have to contend with the basic awkwardness of the postulate that the prin-
ciple of relevance (RP) is exceptionless (Sperber & Wilson 1986), since humor is based
on a violation of the CP or the RP.
For example, Yamaguchi (1988) acknowledges that jokes violate Grices conversa-
tional maxims. He then proposes the Character-Did-It hypothesis, based on the men-
tion theory (Sperber & Wilson 1981), which places the responsibility for the violations
of the CP (at least in part) on one of the characters in the text (Yamaguchi 1988: 327).
Tis hypothesis has been refuted in Attardo (1994: 278282), but Yamaguchis work
is interesting in his elaborate description of the strategies used. Yamaguchi notes also
that the violation of the maxims is hidden away in the text in a paradoxical attempt at
dissimulation bound to failure, since the joke will inevitably foreground the violation
so laboriously dissimulated in the text. In this respect, Yamaguchi joins the research on
the unsaid in humor (see below).
Jodlowiec (1991) presents an IR theory of jokes recast in relevance theoretic terms.
While her model does not depart signifcantly from other IR models, she emphasizes
that the interpretation of the second sense is done on the basis of the principle of rel-
evance, just like the frst one. She also emphasizes the fact that some jokes rely more
than others on implicit information. Unfortunately, Jodlowiecs work is uneven and
some of her claims are patently incorrect (such as those that no generalizations can
be found about verbal jokes or that jokes do not breach the CP). On Jodlowiecs work
see also Chlopicki (1994).
Curcs work (1995, 1996a, b, 1998) is much more carefully presented. She also
presents a two stage IR model formulated in RT terms. In her terminology, the hearer
entertains a key assumption (essentially a proposition consistent with the frst inter-
pretation of the text) and subsequently a target assumption (a proposition consistent
with the second interpretation of the text). By noting that the target assumption typi-
cally represents an attributable thought (1996b:62), Curc manages to connect her
theory of humor to RTs echoic (formerly known as mention) theory of irony. As
mentioned above, Curcs theory is carefully hedged and she calls attention to the fact
that in its present state it is not meant to account for all humorous utterances.
A recent account of RT-based accounts of humor can be found in Yus 2003. Yus
also presents his own IR approach, which is however general, i.e. applies to all types of
humor. Yus is aware that RT cannot accept the idea of a violation of the principe of rel-
evance, which is axiomatically unviolable, he therefore claims that no violation of the
principle of relevance takes place. Tis apparently contradictory claim (since incon-
gruity could only follow by the presence of irrelevant i.e. RP violating material)
is salvaged by the claim that the speaker of a joke, for example, frst uses the RP to
induce the hearer to interpret the joke in a given way and then again uses Relevance to
bring about the second interpretation. With Yus account, any diference between RT
accounts and other IR accounts based on pragmatic inferences (such as the SSTH
pace claims to the contrary) disappears. Yus work consists largely of numerous well-
crafed analyses of how diferent types of assumptions can be retrieved in various
contexts. His is by far the most accomplished RT account of humor.
Humor 141
Let us note that, as one would expect, all RT accounts place more emphasis on the
process of interpretation than on the text itself. Finally, attention has been drawn on
the metarepresentational aspect of irony (i.e. the idea that irony involves a representa-
tion of another representation (Curc 2000; Colston & Gibbs 2002).
4.4 Informativeness approach to jokes
De Palma and Weiner (1990, 1992), Weiner and De Palma (1993), Weiner (1996, 1997)
have developed an account of humor (in riddles, but the model is extensible to many
other types of humor), which is in part based on the SSTH, and assumes the basic IR
model, but seeks to enrich it with the idea of accessibility based on the notions of
prototypicality ( la Rosch 1973), salience (for example, a salient feature of chairs is
that they are used for sitting), and of parallelism (i.e. the tendency to continue in the
same frame be it semantic, syntactical or pragmatic once one has been activated). Te
frst script is highly accessible and based on a neutral context, whereas the second
script is much less accessible and strongly context dependent. Once the frst script has
been activated the tendency to parallelism will tend to keep the hearer/reader within
the frst script until this becomes impossible because of the occurrence of the punch
line. Tis account improves the SSTHs oppositeness requirement by making it more
specifc (high vs. low accessibility, neutral vs. specifc context) and by dispensing with
the list of hardwired oppositions.
Independently from De Palma and Weiners work, and in part against it, since
she is critical of the oppositeness requirement of the SSTH, Giora (1991) sets out to
explain the surprise efect of humor. Her theory states that a well-formed text should
start with the least informative material and gradually introduce more informative
material. Informativeness is defned in terms of reduction in the number of alterna-
tives. Great informativeness will thus correspond to least predictability and hence sur-
prise value. Furthermore, Giora introduces the concept of marked informativeness
to design the marked members of a set (those that ft less well a particular category).
Tus, if, afer declaring that one has seen a bird, one specifes that one saw a penguin
one is being not only informative, but markedly so.
Giora can then defne jokes as texts that do not gradually introduce more infor-
mative material but that in fact end with a markedly informative element. Te rest
of Gioras model corresponds more or less to the IDM, although Giora stresses the
abrupt passage from the frst to the second sense/interpretation of the text (471). If
one flls in the gap between the two interpretations with information that gradually
brings about the second one, the joke is no longer perceived to be funny (475).
Tis point has great signifcance because it may provide a tool to solve the notori-
ously complex issue of how to distinguish humor from other types of texts which also
include a punch line and the ensuing IR efect but are nor perceived as humorous
(e.g. detective stories, cf. Navon 1988: 209; Wenzel 1989; Giora 1991: 482483; Weiner
1997). Also, this is clearly related to the issue of the unsaid in humor (cf. Dascal
1985: 9899; Dolitsky 1992). In order to preserve the element of surprise the text of
the joke must not allow the second interpretation/script to become available until the
punch line.
Gioras analysis is also clearly compatible with the functional sentence perspec-
tive analysis of jokes punch line as the rheme of the last sentence of the text in which
they occur (Attardo et al. 1994). In fact, Giora herself points out the similarities
between her approach and FSP (Giora 1988: 550n). Recently, Gioras approach has
been fully articulated in Giora (2003). Besides Gioras own psycholinguistic work,
empirical support for her theory (and some aspects of the GTVH) has been provided
by Vaid et al. (2003).
4.5 Two-stage processing of humor
It should be noted that the IDM, the SSTH/GTVH, the violation of the CP, most of the
Relevance Teoretic, and the informativeness theory approaches to humor are (just
like the IR model to which they ultimately subscribe) two stage processing theories
(Suls 1972). Recently, two stage approaches to implicature have come under attack by
psycholinguistic research (cf. Gibbs 1994 and references therein). Te approaches to
humor theory presented in this discussion all would fall squarely under Gibbs attack
to the notion of literal meaning.
In fact, Gibbs (1989) has presented (in a marginal discussion, to be fair) an attack
against the two step processing of humor embodied in mainstream humor research,
claiming that the humorous efect would not derive from the shif from a literal to
secondary (sic) meaning in the text (1989: 249) but in the shifing of speakers inten-
tions (Emphasis in the original). It seems clear that Gibbs position is here in error, as
the speakers intentions may be unavailable to the hearer/reader and are in all cases
more or less irrelevant to the perception of humor (see for instance the so-called unin-
tentional humor which consists of the humorous interpretation of actions/utterances
that were specifcally not meant to be humorous). Regardless of Gibbs own position
on the processing of jokes, one can construe a potential attack against the IR models
based on the denial of the central tenet of the IR models that one sense is processed
in isolation prior to the accessing of the second one. However, recent work by Giora
(1995) and Giora and Fein (1996) has shown that Gibbs fndings which led him to
the rejection of two step theories can be analyzed in ways consistent with a two step
theory of processing and thus consistent with the IR models. Gibbs has replied to these
attacks in Gibbs (2002). It should be emphasized that Gibbs work is primarily centered
around irony and other forms of non-literal meaning and thus is not directly relevant
to humor.
Humor 143
5. Conversation analysis
Conversation analysis (CA) here taken as a cover term for discourse analysis and
research in ethnomethodology has dedicated quite some work to humor. Overviews
can be found in Sherzer (1985), Norrick (1993: 139164), Attardo (1994: 293331,
2001a:6169), Kotthof (1996), Glenn (2003), and several papers in a special issue of
the Journal of Pragmatics on humor (Attardo ed. 2003.). To the publications reviewed
in those sources can be added the following: Fillmore (1994), Eggins and Slade
(1997: 155167), Priego Valverde (1998, 2003), Nardini (2000), Buttny (2001), Viana
(2001), Scheglof (2001), and Branner (2003).
When considering the CA of humor a distinction between canned and conversa-
tional jokes is ofen introduced: essentially, a canned joke is a joke that has less contex-
tual ties than a conversational one and thus can be recycled more easily (see Attardo
1994: 295299 for references). Canned jokes prototypically exist in written or oral
repertoires, while conversational jokes are improvised. It has been argued convinc-
ingly (Zajdman 1991) that the distinction is in fact a continuum that can range from
(almost) total lack of contextual ties to the point where the joke is indistinguishable
from the rest of the non-humorous discourse (in this case, obviously, the speaker does
not signal that humorous nature of the joke).
5.1 Canned jokes in conversation
Te most infuential work in the CA of jokes is without discussion Sacks. Sacks
describes what he calls the sequential organization (1974: 337) of joke-telling in con-
versation. He fnds that it breaks down into three parts: the preface, the telling, and
the reaction.
5.1.1 Preface
Canned jokes are (mostly) narratives and hence they need to be introduced in the
fow of conversation, for example so that the narrator may be granted the foor for an
extended turn. Te preface has three main functions: securing the acceptance of the
joke-telling form the audience, negotiating the acceptability of either the joke or the
telling of one, and clueing the audience on the correct (i.e. humorous) interpretation
of the text.
5.1.2 Telling
Te telling being a narrative, consists of only one turn, by one speaker. Other turns within
this part will generally be perceived as interruptive (although there are exceptions, e.g.
interactive jokes, such as knock-knock jokes, which require the audiences interven-
tion). Sacks sees the telling of the joke as the setting up of a puzzle for the audience,
a sort of understanding test. Tis idea, albeit widely accepted in the literature (e.g.
Norrick 1993; Glenn 2003) seems to be in error (Attardo 1994: 305307; 1995: 8182).
5.1.3 Response
Te literature in CA analyzes three classes of responses: laughter, delayed laughter,
and silence. Of these, laughter has received by far the greatest amount of attention
(see Attardo 1994: 307309 for references). Signifcantly, laughter is not an uncon-
trolled reaction to the humorous text (Jeferson 1985) and can be used for a variety
of purposes. For example, laughter is used to negotiate the humorous interpretation
of the text by the speaker, who invites laughter by laughing at the end of the telling
(Jeferson 1979).
Delayed laughter is explained by the presence of two conficting desires in the
audience: display understanding of the text and check the rest of the audiences reac-
tions. Silence can manifest the audiences disapproval of the materials in the telling
or can be used by the joke teller to keep in character with the narrative convention
whereby a narrator should believe what he/she tells. In other words, the teller is giving
his/her audience time to get the joke.
5.2 Conversational humor
Conversational humor has been the most active feld of humor research in linguistics,
since the mid-nineties to date. Two strands of research have emerged: quantitative and
functional. Sacks original work tipifes the functional approach, which tries to describe
the various functions to which humor can be put and emphasizes detailed analysis of
single cases. More quantitatively oriented work, such as Tannens and Holmes (see
below), relies on larger corpora and tends more toward generalizations. Needless to
say, this distinction is purely classifcatory and authors incorporate quantitative data
as they are available (although see Scheglof 1993: 103104 and Glenn 2003: 3839 for
an argument against quantitative data, or at least indiscriminate quantitative data).
5.2.1 Functional conversational analyses
As opposed to canned jokes, conversational humor is improvised and with great con-
textual ties. Puns have been the object of some research in CA. Sacks (1972) noted that
puns may occur in proverbials (i.e. formulaic expressions). Sherzer (1978, 1985, 1992,
1993) has much broadened the scope of analysis to include anthropological, psycho-
logical, and ethnological considerations that put puns in an area that includes speech
play and verbal art (see Kirschenblatt-Gimblett 1976 and Sherzer 1992). Sherzers
(1978) analysis is based on the idea that puns are cohesive since they ofen sum up,
or conclude a topic (given their occurrence in proverbials). While unintentional puns
tend to be cohesive and do not imply linguistic manipulation, intentional puns will
Humor 145
be disjunctive and manipulative. On puns in CA, see Attardo (1994: 311317) and
Norrick (1993: 161).
Norrick (1993) analyzes a corpus of familiar conversations. His most interest-
ing point is that spontaneous joking serves many functions in our everyday talk
(129) and is thus an important and signifcant component of conversational inter-
action. Overall, his analysis of the social functions of humor follows the humor as
social facilitator and as social corrective lines of thought (cf. Long & Graesser 1988;
Bergson 1901). For a recent overview of Norricks work, which has developed steadily
in the intervening decade, see Norrick (2003). Glenn (2003) departs somewhat from
the typical perspective of humor research in CA because it focuses on laughter and
not humor. Glenn embraces a social/communicative theory of laughter, i.e. that sees
laughter as controlled, at least in part, by the laugher and exploited for communica-
tive purposes. For example, shared laughter is initiated via an invitation-acceptance
sequence. Te speaker may place laughter particles within or afer his/her turn. Te
hearer, in turn can accept or decline the invitation to laugh. Shared laughter tends to
be short lived and that speakers need to renew shared laughter. Methods for doing
so include, renewing laughter, repeating the cause of laughter, or saying something
else funny (cf. Hay humor support, below). Turning to the more strictly functional
approach, Glenn fnds that laughing along, while not outright afliative, makes more
of the same laughable (teasing or impropriety) relevant and may lead to clearer dis-
plays of afliation (131). However, laughter can also be used to resist afliation (i.e. to
reject the implicit ofer of in-group membership extended by the joker). Analyses of
humor along the functional paradigm have been performed also on German texts, see
Kotthof (1986, 1996, 1998, 2000, 2003), Branner (2003).
Along the same lines of inquiry, Everts (2003) introduces the idea of family
humor style (see below on humor style) in which family members are shown to
share characteristics in humor, such as imitation and impersonation, to establish
and foster intimacy. On the functions of humor in interactions, see also Priego-
Valverde (2003), and also Attardo (1994: 322330), Long and Graesser (1988) and
Graham et al. (1992). A more sociological, but very signifcant work on the func-
tions of humor is Mulkay (1988). Finally, it should be noted that Attardo (2002b) has
claimed that all linguistically accessible functions can be performed by humorous
means, except of course when prohibited by context, a conclusion also reached by
Priego-Valverde (2003).
5.2.2 Quantitative conversational analyses
In contrast to Sacks analysis of one single occurrence of a joke in a conversation,
Tannen (1984) records and analyzes all the humorous occurrences in the conversa-
tions held at a Tanksgiving dinner (it should be noted, in passing, that humor was not
her central focus). Tannen (1984: 130) claims that One of the most distinctive aspects
of any persons style is the use of humor and provides a detailed analysis of the style
of humor of each of the dinner guests.
Quantitative analyses of the Tanksgiving conversation reveals that humor has a
larger role than expected in conversation. It appears that two speakers had 11% of their
turns counted as humorous. Te lowest fgure, i.e. the speaker with the least humorous
conversational turns, was 2%. Tese results are somewhat surprising, as it seems that
the presence of humor in conversation has been downplayed in analyses of humor as a
deviation from a serious norm. Te use of humor in conversation is found to stand
out. Tannen sums up by saying that humor makes ones presence felt (1984: 132).
Te most signifcant development in the feld has been the work of the New
Zealand-based sociolinguists, associated with the work of Janet Holmes. Both Hays
work and Holmes are based on large corpora consisting of hundreds of instances of
spontaneous humor and many hours of recordings. Hay (1994) is one of the few stud-
ies that deals with failed humor, but most analysts merely ignore the issue or when
they are aware of it, dismiss it for simplicitys sake (cf. Holmes 2000: 163). Hay (2000)
fnds women more likely to use humor for solidarity than men; men use humor to
solve conficts. Hay (1995, 2000, 2001) has investigated humor support i.e. conversa-
tional strategies used to acknowledge and support humorous utterances, among which
fgures prominently the production of more humor.
Holmes work on humor revolves around a large study of conversation in the work-
place (see Holmes & Stubbe 2003, for an overview). Her approach is functional, for
example she concludes that in work contexts humour can be used by subordinates as a
subtle (or not so subtle) license to challenge the power structure, as well as by those in
power to achieve the speakers goal while apparently de-emphasizing the power diferen-
tial. (Holmes 2000: 176). Te same subversive function of humor, along with the more
conventional expression of friendly feelings and positive politeness is found in another
study (Holmes & Marra 2002). However, because of the quantititive focus Holmes can
make extremely signifcant generalizations: the overall amount of humour produced
by the women is greater than that produced by the men (Holmes 2001: 93). Tis fnd-
ing fatly contradicts folk humor theorists and two decades of feminist research which
both contended that women produce less humor (since there is no reason to believe
that New Zealand women are funnier than their American or European counterparts).
Similarly signifcant fndings concern to social status: the chair [of a meeting] makes a
disproportionately high contribution to the humour in most meetings (2001: 96) (32%
of instances, overall) Tis is true also of multi-turn humor.
Strangely, CA of humor is largely unafected by the research in humor studies,
including in linguistics. Exceptions are Antonopoulou and Sifanou (2003) a particu-
larly interesting study, based on spontaneous humor in Greek telephone conversa-
tions, which is informed by recent theoretical work in humor research, and Norrick
(2003) which attempts a synthesis precisely of CA and humor research.
Humor 147
6. Sociolinguistics of humor
As Gasquet-Cyrus (2004) rightly argues, the relationship between sociolinguistics and
humor research can be characterized as having mutually missed the boat. It is only
very recently that research aware of the advances of humor research has begun to
appear and that competence (in the Chomskian sense) theorists have begun to take
notice of ethnomethodological and sociolinguistic work. In fact, one could argue that
before Norrick (1993) and Attardo (1994: ch. 10) the interplay had been virtually
nonexistent (with the obvious exceptions of anthropological work, e.g. Basso 1979;
Beeman 1981a/b; see also Beeman 2000). Te situation is however changing fast. How-
ever, things are beginning to change, witness Gasquet-Cyrus own work on urban dia-
lectology in Marseilles, in which humor is central (2004), Crawfords work on gender
(1989, 1995, 2003) and a recent crop of researchers (Georgakopulou 2000; Rutter 2000;
Liao 2001) whose work is beginning to appear.
6.1 Gender diferences
Lakofs provocative claim that women have no sense of humor (1975: 56) opened
the way for serious study of gender-related diferences in humor appreciation and pro-
duction. Most of the research on gender-based diferences on the use of humor in
conversation seems to converge on the idea that the type of language used by women
directly infuences the type of humor that they use and appreciate.
An excellent synthesis of research on humor and gender comes from anthropol-
ogy. Apte (1985: 6781) argues that women are prevented from engaging in the same
range of humorous exchanges by men, who thus emphasize their need for superiority
(1985: 81) and that restrictions on the kinds of humor in which women can engage
ofer an important avenue to social control (Ibid.). Crawford (1989, 1995: 129169)
reviews the scarce literature on the subject and presents an interesting discussion criti-
cizing the bulk of psychological research on the gender-based diferences in humor
and advocating the use of CA (among other tools) as a way to overcome the andro-
centric approach of much psychological research on humor and gender.
Jenkins (1985) and Ervin-Tripp and Lampert (1992) have noted that women
are more likely to create cooperative humor (stacked humor) and that both male
and female speakers change their joking styles in mixed-gender groups. Specif-
cally, women decrease the number of self-targeted humor (Ervin-Tripp & Lampert
1992: 116). Kothof (1986) and Tannen (1986, 1990, 1994) report various examples
of gender-based diferences in the use of humor. For example, women tend to tell
less jokes than men in mixed-gender groups and prefer smaller audiences than men
(Tannen 1990: 8990). Tere is a growing literature on humor and gender from
a feminist but not necessarily linguistic perspective, collected and reviewed in
Nilsen (1993: 2328). A recent review of the research on gender and humor can be
foun in Crawford (2003), which is however unaware of the fndings in Holmes (2001),
discussed above. A re-evaluation of these theoretical approaches based on these data
has yet to take place.
6.2 Ethnicity and humor
Under this heading we can classify two strands of humor research: one that investi-
gates the targets of humor (and this is the common understanding of the phrase eth-
nic humor) and one that investigates the diferences in sense of humor across cultures.
Te study of ethnic humor is the province of folklorists, anthropologists, and
sociologists. Beyond the synthesis in Apte (1985: 108148), the classical reference is
Davies (1990) which presents the stimulating hypothesis that ethnic humor is tar-
geted at geographical outsiders, i.e. groups of people who live at the boundaries of
whatever happens to be the homeland of the group doing the joking (e.g. in France
the jokes about Belgians, in Canada about Newfoundlanders, etc.). Apte notes that
ethnic humor is the result of urbanization, since prevalently rural societies do not have
enough contacts with other ethnic groups to make such humor possible, for example
because the necessary stereotypes will be missing. Incidentally, it should be noted that
the stereotypes on which ethnic humor is based are entirely mythical (on the concepts
of ethnic group and stereotype see Apte 1985).
Te study of cross-cultural variation in sense of humor is largely anecdotal and
consists of isolated descriptions of the humor of a given society. Ziv (1988) presents
a collection of articles covering mostly European states. Morain (1991) describes a
study contrasting ESL students and American students ratings of New Yorkers car-
toons and underscores the necessity to possess a given cultural script to be able to
understand the humor, let alone appreciate it. To date there are no large scale cross-
cultural comparisons of humor available. Tis state of afairs is regrettable, as the claim
of universality of humor (and of its mechanisms) could be verifed by such a study.
An exception is the study of puns, which are widely documented in a great variety of
languages (cf. Hausmann 1974 and Sobkowiak 1991 for a bibliography). However, the
universality of puns has recently been challenged (Sherzer 1996: 134).
7. Computational humor
A very recent development is the emergence of research in the computational treat-
ment of humor. A (critical) bibliography of early eforts in that direction can be found
in Raskin and Attardo (1994). For the time being, it appears that little signifcant arti-
fcial intelligence is used in any of the systems.
Humor 149
Te best defned and implemented system is JAPE a system that can generate a type
of punning riddles (Binsted & Ritchie 1994, 1997). An attempt has been done at inte-
grating JAPE in a real-life on-line communicative situation with mixed results (Loehr
1996). Te papers presented at the frst conference on computational humor (held in
1996 at Twente University in the Netherlands) are collected in Hulstijn and Nijholt
(1996). It is too early to determine in what directions the research in this area will
engage and if the results will be theoretically signifcant. An interesting issue, which
caused already some debate, is whether it is necessary to have a full-fedged natural
language processing system before attempting the generation of humor. Raskin and
Attardo (1994) have taken a positive stance on this point, while Binsted and Ritchie
take the opposite stance.
Recent work in the feld has seen further conferences (Stock et al. 2002) and sig-
nifcant publications (Stock 1996; Stock & Strapparava 2003; Ritchie 2003).
8. Cognitive linguistics and humor
Recently, work in cognitive linguistics has begun to appear which seeks to challenge
the traditional linguistc analyses of humor, from the standpoint of cognitive linguis-
tics (see Brne and Feyaerts [2004] for discussion). For example, blending, a recent
development of cognitive linguistics (see Coulson & Oakley 2001), has been used
to analyze humor (Coulson 1996, 2001, in press). It is clear that blending, i.e. the
creation of a new mental space (domain, idea) out of existing, and not necessarily
related, other mental spaces, can account for some aspects of some types of humor
(insofar as it corresponds to the script overlapping aspect of the SSTH). However, it
is not clear, and no claims have been made to that efect, that it can provide a general
account of humor. Relatedly, neurological analyses have been used to argue for and
against the IR models, with the evidence seeming to support IR models (e.g. Derks
et al. 1997; Goel & Dolan 2001; Coulson & Kutas 2001). Te problem for cognitive
linguistic approaches to humor is to show that they are making claims that are empir-
ically and/or theoretically distinguishable from those of the previous theories they
seek to challenge.
9. Conclusion
While humor research has accomplished much in the last 30 years, it is clear that
there is still much to be done and that, while some of the central questions have
been tackled, we are still far from having a complete, let alone satisfactory, view of
the problem. But if we look at how much new research has appeared in the ten years
that have passed between the writing of the frst version of this entry and its update,
which has seen, for example, the emergence of the cognitive linguistic and variationist
sociolinguistic approaches, the challenging of some deeply entrenched beliefs in the
gender diferences of humor, and some signifcant results in the broadening of the
linguistic approaches to texts other than jokes, it becomes clear that the feld of humor
research is defnitely taking a pragmatic perspective and that signifcant fndings can
be expected in the future.
References
Antonopoulou, E. & M. Sifanou (2003). Conversational dynamics of humour: the telephone game in
Greek. Journal of Pragmatics 35(5): 741769.
Apte, M. (1985). Humor and Laughter. Cornell University Press.
Attardo, S. (1988). Trends in European humor research: towards a text model. HUMOR. Interna-
tional Journal of Humor Research 1(4): 349369.
(1994). Linguistic theories of humor. Mouton De Gruyter.
(1995). Review of Norrick (1993). HUMOR. International Journal of Humor Research 8(1):
7983.
(2000). Irony as relevant inappropriateness. Journal of pragmatics 32: 793826.
(2001). Humorous Texts. Mouton De Gruyter.
(2002a). Cognitive Stylistics of Humorous Texts. In E. Semino & J.Culpeper (eds.) Cognitive
Stylistics. Language and Cognition in Text Analysis: 231250. Benjamins.
(2002b). Semiotics and Pragmatics of Humor Communication. BABEL Aspectos de Filoloxia
Inglesa e Alemana: 2566. Rpt. in O. Leontovich (ed.) (2003). Communication Studies 2003.
Modern Anthology: 209252. Peremena.
(ed.) (2003). Te Pragmatics of Humor. Journal of Pragmatics 35(9) (Special issue).
(ed.) (2004). Festschrif for Victor Raskin. HUMOR: International Journal of Humor Research
17(4) (Special issue).
Attardo, S., D.H. Attardo, P. Baltes, & M.J. Petray (1994). Te linear organization of jokes: analysis of
two thousand texts. HUMOR. International Journal of Humor Research 7(1): 2754.
Attardo, S. & V. Raskin (1991). Script theory revis(it)ed: joke similarity and joke representation
model. HUMOR. International Journal of Humor Research 4(3/4): 293347.
Aubouin, E. (1948). Technique et psychologie du comique. OFEP.
Basso, K.H. (1979). Portraits of the whiteman: Linguistic play and cultural symbols among the west-
ern Apache. Cambridge University Press.
Beeman, W.O. (1981a). Why Do Tey Laugh? An Interactional Approach to Humor in Traditional
Iranian Improvisatory Teatre. Journal of American Folklore 94(374): 506526.
(1981b). A full arena: Te development and meaning of popular performance traditions in
Iran. In M.E. Bonine & N.R. Keddie (eds.) Modern Iran: the dialectics of continuity and change:
361381. State University of New York Press.
(2000). Humor. In Linguistic Lexicon for theMillenium. Journal of Linguistic Anthropology
9(1/2): 103106. Rpt. in A. Duranti (ed.) (2001). Key Terms in Language and Culture: 98101.
Blackwell.
Bergson, H. (1901). Le rire. Essai sur la signifcation du comique. Presses Universitaires de France.
Humor 151
Binsted, K. & G. Ritchie (1994). An implemented model of punning riddles. Proceedings of the
Twelfh National conference on Artifcial Intelligence (AAAI-94), Seattle.
(1997). Computational rules for generating punning riddles. HUMOR. International Journal
of Humor Research 10(1): 2576.
Branner, R. (2003). Scherzkommunikation unter Mdchen. Lang.
Brne, G. & K. Feyaerts (2004). Assessing the SSTH and GTVH: a view from cognitive linguistics.
HUMOR. International Journal of Humor Research 17(4): 361372.
Buttny, R. (2001). Terapeutic humor in retelling the clientstellings. Text 21(3): 303326.
Chlopicki, W. (1987). An application of the script theory of semantics to the analysis of selected
Polish humorous short stories. M.A. Tesis. Purdue University.
(1994). Review of M. Jodloviec. Te role of Relevance in the Interpretation of Verbal Jokes:
A Pragmatic Analysis. Ph.D. Jagiellonian University. (1992). HUMOR. International Journal of
Humor Research 7(1): 8793.
(1997). An approach to the analysis of verbal humor in short stories. Humor: International
Journal of Humor Research 10(3): 333347.
(2001). Humorous and non-humorous stories: Are there diferences in frame based recep-
tion? Stylistika X: 5978.
Colston H.R. & R.W. Gibbs (2002). Are Irony and Metaphor Understood Diferently? Metaphor and
Symbol 17(1): 5780.
Coulson, S. (1996). Te Menendez Brothers Virus: Blended Spaces and Internet Humor. In
A. Goldberg (ed.) Conceptual Structure, Discourse, and Language: 6781. CSLI.
(2001). Semantic Leaps. Cambridge University Press.
(In Press) Whats so funny: Conceptual Blending in Humorous Examples. In V. Herman (ed.)
Te Poetics of Cognition: Studies of Cognitive Linguistics and the Verbal Arts. Cambridge Uni-
versity Press.
Coulson S. & T. Oakley (2000). Blending Basics. Cognitive Linguistics 11(3/4): 114.
Coulson S. & M. Kutas (2001). Getting it: Human event-related brain response to jokes in good and
poor comprehenders. Neuroscience Letters 316: 7174.
Crawford, M. (1989). Humor in conversational contexts: beyond biases in the study of gender and
humor. In R.K. Hunger (ed.) Representations: Social Constructions of Gender: 155166. Baywood.
(1995). Talking Diference: On Gender and Language. Sage.
(2003). Gender and Humor in Social Context. Journal of Pragmatics 35(9): 14131430.
Curc, C. (1995). Some observations on the pragmatics of humorous interpretations: a relevance
theoretic approach. UCL Working papers in Linguistics 7: 2747.
(1996a). Te implicit expression of attitudes, mutual manifestness, and verbal humor. UCL
Working papers in Linguistics 8: 8999.
(1996b). Relevance theory and humorous interpretations. In J. Hulstijn & A. Nijholt (eds.):
5368.
(1998). Indirect echoes and verbal humor. In V. Rouchota & A.H. Jucker (eds) Current issues
in Relevance Teory: 304325. Benjamins.
(2000). Irony: Negation, echo and metarepresentation. Lingua 110(4): 257280.
Dascal, M. (1985). Language use in jokes and dreams: sociopragmatics vs psychopragmatics.
Language and Communication 5(2): 95106.
Davies, C. (1990). Ethnic Humor around the World. Indiana University Press.
De Palma, P. & E.J. Weiner (1990). When a riddle is not a riddle? Paper, International Pragmatics
Conference, Barcelona.
(1992). Riddles: Accessibility and Knowledge Representation. Proceedings of COLING 92:
11211125. Nantes.
Derks, P., L.S. Gillikin, D.S. Bartholome-Rull & E.H. Bogart (1997). Laughter and electroencephalo-
graphic activity. HUMOR: International Journal of Humor Research 10(3): 285300.
Dolitsky, M. (1992). Aspects of the unsaid in humor. HUMOR. International Journal of Humor
Research 5(1/2): 3343.
Drew, P. (1987). Po-faced receipts of teases. Linguistics 25: 219253.
Eggins, S.& D. Slade (1997). Analyzing Casual Conversation. Cassell.
Ervin-Tripp, S. & M.D. Lampert (1992). Gender diferences in the construction of humorous talk.
In K. Hall, M. Bucholtz & B. Moonwomon (eds.) Locating Power: Proceedings of the Second
Berkeley Women and Language Conference: 108117. Berkeley Women and Language Group.
University of California.
Everts, E. (2003). Identifying a particular family humor style: a sociolinguistic discourse analysis.
Ferro-Luzzi, G.E. (1990). Tamil jokes and the polythetic-prototype approach to humor. HUMOR.
International Journal of Humor Research 3(2): 147158.
Fillmore, C.J. (1975). An alternative to checklist theories of meaning. In C. Cogen et al. (eds.)
Proceedings of the First Annual Meeting of the Berkeley Linguistics Society: 123131. Berkeley
Linguistics Society.
(1985). Frames and the semantics of understanding. Quaderni di semantica 6(2): 222254.
(1994). Humor in academic discourse. In A.D. Grimshaw (ed.) Whats Going On Here? Com-
plementary Studies of Professional Talk: 271310. Academic Press.
Forabosco, G. (1992). Cognitive aspects of the humor process: the concept of incongruity. HUMOR.
International Journal of Humor Research 5(1): 4568.
(1994). Seriality and appreciation in jokes. HUMOR. International Journal of Humor
Research 7(4): 351375.
Freud, S. (1905). Der Witz und seine Beziehung zum Unbewussten. Berlin. Engl. tr. J. Strachey (1960).
Jokes and Teir Relation to the Unconscious. Norton.
Gasquet-Cyrus, M. (2004). Pratiques et reprsentations de lhumour verbal. Etude sociolinguistique
du cas marseillais. Ph.D. University of Aix-Marseille.
Georgakopoulou, A. (2000). On the sociolinguistics of popular fims: Funny characters, funny voices.
Journal of Modern Greek Studies 18: 119133.
Gibbs, R.W.J. (1989). Understanding and literal meaning. Cognitive science 13: 243251.
(1994). Te poetics of mind. Cambridge University Press.
(2002). A new look at literal meaning in understanding what is said and implicated. Journal
of Pragmatics 31: 457486.
Giora, R. (1988). On Te Informativeness Requirement. Journal of Pragmatics 12(5/6): 547565.
(1991). On the cognitive aspects of the joke. Journal of Pragmatics 16(5): 465485.
(1995). On irony and negation. Discourse processes 19: 239264.
Giora, R. & O. Fein (1996). Irony comprehension: the graded salience hypothesis. Paper, 12th Twente
Workshop on Language and Technology. Automatic Interpretation and Generation of Verbal
Humor, Enschede.
Glenn, P. (2003). Laughter in Interaction. Cambridge University Press.
Goel, V. & R.J. Dolan (2001). Te Functional Anatomy of Humor. Segregating Cognitive and Afec-
tive Components. Nature Neuroscience 4(3): S.237238.
Graham, E., M.J. Papa & G.P. Brooks (1992). Functions of humor in conversation: Conceptualiza-
tions and measurement. Western Journal of Communication 56(2): 161183.
Greimas, A.J. (1996). Smantique structurale. Larousse.
Grice, P. (1989). Studies in the way of words. Harvard UP.
Humor 153
Hausmann, F.J. (1974). Studien zu einer Linguistik des Wortspiels. Das Wortspiel im Canard Enchain.
Niemeyer.
Hay, J. (1994). Jocular abuse patterns in mixed-group interaction. Wellington Working Papers in Lin-
guistics 6: 2655.
(1995). Gender and Humor: Beyond a Joke. MA Tesis. University of Victoria.
(2000). Functions of humor in the conversation of men and women. Journal of Pragmatics
32(6): 709742.
(2001). Te pragmatics of humor support. HUMOR. International Journal of Humor Research
14(1): 5582.
Holmes, J. (1998). No joking matter! Te functions of humor in the workplace. Proceedings of the
Australian Linguistics Society Conference. Brisbane University of Queensland.
(2000). Politeness, power and provocation: how humor functions in the workplace. Discourse
Studies 2(2): 117.
Holmes, J. & J. Hay (1997). Humour as an Ethnic Boundary Marker in New Zealand Interaction.
Journal of Intercultural Studies 18(2): 127151.
Holmes, J. & M. Marra (2002). Over the edge? Subversive humor between colleagues and friends.
Holmes, J. & M. Stubbe (2003). Power and Politeness in the Workplace. Pearson ESL.
Hulstijn J. & A. Nijholt (eds.) (1996). Automatic Interpretation and Generation of Verbal Humor.
University of Twente.
Jeferson, G. (1979). A technique for inviting laughter and its subsequent acceptance declination. In
G. Psathas (ed.) Everyday Language. Studies in Ethnomethodology: 7996. Irvington.
(1985). An exercise in the transcription and analysis of laughter. In T.A. van Dijk (ed.) Hand-
book of Discourse Analysis, vol. 3: 2534. Academic Press.
Jenkins, M.M. (1985). Whats so funny?: Joking among Women. In S. Bremmer, N. Caskey &
B. Moonwomon (eds.) Proceedings of the First Berkeley Women and Language Conference:
135151. Berkeley Women and Language Group. University of California.
Jodlowiec, M. (1991). What makes jokes tick. UCL Working Papers in Linguistics 3: 241253.
Keith-Spiegel, P. (1972). Early conception of humor: varieties and issues. In J. H. Goldstein &
P.E. McGhee (eds.) Te Psychology of Humor: 339. Academic Press.
Kirshenblatt-Gimblett, B. (ed.) (1976). Speech Play: Research and resources for studying linguistic cre-
ativity. University of Pennsylvania Press.
Kotthof, H. (1986). Scherzen und Lachen in Gesprachen von Frauen und Mannern. Der Deutschun-
terricht: Beitrage zu Seiner Praxis und Wissenschaflichen Grundlegung 38(3): 1628.
(ed.) (1996). Scherzkommunikation. Westdeutscher.
(1998). Spa Verstehen: Zur Pragmatik von konversationellem Humor. Niemeyer.
(2000). Gender and joking: on the complexities of womens image politics in humorous
narratives. Journal of Pragmatics 32(1): 5580.
(2003). Responding to irony in diferent contexts: On cognition in conversation. Journal of
Pragmatics 35: 13871411.
Lakof, R. (1975). Language and Womans Place. Harper and Row.
Liao, Chao-Chih (2001). Taiwanese Perceptions of Humor: A sociolinguistic perspective. Crane
Publisher.
Loehr, D. (1996). An integration of a pun generator with a natural language robot. In J. Hulstijn &
A. Nijholt (eds.): 161171.
Long, D.L.& A.C. Graesser (1988). Wit and humor in discourse processing. Discourse Processes 11:
3560.
Mcghee, P. (1979). Humor: Its Origin and Development. Freeman.
Morain, G.G. (1991). X-raying the international funny bone: A study exploring diferences in the
perception of humor across cultures. Georgetown University Round Table on Language and Lin-
guistics: 397408.
Morin, V. (1966). Lhistoire drle. Communications 8: 102119.
Morreall, J. (1983). Taking Laughter Seriously. State University of New York.
(1987). Te Philosophy of Laughter and Humor. State University of New York.
Mulkay, M. (1988). On Humor. Its Nature and its Place in Modern Society. Blackwell.
Nardini, G. (2000). When husbands die: Joke telling in an Italian ladies club in Chicago. Pragmatics
10(1): 8797.
Navon, D. (1988). Te seemingly appropriate but virtually inappropriate. Note on characteristics of
jokes. Poetics 17: 207219.
Nilsen, D.L.F. (1993). Humor Scholarship: A Research Bibliography. Greenwood.
Norrick, N.R. (1993). Conversational Joking: Humor in Everyday Talk. Indiana University Press.
Priego Valverde, B. (1998). Lhumour (noir) dans la conversation: jeux et enjeux. Revue de Sman-
tique et Pragmatique 3: 123144.
(2003). Lhumour dans la conversation familire: description et analyse linguistiques.
LHarmattan.
Raskin, V. (1979). Semantic mechanisms of humor. In C. Chiarello et al. (eds.) Proceedings of the
Fifh Annual Meeting of the Berkeley Linguistics Society: 325335. Berkeley Linguistics Society.
(1985). Semantic mechanisms of humor. Reidel.
(ed.) (1985b). Round Table on Frame Semantics. Quaderni di Semantica 6(2/3).
(1986). Script-based semantics. In D.G. Ellis & W.A. Donohue (eds.) Contemporary Issues in
Language and Discourse Processes: 2361. Erlbaum.
Raskin, V. & S. Attardo (1994). Non-literalness and non-bona-fde in language. An approach to for-
mal and computational treatments of humor. Pragmatics and Cognition 2(1): 3169.
Ritchie, G. (1998). Prospects for Computational Humour. Proceedings of 7th IEEE International
Workshop on Robot and Human Communication (ROMAN-98): 283291. Takamatsu.
(1999). Developing the Incongruity-Resolution Teory. Proceedings of AISB Symposium on
Creative Language: Stories and Humour: 7885. Edinburgh.
(2000). Describing Verbally Expressed Humour. Proceedings of AISB Symposium on Creative
and Cultural Aspects and Applications of AI and Cognitive Science: 7178. Birmingham.
(2001). Current Directions in Computational Humour. Artifcial Intelligence Review 16(2):
119135.
(2003). Te Linguistic Analysis of Jokes. Routledge.
Rosch, E.H. (1973). On the internal structure of perceptual and semantic categories. In T.E. Moore
(ed.) Cognitive Development and the Acquisition of Language: 111144. Academic Press.
Ruch, W., S. Attardo & V. Raskin (1993). Towards an empirical verifcation of the General Teory of
Verbal Humor. HUMOR. International Journal of Humor Research 6(2): 123136.
Rutter, J. (2000). Te stand-up introduction sequence: Comparing comedy compres. Journal of
Pragmatics 32: 463483.
Sacks, H. (1972). On some puns: with some intimations. In R.W. Shuy (ed.) Sociolinguistics: Current
Trends and Prospects: 135144. Georgetown University Press.
(1974). An analysis of the course of a jokes telling in conversation. In R. Bauman & J. Sherzer
(eds.) Explorations in the Ethnography of Speaking, 2nd edition (1989): 337353, 467, 490. Cam-
bridge University Press.
Scheglof, E.A. (1993). Refections on quantifcation in the study of conversation. Research on Lan-
guage and Social Interaction 26(1): 99128.
Humor 155
(2001). Getting serious: Joke -> serious no. Journal of Pragmatics 33(12): 19471955.
Sherzer, J. (1978). Oh! Tats a pun and I didnt mean it. Semiotica 22 (3/4): 335350.
(1985). Puns and jokes. In T.A. van Dijk (ed.) Handbook of Discourse Analysis, vol. 3: 213
221. Academic Press.
(1992). Verbal play. In W. Bright (ed.) International encyclopedia of linguistics, vol. 4: 220222.
Oxford University Press.
(1993). On puns, comebacks, verbal dueling, and play languages: Speech play in Balinese
verbal life. Language in society 22: 217233.
(1996). Review of Attardo (1994). Language 72(1): 132136.
Sobkowiak, W. (1991). Metaphonology of English Paronomasic Puns. Peter Lang.
Sperber, D. & D. Wilson (1981). Irony and the use-mention distinction. In P. Cole (ed.) Radical Prag-
matics: 295318. Academic Press.
(1986). Relevance. Harvard University Press.
Suls, J. (1972). A two-stage model for the appreciation of jokes and cartoons. In J.H. Goldstein & P.E.
McGhee (eds.) Te Psychology of Humor: 81100. Academic Press.
Stock, O. (1996). Password Swordfsh: Verbal Humor in the Interface. In J. Hulstijn & A. Nijholt
(eds.): 18. Republished in HUMOR. International Journal of Humor Research 16(3): 281295.
Stock, O. & C. Strapparava (2003). HaHacronym: Humorous Agents for Humorous Acronyms.
Stock, O., C. Strapparava & A. Nijholt (eds.) (2002). Te April Fools Day Workshop on Computational
Humour. Proceedings Twente Workshop on Language Technology 20. ITC-IRST.
Tannen, D. (1984). Conversational Style: Analyzing Talk Among Friends. Ablex.
(1986). Tats Not What I Meant! Ballantine.
(1990). You Just Dont Understand. Ballantine.
(1994). Talking from 9 to 5. Morrow.
Tsakona, V. (2003). Jab lines in narrative jokes. HUMOR. International Journal of Humor Research
16(3): 315329.
Vaid, J., R. Hull, R. Heredia, D. Gerkens & F. Martinez (2003). Getting a joke: the time course of
meaning activation in verbal humor. Journal of Pragmatics 35(9): 14311449.
Viana, A. (2001). Temes difcil, estils cmics: el cas de lhumor mdic. Discurso y sociedad 3(3):
117144.
Weiner, E.J. (1996). Why is a riddle not like a metaphor? In J. Hulstijn & A. Nijholt (eds.): 111119.
(1997). Te incongruity of jokes, riddles and humorous situations. In G. Guy, J. Baugh,
D. Schifrin & C. Feagin (eds.) Towards a Social Science of Language. Papers in Honor of William
Labov, vol. 2: 139151. Benjamins.
Weiner, E.J. & P. De Palma (1993). Some pragmatic features of lexical ambiguity and simple riddles.
Language and Communication 13(3): 182193.
Wenzel, P. (1989). Von der Struktur des Witzes zum Witz der Struktur: Studien zur Technik der Point-
ierung. Winter.
Yamaguchi, H. (1988). How to pull strings with words. Deceptive violations in the garden-path joke.
Journal of Pragmatics 12: 323337.
Yus, F. (2003). Humor and the search for relevance. Journal of Pragmatics 35(9): 12951331.
Zajdman, A. (1991). Contextualization of canned jokes in discourse. HUMOR. International Journal
of Humor Research 4(1): 2340.
Ziv, A. (1984). Personality and Sense of Humor. Springer.
(1988). National styles of humor. Greenwood.
Intertextuality
Stef Slembrouck
University of Ghent
1. From literature to text as a productivity which inserts
itself into history
Intertextuality has been one of the most popular and infuential concepts in language,
literary and discourse-oriented research of recent decades. Te term was coined in the
1960s by Julia Kristeva, who assigned it a pivotal role in the formulation of a semio-
logical project in which text becomes the primary object of literary enquiry. Kristeva
(1968: 300) defnes the text as:
[S]ans vouloir rduire le texte la parole orale, mais en marquant que nous ne pouvons
pas le lire en dehors de la langue, nous dfnirons le texte comme un appareil trans-
linguistique qui redistribue lordre de la langue, en mettant en relation une parole
communicative visant linformation directe, avec difrents types dnoncs antrieurs
ou synchroniques. Le texte est donc une productivit ce qui veut dire: (i) son rapport
la langue dans laquelle il se situe est redistributif (destructivo-constructif), par
consquent il est abordable travers des catgories logiques et mathmatiques
plutt que purement linguistiques, (ii) il est une permutation de textes, une inter-
textualit: dans lespace dun texte plusieurs noncs pris dautres textes se croisent et
se neutralisent. (Kristeva 1968: 300)
Te introduction of the term intertextuality marked one of the key moments in the
post-structuralist arrival
1
through its challenge of text-immanism and, with it, the
irrevocable politicization of literary production and critical reception. Te literary text
as an autonomous, bounded unit became a textual moment in a longer chain of ante-
cedences and simultaneities, a result of transformation accomplished within textual
spaces. In the background is also a distinction between discourse (the exchange of
meanings) and text (the production of meaning). Text is what links an instance of
language use (une parole communicative) to the system (langue). Kristeva is keen here
to avoid that the analysis of signifying practice collapses into a study of parole (said
to be the object of the study of communication within structural linguistics). Instead,
it is a historical-materialist perspective which invites a premium on the production
:. However, this did not mean the end of structuralism as such. For instance, Kristeva continues
to dene literature as a type of linguistic structure.
Intertextuality 157
of meaning. Intertextuality is thus seen as a processual state which is refected in a
text. It is multi-directional, unresolved (hence, a preference for structuration over
structure) and it indexes how that text inserts itself into history. One relevant history
is that of successive, never-ending redistributions in the order of language, but that
history cannot ultimately be dissociated from the history of society itself.
2
tudier le texte comme une transformation sur laxe diachronique signife que
nous labordons comme une structuration plutt que comme une structure. La
mthode consisterait considrer les difrentes squences (ou codes) dune struc-
ture textuelle prcise comme autant de transformations de squences (codes) prises
dautres textes.
Ainsi la structure du roman franais au 15
e
sicle peut-elle tre considre comme
le rsultat dune transformation de plusieurs autres codes: la scolastique, la posie
courtoise, la littrature oral (publicitaire) de la ville, le carnaval. La mthode trans-
formationnelle nous mne donc situer la structure littraire dans lensemble sociale
considr comme un ensemble textuel. Nous appellerons intertextualit cette
inter-action textuelle que se produit lintrieur dun seul texte. Pour le sujet con-
naissant, lintertextualit est une notion que sera lindice de la faon dont un texte lit
lhistoire et sinsre en elle. La mode concret de ralisation de lintertextualit dans un
texte prcis donnera la caractristique majeure (sociale, esthtique) dune structure
textuelle. (Kristeva 1969: 443).
2. Text linguistics on textuality
Concepts with a wide appeal ofen fall fat when generalized and made to serve
elsewhere. Tis was also the case with early text linguistic uses of intertextuality in
the 1980s. De Beaugrande & Dressler (1981: 10f.) see intertextuality as one of the
seven constitutive features of textuality which can be used to the defne a text as a
communicative occurrence. Te other standards of textuality are: cohesion, coher-
ence, intentionality, acceptability, informativity and situationality. Intertextuality,
for de Beaugrandre & Dressler, concerns the factors which make the utilisation of
one text dependent upon knowledge of one or more previously encountered texts
(p. 10) especially texts from the same type or stemming from the same domain (e.g.
a trafc sign which reads resume speed presupposes a previously encountered and
interpreted sign of the kind slow children at play). Intertextuality is also behind
the evolution of text types, the reproduction of classes of texts with typical patterns of
i. Not surprisingly, then, the question how limited/limitless intertextual productivity is will con-
tinue to haunt debates between advocates and opponents of the post-structuralist turn in the
social sciences.
158 Stef Slembrouck
characteristics, and within certain text types (e.g. parodies, critical reviews, reports),
intertextuality may be more prominent as the text producer must consult the prior
text continually and actively. Intertextuality here becomes a cognitive condition in
the production and reception of actual texts. Its link with the evolution of genres is
one which favours continuity within text types, rather than seeing the development
of genres or text types as also originating in hybridity and creative assemblage of con-
ventions and sayables with diverse ancestry (in Kristevas work, intertextualities are
multi-directional in pointing to antecedences across received felds of social activity
or societal domains). Finally, with de Beaugrande and Dresslers generalized use of
the term, explicit indications that a texts state of intertextuality is instrumental to an
understanding of its positioning vis--vis the social and the historical-in-discourse
have also largely disappeared from view. Teir evolution of text types is a functionally-
motivated history of forms, not one of ideologically-disposed sayables. Ironically, in
the course of their reception of the term intertextuality, Kristeva (1968) is literally
moved to a footnote. Hers is said to be a narrower use of the term (1981: 13).
3. Dialogism and heteroglossia in a social-diachronic theory of discourse
If Kristeva is to be accredited for having coined a term and developed a concept which
stresses the accumulative, partly self-disguising and transformative nature of discourse
throughout history, and, by doing so, situates the literary irreplacibly among the more
mundane discourses that populate our lifeworlds, then she is also to be acclaimed for
guiding us to the discovery of the writings of Mikhail Bakhtin and his circle. With the
growing popularity of the term and concept of intertextuality came also the grow-
ing importance of a set of other concepts which accompanied Bakhtins belated admit-
tance to the halls of social scientifc fame: dialogism, heteroglossia, double-voicedness,
polyphony, the carnivalesque.
3
With critical discourse analysis, language research restored the link with Kristeva,
but much more than that, it embraced the writings of Bakhtin (and Voloinov) with
considerable debt to mediation by M. Foucault. Fairclough (1992b:193194) sees inter-
textuality as quintessential to a discourse analysis which seeks to map systematic anal-
yses of spoken or written text onto systematic analyses of social contexts. He develops
the argument in detail in a review article which focuses on recent studies published in a
leading discourse analytical journal. For Fairclough, textual analysis subsumes two com-
plementary types of analysis: linguistic analysis (used here in an extended, functional
sense, in which content/meaning-dichotomies have been dissolved) and intertextual
analysis. Tree imperatives follow: (i) Intertextual analysis shows how texts draw upon
. Bakhtin (1986) uses the term trans-linguistic but never used the term intertextuality.
Intertextuality 159
orders of discourse, the particular confguration of conventionalized practices which
text producers and consumers have available in particular circumstances. (ii) Intertex-
tual analysis draws attention to how texts depend upon society and history in the form
of the resources made available within an order of discourse. Here Fairclough echoes
Bakhtin in viewing genres as the drive belts from the history of society to the history
of language. (iii) Intertextual analysis also draws attention to how texts transform the
social and historical resources, how texts mix and re-accentuate genres (discourses,
narratives, registers).
Earlier, in Fairclough (1992a), the author introduced a distinction between
manifest intertextuality (when specifc other texts are drawn upon in a text) and con-
stitutive intertextuality (how a discourse type is constituted through a combination of
elements of orders of discourse; Fairclough renames this type as inter-discursivity).
Tis distinction results from a certain ambiguity in Kristevas writings (indeed, one
must pose the question at what level exactly does one situate les difrents noncs
que notre analyse retrouve a posteriori dans les difrents discours antrieurs ou syn-
chronique which Kristeva (1968: 313) talks about). Te distinction also echoes dif-
ferent uses of dialogism in Bakhtins own writing as the concept occurs a number
of times to signify diferent but related phenomena at diferent levels of generaliza-
tion. Te frst lead (that of manifest intertextuality) surfaces within linguistic enquiry
in analyses of local utterance-based phenomena such as irony, presupposition,
appraisal, reported speech, etc. as intertextual or double-voiced
4
in nature, whereas
the second lead (constitutive intertextuality) points in the direction of an impersonal
history of generic constitution with diferent degrees of naturalization (which keep
intertextual traces from view). One may be tempted to conclude that the second vari-
ant is better attuned to Kristevas defnition of the text. It is also the one which, as I
discuss in greater detail below, returns in critical discourse analysis as instrumental
to an understanding of the dynamics between hegemonic struggle, crises in socio-
cultural values and forms of social linguistic creativity at the level of genre conven-
tions. However, a one-sided concern with constitutive intertextuality comes with a
risk of underestimating the heuristic condition that the dynamics of discursive het-
erogeneity is studied through actual instances of discourse and that, as Fairclough
notes, texts difer in the extent to which their heterogeneous elements are integrated,
and so in the extent to which their heterogeneity is evident on the surface of the
text (1992a: 104). Additionally, the erasure of the traces of hic et nunc assemblage is
|. Te synonymous use of the term polyphony is a post-Bakhtinian extrapolation. As Bjrklund
(2000: 910) points out in detail, Bakhtin borrowed the term polyphony from Dostoevsky to refer
to a particular type of dialogism found in some novels when character voices are on an equal
footing with the authorial voice. As in music, polyphony is an eect of entirety.
160 Stef Slembrouck
intrinsic to certain social processes and to history, more generally, as visibly creative
mixings become established traditions and institutions have an active interest in
maintaining perceptions of stability in practice.
Lemke (1995: 23), in my view, observes correctly that the dialogic conception of
the utterance for Bakhtin is a bridge between the linguistic and social, the event-
meaning and the larger social systems in which that event has its meaning for us.
Bakhtin does not give up the idea of signifcation (meaning-making) as understood
against the background of language,
5
only he saw (and stressed) diversity in language
in such a way that it renders a homogeneous version of it as a grammatical abstraction
which one can only arrive at by suppressing the occurrence of a multitude of bounded
verbal-ideological belief-systems which fll it (codes in Kristevas diction). It is this
heteroglossia (raznorechie, translated literally: varied-speechness) which necessi-
tates a continual analytical return to the level of dialogic relationships between actual
utterances and, not surprisingly, the uptake of Bakhtins work has been characterized
by constant slippage from hetoroglossia as a societal state of being to heteroglossia
being used to refer to the heterogeneic make-up of a specifc piece of text or discourse.
Bakhtin lists as examples of heteroglossia the co-occurrence of:
social class dialects, languages of special groups, professional jargons (), genre
languages, the languages of generations and age groups, of the authorities, of literary
and political movements, historical epochs, etc. (Bakhtin 1981: 2623)
Heteroglossic relationship are ultimately sociological relationships in that they are
about
specifc points of view on the world () each characterised by its own objects,
meanings and values. As such they may all be juxtaposed to one another, mutually
supplement one another, contradict one another, and be interrelated dialogically.
(Bakhtin 1981: 2912)
It is this kind of fusion of the social and the linguistic which informs Kristevas defni-
tion of the idologme as the intertextual function.
que lon peut lire matrialise aux difrents niveaux de la structure de chaque texte,
et qui stend tout au long de son trajet en lui donnant ses coordonnes historiques et
sociales. (Kristeva 1968: 313314)
,. Similarly, Kristeva continues to see langue as the explanatory backdrop but redenes it in the
course of her formulation. For her, the question how a communicative occurrence inserts itself
into history cannot be dealt with in isolation from its constitutive eect on langue. Te two views
are by no means identical: inasmuch as Bakhtin rejects the Saussurean linguistic project, Kristevas
formulation of intertextuality counts as a reformulation of it.
Intertextuality 161
It takes just a small step now to see how for Fairclough argues the importance of link-
ing up intertextuality and hegemony:
Te combination of hegemony theory with intertextuality is particularly fruitful. Not
only can one chart the possibilities and limitations for intertextual processes within
particular hegemonies and states of hegemonic struggle, one can also conceptualise
intertextual processes and processes of contesting and restructuring orders of
discourse as processes of hegemonic struggle in the sphere of discourse, which have
efects upon, as well as being afected by, hegemonic struggle in the wider sense.
(Fairclough 1992a: 103)
Two points deserve mention here. My frst point is explanatory. Faircloughs specifc
appropriation of Bakhtin is one in which the state of play of heteroglossia (in a
specifc social, cultural, historical, institutional context) is viewed as ordered hier-
archically in what he calls orders of discourse (these may be local, institutional,
domain-based, or more widely societal, cultural) and, in Bakhtinian fashion, orders
of discourse are necessarily social orderings. Tis picture informs for instance expla-
nations of commodifcation, conversationalization, etc. in terms of the permeability
of domain-specifc orders of discourse by practices and discourse conventions which
originate in other domains (one example is: public authorities who resort to adver-
tisements on hoarding posters to bring home essentially bureaucratic messages, say,
to combat tax fraud; another example is the widespread adoption of informal conver-
sational modes of communication in the workplace which accomponies transitions
to post-Fordist models of economic production). In this view, hybridization helps
explain language users creative responses to crises in social outlooks. My second
point is that the addition of hegemony theory, while underscoring the sociological
(Bakhtin) or ideological (Kristeva) nature of heteroglossia, seeks to lend a systematiz-
ing framework to the analysis and weighting of competing voices and their social
outlooks: for instance, does conversationalization lead to more democratic work-
ing relationships or is it just an apparency which seeks to enhance worker motiva-
tion but leaves power relationships untouched because it banks on the managerial
groups capacity to adjust their speech strategically? One of the interesting questions
indeed now is how far a (singular, unitary?) hegemonic framework of social analysis
will actually take us in charting out the possibilities and limitations of intertextual
processes.
6
o. Part of the attraction is no doubt that the application of the concepts of intertextuality or
heteroglossia can be repeated at various levels of analyis. Tis is also true for the concept of orders
of discourse. Tis repeatability comes with a risk of vagueness.
162 Stef Slembrouck
4. Voloinov, pragmatics and conversation analysis: Sequential
implicativeness and the translation of the others perspective
Te detection of dialogisms at diferent levels of enquiry also helps us understand
dualities in the reception of Voloinov (1929, trnsl. 1986). When Voloinov proposes
in his Exposition of the Problem of Reported Speech that
the productive study of dialogue presupposes () a more profound investigation
of the forms used in reported speech, since these forms refect basic and constant
tendencies in the active reception of other speakers speech.
(Voloinov 1986: 117, the italics are the translators)
then he is using the problem of reported speech as a vehicle for developing an argu-
mentation about the priorities of language study itself, one with consequences for pre-
ferred sites of enquiry. For him, language does
not refect subjective, psychological vaccilations, but stable social interrelationships
among speakers () Everything vital in the evaluative reception of anothers utterance,
everything which is of ideological value, is expressed in the material of inner speech ()
Te context of this inner speech is the locale in which anothers utterance is received,
comprehended, and evaluated, it is where the speakers active orientations take place. ()
[L]ines of reception fnd their expression, are objectifed, in the authorial context
which is surrounding the reported speech. (Voloinov 1986: 117)
Tus, the focus of enquiry should be the various linguistic forms of these
interrelationships which prevail in diferent periods, in diferent languages, within dif-
ferent social groups and under the efect of diferent contextual aims. Whats more, in
Voloinovs book we arguably come very close to the formulation of a kind of conver-
sation analytic project and it is in this direction that one can see a further extension of
the relevance of intertextuality for language study.
Voloinov (like Bakhtin) takes the dialogic to the most general level of positing
an ontology of language: for him, the utterance is always a reception of and a response
to another, previous utterance, and this happens along fundamentally social hori-
zons of reception. At the basis of this is fundamentally a pragmatic, utterance-
oriented
7
and context-sensitive theory of language. His proposal is to examine how
regularities in parole (systematic tendencies in reception and response as revealed
in reported speech) are grounded in the socio-economic being of a community of
. Note how the disclaimer, language does not reect subjective, psychological vaccinations,
alludes to the Saussurean view which saw parole as chaotic, unsystematic, individual, and therefore
not worthy of linguistic attention. For Voloinov, parole is social.
Intertextuality 163
speakers.
8
Indeed, one can only speculate about how orientations and objects of
enquiry in this type of linguistics would have evolved, had the socio-technological
means of the time aforded traditions of feld work and the use of a tape recorder and,
with these, opportunities to access dialogisms in a form that could be made to stand
still for detailed examination other than those of reported speech and dialogue in
the context of the novel (and a handful of other, written text types). It is fascinating
to observe how Voloinov (very briefy, though) touches on the distinct teleologies
of reporting activities such as courtroom proceedings, newspapers, etc., whereas in
fact most of the book deals with reported speech in the novel (but then, reported
speech is the primary means available to display the conversational in a novelis-
tic context). It is also interesting to observe how Bakhtins essay on the problem of
speech genres opens with a wide perspective on non-literary genres, but returns to
the literary out of necessity, because literature was the best/only data around?
One is tempted to push this speculative discussion even a little bit further.
Voloinovs defnition of the utterance reveals a conversational principle, arguably a
very early defnition of the adjacency pair the utterance is always a response to a
previous utterance. To begin with, note the immediately apparent afnity with the
conversation analytic concept of sequential implicativeness (Heritage 1997), which
reads: each move in a conversation is essentially a reponse to the preceding talk and
an anticipation of the kind of talk which is to follow; in formulating their present turn,
speakers show their understanding of the previous turn and reveal their expectations
about the next turn to come. Note also how Sackss claim that it takes two turns to
have a conversation (Sacks et al. 1974) is equally a claim with axiomatic implications
for preferred methods and sites of enquiry: his aim is to arrive at an understanding
of how social order is brought about through the detailed actions accomplished by
interactants engaged in everyday talk. Despite the obvious diferences in what is meant
by an understanding of the social, it is not difcult to see how this strikes a chord
with Bakhtins claim: two voices is the minimum for life, the minimum for existence
(Bakhtin 1973: 213).
Aiming at a fusion between the two senses of dialogism utterance-within-
utterance and turn-following-turn, one can detect a complementarity of perspec-
tives which can be expressed in a layered defnition of sequential implicativeness:
when the present turn involves a quotation of what somebody else said on a diferent
ocassion, then the speaker shows his/her understanding of the interlocutors imme-
diately preceding turn by displaying his/her understanding of another turn distant
in place, time and ocassion. Tis other turn can shade over into two directions.
8. Note how Voloinovs proposal redenes grammar as an eect of language/ideology, i.e. gram-
mars of reported speech count as crystallizations of prevailing receptive orientations.
164 Stef Slembrouck
One: it can be located in the immediately preceding turn. Tis is the case in echoic
turn design, where literally part of the previous utterance is being repeated. Tings
can also go in the other direction: there is no reason why another turn should not
be equated here with any piece of discourse and it can come with various degrees
of explicit attribution (or go entirely unnoticed in that respect). For instance, a pupil
who repeats parts of a teachers question while responding to it, and, in doing so,
positions herself vis--vis a received way of talking about knowledge in a specifc
area, arguably accomplishes the three possibilities at the same time if one accepts
that the utterance belongs both to this particular teacher and to a transpersonal
history of practices (cf. Maybin 1999). It is this double orientation toward other
voices of speakers co-present and not and voices individualized and not which is
behind for instance Wetherells (1998) suggestion that a conversational analytic con-
cern with local sequential dynamics must be complemented with a post-structuralist
orientation to the efectuation of interpretative repertoires (read: discourses of).
Tis double-orientation provides an obvious way in which conversational analysis
can reclaim the social-cum-ideological, looking for sequential implicativeness, as it
were, by raising their eyes from the immediately previous/next turn
9
to the other
voices, and, with it, towards the social perspectives which these index. Participation
in interaction takes place in rapidly evolving situations of co-presence. In the course
of these, talk by other people, or social outlooks, more generally, may be reported,
commented upon, presupposed, etc. and become part of the current and transient
here/now. At the heart of the theoretical fusion just outlined is indeed another
bridge from event-meaning to larger social systems though with equal sensitiv-
ity to the production of frames/activity-types (Gofman, ethnomethodology) and
ideological outlooks (Bakhtin, Voloinov).
Te suggested combination of perspectives also brings into view the social
transformative functioning of discourse, or, put diferently, it is through socially-
situated activities that the dialogic-in-discourse can become consequential for indi-
viduals. Let me give a couple of examples here: a decision to grant asylum in Belgium
to an economic refugee from Kazachstan which is based on a condensed, word-
processed and translated report of an oral interview with the applicant conducted
in a language which counts as non-native for both interviewer and interviewee; a
language dictation in a classroom as a regularly instituted educational activity where
each instance counts as one of the many steps on the road to the legitimate acqui-
sition of an educational qualifcation; the social worker who quotes the voice of a
16-year-old client of the social services to justify before a policy reviewer why in
this particular case foster care was not preferred over residential care (the written
. Here I am echoing Wetherell(1998:402).
Intertextuality 165
fles help to freshen up his memory); etc. In all of these cases we are dealing with
social facts which involve the intertextually-mediated reception of (and a position-
ing reply to) discourse along fundamentally social horizons but the positioning
that goes on is just as much activity-type oriented (as part of an institutional set of
procedures, accomplished sequentially) as it is about social outlooks embedded at
various levels of double-voicedness. In each of these instances, the events are quite
ordinary, but they are consequential for the participants involved. In each case, the
intertextualities involved are of considerable complexity and they defy a reduc-
tion to received themes and traditions of enquiry: reported speech, conversational
exchanges, translation, interpretation, etc. By inviting us to tease out its various
dimensions, intertextuality can help us to take the dialogic beyond the boundedness
of recognized domains of language study.
5. Synoptic and participatory views of human activity: Bakhtin, Bourdieu,
sociolinguistic legitimacy (and the body)
For Maybin (1999), the accomplishment of local interactional intentions through a
speakers positioning vis--vis the repeated, appropriated and reported voices of oth-
ers forms part of an explanation of how children in primary school contexts manage
institutional frames while being dialogically induced (in the two senses described
above) into institutional roles, procedures and discourses. In this respect, it is
important to remind ourselves of the diference between postioning being inevitably
intertextual in nature and situated struggles to make the words of the other ones
own in the many diferent senses that may apply: learning how to by adopt-
ing, getting others to adopt, stopping others from continuing to adopt. So, if
to quote or to translate equals to make ones own, then it will also mean: to render
something legitimate or to dispell it. Note (in passing) that when language users
contest the ways in which they have been quoted between inverted commas, this is
hardly ever so because the wording was changed but because their social outlook
has been misrepresented (the wording is usually the evidential ground on which
the quotation will be disputed). I think this is an appropriate point for a digression
which ventures into the work of Pierre Bourdieu.
10
Tis should not surprise my
reader, if one recalls for a moment that it is not just variedness-in-language (and
dialogic contact) which separated Bakhtin and Voloinov from de Saussure but also
a radically diferent view on the essence of the linguistic sign:
:o. I am grateful to Lemke (1995: 25.) for putting me on track here in developing a number of
parallels between Bourdieu and Bakhtin. See also Footnote1 in Bourdieu (1977: 17).
166 Stef Slembrouck
A single, concrete utterance is always given in a value-and-meaning context,
whether it is scientifc, artistic, poetical, etc., or in the context of a situation from
everyday personal life. Each separate utterance is alive and has meaning only
within these contexts: it is true or false, beautiful or ugly, sincere or deceitful,
frank, cynical, authoritative, etc. there are no neutral utterances, nor can there be.
(Bakhtin 1990: 292)
Tis quotation can just as easily be read as a preamble to a treatise on the relevance
of language study for a domain-based sociology of taste and distinction. Let me draw
up this parallel more fully. As any situation in which language is used is defned as a
linguistic market (e.g. in Bourdieu 1984: 132), Bourdieu leaves little doubt over how
and where speaker legitimacy is being (re)produced: it happens in the situation of
contact between forms of symbolic capital and with variable efects e.g. a proftable
return in social capital (such as improving ones positioning in a prestigious network
of social relationships), enhancing solidarity in the case of le franc parler (in situa-
tions where the laws of the linguistic market are temporarily put between brackets,
un lot arrach aux lois du march),
11
self-censured and respectful silence (when
popular speech is made to confront ofcial diction, condamns au silence, a silence
que lon dit respectueux). In the background here is not only the shif from a primary
concern with communicative understanding to the need to recognize that speakers
are driven by a desire to be believed, but also a view in which the production of legiti-
macy or symbolic violence are efects of situations of enacted heteroglossia (contact
between discursive dispositions in a transactional space which is never value-free).
Bourdieus outline of a theory of linguistic practice which avails itself of the con-
cepts of capital, habitus and market is arguably a project of linking an analysis of
participant-oriented event-meaning with an analysis of objective conditions and rela-
tions between social groups as indexed by constituents of heterglossia:
La description interactionniste des rapports sociaux, qui est en soi trs intressante,
devient dangereuse si lon oblie que ces relations dinteraction ne sont pas comme
un empire dans un empire; si on oublie que ce qui se passe entre deux personnes,
entre une patronne et sa domestique ou entre deux collgues ou entre un collgue
francophone et un collgue germanophone, ces relations entre deux personnnes sont
toujours domines par la relation objective entre les langues correspondantes, c est--
dire entre les groupes parlant ces langues. (Bourdieu 1984: 127)
In this way, a market presupposes both a situated activity and a set of objective
conditions which entail a number of price-determining mechanisms which will bear
on any situated transaction and its outcomes. Similarly, the concept of habitus will
receive a number of defnitions (Bourdieu 1984: 133136): it is discourse adjusted to
::. Note how lot can be translated into English both as island and as hotbed of resistance.
Intertextuality 167
a situation (a market, a feld) but it is also transsituational capital. Habitus brings out
the dimension of the speaker as a strategic player, but it is also being defned as a set of
internalized bodily dispositions, acquired over time and in response to the sanctions
and prohibitions of the market. Habitus is schematic knowledge which generates prac-
tices as well as the evaluative perception of these practices. It encompasses ethos (this
underscores both the depth of confict or the extent of solidarity in situated contacts
between diferent/identical ideological viewpoints; ds que nous disons blanc ou noir,
nous disons bien ou mal). Finally, habitus is hexis it indexes an overall positioning
of the body. Tere are three types of consequences from this which I would like to
highlight.
One obvious area of application lies in analyses of sociolinguistic contact inviting
full attention to the diversity of local dialogisms of contact between strictly sepa-
rated languages, between class dialects, as entailed in forms of code-switching and
code-mixing. Tese are all very much part of an established sociolinguistic agenda,
but one can also add to this agenda (encouraged by Bakhtin): the role of playful or
derogatory citations, parody, mimicry and fantasy-displays of multilingualism and
sociolectal diversity. For instance, by marking of a phrase as cited from another class
dialect, speakers may draw normative boundaries around their own speech and that of
absent and excluded others, even when the cited phrase appears to mark a temporary
adherence to a value embodied in the other group (borrowing a perspective while
drawing a boundary). Conversely, in some conditions of contact, a failure to mark of
a phrase as cited from someone elses habitus may be earmarked as evidence of a lack
of competence.
12
Secondly, for Bourdieu, issues to do with the nature of sociolinguistic contact
cannot be dealt with separately from issues to do with domain-based legitimacy (the
domain of professional or institutional processes). Remember in this respect, that
Bakhtin, in a very similar fashion, mentions professional languages in the same breath
as social languages when defning heteroglossia. Bourdieu will time and again draw
attention to the role of speaking correctly in the production of legitimate speech,
but he equally cautions that it cannot exhaust le bien dit. As a result, it cannot tell
the whole story of dominance or legitimacy. Consider for instance Hellers (2001)
point that the production of legitimate language in the Franco-Ontarian schools
which she studies is not just a matter of French versus other languages or what sort
of French, it also depends on the implicit recognition of distributional values and the
structuring that is inherent in competing models of turn-taking: there is the ofcial,
on-record, French foor of orderly classroom talk controlled by the teacher and there
:i. In Voloinovs terminology, one could start talking about the sociolinguistics of high and low
boundary maintenance.
168 Stef Slembrouck
is the competing multivocal and nonconsequential foor which usually takes place in
a language other than French and where many people can talk at once, overlap and
where the teacher exercises no control.
Tirdly, Bakhtins and Bourdieus programmes both mark a return to the body
seeing posture as embodying a social and a moral position. One defnition of linguistic
habitus is that of hexis: a set of internalized, bodily dispositions (both eidos and
ethos). Te penultimate section of Bourdieu (1978: 31) opens as follows:
Mais il faut encore dgager les consquences du fait que le capital linguistique est un
capital incorpor et que lapprentisage de la langue est une dimension de lapprentisage
dun schema corporel global qui est lui-mme ajust un systme de chances objective
dacceptabilit. Le langage est une technique de corps et la comptence proprement
linguistique, et tout spcialement phonologique, est une dimension de lhexis
corporelle o sexprime tout le rapport au monde social. (Bourdieu 1977: 31)
He continues with a set of observations which underscore the extent to which the
social reinterpretation of biological determinisms afects linguistically gendered prac-
tices (and their representations) in a way which is structured along a homology of
oppositions which shapes the image of the sexes and the classes.
L opposition entre le rapport populaire et le rapport bourgeois au langage se
rsume dans lopposition entre la bouche, fminine, recherche, distingue, et la
gueule, typiquement masculine, en tant que rsum de tout le corps masculin (bonne
gueule, sale gueule). Dun cot, les dispositions bourgeoises ou, dans leur forme
caricaturale, petites-bourgeoises, hauteur et ddain (faire la petite bouche, bouche
fne, pince, lvres pinces, serres), distinction et prtention (bouche en
cur, en cul de poule) ; de lautre, les dispositions virile telles que les conoit la
rpresentation populaire, disposition la violence verbale (fort en gueule, coup
de gueule, grande gueule, engueuler, sengueuler) ou la violence physique
( casser la gueule , mon poing sur la gueule ), le sens de la fte comme ripaille
(sen mettre plein la gueule, se rincer la gueule) et franche rigolade (se fendre la
gueule). Du point de vue des membres des classes domines, les valeurs de culture et
de rafnement sont perues comme fminines et lidentifcation la classe dominante,
en matire de langage par exemple, implique lacceptation dune manire de tenir son
corps qui apparat comme efmine ( faire des chichis , des manires , des
mines, minauder, faire du genre), comme un reniement des valeurs viriles.
(Bourdieu 1977: 32).
Bourdieu here pinpoints some of the pains that come with class and gender:
social mobility is received in compensation for docility in the relationship to the body:
unlike men, women can identify with the dominant culture without cutting themselves
of from their class and without the risk of their transformation being interpreted as
a change in social and sexual identity; the dominant properties also entail a double
Intertextuality 169
denial of virility: their acquisition requires docility which is imposed on women in
the gendered division of labour (and the division of sexual labour), while docility is
required particularly in the domain of feminine dispositions.
Tis rather long and detailed digression is worth making because it evokes a
comparable dimension in Bakhtins work, viz. his writings on the carnivalesque, which
equally point in the direction of a need to take analyses of heteroglossia to the level of
relations between body-image and collective identities. Tis is perhaps a dimension of
Bakhtins writings which is rarely thematized in sociolinguistic and pragmatic enquiry
(with notable exceptions, see below) but it has been picked up more elaborately in
cultural studies for instance, in Stallybrass & Whites (1996) seminal enquiry into the
politics and poetics of transgression.
Carnival in Bakhtins study of Rabelais is characterized as heteroglot exuberance;
it is full of transcodings and displacements efected between the high/low image of the
physical body and other social domains. According to Stallybrass & White (1996), a set
of oppositional correlations can be derived from this, between high, ofcial and the
(desembodied) classical body with its associated values of reason and contextual dis-
tance, on the one hand, and low, popular and the grotesque body with its associated
values of mobility, physical needs and pleasure, impurity and as well as a refusal to
close the body of from its ecosystem. Stallybrass & White interpret the carnivalesque
as an instance of a wider phenomenon, transgression, and this can be developed into
what they call, a political anthropology of binary extremism in class societies. Teir
claim is that the underlying structural features of carnival operate far beyond the
strict confnes of popular festivity and are intrinsic to the dialectics of social classif-
cation as such. (1996: 26). For them, the history of contact between high and low is
not just one of constant scrutiny, erasure and suppression, it is also one of fascination
and desire and this is a premise which can be used fruitfully to explore the nature and
functioning of situated forms of hybridity, inmixing and boundary transgression in
language use (and other behaviours). For instance, Rampton (2003) observes in his
data, encounters in the London area, how teenagers refexively play out gradations
between and exaggerations of perceived points of language/class-contrasts:
ultra-posh and Cockney emerged at moments when in one way or another, participants
were attending to issues of sexuality, territory, reason and emotion, demeanour and
deportment; they featured in sound play, self-talk, response-cries, greetings, taunts,
commands, rebukes, and summonses; they were used to address, mock and caricature
teachers, to fetishise particular individuals and to portray a variety of social types.
(Rampton 2003: 40)
Tese apparent fexibilities in conduct (ranging from pretty stable style-shifs to
loosely performed stylizations) foreground social class as a frame relevant to the
fux of experience (2003: 46). Here what is traditionally perceived as a macro-issue
170 Stef Slembrouck
(language/class) which is traditionally examined through large-scale quantifying
variationist research is being explored through the nitty-gritty interactional analysis
of local, turn-based mini-citations. Tis is more than just a valuable addition to
macro research, because it simultaneously raises questions of objective positionings
and subjectifying perceptions of these (with partially denaturalising enactments in
interaction), and in doing so, takes us to the very heart of some of the key questions
posed by late modern conditions: e.g. do the growing diversity, apparent contradic-
tions and apparent transcience which characterize contemporary discourse forms
mark a breakdown or instead an intensifcation of old hierarchies, binarisms and
social oppositions?
Concluding this section, note that Bourdieu, who tends to be associated somewhat
too easily with a one-sided analysis of symbolic violence and silencing, concludes
lconomie des changes linguistiques with the observation that all linguistic phenom-
ena lie between the poles of highly-censured discourse which reaches its limit in the
silence of those who do not possess the resources of euphemization and the unbridled
language use of the revolutionary crisis or the popular festivity as described by Bakhtin:
On voit quil est la fois vrai et faux de rduire lopposition entre les classes
lopposition entre la distinction, censure devenue nature, et le franc-parler, qui ignore
les interdits de la langue commune, rgles de grammaire et de la politesse, et les
barrires hirarchiques (usage du tutoiement, de diminutifs, de sobriquets, dpitethes
injurieuses, dinjures afectueuses), et qui se dfnit par le relchement de la tension
articulatoire (comme dit Guiraud) et de toutes les censures que la biensance fait
peser, en particulier sur le corps tabou, ventre, cul et sexe, et surtout, peut-tre sur
le rapport au monde social quil permet dexprimer, renversement des hirarchies
(cul par-dessus tte) ou rabaissement de ce qui est lev (boufe, tripes, merde).
(Bourdieu 1977: 33)
6. Natural histories of discourse: Recontextualization/entextualization
and textual ideologies
Tere is one remaining, loose thread in this overview one not unimportant for
pragmatic theory. It is that of the speaker, or rather, what is lef of him/her. If Maybins
(1999) argument above means the end of the individual communicative intention,
then she is only repeating an exercise performed about 20 years earlier, when the
rapid ascent of intertextuality meant a further challenge to the Romantic concept of
the author as a source of meaning and signifcation.
13
Te implications of this for the
:. Structuralism had already replaced the author as a source of meaning by structural relation-
ships in an autonomous, bounded textual space.
Intertextuality 171
ways in which institutions work reach further than one would imagine at frst sight.
Arguably, they take us beyond Kristevas proclamation of the centrality of the text.
Tere are also metadiscursive implications for the study of language use. To develop
these points, I will make use of the examples of literary translation and institutional
self-reporting. Both are intrinsically bi-focal and dialogic activities, but so far I have
not talked about this as raising questions about originals, derivations and their
authorship. Developing the implications of intertextuality in this direction will again
lead us to address issues to do with communicative legitimacy but with more of an
explicit focus on institutions. Whats more, what follows is a necessary venture into
(re)contextualization and ideologies of linguistic and textual reproduction.
Consider frst a few practical details about the workings of one or two
institutions. For all purposes of parliamentary activity (future references to ear-
lier debates, political history, etc.), the Hansard report counts as th source of what
happened in Britains House of Commons.
14
It is the printed report which is the
institutionally-endorsed, and hence also the authoritative account of what hap-
pened. In that sense, much more than its antecedent spoken debates, it counts as
an original. Similarly, Flemish literary audiences reading Les feurs du mal in the
Dutch translation by Petrus Hoosemans, will say they are reading Baudelaire not
Hoosemans. Tey may not even remember the translators name. Insofar as a user
is in a position where s/he cannot access the source text (because it is not available
or accessible or its authority cannot be endorsed publicly), intertextually-established
re-enactments function as an original. It is the widespread but counterfactual belief
in reproduction accomplished through a verbatim rendition or translational equiva-
lence which makes it possible for a derived version to function as an original. Such
can only happen at the expense of the intertextual transformation itself, denying
its relevance and impact (and with this, translators names and the names of Han-
sard scribes and editors are forgotten or thought of as irrelevant). Note that this is
more than an imaginative leap which neutralizes and suppresses inevitable condi-
tions of intertextual mediation which are traceable in textual practice (e.g. when
one compares the literary text and its translation; when one compares the printed
Hansard with audio-taped recordings). Without this suspension of disbelief, the
very claims to institutional authority and legitimacy collapse: the Hansard becomes
a systematically distorted rendering, newspapers can no longer be called factual;
translators become co-authors and become likely to sue publishers in demand of
:|. I am taking the example of the British Parliamentary records here because it is a case I have
studied in detail (Slembrouck 1992).
172 Stef Slembrouck
a greater share in royalties.
15
In this sense, it is probably more accurate to observe
that intertextuality stands for the death of some authors/speakers and that, detect-
ing the details of its workings, also entails the rediscovery/unearthing of other
authors/speakers, whose visibility can ultimately become quite threatening to insti-
tutional legitimacy itself (invisible as they ideally must remain when performing
adequately in the service of authors/speakers more important). Having said this,
something else will also become apparent to my reader: the need to understand
intertextualities as contextualities. (a) Intertextual transformation always comes
with recontextualization. (b) Translation and direct quotation entail (ofen) divided
loyalties to two (or even more) contexts. (c) contexts of intertextual mediation are
being shaped by (and help shape) cultural, societal, institutional, etc. ideologies of
language, text and textual practice (Schiefelin et al. 1998; Collins 1996).
Tis is an appropriate point to refer to linguistic anthropological traditions, in
which the concept of intertextuality itself appears to be rather absent as a key term,
16

whereas most of the processes I have been engaged with so far are dealt with through
such concepts as (re)contextualization, (de)contextualization and (re)entextualiza-
tion. A benchmark publication in this respect is Silverstein & Urbans (1996) edited
volume on natural histories of discourse. Programmatically, the perspective of natural
histories of discourse amounts to an enquiry into the make-ability of text text as an
artefact for analysis and cultural transmission, text as a reifed category attached to
which is the fundamental consideration that the student of cultures engagement with
texts forms no exception to these processes. In their editorial introduction, Silverstein
and Urban connect the concept of natural histories of discourse to a focus on
contextually contingent semiotic processes involved in achieving text and
culture. Tese are recoverable in some measure only by analytically engaging with
textual sedimentations. Each chapter in our collective natural history thus focuses
on certain analytic moments in the entextualising/co(n)textualizing process.
(Silverstein & Urban 1996: 23)
:,. Note at the same time how translation studies has failed to take intertextuality to its full socio-
political consequences. While Hatim & Masin (1990) do devote a complete chapter to intertextu-
ality, they fail to take on board the consequences of an obvious (but politically risky) observation:
translation is intertextual practice.
:o. Even though the impact of the work of Bakhtin and Voloinov is quite considerable within
linguistic anthropological enquiry, the term intertextuality (unlike the concept heteroglossia)
does not appear among the keywords of language and culture in Duranti (ed. 2001a), in the same
authors linguistic anthropological handbook (Duranti 1997), in Hanks (1996) or in the authorita-
tive reader (Duranti ed. 2001b). Signicantly, Silverstein & Urban (1996: 6) in their editorial intro-
duction refer to intertext only once, at a point where they mean co-text.
Intertextuality 173
Natural history of discourse takes issue with the anthropological idea of
text-as-culture to the extent that such a view encourages the analyst of culture to
extract a portion of ongoing social action and draw a reifying boundary around it,
before enquiring into its structure and meaning (compare with the rather common
perception of culture in terms of texts handed down from generation to generation).
To equate culture with its resultant texts is to miss the point that the thingy-ness of
texts is but one stage in ongoing cultural processes and although on the surface that
thingy-ness may appear to have a quasi-permanent shape (which is by no means
always the case), it travels from context to context and as a result, it will not only
enter into new diferential orderings between textual artefacts and be surrounded
by changed conditions of response, uptake, commentary and explanation; its textuality
itself will at the same time be contextually contingent (for instance, a textual artefact
may be a set of anticipatory notes awaiting performance and therefore not really a
text; one and the same textual artefact may enter into diferent contexts accompanied
by diferent scripts for performance).
From this point of view, then, text is a metadiscursive notion, useful to participants
in a culture as a way of creating an image of a durable, shared culture immanent in
or even undiferentiated from its ensemble of realised or even potential texts. It is a
metadiscursive construct () that grows out of and refers to actual cultural practices,
which themselves are presumably to be studied ethnographically, in addition to con-
stituting the essence of ethnographic method itself. (Silverstein & Urban 1996: 2).
Te natural histories perspective invites attention to the full range of strategies and
modes of (re)entextualization (e.g. transcription, copying, dictating, editing, enacting,
translating, commenting, etc.) and seeks to examine the specifc ways in which such
performative operations connect with particular textualities which are constitutive of
social processes. For instance, Mertz (1996) examines how overtly authoritarian forms
of classroom interaction and interrogation in US law schools play a role in inducing
students into a pragmatic ideology of text in which past cases are no longer read as
narratives but as defning an anticipatory space for pragmatic leeway, promoting an
ideology of text which is characteristic for an elite professional ethos. Collins (1996)
draws attention to how the diferential insertion of textual materials into low-ranked
and high-ranked reading classes goes together with conceptions of the text as contain-
ing a set of phonetic instructions or, semantically, as containing a narrative; he also
explores the links with the conclusions about literacy which pupils draw for them-
selves as a result of sustained exposure a particular type of recontextualization practice
and its embeddedness in routine forms of classroom management by the teacher.
In Kristevas original article, intertextuality represents an impersonal state-of-play
and it is central to a defnition of what is a text. Although essentially dealing with the same
phenomena (but twenty or more years later), (re)contextualization/ entextualization
stresses the contextual dynamics of what is text in relation to intertextual spaces. In
174 Stef Slembrouck
this view, (inter)text is a contextually grounded process, publicly recognized textualities
count as by-products of situated practice and cannot be divorced from language ideolo-
gies. On top of that, the very idea of text is exposed as ethnocentric (and inadequate),
unless it is itself recognized as a stake in social and cultural processes and studied eth-
nographically, as linked to the metadiscursive orientations of actual discourse partici-
pants. Interestingly, the concept of (re)contextualization/entextualization then does not
demolish or bypass the concept of the author/speaker.
Intertextuality is about how history is inscribed into situated practice. As I hope
to have demonstrated in the course of my explorations, it suggests a kind of historicity
that calls for inter-disciplinarity.
References
Bakhtin, M. (1968 [1929, 1965]). Rabelais and his World. MIT Press.
(1973). Problems of Dostoevkys Poetics. Ardis.
(1981 [19345]). Discourse in the Novel. In Te Dialogical Imagination: 259422. University
of Texas Press.
(1986 [19523]). Te problem of speech genres. In Speech Genres and Other Late Essays:
60102. University of Texas Press.
(1990 [1924]). Te problem of content, material, and form in verbal art. In Art and Answer-
ability: Early Philosophical Essays: 257325. University of Texas Press.
De Beaugrande, R. & W. Dressler (1981). Introduction to Text Linguistics. Longman.
Bjrklund, M. (2000). Mikhail Bakhtin. In this Handbook of Pragmatics. John Benjamins.
Bourdieu, P. (1977). L economie des changes linguistique. Langue franaise 34: 1734.
(1984). Questions de Sociologie. ditions de Minuit.
Collins, J. (1996). Socialisation to text: structure and contradiction in schooled literacy. In
M. Silverstein & G. Urban (eds.): 203228.
Duranti, A. (1997). Linguistic Anthropology. Cambridge University Press.
Duranti, A. (ed.) (2001a). Keyterms in Language and Culture. Blackwell.
Duranti, A. (ed.) (2001b). Linguistic Anthropology: A Reader. Blackwell.
Fairclough, N. (1992a). Discourse and social change. Polity Press.
(1992b). Discourse and text: linguistic and intertextual analysis within discourse analysis.
Discourse and Society 3(2): 193217.
Hatim, B. & I. Mason (1990). Discourse and the Translator. Longman.
Heller, M. (2001). Legitimate Language in a Multilingual School. In M. Heller & M. Martin-Jones
(eds.) Voices of Authority: 381402. Ablex.
Heritage, J. (1997). Conversational analysis and institutional talk. In D. Silverman (ed.) Qualitative
Research: 161182. Sage.
Hanks, W. (1996). Language and Communicative Practices. Westview Press.
Kristeva, J. (1968). Problmes de la structuration du texte. In Tel Quel. Torie densemble. Editions
du Seuil.
(1969). Narration et transformation. Semiotica 1(4): 422448.
(1974). La rvolution du langage potique. Editions du Seuil.
Intertextuality 175
Lemke, J. (1995). Textual Politics. Discourse and Social Dynamics. Taylor & Francis.
Maybin, J. (1999). Framing and evaluation in ten- to twelve-year-old school childrens use of repeated,
appropriated, and reported speech in relation to their induction into educational procedures
and practices. Text 19(4): 459484.
Mertz, E. (1996). Recontextualization as socialization: text and pragmatics in the law school class-
room. In M. Silverstein & G. Urban (eds.): 229249.
Rampton, B. (2003). Hegemony, social class and stylisation. Pragmatics 13.
Sacks, H., E. Scheglof & G. Jeferson (1974). A simplest systematics for the organisation of turn-
taking for conversation. Language 50: 696735.
Schiefelin, B., K. Woolard & P. Kroskrity (1998). Language Ideologies. Oxford University Press.
Silverstein, M. & G. Urban (1996). Te natural history of discourse. In M. Silverstein & G. Urban
(eds.): 117.
Silverstein, M. & G. Urban (eds.) (1996). Natural Histories of Discourse. Te University of Chicago
Press.
Slembrouck, S. (1992). Te parliamentary Hansard verbatim report: Te written construction of
spoken discourse. Language and Literature 1(2): 101119.
Voloinov, V. (1985 [1929]). Marxism and the Philosophy of Language. Harvard University Press.
Wetherell, M. (1998). Positioning and interpretative repertoires: conversation analysis and post-
structuralism in dialogue. Discourse and Society 9(3): 387412.
Manipulation
Paul Chilton
University of Lancaster
Manipulation has a bad name, unless you are a physiotherapist by profession. In
English the word was not current in even its basic sense (to fashion something, to
change its shape by hand) until the middle of the nineteenth century. But it soon
developed a more abstract sense the sense of treat unfairly, by skilful means to ones
own advantage. Te skilful means came to include, perhaps predominately, verbal
means. I take it that our notion of verbal manipulation is closely related to what we
also call persuasion. Persuasion was directly associated with the teaching and practice
of rhetoric whether it was regarded as manipulation depended on how you viewed
rhetoric and on your theory of rhetoric itself. Most people assume that verbal manipu-
lation exists and that it works. In this paper I want to present some of the ways in
which manipulation and persuasion have been viewed, but also to ask whether there
is such a thing as manipulative persuasion that is intrinsically verbal.
1. The ancient technique of rhetoric
Te art of persuasion is traditionally said to have been founded by Gorgias, in Sicily,
in the ffh century BC. It is said to have been subsequently developed in Greece
by Isocrates, whose successors, not least Aristotle, bequeathed the discipline to the
Roman rhetors, best known of whom are Cicero and Quintilian. Gorgias is said
to have had an attitude akin to that of a modern publicity agent persuasion is a
transferrable technique, morality or truth irrelevant. Isocrates on the other hand is
credited with the opposite approach, an approach transmitted to the culture of Renais-
sance Europe. Until the eighteenth century at least, schoolchildren in Europe were
taught the art of persuasion. Rhetoric was a detailed, prolifcally classifed collec-
tion of tricks for professional persuaders, politicians and lawyers. It ofered linguistic
devices for example, how to remember your speech, what gestures to use and how
to project your voice. Rhetorical manuals gave instructions for thinking up topics
for speeches and for diferent parts of speeches. It provided recipes for diferent pat-
terns of lexical repetition, for trope or semantic manipulations such as irony, hyper-
bole and, most notoriously, metaphor. And the manuals told you which ones were
best for which type of speech or occasion of use legal, political, ceremonial, and so
Manipulation 177
on and how to distribute diferent tricks throughout the speech, so that, for instance,
you could conclude with a rousing verbal fnale (or peroration).
We do not know for sure if, or how, peoples minds were indeed persuaded. At
any rate, rulers were prepared to employ professional persuaders, and people were
prepared to spend time and money learning their secrets. Rhetoric did not, how-
ever, always have an easy ride. It seems that Gorgias, and his teachers Corax and
Tisias, made their names defending and prosecuting legal cases. Gorgias acquired
a reputation for highly decorated and specious language, while having no concern
whatever about the truthfulness of its content. Persuasion was all, and the clients
interest paramount. Isocrates is credited with higher moral standards in the verbal
arts, and his high-minded verbal ethics was continued centuries later by the famous
orator Cicero.
When empirical science, and the frst shoots of modern democracies began to
appear in eighteenth-century Europe, the art of rhetoric fell from its pedestal. Philoso-
phers such as Hobbes and Locke regarded rhetoric as a threat to right thinking. In Eng-
land, the Royal Society the newly founded scientifc academy equally condemned
ornamental, emotive and persuasive uses of language.
2. The twentieth-century nightmare of thought control
It would be absurd to think that oratorical skills and linguistic persuasion then
dwindled and died. Indeed, rhetoric may have done no more, all along, than crudely
record the natural devices used by individual speakers, with lesser or greater degrees
of skill. Anyone who has brought up children will suspect that verbal manipulation
is a part of the linguistic interaction engaged in by all human individuals. Te social
proscription of lying, and the fact that lying itself is not a hard-and-fast category but a
spectrum of distortions and twistings and (as one English politician has put it) being
economical with the truth, shows that the use of language to get people to do or believe
what you want them to do or believe has a natural life of its own.
One factor is of high signifcance as the age of rhetorical pedagogy ends and the
industrial and post industrial ages begin: the huge and increasing size of the public
forum. At the same time, the transport networks develop and the communication
networks explode. Te frst mass circulation newspapers appear to create a public
sphere that could in principle encompass a mass citizenry. Face-to-face persuasive
communication becomes, relatively speaking, a thing of the past. Mediated mass com-
munication the mass media becomes central to social and political functioning. Te
use of language to manipulate thus becomes qualitatively and quantitatively diferent
from anything that had preceded it.
178 Paul Chilton
Te theoretical question at the bottom of the topic of manipulation is: does or
can language determine the thoughts you have? In looking at questions formulated
with the word language in them, one needs to be careful. One needs to decide
whether we are talking about language in use used by someone to someone in
a defnite situation. Or are we talking about language in general, or about a lan-
guage in particular? Since these are complicated matters all we can say here is the
following. Language and thought can be separated. A particular language (Welsh,
Hopi, Korean) may lexicalize and to a lesser extent grammaticalize thoughts dif-
ferentially, though this does not prevent speakers thinking thoughts not so lexical-
ized or grammaticalized. And, with respect to speakerhearer communication, a
speaker may, or at least so one tends to think from informal practical evidence,
persuade a hearer to change their thoughts but if so there are many ancillary con-
ditions required for efective persuasion to take place.
Twentieth-century writers, however, took harder positions on such questions,
a fact that explains the emergence of the notion of linguistic manipulation. Perhaps
unfairly Sapir, and perhaps less unfairly Whorf, became associated by linguists with
the notion of linguistic determinism. Te particular language you happen to speak
(Welsh, Hopi, Korean) determines the way you think: you think Welsh, Hopi,
Korean thoughts. Tis claim the Sapir-Whorf hypothesis is notoriously difcult to
defne, let alone demonstrate experimentally. Its key feature is that it denies the sepa-
rability of language and thought. Te importance of this extreme form of this fantasy
for the present exposition is that it lays the foundation for George Orwells nightmare.
In Nineteen-eighty-four, Orwell gave us a novelistic, albeit unusually detailed, story of
linguistic mind control.
Orwells nightmare was that in some future dystopia the rulers would be so
powerful that they would be able to engage in a form of language planning that would
make the entire population speak a new variant of English Newspeak. Tis new lan-
guage would be identical with a set of thoughts, the thoughts of the approved political
ideology (which Orwell called Ingsoc):
Te purpose of Newspeak was not only to provide a medium of expression for the
world-view and mental habits proper to the devotees of Ingsoc, but to make all other
modes of thought impossible.
(Orwell 1949: 305, in Te Principles of Newspeak, the appendix to his novel)
Whether this should be read as fanciful fction or not, Orwell clearly considered that
thought could be determined by substituting one word, or set of words, for another
word or set. His position represents a limiting case for the theory of linguistic manipu-
lation that haunted theorists who came from the academic world of linguistics in the
second half of the twentieth century.
Manipulation 179
3. Manipulation is not inherent in language structure
Taking their cue from Orwell, Fowler, Kress and Hodge adopted theoretical
perspectives that come very close to equating, if they do not entirely equate, language
on the hand and thought on the other. Te titles of their books have an Orwellian
cast Language and Control (1979) and Language as Ideology (1979, 1993). Te theo-
retical perspectives came mainly from the version of generative grammar prevailing at
the time, although later they were to espouse Hallidays functional-systemic grammar.
Generative grammar provided the apparatus of transformation, which in Chom-
skys formulation was an entirely syntactic operation thus actually divorced from
thought and meaning. What it gave the Critical Linguists as this group came to be
known was a tool for arguing the claim that syntax could be used for manipulation.
Te argument, schematically speaking, runs as follows. A transformation takes an
input structure, known at that time as deep structure, and gives as its output another
grammatical form, its surface structure. For the Critical Linguists, the input structure
could be seen not as a mere syntactic form (a string of syntactic elements) but as a
meaning. Typically, such a meaning would be an explicit representation of an action: X
does Y to Z, for example. Such a meaning structure could be transformed. And such
a transformation could be ideologically motivated. Tis motivates the notion of lan-
guage as ideology: language is ideology. Two transformations were and continued to
be frequently mentioned: passivization and nominalization. Consider (cf. Trew 1979):
(1) Te army destroyed the house.
(2) Te house was destroyed by the army.
(3) Te house was destroyed.
(4) Te destruction of the house by the army
(5) Te destruction of the house
Sentences (1) and (2) are transformationally connected (passivization). Te afected
participant (demonstrators) could be said to be manipulated into subject position.
Te speakers or writers motives for doing this might be not at all ideological; they
might be to achieve consistency of discourse perspective, if, for example the theme of
the paragraph is a particular house. But (3) lent itself more to the claim that something
was being suppressed namely, the identity of the agent of the action, the identity of
the person or persons responsible for an action. A similar argument can be applied
to nominalization, illustrated in (3) and (4). Deverbal nouns are in parallel with verb
phrases in that agentive role players may be lef unmentioned. Prima facie, we have a
clear case of linguistic manipulation here: an action is being represented linguistically
with part of the story missing.
180 Paul Chilton
Te trouble with these analyses is that one cannot be sure that the hearer or reader
is incapable of inferring the identity of the agent. Is the claim that syntax is inherently
ideological because it enables speakers to manipulate by omission? Tis cannot be
the case because agent-less passives ofen seem to occur precisely because the agent is
manifest in the context or can reasonably be expected to be inferred. We are lef with
the possibility that these constructions may be used, given certain, circumstances to
exclude, or occlude, the agent. But this is far from identifying language and thought
and requires many circumstantial factors to explain how a hearer may be taken in.
Tis kind of criticism, and others, has been made by Pateman (1987).
A framework that does bring into play some of these other factors is Critical
Discourse Analysis, best represented in the work of Fairclough (e.g. 1989) and also
Wodak (1996). Te extra factors are political and social. Tis is not to say that such
factors are absent in the earlier work, merely that in the later work, the emphasis is
less on linguistic mechanisms, and more on underlying frameworks of political analy-
sis. Fairclough, for example, has as his starting point concepts of power and exploita-
tion, and the critical involvement in language analysis is wider and more difuse. In
particular, it involves Foucaults concept of discourse and discourse formations. Te
point is that, manipulation is seen now as part of a complex web of interacting loci of
discourse forces of power and resistance in diferent settings, with varying degrees
of institutional sanction. An important distinction (Fairclough 1989) is that between
power behind discourse and power in discourse. Power behind discourse includes, for
example, the physical sanctions underpinning an order (itself an linguistic act) given
by a policeman. Power in discourse might be instantiated in the use of an agent-less
passive, as described above, in circumstances (a) where the hearer could not possi-
bly identify the identity of the agent and (b) where the same hearer had a signif-
cant (social, political, economic, ethical) interest in the identity of the agent. But it
could equally be claimed to be instantiated in a whole range of manipulative devices.
For example: alliteration, intonation, regional or class accent switching, passiviza-
tion, nominalization, pronouns, metaphors, antonymy, tripartite co-ordinated noun
and verb phrases, presupposition. And then there are strictly pragmatic phenomena:
interruption, back-channelling, body language, latching, questioning, irony, rhetori-
cal questions, evasive responses in question-answer pairs, defocusing by syntactic and
semantic selections as in presupposition and accommodation, etc.
However, linguistic structures are not in my view inherently deceptive or
manipulative, though there have been evolutionary psychologists who claim that the
origin of language, like that of social intelligence itself, is essentially machiavellian
manipulation (Knight 1998). Linguistic structures are just linguistic structures. Human
users can manipulate them (like one manipulates a lump of clay, for example) with
goal-directed intentions. Examples of such intention might be to deceive a hearer as to
physical or social facts, which might involve getting a hearer to form new thoughts about
Manipulation 181
social and physical facts. But there are many other possible intentional deployments of
language. All this being so, it is more logical to investigate strategies (i.e. goal-directed
plans of action, here verbal action) directed at deception or at changing a hearers rep-
resentation of social and physical reality.
1
An example of such an approach is Haber-
mass notion of strategic communicative action (e.g. Habermas 1979, 1981). Within the
context of language and politics Chilton & Schfner (2002: 141) propose strategies,
recognisable as political: legitimization of the self, delegitimization of others, coercion,
and dissimulation. To realise such strategic ends speakers may use a variety of linguistic
structures in interaction with context; there is no necessary correspondence between
strategy and structure. Moreover, whether an utterance is strategic or not is dependent
upon the ability of human language users to detect and metarepresent strategic intent.
4. So lets look at thought and social action
In all approaches to manipulation the key ideas appear to be as follows. (1) A persons
thoughts and possibly actions are changed as a result of some controlling use of lan-
guage. Tis includes getting people to accept a representation of states of afairs that
are diferent from what they really are. (2) Some information may be omitted or
partially obscured. (3) Te manipulator intends (1) and/or (2). And (4), the person
manipulated is not aware of what the manipulator has done to their thoughts.
2
I con-
sider key aspects of these assumptions below. In particular, what it might mean to
have ones thoughts changed, whether that change involves the modifcation of exist-
ing thoughts, the elimination of existing thoughts or the addition of thoughts not
before entertained? I shall consider not simply the individual case but how it might be
possible for a population to be manipulated into changing its mental representations.
4.1 Drumming it in
Perhaps a linguistically formulated idea can be repeated so ofen that the mind stores
it as self-evidently true? According to such a view, repeated activation of mental rep-
resentations especially those that are connected with basic (innate or at least early
:. I do not wish to deal with epistemological issues here, but I should mention that when I use
the term reality here, I am indicating a group-coordinated mental representation of perceptions.
i. We might also consider the case where the person believes consciously that their change of
mind is reasonable and justied. But have they been manipulated into that conscious belief? I do
not pursue the implications of this question in this paper.
182 Paul Chilton
acquired) cognitive representations, including metaphor structures, can lead to change
in peoples minds. Such changes may be similar across a population under the appropri-
ate conditions. Appropriate conditions here will include exposure of a population to a
repeated message that is, a repeated mental representation stimulated by a linguistic
expression that sufciently large numbers of individuals are mentally processing. How-
ever, while repetition may be a necessary condition for the widespread fxing of a men-
tal representation in a population, it surely cannot be sufcient. For example, to process
the incoming linguistic stimulus into a mental representation, and then to store it into
long-term memory, the receiving mind must, in the frst place, fnd that representation
relevant. Tis is a factor that can be further understood in terms of Relevance Teory
(Sperber & Wilson 1995), the core idea being that individuals expend processing efort
in order to construct meaning on the basis of incoming linguistic material but they
will only do so to the extent that the processing cost is worth the meaning beneft. What
determines the beneft will, of course, depend on individual circumstances and, for
populations, the socio-historical contingencies of the context.
4.2 Ideas that spread
Tese considerations suggest that it is not possible to predict easily how and when
thought manipulation by means of language could occur. Te question can in
fact be viewed the other way round: how likely is it that thoughts change without
manipulation? Here we can think of change in spatial terms, as well as temporal, as
the spreading of mental representations. Tis is precisely what Dawkins (1976) has
mooted with his notion of memes, and Sperber (1996) with his notion of epidemiol-
ogy. Dawkins thinks that ideas can be regarded as not just compared to genes
which replicate themselves and sometimes mutate. He also compares some memes
(the ones he especially dislikes, such as religious ideas) to viruses. Because Dawkins
thinks of genes as replicating themselves blindly, it should follow that he also thinks
of ideas as replicating themselves blindly, without human deliberation, the success-
ful ones just reproducing themselves because it is their property to do so. But the
odd thing about Dawkinss exposition of this theory is that it frequently seems to
involve what he regards as wicked indoctrinators manipulators of the mind (e.g.
Dawkins 1982, 1993). Te problem with this is that manipulators must be presumed
to have intentions, whereas the meme idea seems to imply that people are merely the
medium for the self-replicating memes.
Dawkins (1982: 57, citing Dawkins & Krebs 1978) regards all of animal
communication as manipulation of signal-receiver by signal-sender. But also, he
argues, an arms race of communicative counter-measures and counter- counter-
measures will ensue. And humans do not entirely interact in stimulus-response fashion.
Manipulation 183
At the end of his book, Te Selfsh Gene, Dawkins says (1976: 200) that we have
the power to defy [] the selfsh memes of our indoctrination. Dawkinss idea has
inspired so-called memetics, whose proponents see it as their mission to diagnose
and treat invading thought viruses. Note too that if Dawkins happened to be right
in his claim that memes replicate themselves, like genes, for no ones advantage but
their own, then they cannot originate in the minds of human manipulators for the
purposes of the manipulator.
Dawkins gives examples of human manipulation: advertisements, nuns, teachers,
brainwashing of prisoners, hypnotism, electrical stimulation of the brain. It is important
to note that these examples involve forms of non-linguistic control by powerful per-
sons over contextually weak persons readers of adverts cannot easily check claims in
a hurry, nuns and teachers (according to Dawkins) have direct access to undeveloped
young minds, hypnotists have by one means or another persuaded their subjects to
suspend their consciousness. Human agents are certainly involved in such examples,
as are complex social institutions within which they work. Tis means that there is
more to manipulation than memes blindly self-replicating. Te idea of memes is in
fact self-contradictory: memes cannot be both blindly self-replicating and deliberately
deployed by human beings. Te jump from genetics to memetics is not scientifcally
warranted and is used to disparage, on pseudo-scientifc grounds, a range of targets
that should not be lumped together.
Sperbers (1996) use of the Darwinian metaphor leads down a slightly diferent path.
Considering culture rather than manipulation, he sees (linguistic) communication as
the means by which (cultural) ideas spread through a population, resulting in stabili-
zation of ideas across its members and over generations. But for Sperber (2000) this is
not necessarily a pathological process, though it is like an epidemic. More to the point,
his view of communication not only builds on relevance theory, as outlined above, but
also the notion of a logico-rhetorical module, an evolved mental mechanism which
enables addressees to check for internal and external consistency of incoming lan-
guage. Why should such a biological adaptation have evolved? Sperbers claim belongs
to a background of thought that has direct bearing on the notion of manipulation.
According to Byrne & Whiten (1988) human communication, bound up with social
intelligence, is inherently machiavellian, and thus manipulative. Communication nec-
essarily involves the possibility of lying and deceiving, and so a consistency-checking
device is a natural adaptation that is part of the human communicative set-up. Sperber
draws in particular on Cosmides (1989), who postulates a cheater detection device
in the human mind-brain. Communication has to be co-operative or no one would
want it it would not evolve. If it does evolve, some people might still want to cheat
manipulate the co-operative folk. But a point would be reached where everyone
would end up being cheaters, so humans must have acquired some cheater-detection
184 Paul Chilton
ability. Tis is consistent with what Dawkins (1982) and Dawkins & Krebs (1978) call
an adaptive arms race in evolutionary time.
3
What Sperber proposes is that because humans are so communication-dependent,
evolution has given humans a consistency-checking module to check the
internal logical consistency of what people say and to check the consistency of its
content with their existing beliefs about what is true or real. Of course, here the
persuasion-counter-persuasion war escalates Quite likely we leave innate endow-
ments and enter the cultural. Te ability to check arguments and rhetoric through
scrutiny of the linguistic sits on top of an innate cheat-detection disposition. But its
phenotypes are, or can be, culturally honed through education, critical discourse
analysis and such cultural forms.
Te implications of this story for the notion of manipulation are considerable. If
everyone has a consistency checking device, then nobody need be manipulated. Tis is
an extremely important principle. However, one might add a proviso: nobody need be
manipulated if their device is functioning, and functioning well.
5. What might override the cheat-checker?
Relevance theory can supply some of the answer. For example, what is relevant to a
hearer will be afected by the degree of trust in the speaker. If you trust the speaker you
are more likely to process what they tell as being true. Tis is an extremely complex
matter, for it involves factors external to the hearer as well as the interests of the hearer
herself in believing that what the speaker says is true. Tis is as much as to say that her
consistency-checking device might be swamped by para-linguistic and extra-linguistic
factors (stirring music, promises, handsome appearance), and, more seriously
for manipulation theory, it is as much as to say that she may be deceiving herself in
her own interests. In short, I am suggesting that the responsibility of the hearer for
being manipulated could be considerable. Te issue of manipulation is rarely turned
back-to-front in this fashion, but it provides a heuristic for probing further.
What then are the factors that might override the natural functioning assuming
that such is the case of the consistency-checking module? Tis question provides us
with the shape of a whole research programme, but there are three paths we might think
worth going down. One is psychological, specifcally cognitive. We need of course to
. Another important line of argument is that manipulative deception is closely bound up with
the basic human ability to detect symbolic ctions with a deeper meaning that are distinct from
truth claims about physical and social reality, such symbolic ctions having the adaptive advantage
that they underpin collective representations of group identity (Knight 1998).
Manipulation 185
know whether the hypothesized logico-rhetorical or consistency- checking module
actually exists and eventually what are its neuroanatomical substrates. Psychological
and psycholinguistic experimentation might establish this and reveal something of its
workings. In particular, we need to know what factors interfere with its operations. If
conscious inattention is involved, for example, we need to know what distraction fac-
tors can lead people to accept representations without checking them. As suggested
above, these might be linked to the nature of trust, to interference with the reception
channels by such means as musical or pictorial stimulation (cf. OKeefe 2002). Te
limiting case of manipulation is perhaps torture. Does it work? Although Nineteen
eighty-four tells the tale of one mans mental representations being changed in this way,
it is by no means obvious or acceptable that they can be changed as distinct from
verbally disavowed.
Te second type of override could be sociological. Habermas, for example, has
asserted that the distribution of power distorts communication in line with the stra-
tegic interests of powerful communicators (cf. Habermas 1998), and has imagined
the counterfactual (ideal) communicative situation in which manipulation would
not be possible because all interlocutors had equal rights and powers of expression
and understanding. We might re-interpret this in the light of our earlier discussion:
the ideal situation would be one in which everyones machiavellian intelligence and
logico-rhetorical modules were equally free to function. To be concrete, what this
might mean is that everyone has the freedom to develop their communicative skills.
Most people can do this, one might guess, in everyday micro-settings, perhaps within
some gullibility range, but human needs in large-scale societies are not well served by
innate mental abilities evolved in small-scale communities.
Te third, and the most obvious and controllable extension of these ideas is
physical and political limitation and control of the means of communication. Tere
are two ways in which we can pursue this. On the one hand, there is what Fairclough
(1989) calls power exercised in discourse. Particularly skilful or trained speakers can
use language in such a way that hearers critical consistency-checking modules do not
work well. At the most obvious, such speakers can lie or withhold information. Tey
may also use techniques that induce hearers to incorporate unspoken assumptions in
their mental representations without being aware of it. Such claims are, however, dif-
cult to sustain. Tey may be able to use emotive means through language to induce the
acceptance of propositions. Tese last two claims, though attractive, and traditional,
are difcult to demonstrate and in all probability cannot work without the sort of
overrides mentioned in connection with the frst path.
On the other, there is a manifestly obvious form of manipulation controlling the
channel of communication. Tis may involve social and economic control limiting
access to education, limiting education, banning TV sets or restricting opportunities
to speak or write in public media. If access in some degree is granted, then what is
186 Paul Chilton
uttered may be restricted. Tese are forms of negative control: essentially, censorship.
Te positive (not good, of course) face of channel-control is propaganda that is the
reservation of the right to speak in public to certain spokespersons and to constrain
their utterance-meanings within certain semantic and pragmatic limits. Both dimen-
sions of channel-control are manipulative in the sense that mental representations
can perhaps somewhat crudely be confned. Tis holds in particular for represen-
tations of states of afairs that are not witnessable, such as government deliberations
behind closed doors, events in remote regions, events that have taken place in the past.
If people do not have multiple sources of empirical knowledge, if their only source
is provided by, say, a government ofce of information, then their ability check for
consistency has nothing to work on. A further signifcant form of channel-control in
this sense would be the denial of language rights: if a community in a polity cannot
understand the broadcast language it is in the worst possible situation.
6. Conclusion: Manipulation and counter-manipulation
Te core question is this. How can a speaker control the meanings that enter the
mind of a hearer? Of course they do not go into the mind from outside the hearers
head they are not put there by the speaker. Tey are made in the mind-brain
of the hearer and by the mind-brain of the hearer in response to the speakers
utterance but this is not to say we are looking at a determinate causal stimulus-response
efect. Rather, what happens is that the speakers utterance prompts the hearers mind
to access her own knowledge bases in order to build an integrated conceptual repre-
sentation out of the incoming linguistic material, and perhaps respond emotionally in
the process. Tis is a very complex process and it is beyond the control of the speaker.
But, we have asked, how far is it beyond the control of the hearer? Well, it is doubt-
less true that human brains do automatically process linguistic input, at least to some
degree. But if that is so, then it applies to all communication and not just to some
particular kind of communication that we call manipulative. Te general point is that
speakers cannot be sure about the precise mental representation their words may lead
to in their interlocutors brain; they can be sure enough for social co-ordination to
take place, but there is a margin of unpredictability. If meaning escapes the control of
speaker intention (Lee 2001: 207), we need to be circumspect regarding the question
of manipulation, or thought-control-by-linguistic-means.
If representations (call them ideas, notions, beliefs, attitudes, if you wish) are not
chosen by individuals, or indeed by groups of individuals then the individuals are not
responsible for holding them. So it might seem. Note that if individuals do not choose
to have the representations they have, then they cannot, according to our usual ethical
Manipulation 187
calculus, be held responsible for them. So it might seem. But is it really so? For, while
it may be the case that people may grow ideas, beliefs etc. without their being aware of
it, it remains the case that the human mind has the capacity or the potential to inspect
or introspect their beliefs, etc. Broadly, this idea corresponds to what are called meta-
representations in the cognitive science literature; language systems themselves have
metalinguistic mechanisms that support the evaluative representation of truth claims.
Taking the key points of this line of thought further, and considering the uti-
lizations of the unconscious knowledge of language, i.e. utterances produced and
processed in contexts of human interaction, human language-users might have a
tendency to assume the truth of what people tell them, but that does not intrinsically
absolve them from the responsibility of checking for veracity indeed, humans do
have this capacity, and use it more or less automatically, though there may be limita-
tions on their tendency to do such checking under certain social or psychological
conditions.
Now, we can look at the spread of representations the other way round, focusing
not on the receiving minds but on the hypothetical sending mind and communicator.
If it is the case that representations in some sense reproduce themselves by being
attracted to cognitive formats already existing in the mind-brains of individuals and
groups, then is it the case that communicators communicating such adaptable rep-
resentations are responsible for that fact? I guess, they would be if they knew what
they were doing. Let us consider what knew what they were doing might mean. It
seems implausible that a communicator passing on some infectious representation
has calculated the process in detail. Did Hitler, for example, say consciously to himself,
I will call Jews parasites because that conceptual chunk will attach itself to a natural
pre-existing mental schema in the minds of my listeners? Tis is not of course the
same as saying that such a communicator does not know, from empirical experience
of the efectiveness of such communications, that he is likely to promote the represen-
tation in question, and his purpose may indeed be to change recipients beliefs, ideas,
attitudes, etc.
Tere remains a problem, however, if one now asks how such representations
actually spread. One possibility is that they are planted simultaneously in the minds of
a population of individuals receiving a particular broadcast communication, or series
of broadcasts. Tis cannot be the whole picture, because people talk to one another.
If then, their linguistic communications express the infectious idea (say, Jews are
parasites), it can be picked up by many hearers in the communicative networks of
the speech community without the spread of the idea being the express purpose
of the communicators. Even if it is not their express purpose, however, would one
hold them responsible for a manipulative act? Here is perhaps the nub of the matter.
As already noted, human mind-brains certainly have the capacity to refect on their
unconscious cognitions, communicative behaviour and physical action. It is arguable,
188 Paul Chilton
however, that the tendency is not to do so (it would not be helpful to be always intro-
specting while out hunting, for example), unless trained, educated or encouraged by a
whole variety of complex factors.
Having said that, does any more follow? Probably. Introspections of all kinds
take time and technique: one way to acquire the technique of critical introspection
is through education. Public education may not provide it, either because a priest or
dictator, or for the matter of that a teacher, wishes to suppress such critical techniques,
or through sheer force of circumstances and lack of opportunity. Critical Discourse
Analysis, which is less a linguistic discipline than a social movement, seeks to do just
this (cf. Fairclough & Wodak 1997). Te material point, for making sense of the term
manipulation is that it does not occur if communicative receivers evolve and hone
their checking devices. A key implication of this point is that manipulation will take
place not merely by way of deceptive use of language or infective distribution of ideas
that take hold in the mind: it is necessary also for the target individuals to be beref of
critical defences.
Tere is not much that is novel in this line of argument, which has existed since
the Enlightenment, but is now out of favour since technical training of various kinds
has replaced education as an ideal education in the sense of drawing out innate abili-
ties, including critical abilities. At least the teaching of rhetoric potentially gave some
small purchase on the operations of verbal manipulation.
References
Byrne, R.W. & A. Whiten (eds.) (1988). Machiavellian Intelligence: Social expertise and the evolution
of intellect in monkeys, apes and humans. Clarendon Press.
Cicero, M.T. (1942). De oratore. Heinemann.
Chilton, P. & C. Schfner (2002). Temes and Principles in the analysis of Political Discourse. In
P. Chilton & C. Schfner (eds.) Politics as Text and Talk: 141. John Benjamins.
Dawkins, R. (1976). Te Selfsh Gene. Oxford University Press.
(1982). Te Extended Phenotype. Oxford University Press.
(1993). Viruses of the Mind. In B. Dahlbohm (ed.) Dennett and his Critics: Demystifying the
Mind: 1327. Blackwell.
Dawkins, R. & J.R. Krebs (1978). Animal Signals: Information or Manipulation? In J.R. Krebs &
N.B. Davies (eds.) Behavioural Ecology: 282309. Blackwell.
Fairclough, N. (1989). Language and Power. Longman.
Fairclough, N. & R. Wodak (1997). Critical Discourse Analysis. In T. van Dijk (ed.) Discourse as
Social Interaction: 258284. Sage.
Fowler, R., B. Hodge, G. Kress & T. Trew (1979). Language and Control. Routledge Kegan Paul.
Habermas, J. (1979). Communication and the Evolution of Society. Heinemann.
(1981). Teorie des kommunikativen Handelns. Suhrkamp.
Hodge, R. & G. Kress (1979). Language as Ideology. Routledge. (2nd ed. 1993).
Manipulation 189
Knight, C. (1998). Ritual/Speech Coevolution: a Solution to the Problem of Deception. In
J.R. Hurford, M. Studdert-Kennedy & C. Knight (eds.) Approaches to the Evolution of Language:
6891. Cambridge University Press.
Lee, D. (2001). Cognitive Linguistics. Oxford University Press.
Okeefe, D.J. (2002). Persuasion: Teory and Research. Sage.
Orwell, G. (1949). Nineteen eighty-four. Secker & Warburg.
Patemen, T. (1987). Language in Mind and Language in Society. Clarendon Press.
Plato (1925). Gorgias. Heinemann.
Quintilian, M.F. (192022) De institutione oratoria. Heinemann.
Sapir, E. (1970). Culture, Language and Personality. Selected Essays. University of California Press.
Sperber, D. (1996). Explaining Culture: A Naturalistic approach. Blackwell.
Sperber, D. (ed.) (2000). Metarepresentations. Oxford University Press.
Sperber, D. & D. Wilson (1995). Relevance. Blackwell.
Trew, T. (1979). Teory and Ideology at Work. In R. Fowler et al. (eds.): 94116. Routledge & Kegan
Paul.
Whorf, B.L. (1956). Language, Tought and Reality. Selected Writings of Benjamin Lee Whorf. MIT
Press.
Wodak, R. (1996). Disorders of Discourse. Longman.
Narrative
Alexandra Georgakopoulou
Kings College London
1. Narrative as a mode of communication
Narrative is viewed by numerous linguistic and non-linguistic disciplines as
inescapably fundamental in human life, central to the (re)constitution and inter-
pretation of personal, social, and cultural reality. As Bruner has asserted, we orga-
nize our experience and our memory of human happenings mainly in the form of
narrative stories, excuses, myths, reasons for doing and not doing and so on (1991: 4).
Te narrative texts which we are engaged in as part of our everyday life form an
endless list that includes personal stories, comic strips, jokes, autobiography, personal
letters, gossip, legal testimonies, novels, news reports, fairy-tales, retellings of plots of
flms and TV series, etc. Tese texts, popular or artistic (literary), not only crosscut the
spoken, written and electronic modality, but are nowadays increasingly multi-modal
(e.g. verbal-visual-musical).
Te importance of narrative as a mode of communication has been eloquently put
forth by Bruner who identifes (1986, 1990) two major ways of knowing and discur-
sively interacting with reality: (a) the narrative mode, which encodes and interprets
human reality, experiences, beliefs, doubts and emotions; and (b) the paradigmatic
(logicosemantic) mode, which deals with natural (physical) reality, truth, observation,
analysis, proof and rationality. If we consider discourse as a symbolic activity by which
we relate to the world, we can rephrase Bruner by claiming that the narrative mode
performs the symbolic function of recreating or reconstructing reality through a story
(taleworld, Young 1987). Bruner has also contended that the paradigmatic mode has
been privileged over narrative throughout centuries of western rationality. Te bal-
ance has, nonetheless, been redressed by the recent canonizing and sanctioning of nar-
rative within numerous paradigms that advocate its primacy as an epistemology and a
mode of communication. Specifcally, narrative has emerged within personality studies,
psychoanalysis, and psychotherapy as the main if not the only way of discovering
and being ourselves (Kerby 1991). Similarly, numerous (mainly non- western) societies
have been proven to encode their claims about knowledge and its sources (evidentiality)
as well as their views about the world almost exclusively in narrative mode (see
Maranho 1993). Furthermore, narrative (or the lay term story) is widely held as a uni-
versal and archetypal way of applying an order and a perspective to experience. In its
various forms and shapes, the making and appreciation of shared narratives is viewed
Narrative 191
as a particularly crucial component for the cohesion of a culture (Lakof 1997). Tus,
within social and cultural anthropology, narrative has increasingly become an object
of ethnographic analysis with much to contribute to a range of theoretical and meth-
odological issues of interest (e.g. Moerman 1988). Finally, developmental studies have
consistently emphasized socialization through the narrative mode, embracing the ways
in which children are introduced into their parents culture through narrative as well
as trained in the rhetorical skills needed for constructing coherent and culturally tel-
lable narratives (e.g. Bamberg 1997a). In a post-Vygotskyan (1978) manner of locating
developmental processes in the dynamics of social interaction, recent work has focused
on childrens various forms and politics of participation in narrative events that occur
within family life (e.g. Georgakopoulou 2002a; Ochs & Taylor 1992).
Te current fascination with narrative is in many respects a rediscovery: narrative
has fgured in all categorizations of communication on the basis of rhetorical mode (or
stance) from Aristotles Poetics to modern day rhetorical theory. More importantly,
it has been posited as a pre- or meta-genre that cannot be put at the same level as
ordinary genres (Swales 1990), or as a basic text-type (i.e. underlying or overriding
structure instantiated by diferent surface, textual forms); this means that it realizes
but is not realized by other text-types, such as argument, description, and exposition:
e.g. we can tell a story in order to present an argument but the discourse function or
goal of narrating cannot be realized by argumentative texts (Virtanen 1992).
In the light of the above, it is not surprising that narrative analysis is one of the
best and most extensively researched areas of the multidisciplinary study of discourse
(van Dijk 1993: 121), embracing a wide spectrum of disciplines (e.g. linguistics, psy-
chology, history, law, anthropology, semiotics, literary criticism, etc.) and theoretical
and methodological approaches. Tere are, however, downsides to this diversity: frst,
there is is the lack of cross-fertilization, even within linguistics; as a result, similarities
in aims, method, and analysis are too frequently obscured. Similarly, the proliferation
of theories on narrative has blurred boundaries, threatening to subsume all texts and
discourse activities under an all-encompassing narrative which variously embodies
a communicative mode, an epistemological perspective, a set of genres, a discourse
meta-function, or a style (see Georgakopoulou & Goutsos 2000: 65f). All this has
resulted in an idealization and essentialization of narrative that has hampered empiri-
cal research of the generic variability of narrative (idem). Any attempt to typologize
narrative analysis is thus bound to be schematic.
On this understanding, following Mishler (1995: 87123), we can distinguish
between models which focus on (i) reference (relationship between real-world and
narrative events) and temporal order, (ii) textual coherence and structure (narrative
strategies), and (iii) narrative functions (Mishler 1995: 87123). Tis distinction will
be rephrased in the present discussion in terms of the following three confgurations
of narrative properties: referential, textual, and contextual. Te referential properties
192 Alexandra Georgakopoulou
are concerned with the organization of the world that a narrative reconstructs, thus
capturing the texts external relations (Martin 1992). Te textual properties refer to
the organization of the text itself (i.e. text-internal relations); fnally, the contextual
properties refer to the interface between internal and external, that is, with the interac-
tion between the text and its immediate (situational) and wider (sociocultural) context
of occurrence. All three sets of properties will be tackled here on the understanding
that there are inevitable overlaps, intersections and indeed cross-fertilizations amongst
them; also that any critical enquiry into linguistic studies of narrative is a daunting task
with a forbiddingly ample domain. Te present discussion will be mainly concerned
with sociolinguistic and discourse narrative studies. Tese have focused on oral,
non-literary narratives, which are in turn either spontaneously occurring in conversa-
tions or elicited by the analyst; the latter mainly involve oral histories or life stories
(e.g. Josselson & Lieblicht 1993; Linde 1993).
1
2. Referential properties
Te least controversial of the minimal defnitions of narrative seems to converge on
its chronological dimension (cf. Ricur 1988), or in Chatmans terms (1991: 9), the
chrono-logic, which involves the depiction of a temporal transition from one state of
afairs to another. Tis temporal unfolding may concern past, present, future, hypotheti-
cal, or habitual events, everyone of which is defned as an occurrence in some world
:. Literary narrative (cf. narrative ction) has attracted an impressive volume of studies within
(discourse) stylistics and literary pragmatics that has been informed by classic narratological pre-
occupations and concepts (e.g. see Barthes 1977; Genette 1980; Prince 1982). A main strand of
research has analysed the devices which establish perspective or point-of-view; the focus here
has been on the various textual roles or personae mediating between the authorial voice and the
characters (e.g. implied author, narrator, focalizer), in particular as encoded in instances of speech
and thought presentation (Leech & Short 1981). Tropes, gures, and generally, the poetics (tech-
nical choices) or rhetoric (technical choices and ends, Chatman 1991) of narrative ction have
also been systematically researched, mainly with reference to the distinction between surface and
deep narrative structure (cf. histoire vs. discours, fabula vs. sjuzhet, Bals fabula-story-text, 1991), or
between the presentational modes of mimesis (showing) and diegesis (telling, see Rimmon-Kenan
1983). Te series of modications to which narrative action is subjected at the discourse level
have been investigated in relation to time structure (i.e. order and duration), informational selec-
tion and distribution, and point-of-view (see Genette 1980). Te discussion of what constitutes
the property of literariness and its relation with narrative ctionality, an ongoing preoccupation
within narratology, has also penetrated linguistic studies of literature. Relevant approaches are
increasingly pragmatically oriented, focusing on the contextualized functions, uses, and eects of
textual choices within literary enunciation (cf. Fludernik 1996).
Narrative 193
which is encoded in a proposition which receives an instantaneous rather than durative
interpretation (Polanyi 1982: 510). In practice, it is the stories of past events that have
come to be recognized as the prototypical data for narrative analysis. Tis may be related
to the infuence of Labovs study of narrative, in which tellingly the sequence of past
events is seen as being matched by a verbal sequence of (temporally ordered) clauses that
make up a narratives skeleton (1972: 359): in this way, the order of the telling recapitu-
lates the order of the told. Tis question of representation, that is, the relation between
the events represented and the structure of their representation, has been one of the most
debated issues of narratology (cf. fabula vs. sjuzhet).
Central to this inquiry is the very relationship between narrative and real-
ity: the extent to which we can perceive the world in anything else other than a
narrative form or put diferently, whether life is itself narratively structured
or not, is a much debated issue (e.g. Carr 1986, MacIntyre 1981). Te debate is
based on the premise that narrative shapes experience, providing it with mean-
ing, structure and perspective, in the process of constructing a plot (cf. Aristotles
mythos). In this way, they make a point about the world which teller and story
recipients share (Polanyi 1985: 16). In Labov & Waletzkys terms (1967), it is
thus no accident that narratives tend to recount noteworthy events, that is, sur-
prising, disturbing, interesting, or, otherwise, tellable events. In so doing, they
can serve to highlight the restoration of a disrupted order (equilibrium) or ren-
der the exceptional comprehensible (Bruner 1990: 52). Te view of a series of
discrete events (real or fctitious) as the core of a narrative which is elaborated
on, embellished, and manipulated (i.e. perspectivized) has informed both tex-
tual and contextual narrative analyses, as we will see below. What is worth noting
here is that earlier studies tended to assume a correspondence or mirror relation
between narrative and extra-textual worlds, or, in the case of fctional narrative,
a relation of verisimilitude (vraisemblance). Tis was, however, duly problematized
by a constructionist view of narrative, heavily infuenced by post-structuralism
and, further back, by Bartletts seminal work on memory (1932), which saw stories
as selective (re)playings and (re)constructions rather than as accurate refections
and representations of events. As such, their tellings are shifing, dynamic, and
contextually grounded, rather than given and pre-determined. Tey are produced
as part of social interactions in specifc situations and for specifc purposes.
3. Textual properties
3.1 Narrative organization
Aristotles view of plot as a linear (i.e. with a beginning, middle, and end), yet
complicated (i.e. with a peripeteia) progression of events has inescapably infuenced
studies of narrative structure.
2
Within story grammars, narrative organization consists
of episodes (superordinate units) described in terms of spatiotemporal sequences
of the following recursive categories: (i) setting, (ii) initiating event (marking some
type of change in the protagonists environment), (iii) responses or reactions to it,
(iv) attempt (a set of overt actions in the service of the protagonists goal), (v) conse-
quence of the attempt (e.g. success or failure to attain the goal), and (vi) (protagonists)
reaction to these consequences (Rumelhart 1975; Stein 1982). Tese categories are to
be found in the canonical form or underlying story schema from which individual
stories are generated (cf. Propps work on the Russian folktale, 1986). Te schema is
a goal-based structure, consisting of a setting, participants, including the protago-
nist, the desired goals, and the actions for achieving them. Tis episodic structure is
postulated as a conceptual framework that enables addressers and addressees, in the
course of the production and comprehension of stories, to retrieve and strategically
employ information about story characters and events (Johnson-Laird 1983; van Dijk
1988). Experimental evidence has suggested that certain categories (e.g. protagonists
goal-oriented attempts) as well as causal relationships between events are particularly
critical for story comprehension and recall (Trabasso & Sperry 1985). It is, however,
recognized that there is still a long way to go before fully documenting the universal
properties of story grammars.
Unlike story grammars that focused on traditional tales and by and large
disregarded narrative communication in real-world contexts, Labovs narrative struc-
ture was developed for conversational albeit elicited personal stories (1972; revised
in 1997). Te study has been instrumental in legitimizing oral narrative as an object
of study and bridging the gap between vernacular and non-literary narrative. Labovs
description of a fully-formed (or classic) narrative is as follows:
It begins with an orientation, proceeds to the complicating action, is suspended at the
focus of evaluation before the resolution, concludes with the resolution and returns
the listener to the present time with the coda. (Labov 1972: 369)
Both orientation and evaluation were subsequently assigned a free position in the nar-
rative structure. While Labovs formal clausal analysis was rapidly seen as unworkable,
the structural components, evaluation in particular, have proved extremely robust
and have had extensive use for many and varied purposes by researchers in a very
wide range of disciplines (Holmes 1997: 95). As such, they have formed the basis of
various revisions and elaborations. Te most important of those involves their situ-
ational and cultural context-dependency, as these have been attested to by the study
i. It is notable however that afer a longstanding tradition of temporality as the main organizing
principle of narrative, the balance has been duly redressed and the vital role of space increasingly
recognized (e.g. Georgakopoulou 2003b; Herman 2002).
Narrative 195
of spontaneous conversational storytelling: e.g. the degrees of orientation and evalu-
ation are widely variable; non-obligatory categories (abstract, coda) are dispensed
with in conversational contexts more ofen than in Labovs elicited stories. More
importantly, stories -particularly in relaxed and informal settings- tend to be a joint
enterprise, with the audience playing an active role in the shaping of the storys struc-
ture (Duranti 1986).
While Labovs narrative structure captures a storys plot development, another
infuential line of enquiry into the schematic organization of oral narrative has posited
units (e.g. episodes, stanzas, verses) which rest on a combination of thematic, syntactic
and intonational criteria (e.g. Chafe 1987; Longacre 1989). It is generally conceded
that these units are delineated by shifs in action, time, location, and participants: the
higher-level the units, the more salient the shifs and their linguistic signals. Supra-
segmental features of pause, stress and pitch are also important demarcatory signals.
Work on American Indian narratives within ethnopoetics has attested to rhythmic,
poetic patterns in narrative organization, shaped in the magic numbers of each com-
munity (i.e. twos and fours or threes and fves, e.g. Hymes 1981, Sherzer & Woodbury
1987). In this way, narrative structure refects sociocultural modes of interpreting the
world and making sense of experience. Similar rhythmic patterns have been found
in narratives of other cultures with a strong oral tradition. Gees (1990) comparative
analysis of narratives from black working-class and white middle-class children in
America demonstrated that a poetic structure is very evident in the former by way
of relatively short and quite evenly balanced segments across the text as a whole;
these segments are interlocked through numerous poetic patterns such as sound play,
repetition, parallelism, and prosodic changes.
Research into the organizational aspects of storytelling has gone hand in hand
with developmental studies of linguistic forms (e.g. markers of tense/aspect, connec-
tivity) and their corresponding discourse functions as part of the ability to organize
a narrative. Bermans & Slobins (1994) volume has produced a fne-grained devel-
opmental map of linguistic forms and functions across languages (English, German,
Hebrew, Spanish, Turkish) whilst demonstrating the impact of typological features of
a particular language on the development of rhetorical options for narrativization.
Last but not least, another line of research sees narrative structure as emergent
in the moment-by-moment unfolding of talk, rendering the notion of an abstract
narrative structure the forms and functions of which can be posed a priori of actual
narratives-in-interaction problematic (e.g. Scheglof 1997). Tis conversation-analytic
work places emphasis on how stories come about in the process of other conversational
activities and how they both orient to prior talk and shape upcoming talk. Endpoints
(i.e. entry into and exit from a story) are very important in this respect (Jeferson 1978).
Tere is still much scope for reseach here: how do storytelling participation roles (i.e.
contributions by various participants) shape narrative structure? How do the diferent
types of entry into a story relate to subsequent tellings? Finally, it is remarkable how
little explored the part(s) in-between a storys entry and exit are from a sequential point
of view.
3.2 Narrative evaluation
Research on narrative text properties has focused not only on the devices which
provide a storys structure, but also on signals of a storys point, of its worthiness of tell-
ing or tellability. As suggested, stories discursively construct and evaluate experience;
they encode the storytellers selection and interpretation of what happened, their feel-
ings, attitudes, and emotional interest in the tale, telling, and/or audience. Subjectivity
encodings are thus pervasive in narrative. Consequently, their study has been marked
by a diversity of approaches and a proliferation of terms, with various degrees of resem-
blance and overlaps but, nonetheless, little dialogue between them. Such terms include
involvement strategies (Tannen 1989), performance devices (Bauman 1986; Wolfson
1982), tellability, afect (Ochs & Schiefelin 1989), expressivity (Fleischman 1990), and,
last but not least, evaluation (internal vs. external, cf. showing vs. telling). All of them
have been confusingly used to capture tellers intentions, linguistic (textual) strategies,
and/or audience response/efects (Caf & Janney 1994). Teir study has, however, made
apparent that a universal defnition and postulation of subjectivity devices is a problem-
atic enterprise. Tose should be discovered empirically and ethnographically, i.e. linked
to specifc stories and their contexts. It is nonetheless notable that certain devices are
recurrent in cross-cultural studies of narrative subjectivity. Tese are as follows:
Repetition in its various forms
Narrative present
Characters direct speech/thought, and dialogues
Proximal deictics
Imagery
Details
Ellipsis
Paralinguistic devices (e.g. intonational variation)
All the above devices have been time and again put forth as displays of the tellers verbal
artistry aimed at involving the audience in the storytelling. In terms of categories of
emotive communication, they tend to be described as markers of iconicity, proximity,
or less frequently, intensity. Teir common denominator thus seems to be the empha-
sis on the experiential element of storytelling, which involves (emotional, perceived)
proximity between teller, audience, and events. Tis prominence of the properties of
immediacy and vividness is traceable to the concept of verisimilitude (vraisemblance)
as the essence of literature, which has preoccupied literary scholars since Aristotle.
Tese properties also typify the concept of narrative performance that has sprung out of
Narrative 197
ethnographic work on the verbal artistry of oral literature. According to Bauman (1986),
for a narrative to qualify as a more or less sustained performance, the teller needs to
assume responsibility to an audience for display of communicative skill and efciency.
Bauman also postulated that the constellation of devices that defne performances
are culturally variable and are to be discovered empirically. My own study of Greek
storytelling (1997) has attested to the cultural specifcity not just of the actual perfor-
mance devices but also of the participation modes and local functions of performances.
Te following story from my data exemplifes the use of performance (cf. evaluative,
involvement) devices:
Imaste tora me ton Gosta Focionos Neri
eci sto romo me tis kafeteries.
Now were on Fokionos Negri with Kostas
the street with lots of cafeterias right?
Lipon mesimeraci peripu
piname kafe ekso tora.. jati ekane zesti
aramena tafocinita mas ce ta lipa.
So it was some time in the afernoon
we were having our cofees outside cos it was hot
the cars parked in a row and so on..
Ce se kapa fasi skai miti apo to romo eci.. apo ti onia
ena fortio kocino
vameno olo kocino
And at some point there shows up out of the sidestreet
a lorry red
painted all red
Ie vali meafona pano
cie vali ena traui
afo pu lei pa ra tu ru ru ru rum (sings).
it had speakers on top
and it was playing this song
the one that goes pa ra ra ra tu ru ru r rum (sings).
Zu anelis pandos colos o iasos
sto cenurjo cedro pu a trauusane
ce perasane apo ci ja kapa iafmisi.
Anyway (it was) Zuganelis and the whole troupe
at the club where they were doing the show
and theyd passed for some free publicity.
Ce stamatane brosta safo to romo
brosta i kafeteria to Sale
nosto.. poli nosto.
And they stop in front of us at the street
in front of the cafeteria (street cofee-place) the Chalet
well-known.. very well-known.
Ce to me afono na lei afo to traui
para ra ra ra tu ru ru rum
co Zuanelis apo to mikrofono na lei papa::res ipokofa
A(udience): heh heh
And the (lorrys) loudspeaker (na) playing this song
para ra ra ra tu ru ru rum (humming the tune)
and Zuganelis (famous Greek satyrist) saying over the mike papa::res (word-play on
the previous tune: the word papares is slang for testicles) in a deep voice.
Ce na pefune i sokolates tora apo to forti o
ce na tis petai kato se mas sta trapezaca.
And now chocolates (were) being thrown of the truck
and he (was) throwing them at us at the tables.
Ce perni enas tis sokolates
ce tis petai piso
emis imaste eci pera.
And somebody takes the chocolates
and throws them back (at the lorry)
we (emphatic) were there.
Ce na jinete tora enas polemos me tis sokolates
ce na vlepume tis sokolates na pefune sto fortio
ce to Zuaneli na lei afo to papa::res
ce mu:: sa voi.
A: hehhhh hhhh
And (we) seeing a rain of chocolates falling on the lorry
and Zuganelis saying this papa::res thing
and mooing like a cow.
A: hehhhhh hhhh
Jeljo::! ti jeljo epese ecino to proi!
Laughter::! Tat was a really good laugh!
(Story included in Georgakopoulou 1997: 217218)
As can be seen, the storys complicating action is rendered in narrative present and the
climax in narrative na (a language-specifc device, roughly equivalent to imperfect).
Narrative na sets of the high point in performed stories thus marking heightened
tension and enduring vividness (Georgakopoulou 1997: 124f). In this case, its efect
is reinforced by the use of the proximal deictic now. In addition, the story owes its
dramatization to the animation of voices: thus, its humorous punchline is provided by
an instance of direct speech. Finally, the story is delivered in short segments, mostly
comprising three lines (see above visual display). Tese are interrelated by means of
repetition patterns. For instance, the climactic scene is delivered in two similarly built
segments: segment 1: chocolates falling of the truck, segment 2: chocolates being
Narrative 199
thrown back; the comedians humorous punchline is uttered in both segments. Te
storys segments are also brought together by rhythmic patterns based on expressive
phonology. Te teller puts on a visual and aural show by humming a tune, lengthening
the pronunciation of vowels for emphasis, exploiting the word-play between the tune
and the rude Greek word which provides the storys punchline, and above all, by
creating an intricate and consistent alliteration of /p/ combined with the assonance of
/e/, which highlights the climactic scene (bold in the original):
Ce to meafono na lei afo to traui, para ra ra ra tu ru ru rum, co Zuanelis
apo to mikrofono na lei papa::res ipokofa. Ce na pefune i sokolates tora apo to
fortio, ce na tis petai kato se mas sta trapezaca. Ce perni enas tis sokolates, ce tis
petai piso, emis imaste eci pera. Ce na jinete tora enas polemos me tis sokolates, ce na
vlepume tis sokolates na pefune sto fortio, ce to Zuaneli na lei afo to papa::res, ce
mu:: sa voi. Jeljo::! ti jeljo epese ecino to proi!
Tough devices are employed which have also been attested to in other communities,
the above storys display of subjectivity cannot be detached from its immediate context
of telling, i.e. who tells it to whom and why, what precedes and follows (not looked
at in the above example) and what larger (social, cultural) contexts may be relevant.
As has been demonstrated (Georgakopoulou 1997), conversational narrative perfor-
mances in Greece are inextricably bound to cultural concepts of sociability and inti-
macy (re)afrmation. Tey are also dependent on local defnitions of and expectations
about narrative functions and tellability. Te culture-specifcity of storytelling has
been amply documented by numerous studies, one of the most infuential of which
has been Chafes (1980) edited volume on the retellings of a silent flm, known as Pear
Stories. Such studies have cast serious doubt on the Labovian intrinsic defnitions of
narrative reportability as well as the association of tellability with unexpectedness.
Unexpectedness has thus been redefned as a contextual concept, relative to tellers,
audiences, situations, and cultures. More importantly, it has been demonstrated that
on occasion, it is the narration of the ordinary and/or the familiar which is called
for and tellable (cf. family fables, Blum-Kulka 1997). In sum, studies of subjectivity
displays in narrative have brought to the fore their individual, social, and cultural vari-
ability (e.g. Gee 1990; Heath 1983; Michaels 1981; Scollon & Scollon 1981). To draw on
Blum-Kulkas (1997) three interrelated but analytically separable dimensions of narra-
tivity, subjectivity is ultimately dependent as well as contingent on the dimensions
of teller-tale-telling.
A further revisiting of the Labovian evaluation involves its textual role. Evalua-
tion is currently recognized as a text-building mechanism of vital importance for a
storys global discourse structure. As developmental studies of narrative have shown,
evaluation points to the hierarchical perspective from which a narrative gains coher-
ence (Bamberg & Damrad-Frye 1991) thus giving meaning to the individual events
and actions. Specifcally, it was found that the ability to employ evaluation strategi-
cally in order to signal important events is lacking in younger children in favour of
locally confned uses of evaluation. Tis view of evaluation as a mechanism for a sto-
rys coherence is informed by multi-level discourse analyses, which postulate diferent
levels in narrative (e.g. plot level, background information, commentary, evaluation)
and shifs or moves between these levels as organizational strategies. Multi-level anal-
yses are committed to the premise of form-function anisomorphism (i.e. one linguis-
tic form more than one discourse function and vice versa). Tis means that instead
of seeking simple, one-to-one associations between textual devices and evaluation,
they bring to the fore how the same devices can be both organizational and evalu-
ative (expressive). Tis multifunctionality has been amply demonstrated for the use
of various devices, including narrative present (Fleischman 1990; Georgakopoulou
1997; Schifrin 1981).
An important spin-of of the above reconceptualization of evaluation involves its
relation with a narratives foreground. In Labovs model, the underlying assumption
was that evaluation belonged to a storys background, as it was not part of the sequen-
tial series of narrative clauses (see Toolan 1988: 159). More recently though, the defni-
tion of foreground has been disassociated from the criterion of sequentiality (Reinhart
1984). Tus, as a textual pointer to how the parts relate to one another and to a sto-
rys whole, evaluation gains a place in the repertoire of foregrounding devices. More
importantly, within a contextual defnition of foreground-background, evaluation is
seen as a scalar category which assigns various degrees of prominence and diferent
weights to various textual elements (Polanyi 1985: 22).
Finally, it is expected that studies of narrative as talk-in-interaction will increasingly
move away from aesthetic and verbal artistry criteria in the defnition of narrative
evaluation towards viewing it as a process of the participants joint construction, nego-
tiation and contestation of alternative interpretative viewpoints on the events related
as well as on the local relevance of the storytelling activity itself (cf. Georgakopoulou
2001, Ochs & Capps 2001).
4. Contextual properties
As discussed above, the textual properties of oral narrative have been mainly
studied through contextualized approaches. Tese have frequently employed narra-
tive as a point of entry into the presentation of self : this involves the construction
of personal and sociocultural identities, such as gender (Georgakopoulou 1995;
Johnstone 1990), age (Georgakopoulou 1997; Hausendof & Quasthof 1992), peer
Narrative 201
group membership (Georgakopoulou 2003b; Goodwin 1990; Shuman 1986), etc. In
particular, personal narratives or life histories lend themselves to such an enquiry, as
the prime discourse mode for identity constructions (Schifrin 1996). Furthermore,
identity displays through narrative are held to be very efective in recruiting
audience alliances: the taleworld, deictically distinct from the moment of the storys
telling, is more shielded from testing, justifcation, debate, and proof than, for
instance, an opinion or an argument (Schifrin 1990). Tis is intimately linked with
the prototypical functions of narration: we tell stories to move or convince through
moving; in turn, our audiences critical response and soliciting of proof for truth
claims are deterred in favour of an emotive engagement in the narration (for a dis-
cussion see Georgakopoulou & Goutsos 2004: Ch. 2). Te separation between the
taleworld and the world of telling also serves self-presentation in another way: it
allows the tellers to socialize aspects of their self, as defned by Gofman (1974), by
appearing both as narrators and characters in the taleworld. Te prime site for this
self-lamination is the realm of reported speech. Gofmans notion of footing (i.e.
stances or alignments which tellers take up with regard to themselves and their audi-
ences) is crucial in understanding this process. As Schifrin (1990) has explicated
it, storytellers can present themselves in the capacity of (a) animator (the aspect of
self which physically produces talk), (b) author (the aspect of self responsible for the
content of talk), (c) fgure, the main character in the story, someone who belongs to
the world that is spoken about and not the world in which the speaking occurs, and,
fnally, (d) principal, the self established by what is said, committed to what is said.
Alternatively, they can delegate any of the aspects of author, fgure, and/or principal
to other characters in the taleworld. Trough such manipulations of their kaleido-
scope of selves, storytellers can difuse their agency or responsibility in the social
feld, create a widened base of support for their views and beliefs, or, generally, cast
positive light on them (e.g. see Hill 1995).
Bamberg (1997b) proposes a more explicit scheme for looking into a tellers
construction of identities through narrative. Building on Davies & Harr (1990), he
locates a tellers positioning, that is, the discursive practices for the tellers construction
of self, at three levels:
Level 1: positioning of the characters vis--vis one another in the taleworld (e.g. as
protagonists or antagonists).
Level 2: positioning of oneself as the speaker with regard to an audience in the act
of narrating, involving claims with regard to the tellers identity, which are
locally and situationally achieved.
Level 3: an achievement of levels 1 and 2, this is the positioning of tellers vis--vis
themselves, involving claims that they hold to be true and relevant above and
beyond the local conversational situation.
Te interaction of these three levels indicates that self-presentation in narrative fully
exploits subjective, deeply imaginative and afectionate processes to create alignments
with the audience. Textually, positioning strategies mobilize a wide range of more or
less routinized, patterned and nuanced resources, among which the presentation of
characters speech. For instance, as we can see in the following anecdote (translated
from Greek), the tellers positioning is mainly constructed by means of his reported
dialogue with a male friend who assumes the roles of the author, fgure, and principal.
Tese three capacities orchestrate to lend validity to the tellers view that the dual goal
of love and fnancial security in ones marriage is feasible:
A: Look Nikos, you cant have it all, you must get your priorities right=
N: you cant say this. I have a friend, Ive known him for years, and his dream was,
well we were talking once, and he tells me, love is good, but the package is good
too, the Delors package. //he he. Well I tell him, you cant have your pie and eat it
too Kostas, you will lose out somewhere, either in money or love. A week later he
tells me, well you know something, only the package is good afer all, if you have
the package, you can have love too. Some time went by, he meets a girl,>without
knowing anything about her <, they just met, accidentally, but it turns out shes
loaded. Tey got married, we have another chat afer that, this is success he tells
me, love and money at the same time.
Te storys main purpose is opinion-expressing and argumentative, as it is related in
order to disprove an interlocutors view. Tus, the tellers positioning with regard to
levels 2 and 3 is chosen to be rendered in narrative rather than argumentative form. In
particular, the main device for putting forth positioning 2 and 3 is positioning at level
1. Tis type of positioning construction is found in numerous Greek stories related by
men, whereby a male friend of the teller is encoded as author, principal, and/or fgure
of the story: this is a device for validating the tellers views or generally promoting a
favourable self-presentation (i.e. self-foregrounding). By contrast, the same participa-
tion framework (i.e. reported speech of male friend or relative) is linked with diferent
patterns of self-presentation in womens stories (see Georgakopoulou 1995). In this
respect, Greek stories show how the textual devices of positioning are not only multi-
functional but also indexical of their context, that is, indirect pointers to social roles,
stances, attitudes, and identities (in this case, gender, cf. Ochs 1992).
In addition to informal conversational storytelling, contextual studies of narrative
have focused on storytelling in institutional contexts (Mumby 1993), such as clinical
encounters (Aronsson & Cedeborg 1994; Hunter 1991), schools, experimental settings
(Mishler 1995), police interviews, courtroom (witness testimonies), television shows,
family interactions (Blum-Kulka 1997), etc. Te emphasis in these lines of inquiry has
varied from documenting narratives as alternative, counter-hegemonic epistemologies
increasingly valued in domains that had been the stronghold of models of rational, legal,
Narrative 203
and/or scientifc reasoning (Abrams 1991)
3
to a more micro-analytic focus on narrative
as an orderly, social activity which situates, constitutes but also resists power roles, ofcial
identities, and institutional agendas and ideologies (e.g. Ochs & Taylor 1992).
In general, research on narrative and identities exemplifes the same tension as
other analyses of discourse between fne-grained analyses on one hand and macro-
analytic accounts on the other hand, both of which juggle in diferent ways and with
a diferent emphasis the hugely complex and intricate connections between local
instances and extra-situational processes. Another issue here concerns the monopoly
of the life-story or, in more micro-scale terms, of the personal experience story, within
identity analysis (see papers in Brockmeier & Carbaugh 2001). Tis type of narra-
tive tends to be elicited in research interview situations and, as such, presents specifc
interactional features (e.g. one-teller accounts, high in tellability, relatively detached
from its discourse surroundings, cf. Ochs & Capps 2001: 57f). Tere is increasing rec-
ognition (papers in Bamberg 1997c, Georgakopoulou 2002b; 2003) however that this
autobiographical in kind narrative is far from being the norm in ordinary conversa-
tional encounters. Its frequent use, however as the basis of numerous empirical stud-
ies in the social sciences in general and in linguistics in particular has had profound
implications about the direction of narrative analysis, creating notions of a narrative
canon and orthodoxy, i.e. what constitutes a story, a good story, a story worth analys-
ing, etc. that in turn dictate a specifc analytic vocabulary and an interpretive idiom.
It is hoped that future research will legitimate a wide range of a-typical stories by
documenting the contexts which encourage their occurrence and by seeking out the
analytical tools appropriate for them.
In conclusion, narrative research has been shaped by three even if overlapping
preoccupations with the referential, textual, and contextual properties of narra-
tive. Te quest of what is specifc or unique to narrative along with an overarching
assumption about its primacy as a mode of communication have been central to these
focal concerns. Te reifcation of narrative has nonetheless increasingly been shied
away from, as attention has shifed from narrative as text to narrative as social prac-
tice. It is hoped that this line of inquiry will lead the analysis to more fexible and
elastic models that will be ft for the inquiry into micro-lived instances of narrative
in diverse contexts. At the same time, it is imperative that the baby is not thrown out
with the bathwater and that the analytical apparatus of narrative is not compromised
in favour of macro-accounts that are not empirically grounded. Finally, the search of
. It is, however, becoming increasingly apparent that access to institutional domains through
narrative does by no means automatically lead to the overturning of an inequitable socio-political
order. Future work in the area will need to illuminate the role of narrative as a resource for ideolo-
gies in a variety of institutional contexts, taking into account the complexities and idiosyncrasies
of, among others, the participation frameworks and roles involved.
how the richness and diversity of existing approaches can be creatively fltered into
new transdisciplinaries and give rise to inclusive research strategies is a desideratum
of discourse analysis in general and narrative analysis in particular. To what extent
this enterprise will come to fruition remains to be seen. For scholars of narrative at
least, it rests on putting to good use the premise that every story is not just a creation
but also a joint creation.
References
Abrams, K. (1991). Hearing the call of stories. California Law Review 79: 9711052.
Aristotle (1995). Poetics. [Trans. & Ed. S. Halliwell] Harvard University Press.
Aronsson, K. & A. Cedeborg (1994). Conarration and voice in family therapy. Voicing, devoicing,
and orchestration. Text 14: 345370.
Bal, M. (1985). Narratology. Introduction to the Teory of Narrative. University of Toronto Press.
Bamberg, M. (ed.) (1997a). Narrative Development. Six Approaches. Erlbaum.
(1997b). Positioning between structure and performance. Oral Versions of Personal Experi-
ence: Tree Decades of Narrative Analysis. Journal of Narrative and Life History 7 (1/4) (Special
issue): 335342.
(1997c). (ed.) Oral Versions of Personal Experience: Tree Decades of Narrative Analysis.
Journal of Narrative and Life History 7(1/4) (Special issue).
Bamberg, M. & D. Damrad-Frye (1991). On the ability to provide evaluative comments: Further
explorations of childrens narrative competencies. Journal of Child Language 18: 689710.
Bartlett, F.C. (1932). Remembering. Cambridge University Press.
Barthes, R. (1977). Image Music Text. Hill & Wang.
Bauman, R. (1986). Story, Performance and Event. Contextual Studies of Oral Narrative. Cambridge
University Press.
Berman, R. & D. Slobin (1994). Relating Events in Narrative. A Crosslinguistic Developmental Study.
Erlbaum.
Blum-Kulka, S. (1997). Dinner Talk: Cultural Patterns of Sociability and Socialization in Family Dis-
course. Lawrence Erlbaum.
Brockmeier, J. & D. Carbaugh (eds.) (2001). Narrative and Identity Studies: Autobiography, Self and
Culture. Benjamins.
Bruner, J. (1986). Actual Minds, Possible Worlds. Harvard University Press.
(1990). Acts of Meaning. Harvard University Press.
(1991). Te narrative construction of reality. Critical Inquiry 18: 121.
Caf, C. & R. Janney (1994). Toward a pragmatics of emotive communication. Journal of Pragmatics
22: 32573.
Carr, D. (1986). Time, Narrative and History. Indiana University Press.
Chafe, W. (1987). Cognitive constraints on information fow. In R. Tomlin (ed.) Coherence and
Grounding in Discourse: 2151. John Benjamins.
Chafe, W. (ed.) (1980). Te Pear Stories. Cognitive, Cultural and Linguistic Aspects of Narrative
Production. Ablex.
Chatman, S. (1991). Coming to Terms: Te Rhetoric of Narrative in Fiction and Film. Cornell
University Press.
Narrative 205
Davies, B. & R. Harr (1990). Position: the discursive construction of selves. Journal of the Teory of
Social Behaviour 20: 4363.
van Dijk, T.A. (1988). Models in memory. Te role of situation representations in discourse processing.
Ms. University of Amsterdam.
(1993). Stories and racism. In D.K. Mumby (ed.): 121142.
Duranti, A. (1986). Te audience as co-author: an introduction. Text 6: 239247.
Fleischman, S. (1990). Tense and Narrativity: From Medieval Performance to Modern Fiction.
University of Texas Press.
Fludernik, M. (1996). Linguistics and literature: prospects and horizons in the study of prose. Journal
of Pragmatics 26: 583611.
Gee, J.P. (1990). Social Linguistics and Literacies. Ideology in Discourses. Falmer Press.
Genette, G. (1980). Narrative Discourse. Cornell University Press.
Georgakopoulou, A. (1995). Women, men and conversational narrative performances: aspects of
gender in Greek storytelling. Anthropological Linguistics 37: 460486.
(1997). Narrative Performances: A Study of Modern Greek Storytelling. Benjamins.
(2001). Arguing about the future: on indirect disagreements in conversations. Journal of Prag-
matics 33: 18811900.
(2002a). Greek children and familiar narratives in family contexts: en route to cultural perfor-
mances. In S. Blum-Kulka & C.E. Snow (eds.) Talking to Adults: 3354. Erlbaum.
(2002b). Narrative and identity management: discourse and social identities in a tale of
tomorrow. Research on Language and Social Interaction 35: 427451.
(2003a). Plotting the right place and the right time: place and time as interactional
resources in narratives. Narrative Inquiry 13: 413432.
(2003b). Looking back when looking ahead: adolescents identity management in narrative
practices. In J. Androutsopoulos & A. Georgakopoulou (eds.) Discourse Constructions of Youth
Identities: 7591. John Benjamins.
(2004). Same old story? On the interactional dynamics of shared narratives. In T. Baker &
U. Quasthof (eds.) Narrative Interaction: 223241. John Benjamins.
Georgakopoulou, A. & D. Goutsos (2000). Revisiting discourse boundaries: Te interaction of nar-
rative and non-narrative modes. Text 20: 6382.
(2004; 1st ed. 1997). Discourse Analysis. Edinburgh University Press.
Gofman, E. (1974). Frame Analysis. Harper & Row.
Goodwin, M.H. (1990). He Said-She Said: Talk as Social Organisation among Black Children. Indiana
University Press.
Hausendof, H. & U. Quasthof (1992). Childrens storytelling in adult-child interaction: Tree
dimensions in narrative development. Journal of Narrative and Life History 2: 293306.
Heath, S. (1983). Ways with Words: Language, Life and Work in Communities and Classrooms.
Herman, D. (2002). Story logic: problems and possibilities of narrative. University of Nebraska Press.
Holmes, J. (1997). Struggling beyond Labov and Waletzky. In M. Bamberg (ed.) Oral Versions of
Personal Experience: Tree Decades of Narrative Analysis. Journal of Narrative and Life History
7(1/4) (Special issue): 9196.
Hunter, K.M. (1991). Doctors Stories: Te Narrative Structure of Medical Knowledge. Princeton Uni-
versity Press.
Hill, J. (1995). Te voices of Don Gabriel: responsibility and self in a modern Mexicano narrative.
In D. Tedlock & B. Mannheim (eds.) Te Dialogic Emergence of Culture: 97147. University of
Illinois Press.
Hymes, D. (1981). In vain I tried to tell you: Essays in Native American Ethnopoetics. University of
Philadelphia Press.
Jeferson, G. (1978). Sequential aspects of storytelling in conversation. In J. Schenkein (ed.) Studies in
the Organization of Conversational Interaction: 219248. Academic Press.
Johnson-Laird, P. (1983). Mental Models: Towards a Cognitive Science of Language, Inference and
Consciousness. Harvard University Press.
Johnstone, B. (1990). Stories, Community and Place: Narratives from Middle America. Indiana
University Press.
Josselson, R. & A. Lieblich (eds.) (1993). Te Narrative Study of Lives, vol. 1. Sage.
Kerby, A. (1991). Narrative and the Self. Indiana University Press.
Labov, W. (1972). Language in the Inner City. Blackwell.
(1997). Some further steps in narrative analysis. In M. Bamberg (ed.) Oral Versions of Per-
sonal Experience: Tree ecades of narrative analysis. Journal of Narrative and Life History 7(1/4)
(Special issue): 395415.
Labov, W. & J. Waletzky (1967). Narrative analysis: oral versions of personal experience. In J. Helm
(ed.) Essays on the Verbal and Visual Arts: Proceedings of the 1966. Annual Spring Meeting of the
American Ethnological Society: 1244. University of Washington Press.
Lakof, R. (1997). Te O.J. Simpson case as an exercise in narrative analysis. Discourse Processes 23:
547566.
Leech, G. & M. Short (1981). Style in Fiction. Longman.
Linde, C. (1993). Life Stories. Te Creation of Coherence. Oxford University Press.
Longacre, R. (1989). Two hypotheses regarding text generation and analysis. Discourse Processes 12:
413460.
Maranho, T. (1993). Recollections of feldwork conversations, or authorial difculties in anthro-
pological writing. In J. Hill & J. Irvine (eds.) Responsibility and Evidence in Oral Discourse:
260288. Cambridge University Press.
Martin, J.R. (1992). English Text: System and Structure. John Benjamins.
Macintyre, A. (1981). Afer Virtue. University of Notre Dame Press.
Michaels, S. (1981). Sharing time: childrens narrative styles and diferential access to literacy.
Language in Society 10: 423442.
Mishler, E. (1995). Models of narrative analysis: a typology. Journal of Narrative and Life History 5:
87123.
Moerman, M. (1988). Talking Culture. Ethnographic and Conversation Analysis. University of
Pennsylvania Press.
Mumby, D.K. (ed.) (1993). Narrative and Social Control: Critical Perspectives. Sage.
Ochs, E. (1992). Indexing gender. In A. Duranti & C. Goodwin (eds.) Rethinking Context: 335358.
(1997). Narrative. In T. van Dijk (ed.) Discourse as Structure and Process. Discourse Studies,
vol. 1: 185207. Sage.
Ochs, E. & L. Capps (2001). Living Narrative. Creating Lives in Everyday Storytelling. Harvard
University Press.
Ochs, E. & B. Schiefelin (1989). Language has a heart. Text 9: 725.
Ochs, E. & C. Taylor (1992). Family narrative as political activity. Discourse and Society 3: 301340.
Polanyi, L. (1982). Literary aspects of oral storytelling. In D. Tannen (ed.) Written and Spoken
Language. Exploring Orality and Literacy: 155170. Ablex.
(1989). Telling the American Story: A Structural and Cultural Analysis of Conversational
Storytelling. Ablex.
Narrative 207
Prince, G. (1982). Narratology. Te Form and Functioning of Narrative. Mouton.
Propp, V. (1986). Te Morphology of the Folktale. University of Texas Press.
Reinhart, T. (1984). Principles of gestalt perception in the temporal organization of narrative texts.
Linguistics 22: 779809.
Ricur, P. (1988). Time and Narrative. University of Chicago Press.
Rimmon-Kenan, S. (1983). Narrative Fiction: Contemporary Poetics. Methuen.
Rummelhart, D. (1975). Notes on a schema for stories. In D. Bobrow & A. Collins (eds.) Representa-
tion and Understanding: Studies in Cognitive Science: 211236. Academic Press.
Scheglof, E.(1997). Narrative analysis: thirty years later. In M. Bamberg (1997). (ed.) Oral versions
of personal experience: three decades of narrative analysis. Journal of Narrative and Life History
7(1/4) (Special Issue): 97106.
Schifrin, D. (1981). Tense variation in narrative. Language 57: 4562.
(1990). Te management of a co-operative self during argument: the role of opinions and
stories. In A.D. Grimshaw (ed.) Confict Talk: 241259. Cambridge University Press.
(1996). Narrative as self-portrait: sociolinguistic constructions of identity. Language in Society
25: 167203.
Sherzer, J. & A. Woodbury (eds.) (1987). Native American Discourse: Poetics and Rhetoric. Cambridge
University Press.
Shuman, A. (1986). Storytelling Rights. Te Uses of Oral and Written Texts by Urban Adolescents.
Stein, N. (1982). Te defnition of a story. Journal of Pragmatics 6: 487507.
Swales, J. (1990). Genre Analysis. English in Academic and Research Settings. Cambridge University
Press.
Tannen, D. (1989). Talking Voices: Repetition, Dialogue and Imagery in Conversational Discourse.
Toolan, M. (1988). Narrative: A Critical Linguistic Introduction. Routledge.
Trabasso, T. & L. Sperry (1985). Causal relatedness and the importance of story events. Journal of
Memory and Language 24: 595611.
Virtanen, T. (1992). Issues of text typology: Narrative a basic type of text. Text 12: 293310.
Vygotsky, L. (1978). Mind in Society. Harvard University Press.
Wolfson, N. (1982). CHP: Te Conversational Historical Present in American English Narrative. Foris.
Young, K.G. (1987). Taleworlds and Storyrealms. Te Phenomenology of Narrative. Nijhof.
Polyphony
Eddy Roulet
University of Geneva
1. Preliminaries
Te term polyphony is used in the feld of discourse analysis and pragmatics in order
to describe one important dimension of the organization of discourse, and also of
utterances, i.e. the fact that they can express and combine diferent voices.
Te notion of polyphony has a strange history. Early on, in the thirties, polyphony
was identifed as a major dimension of discourse by the Russian linguist Mikhail
Bakhtin, in the frame of his dialogical approach to discourse, but compared to the
large number of studies on speech acts or implicatures, the body of literature on
polyphony is considerably smaller. Tis is probably due to a conjunction of diferent
facts: frst, for various political and personal reasons, Bakhtins texts were discovered
and translated outside Russia only at the beginning of the seventies (see Todorov
1981); second, Bakhtins thought is ofen complex, elliptic and very difcult to trans-
late; third, the paradigms which have dominated anglo-saxon pragmatics during the
past thirty years (speech act theory, conversation analysis, relevance theory) ignored
Bakhtins sociological approach.
Given this state of afairs, it comes as no surprise that, although the importance of
the polyphonic dimension of discourse is not contested, few linguists have worked on
it. Te only one to have undertaken recently a systematic elaboration of the notion of
polyphony is the French linguist Oswald Ducrot, in the frame of his enunciation the-
ory, but he is more interested in the utterance than in the discourse level.
In what follows, we will frst present briefy Bakhtins and Ducrots notions of
polyphony and evaluate their contribution to our knowledge of this dimension of dis-
course, and then, describe the diferent aspects of the polyphonic organization which
have to be taken into account in discourse analysis. Finally, we will present an analysis
of the polyphonic organization of a short fragment of a letter.
2. Polyphony in Bakhtins work
Te central notion in Bakhtins conception of discourse is dialogism (cf. Discourse in
the novel, henceforth DIN: 279). He thus takes a clear position against what he calls
the one-sided monologism of linguistic science (cf. Marxism and the philosophy of
Polyphony 209
language, henceforth MPL: 99). But Bakhtins dialogism, if we refer to DIN, has to
be understood in a very broad sense of the word, which covers at least four diferent
notions:
a. it frst refers to the constant interplay between socio-ideological languages in
discourse and society, what Bakhtin calls dialogized heteroglossia (273);
b. it also refers to the fact that any discourse is related by a thousand dialogical
threads to all existing discourses of the same type on the same subject (276); this
is what we today call intertextuality (cf. Kristeva 1969);
c. it refers to the fact that any discourse, even a monological written text, is reacting
to preceding discourses and anticipating further discourses (280); this is dialo-
gism in the narrow sense of the term, corresponding to the sequentiality of actual
or potential exchange structures, as they are described in dialogue analysis (cf.
Roulet et al. 1985);
d. fnally, it may refer to the fact that the discourse of other persons occupies
an important place in the discourse, or even in the utterances, of any speaker
(339); this is what we today call polyphony (cf. Ducrot 1984; but Bakhtins use of
polyphony is sometimes broader, covering not only the diferent voices, but also
the diferent language varieties, styles and ideological representations displayed
in a novel).
Bakhtin then considers polyphony as a case of dialogism, in the broad sense of this
term, i.e. an instance of multi-voiced discourse. But in the narrow, compositional use
of the term, the discussion in MPL clearly distinguishes dialogism from polyphony.
Referring to what we have defned as polyphony under (d), Bakhtin specifes:
Tis phenomenon is distinctly and fundamentally diferent from dialogue. In
dialogue, the lines of the individual participants are grammaticaly disconnected: they
are not integrated into one unifed context. (MPL: 116)
In other words, one has to distinguish clearly between two levels of multi-voiced
discourse: dialogues, produced by two or more speakers, and monologues, produced
by one speaker, and the notion of polyphony, as defned above, has to be restricted to
the latter.
Bakhtin has approached the notion of polyphony, without systematically using the
term, mainly in three texts written between 1929 and 1935, and through two diferent
entries. Te frst entry, which is certainly the most familiar to linguists, is the descrip-
tion of reported speech, a well-known case of double-voiced structure, in the third
part of Marksizm i flosofja jazyka (MPL, frst published in 1929 under the name of his
colleague Voloinov). Te second entry, which has more to do with literary theory (see
Fowler 1981, chapter 8), is the study of discourse in the novel, in Problemy tvorchestva
Dostoevskogo (Problems of Dostoevskijs creative art, frst published in 1929) and in
210 Eddy Roulet
Slovo v romane (DIN, written in 19341935 and frst published in 1975 in Voprosy
literarury i stetiki (Te dialogic imagination)).
As far as reported speech (or, as we should rather say today in terms of discourse
analysis, reported discourse or discourse representation) is concerned, Bakhtin shows
that it is not only a grammatical or a stylistic problem, as it is usually dealt with (see,
for example, Banfeld 1982), but a discourse phenomenon, which he defnes in the
following way:
Reported discourse is discourse within discourse, utterance within utterance, and
at the same time also, discourse about discourse, utterance about utterance.[]
Te true object of inquiry ought to be precisely the dynamic interrelationship
of these two factors, the discourse being reported (the other persons discourse)
and the discourse doing the reporting (the authors discourse) (MPL: 115, 119;
following the French translation, I have deliberately replaced speech in the English
1973 translation by discourse).
Bakhtin thus asserts the necessity of going beyond the syntactic problem of the com-
bination of two sentences in order to describe the interrelation between diferent dis-
courses (or utterances), particularly the diferent forms of permeation between the
discourse of the speaker and the discourses of other persons, and all the changes in
meaning and accent that take place [] during transmission (DIN: 340).
Te third part of MPL is devoted to a revised analysis in a discursive perspec-
tive of direct, indirect and quasi direct discourses, but for Bakhtin polyphony, as
defned above, is not restricted to these diferent types of reported discourse, as we
will see below.
Bakhtin fnds numerous examples of polyphonic utterances in the novels of
Russian, German and French authors of the nineteenth century. In his book on Dos-
toevskij, he describes the phenomenon as follows:
An author may utilize the speech of another in pursuit of his own aims and in such
a way as to impose a new intention on the utterance, which nevertheless retains his
own proper referential intention. Under these circumstances and in keeping with
the authors purpose, such an utterance must be recognized as originating from
another addresser. Tus, within a single utterance, there may occur two intentions,
two voices (180).
But Bakhtin goes beyond this general characterization of the notion of polyphony and
he tries do defne it more specifcally from a linguistic and discursive point of view as
what he calls a hybrid construction:
What we are calling a hybrid construction is an utterance that belongs, by its
grammatical (syntactic) and compositional markers, to a single speaker, but that
actually contains mixed within it two utterances, two speech manners, two styles, two
languages, two semantic and axiological belief systems. (DIN, 304)
Polyphony 211
See the following example, from Dickens novel Little Dorrit:
(1) But Mr. Tite Barnacle was a buttoned-up man, and consequently a weighty one.
and the commentary by Bakhtin (DIN: 305):
If judged by the formal markers above, the logic motivating the sentence seems to
belong to the author, i.e. he is formally at one with it; but in actual fact, the motivation
lies within the subjective belief system of his characters, or of general opinion.
Two chapters of DIN are dedicated to the study in various novels of diferent types of
those hybrid constructions, which mainly correspond to quasi-direct discourse.
Bakhtin brings thus a major contribution to the study of polyphony in the novel,
but he also makes insightful remarks on polyphony in everyday conversation; this type
of polyphony difers from the former in the sense that a major role is here attributed
to the voice of the addressee, a person who, unlike others, is a direct partner in the
interaction:
In any actual dialogue the rejoinder also leads such a double life: it is structured and
conceptualized in the context of the dialogue as a whole, which consists of its own
utterances (own from the point of view of the speaker) and of alien utterances (those
of the partner). One cannot excise the rejoinder from this combined context made
up of ones words and the words of another without losing its sense and tone. It is an
organic part of an heteroglot unit. (DIN: 284)
Tis observation brings him to analyze a specifc hybrid construction, which he char-
acterizes as follows:
Te intentional double-voiced and internally dialogized hybrid possesses a syntactic
structure utterly specifc to it: in it, within the boundaries of a single utterance, two
potential utterances are fused, two responses are, as it were, harnessed in a potential
dialogue. It is true that these potential responses can never be fused into fnished
utterances, but their insufciently developed forms are nevertheless acutely felt in the
syntactic construction of the double-voiced hybrid. (DIN: 361)
He thus describes a form of utterance which is current in verbal interaction and which,
although clearly monologic, embodies a kind of crystallized dialogue (to use Ducrots
expression (1984: 8)), as in the following examples:
(2) You say that you have fnished your work, but I dont believe you.
(3) You have fnished your work, I dont believe you.
Tis double-voiced and internally dialogized hybrid construction is a good example
of what Bakhtin calls the internal dialogization of discourse, as contrasted with the
external, compositionnaly marked, dialogue (DIN: 283).
212 Eddy Roulet
As can be seen from this brief presentation, Bakhtins seminal work leads to new
approaches to the study of multi-voiced discourse and sketches insightful analyses,
mainly devoted to literary examples, but it does not propose any systematic description
of external or internal dialogic structures. In fact, Bakhtin, like Saussure, is foremost
a visionary, i.e. a scholar who is able to radically modify our conception of language,
to raise new problems and to create new fundamental concepts. Among the latter are
dialogism and polyphony, which occupy a major place in present research on verbal
interaction and discourse. Bakhtin is more interested in speculative refexions on dis-
course in the novel than in the empirical description of everyday, face-to-face interac-
tion. But his numerous and rich insights in the notions of dialogism and polyphony
have deeply infuenced research on discourse, mainly in the felds of dialogue analysis
(see Roulet et al. 1985), critical discourse analysis (see Fairclough 1992) and the theory
of literary discourse (see Fowler 1981).
3. Polyphony in Ducrots work
Ducrot, following Bakhtin, rejects the postulate of the unicity of the speaking subject
which is dominating the language sciences, but he afrms that Bakhtin has worked
only on the polyphony of discourse and that one has to go beyond his approach in
order to describe the polyphony of utterances (Ducrot 1984: 171). In fact, as we have
seen above, Bakhtin did work on the polyphony of utterances, but he does not go
as far as Ducrot toward a systematic description. We will here succinctly present
Ducrots polyphonic theory of enunciation (Ducrot 1984). It is important to specify
that, beyond their common interest in polyphony, the two linguists work within two
very diferent frames, the sociology of verbal interaction for Bakhtin and what he calls
ideal discourse structuralism for Ducrot (229).
Ducrots frst objective is to analyze, in the framework of an instructional concep-
tion of meaning, how an utterance indicates the superposition of diferent voices. He
shows that it is necessary to distinguish the following entities:
a. the speaking subject, i.e. an empirical being who is the physical agent of the
utterance;
b. the speaker, i.e. a discourse being, who can say I and who is the author of the
illocutionary acts; Ducrot makes an additional distinction between the speaker
as such and the speaker as referent (or human being) (199); see the diference
between the frst and the second I in the following utterance:
(4) I do assure you that I didnt make any error;
c. the enunciator, i.e. a discourse being corresponding to a certain point of view
which has not necessarily been expressed by a speaker (193).
Polyphony 213
According to Ducrot, the meaning of an utterance can enact two distinct speakers as
such, which may be identifed by traces like I and me and which may be subordinated
to each other. In the following example:
(5) Paul told me: I am sick.
we can identify one speaking subject and two speakers, at two diferent levels, with the
speaker of the whole utterance as main and actual speaker.
Te meaning of an utterance can also enact enunciators, expressing diferent
points of views which are not attributed to speakers. In the following examples:
(6) It is sunny this morning, but it is very windy.
(7) (uttered in a situation of evident failure) What a triumph!
there is one speaking subject and one speaker, who echoes voices of one or two
enunciators (i.e. points of view which are not attributed to any speaker). Te speaker
can manifest his own position by identifying himself or not to one of the enunciators
(205). In (6) the speaker evokes two points of view characterized by but as argumen-
tatively opposite, and he tends to identify himself to the second one, if we take into
consideration another instruction given by but. In example (7), the speaker echoes a
point of view which he manifestly does not assume. In other words, the speaker can
assume responsibility for his utterance, without necessarily assuming responsibility
for the attitudes it expresses (208).
Ducrot applies his conception of polyphony to various pragmatic problems,
such as irony, negation and negative polarity, some uses of discourse markers, and
presupposition.
a. irony
According to Ducrot, to speak ironically is, for a speaker S, to present an utter-
ance as expressing the position of an enunciator E, a position that the speaker can-
not assume and which he considers as absurd (211). See example (7) above. Te
enunciator E may be identifed with the addressee, in which case we have a kind of
aggressive irony (211).
b. negation and negative polarity
According to Ducrot, most negative utterances express a clash between two oppo-
site attitudes, one, positive, attributed to an enunciator E1, the other, negative,
attributed to an enunciator E2 to which the speaker identifies himself (215).
Ducrot bases his hypothesis on the observation of the following facts. First, the
use of the French connective au contraire (on the contrary). If one looks at the
following example.
214 Eddy Roulet
(8) Paul is not hard working; on the contrary, he is lazy.
the only opposition is not between not hard working and lazy, but between hard
working and lazy, which, for Ducrot, is a clear indication of the presence of the
positive attitude in the negative utterance (216). Second, cases of negative polarity, as
in the use of French connective pour autant, which is diferent from the use of its near
synonym pourtant (yet), as can be seen, in the following examples:
(9) Paul est grand; pourtant, il nest pas fort (Paul is big; yet, he is not strong).
(10) Paul est grand, pourtant, il est faible (Paul is big; yet, he is weak).
(11) Paul est grand; il nest pas fort pour autant.
(12) *Paul est grand; il est faible pour autant.
For Ducrot, expressions with negative polarity, like pour autant, are characterized by
the fact that they imply the rejection of a positive attitude, and thus require the use of
an explicit negation (219220).
c. diferent uses of discourse markers
Ducrot explains the diferent uses of discourse markers like mais (but) by showing
that they relate utterances which are attributed to distinct enunciators, which may be
assimilated in diferent ways to the speaker, the addressee, or to other points of views
(229230).
Tus, he hypothesizes that utterances constructed with mais enact two successive
enunciators E1 and E2, who argue in opposite directions, and that the diferent uses of
those constructions, as illustrated in the following examples, depend on the assimila-
tion of those enunciators with diferent persons:
(13) [Paul wonders if he should go out for a walk and says:]
It is sunny, but very windy.
In this case, the speaker identifes himself successively with E1 and E2.
(14) Mary: Lets go out for a walk. It is sunny.
Paul: It is sunny, but very windy.
In this case, the speaker assimilates E1 to the addressee and identifes himself
with E2.
(15) Mary: Lets go out for a walk.
Paul: It is sunny but very windy.
Polyphony 215
Te answer may have two diferent interpretations. Either the speaker assimilates E1
to the addressee and thus attributes to him a point of view he did not express, or, as in
example (13), he identifes himself successively to E1 and E2.
d. presupposition
Te notion of polyphony also allows Ducrot to explain in a simpler way instances of
presupposition which he had already described in earlier work. He would explain a
classical example like.
(16) Pierre has ceased smoking.
by saying that this utterance enacts two enunciators E1 and E2 who are respectively
responsible for what is presupposed (Pierre used to smoke) and what is asserted (He
does not smoke), and that the speaker identifes himself with E2 whereas E1 would be
the expression of a collective voice (231).
Ducrots proposals have made the concept of polyphony more precise and enlarged
the range of linguistic and discursive problems to which it may be applied, but the
French linguist is working more on enunciation than on dialogue and, like Bakhtin,
he is not primarily interested in the empirical description of everyday discourse. So
his contribution has to be reexamined in the light of present researches in dialogue
analysis and critical discourse analysis.
4. The description of the polyphonic organization of discourse
It is premature to give a systematic description of the polyphonic organization of
discourse, which is very complex and which interferes with many other more or less
well-known dimensions, like the social, interactional, topical, hierarchical and rela-
tional dimensions, as we will see later. But it is possible, if we refer to Bakhtins and
Ducrots work and if we take into account the contribution of recent research in dia-
logue analysis and critical discourse analysis, to sketch the more important features of
this organization (in what follows, I will not be dealing with the kind of multi-voiced
discourse which is characterized by a direct exchange structure between two speak-
ers; it belongs to dialogism in the narrow sense of the term, not to polyphony; see
Roulet et al. 1985).
As it is difcult to render words like nonciateur et nonciation in English, I
will restrict the technical vocabulary to the minimum in this presentation. In other
words, I will keep to the notions of speaker (and addressee), and voice (a voice may
transmit the discourse of any speaker or express a point of view which is not related
to a specifc speaker).
216 Eddy Roulet
Any description of the polyphonic organization of discourse should take into
account at least the ten following points:
a. the discourse, or even a single utterance, of a speaker can enact diferent voices,
corresponding to diferent discourses or points of view: the actual discourse of
the speaker, past or future discourses of the speaker, of the addressee, or of other
people, and points of view not related to specifc speakers;
b. one has to distinguish the voices which are verbally manifested in the discourse
of the speaker from those which are implicit. See the diference, in the following
example, between two possible reactions to a proposal:
(17) Lets go out for a walk. It is sunny.
a. It is sunny, but very windy.
b. But it is very windy.
In (a) before expressing his own voice, the speaker echoes the voice of the
addressee, whereas in (b) he just refers implicitly to it by the use of but (according
to the instructions given by this discourse marker); discourse markers, when they
open a move, play a very important, and usually neglected, role in the implicit
reference to others discourse;
c. among the voices which are manifested verbally in the speakers discourse, one
has to distinguish between the one which is presently produced (the present voice
of the speaker), and those which are only an echo of other discourses or points of
view (including past discourses and points of view of the speaker). If we consider
the answer (a) in example (17) above, we can analyze the discourse of the speaker
in two segments bearing two voices: the frst segment is an echo of the addressees
voice, the second produces the present voice of the speaker;
d. one has to distinguish two diferent ways to echo other discourses or points of
view: the speaker can reproduce or express them, as in the preceding examples, or
he can just point at them, as in the following examples:
(18) Your protest has been duly registered.
e. it is generally possible to analyze the discourse of the speaker in segments
corresponding to distinct verbally manifested voices at diferent levels of poly-
phonic embedding; for instance, the following utterance can be analyzed this way
(each segment is delimited by square brackets and the voice is indicated in front
of it in bold types: S = speaker, A = addressee, etc.):
(19) S[Paul has written to me that P[Maria had told him that M[she was ill]]]
(20) S[You say that A[it is a bad paper], I think it is excellent]
As noted above, we have to make a distinction between the present voice (S) and
past voices (S) of the speaker, as in the following example:
Polyphony 217
(21) S[I have told you yesterday that S[I was ill]]
As Bakhtin has shown in his work on quasi direct discourse, it is not always pos-
sible to delimit precisely the borders betweeen segments belonging to diferent
voices. Consequently, we must admit the possibility of diferent segmentations,
corresponding to diferent interpretations, for the same discourse, or even of a
complete superposition of the two voices. Tus, the following utterance.
(22) Max has told me that he unfortunately did not pass his exam.
may be segmented in two diferent ways:
S[Max has told me that M[he unfortunately did not pass his exam]
S[Max has told me that M[he] unfortunately M[did not pass his exam]
In the frst case, the adverb expresses Maxs point of view, in the second, it is a
commentary by the speaker.
f. as far as the source of the voice is concerned, it is not enough to distinguish the
speakers present voice from other voices. Bakhtin shows clearly that it is also
important to make a distinction between the report of the voice of the present
addressee and the report of any other voice, as they do not play the same role
in discourse. For the speaker, to reproduce the discourse of his partner, who
may then react to it, is a part of the actual interaction between them; that is not
the case for the report of the discourse of persons who do not participate in the
interaction, which is just an object of reference among others. Tat is the reason
why we have introduced in Roulet et al. (1985: 6984) a distinction, that has
been widely used since, between diaphony, which refers to the (re)production
of the voice of the addressee, and polyphony (in a restricted sense), which refers
to the (re)production of any other voices. In the case of diaphony, it is useful
to make an additional distinction between immediate and difered reproduc-
tions of the addressees discourse, which may have diferent functions, as we will
see later.
g. In addition to the distinction between verbalized and implicit voices mentioned
above, one has to make a distinction between actual and potential voices, if they
are marked in the discourse. See the diference between
(23) You said that I am negligent, but I cant accept this accusation.
(24) You are probably going to say that I am negligent, but I cant accept
this accusation.
In the frst case, where the speaker refers to an actual discourse of the addressee,
we speak of efective diaphony; in the second case, where the speaker refers to a
potential discourse from his addressee, we speak of potential diaphony (one can
make an analogous distinction between efective and potential polyphony).
218 Eddy Roulet
h. the most important point in the analysis of polyphony, as Bakhtin already noticed,
is the the way in which the others voices have been processed. So we have to
describe
how the discourse or the point of view expressed by the voice has been selected,
i.e. what parts of the others discourse or point of view, including the context, are
reproduced or presented (it may be just an isolated word or a whole narrative);
how it is (re)formulated, at the syntactic, lexical and phonetic levels;
how it is qualifed, by the use of diferent forms: intonation, speech verbs, modals
or metadiscursive commentaries, etc.;
how it is integrated in the speakers discourse, as a lexical, grammatical or dis-
course constituent, and how it is linked to the other linguistic or discursive con-
stituents of the speakers discourse by the use of complementizers, discourse
markers, intonation, punctuation, etc. As far as the syntactic integration of direct,
indirect and quasi direct discourses is concerned, see Banfeld (1982).
i. to describe the degree of the speakers distance or adhesion to the others voices
in his discourse, which may vary from complete distance, in the case of irony (see
example (7) above) to complete adhesion, as in the following example
(25) As Paul recommended, we should accept this project.
j. fnally, one has to describe the various possible functions of dia/polyphony in
diferent types of interactions: thus, the diaphonical reproduction of part of the
addressees discourse may be used by the speaker to mark the topic of his answer
and to regain control of the dialogue, to inverse the argumentative orientation
of the discourse of his addressee, to preface and attenuate a dispreferred answer,
to restore the difered link between a question and an answer in an epistolary
exchange or a parliamentary debate, etc. See the very interesting description by
Fairclough (1988), in the framework of critical discourse analysis, of the represen-
tation of discourse in the news media, where he shows that it is linked with social
determinants and social efects.
5. The interrelations between polyphony and other dimensions
of discourse structures
Tis brief analytical presentation we have given of the main features of polyphony,
which has been illustrated by fabricated examples, does not show the interrelations
between the diferent aspects of polyphony, and between polyphony and other dimen-
sions of discourse structures. Terefore I will complete it by presenting an analysis of a
fragment of an authentic discourse which is very rich from a polyphonic point of view,
Polyphony 219
a letter written by George Sand [GS], when she was seventeen, to her mother. (See for
a more detailed presentation and the analysis of other examples Roulet 1999, chap. 9
and 10, and Roulet, Filliettaz & Grobet 2001, chap.10).
Let us briefy comment on the context of the letter, as it is important to understand
the functions of polyphony. GSs mother, who is living in Paris, has sent her daughter,
who is living in the countryside with her grandmother, a letter where she expresses
several reproaches; these reproaches concern GSs behavior, which, according to what
some people have told her mother, seems to difer from the behavior expected from a
girl at that time. In her answer, George tries to reach at least three partially divergent
goals, without threatening her mothers position and face too directly: frst, to refute
the reproaches, in order to reassure her mother; second, to be recognized as a mature
person with an independent judgment; third, to assert the right for a young woman to
adopt a diferent, more liberated way of life.
Te following fragment, which addresses one of her mothers reproaches, is quite
representative of the way GS constructs her answer:
Cest une sottise, dites-vous, ma chre mre, que dapprendre le latin. Je ne sais qui a
pu vous dire que je me livrasse cette tude, en tout cas on vous a trompe, car je ne le
sais, ni ne lapprends, mais quand je le ferais jprouve une extrme surprise que vous,
ma mre, puissiez trouver mauvais que je minstruisisse.
(Correspondance de George Sand, Paris, Garnier, 1964, 7481)
Here is a translation of this fragment:
(26) It is foolishness, you say, my dear mother, to learn Latin. I dont know who may
have told you that I undertook such a study. Anyway you have been misled,
because I neither know nor learn Latin. But, even if I did learn Latin, I am
extremely surprised that you, my mother, might fnd it bad that I would
improve my instruction.
Te polyphonic structure of this fragment, a perfect example of this very elaborated
structure Bakhtin called an internally dialogized hybrid, can be described as follows:
S[you say, my dear mother M[it is foolishness to learn Latin]. I dont know who
may have told you that X[I undertook such a study]. Anyway you have been misled,
because I neither know nor learn Latin. But, even if M[I did learn Latin], I am
extremely surprised that you, my mother, might fnd M[it bad that I would improve
my instruction]].
Tis polyphonic structure must be related to the hierarchical and relational struc-
tures (see our brief presentation of the Geneva model in Roulet 1995). Tose can be
described as follows:
(where M = move, A = act, m = main, s = subordinate, as far as hierarchical
constituents are concerned, and, as for the relations between those constituents: pre =
preparation, arg = argument, c-arg = counter-argument, pot = potential, and reform
= reformulation)
220 Eddy Roulet
sA It is foolishness you say my dear mother to learn Latin
pre
M
mM
mM
sM
c-arg
mM
But
sA I dont know who may have told you that I undertook such a study
pre
mA anyway you have been mislead
sA because I neither know nor learn Latin
arg
sA even if I did learn Latin
pot c-arg
I am extremely surprised that you my mother might
fnd it bad that I would improve my instruction
mA
It must also be related to a feature of the interactional structure which is typi-
cal of epistolary communication, i.e. a situation of difered interaction which obliges
the speaker and the addressee to connect series of initiative and reactive moves in
their respective letters (see Roulet & al. 1985, 7482, and Mulkay 1985). Finally, it
must be related to power relations, as they are described in critical discourse analysis
(see Fairclough 1992), here between a mother and a seventeen-year-old daughter, in a
confictual educational and sociological context.
Taking into consideration those diferent dimensions of the organization of
discourse, we are now able to specify the role played by polyphony in that organiza-
tion. Te speaker starts with a subordinate act of preparation which reproduces one of
the reproaches of her mothers letter; the production of this diaphonical segment has
a double function: frst, to restore the direct link between an initiative and a reactive
move which is broken in the difered interaction of an epistolary exchange; second, to
introduce the discourse object that will be the topic of her move. GS then constructs
a main move in order to reply to the reproach in three steps. First, as her mothers
reproach is evidently based on the saying of others, GS produces a polyphonic con-
struction in order to question the validity of this unidentifed source. Tis construction
also functions as the preface which, according to work in conversation analysis, must
precede any dispreferred answer, in order to save the positive face of the interlocutor
(see Mulkay 1985, for an interesting analysis of this feature in a collection of letters
written by biochemists). Second, as the reproach rests on a mistaken fact, GS frmly
denies this fact in a main act followed by a subordinate act of argument. Tird, as the
mere denying of the fact might lead her mother to think that GS accepts the presup-
position on which the reproach is based (learning Latin is bad), in a diaphonic act GS
repeats the reference to the fact which is the object of the reproach, but as a potential
counter-argument, to condemn, in the main polyphonic act that follows, the conclu-
sion which her mother drew. Tus, GS also rejects the lower and dependent position
Polyphony 221
which is constantly attributed to her by her mothers judgments and reproaches and
she clearly takes a higher position by asserting her point of view and by criticizing
her mothers conception of education. It is interesting to observe how GS manages
to do this without threatening too much her mothers positive and negative faces, by
strongly modalizing her utterance: jprouve une extrme surprise que vous puissiez
6. Conclusion
Te conception of polyphony we have presented above refers to a speaker who
consciously organizes the expression of diferent voices in his discourse. Although
Bakhtin and Ducrot reject the unicity of the speaking subject, they do not seem to
question the control of the speaker on the discourse he produces. But Authier-Revuz
(1996), following work by Saussure (on anagrams), Freud (on play on words and
psychanalytic discourse) and Kristeva (on paragrammatic discourse), suggests a more
radical conception of polyphony, referring to a divided and confictual subject, and
thus to a speaker who may express unconsciously and simultaneously in the same
discourse conficting points of view.
As for other important dimensions of the organization of discourse, a lot of
empirical research still has to be done in order to describe precisely the forms and
functions of polyphony in diferent discourse types, and the interrelations between
polyphony and other discourse dimensions, but thanks to the work of Bakhtin, Ducrot
and recent research in dialogue analysis and critical discourse analysis, we get progres-
sively a better picture of this still underdeveloped feld of pragmatics and discourse
analysis.
References
Authier-Revuz, J. (1996). Ces mots qui ne vont pas de soi. Larousse.
Bakhtin, M. (under the name of V.N. Voloinov) (1973). Marxism and the philosophy of language.
Seminar Press.
(1981). Discourse in the novel. In Te dialogic imagination. Four essays. University of Texas
Press.
Banfeld, A. (1982). Unspeakable sentences. Routledge and Kegan Paul.
Ducrot, O. (1972). Dire et ne pas dire. Hermann.
(1984). Esquisse dune thorie polyphonique de lnonciation. In O. Ducrot Le dire et le dit.
Seuil.
Fairclough, N. (1988). Discourse representation in media discourse. Sociolinguistics 17: 125139.
(1992). Discourse and social change. Polity Press.
Fowler, R. (1981). Literature as social discourse. Indiana University Press.
Kristeva, J. (1969). Semeiotik. Seuil.
222 Eddy Roulet
Mulkay, M. (1985). Agreement and disagreement in conversations and letters. Text 5: 201227.
Roulet, E. (1995). Geneva School. In J. Verschueren, J.-O. stman & J. Blommaert (eds.): 319323.
Roulet, E. & al. (1985). L articulation du discours en franais contemporain. Lang.
Roulet, E. (1999). La description de lorganization du discours. Didier.
Roulet, E., L. Filliettaz & A. Grobet (2001). Un modle et un instrument danalyse de lorganisation du
discours. Lang.
Todorov, T. (1981). Mikhal Bakhtine, le principe dialogique, suivi de Ecrits du Cercle de Bakhtine.
Seuil.
Wodak, R. (1995). Critical linguistics and critical discourse analysis. In J. Verschueren, J.-O. stman &
J. Blommaert (eds.): 204210.
Verschueren, J., J.-O. stman & J. Blommaert (eds.) (1995). Handbook of Pragmatics, Manual. John
Benjamins.
Pragmatic markers
Karin Aijmer & Anne-Marie Simon-Vandenbergen
University of Gothenburg & University of Ghent
1. The tradition and the present state of research on pragmatic markers
Conversations in English contain a lot of small words such as well, you know, I mean,
sort of which have been referred to as pragmatic markers. Although they have recently
received a great deal of attention, our understanding of the class of pragmatic markers
is still incomplete. To start with, it is difcult to choose among diferent approaches,
terminologies and classifcations. A recent monograph describes the state of art in this
feld of research as follows:
For a newcomer to the feld, it is () ofen very difcult to fnd the bits and pieces that
constitute an original model of the meanings and functions of discourse particles.
Moreover, the studies available so far are hardly comparable; the approaches vary
with regard to very many diferent aspects: the language(s) under consideration,
the items taken into account, the terminology used, the functions considered, the
problems focussed on, and the methodologies employed. (Fischer 2006: 1)
In 1995 Jan-Ola stman published an article called Pragmatic particles twenty years
afer reminding us that the present upsurge of interest in pragmatic markers only
goes back to the late 1960s and early 1970s, although we fnd some early landmarks
such as Dennistons Te Greek particles (1934) and Arndts article Modal particles in
Russian and German, published in 1960.
Te real break-through in particle or pragmatic marker studies came with Weydts
Abtnungspartikel (1969). Also other early works have a German focus and deal with
modal particles rather than with what are known in the Anglo-Saxon tradition as
discourse particles (or discourse markers). In another early work Elizabeth Glich
(1970) discussed French discourse markers, which she referred to as Gliederungssig-
nale within the framework of Macrosyntax.
An impetus to study pragmatic markers came from articles showing that there
were grammatical phenomena which seemed to be dependent on context rather than
on rules formulated in syntax. For example Corum (1975) showed that certain prag-
matic markers (adverbs in her terminology) could be inserted almost anywhere in the
sentence depending on pragmatic factors. Another factor in the development of the
feld was the possibility to record and study authentic spoken language, an important
locus for pragmatic markers.
224 Karin Aijmer & Anne-Marie Simon-Vandenbergen
From a slow beginning the study of pragmatic markers has exploded. Several
publications have been infuential in shaping our view of the nature of expressions
variously referred to as pragmatic markers, discourse markers or discourse particles
(see Section 3 below). Te importance of Deborah Schifrins pioneering Discourse
markers (1987) can probably not be exaggerated. It has been followed by other mono-
graphs such as Schourup (1985), Jucker and Ziv (1998), Lenk (1998), Hansen (1998),
Andersen and Fretheim (2000), Fischer (2000), Aijmer (2002). Other monographs
focus on specifc markers, such as Erman (1987) on you know, I mean, you see and
stman (1981) on you know.
Te study by Jocelyne Fernandez (1994) stands out from the other works because
it deals with pragmatic markers (particules nonciatives) in the languages of the
world. Independently of the European work on pragmatic markers there has also been
research on particles with a modal or evidential meaning in non-European languages
(Chafe & Nichols 1986; Aikhenvald 2004).
Tere are several new tendencies in the study of pragmatic markers which will
be discussed below. Present-day research on pragmatic markers is characterised by
a broadening of the feld to include new phenomena, as well as by a wide spectrum
of approaches. Te number of elements which are treated as pragmatic markers is
growing and there is more interest in pragmatic markers which are less prototypical. It
is for instance not unusual to describe connectives and vocatives as pragmatic markers
(Fraser 1996).
Research on pragmatic markers is expanding in diferent directions. Such markers
are found in many languages and are therefore of interest to cross-linguistic and typo-
logical research. Tey are also studied diachronically, although natural spoken data
is not available. Moreover, a synchronic analysis is ofen combined with a diachronic
analysis focusing on the origin of the markers, to show how their pragmatic functions
have evolved as the result of systematic semantic processes associated with the gram-
maticalization of lexical elements or constructions. Finally, there is more interest in
considering sociolinguistic factors: who uses pragmatic markers and in what situations?
2. Defning the feld
Pragmatic marker is an unclear term and many diferent defnitions are found in the
literature.
Schifrin (1987) characterises discourse markers (her term) as deictic and suggests
that they have indexical functions. What discourse markers or pragmatic markers gen-
erally do is to indexically point to features of the context. Te context to which mark-
ers index utterances can be referred to as diferent discourse planes: the ideational
structure (ideas and propositions), action structure (the way in which speech acts
Pragmatic markers 225
relate to preceding and following or intended actions), exchange structure (turns),
information state (management of knowledge and meta-knowledge) and participant
framework (the speaker-hearer relationship). Ochs (1996) further mentions social
identity (the social persona), social act (e.g. request, ofer), activity (e.g. arguing) and
stance (afective and epistemic) as crucial situational dimensions. In contrast with
prototypical deictic elements such as pronouns, temporal and spatial expressions,
however, pragmatic markers index elements which are less concrete and not yet fully
understood in their full complexity (cf. Aijmer et al. 2006).
Another feature characterising pragmatic markers is refexivity, in the sense that
they comment on the utterance and thus assist in the interpretation of that utterance.
Te marker in that sense has a meta status and an understanding of its functioning
involves defning its metalinguistic or metapragmatic role (Aijmer et al. 2006).
Pragmatic markers can also be seen as heteroglossic, in the sense of the term as
used by White (2003). Aijmer et al. (2006) argue that markers have the interactional
and argumentative function in the discourse to signal the speakers position vis--vis
the hearers, his or her expectations or contextual assumptions (see also Aijmer &
Simon-Vandenbergen 2004 on an application to the semantic feld of expectation).
Other scholars have suggested more features, some of which are diferent,
although there is a great deal of agreement and overlap. stman (1995: 100) refers to
the class-identifying function of pragmatic markers as implicit anchoring: they are
windows through which one can make deductions about the speakers attitudes
and opinions.
Hlker (1991) focuses on four features defning pragmatic markers, of which the
frst two are negative. Pragmatic markers do not afect the truth conditions of an utter-
ance and they do not add anything to the propositional content of an utterance. More-
over, they are related to the speech situation and not to the situation talked about. Tey
can also be defned functionally: they have an emotive, expressive function rather than
a referential, denotative, or cognitive function.
Rhlemann (2007) mentions fve partly overlapping features in his defnition
of discourse markers: (1) they indicate how discourse relates to other discourse; (2)
they do meta-lingual work; (3) they are discourse-deictic and indicate how the utter-
ance containing them is a response to preceding discourse; (4) they create discourse
coherence and (5) they are oriented to the hearers needs.
Another line of approach is to characterize prototypical pragmatic markers as
fully as possible using phonological, lexical, syntactic, semantic, functional, sociolin-
guistic and stylistic features.
Te following list of features is based on Brinton (1996) and includes formal
and functional criteria. Te criteria also refer to the type of data where pragmatic
markers are typically found (i.e. in speech and particularly conversation rather
than in writing).
Phonological and lexical features:
a. they are short and phonologically reduced;
b. they form a separate tone group;
c. they are marginal forms and hence difcult to place within a traditional word
class.
Syntactic features:
a. they are restricted to sentence-initial position;
b. they occur outside the syntactic structure or they are only loosely attached
to it;
c. they are optional.
Semantic feature:
a. they have little or no propositional meaning.
Functional feature:
a. they are multifunctional, operating on several linguistic levels simultaneously.
Sociolinguistic and stylistic features:
a. they are a feature of oral rather than written discourse and are associated with
informality;
b. they appear with high frequency;
c. they are stylistically stigmatised;
d. they are gender specifc and more typical of womens speech. (cf. Hlker 1988;
Jucker & Ziv 1998; stman 1982)
3. The terminology: Pragmatic marker or discourse marker?
Te question of terminology is important because the terms ofen refect diferent per-
spectives on the functions and status of the markers and on what to include in and
exclude from the class. It appears, though, that there is little consensus here, which
is partly due to the fact that pragmatic markers are analysed from many theoretical
perspectives. Not surprisingly, pragmatic markers have been called by many diferent
names including hesitation marker and fller when describing specifc functions that
they can have in context. Some of the best known terms are pragmatic marker (used e.g.
by Brinton 1996), discourse marker (used e.g. by Schifrin 1987 and Jucker & Ziv 1998),
discourse particle (Hansen 1998; Aijmer 2002), and pragmatic particle (stman 1995).
Discourse marker is probably the most frequently used term and is therefore also
found as a broad covering term (Lewis 2006; Jucker & Ziv 1998). However the term
is tricky because it is also used narrowly to defne markers which are sequentially
dependent elements which bracket units of talk (Schifrin 1987). Cf. also Fraser (1990,
1996, 1999) for a similar defnition of discourse markers.
Pragmatic marker is preferred to discourse marker when the markers have a
pragmatic rather than a discourse-marking function. Tis is the case when markers
serve to mark illocutionary force or have an interactional function, for instance taking
the turn or yielding it.
Pragmatic marker is, however, most commonly used as a general or umbrella term
covering forms with a wide variety of functions both on the interpersonal and tex-
tual levels. Here also there is a notable lack of agreement on what to include in this
class. Tis is particularly the case with pauses and hesitation markers that do not seem
to qualify as words (uhm, erm). In contrast, Stenstrm (1990a) has shown that both
pauses and verbal fllers serve as interactional devices in conversation with functions
such as turn-taking, turn-holding and turn-yielding.
Both marker and particle have given rise to discussion. Te term marker captures
the fact that an element functions as a signpost or signal instructing the hearer how the
message should be interpreted. It is used more broadly than particle, which is above all
a well-established grammatical term for a part-of-speech. Marker also has the advan-
tage of not suggesting the formal restrictions of the term particle, which tends to refer
to short monosyllabic items (Andersen & Fretheim 2000: 1).
4. Classifcation
Pragmatic markers have little in common formally. Many elements which have been
categorized as belonging to other word classes can be categorized as pragmatic mark-
ers when they are not part of the propositional content. As a result the class of prag-
matic markers is large. It includes connectives, modal particles, pragmatic uses of
modal adverbs, interjections, routines (how are you), feedback signals, vocatives, dis-
juncts (frankly, fortunately), pragmatic uses of conjunctions (and, but), approximators
(hedges), reformulation markers (Glich & Kotschi 1983; Rossari 1994).
We can expect pragmatic markers to function diferently depending on their origin.
It is therefore helpful to distinguish between diferent types. Fraser (1996, 2006) has pro-
posed a taxonomy of pragmatic markers which distinguishes between basic pragmatic
markers (signalling the illocutionary force of the utterance), commentary markers (sig-
nalling a comment on the basic message), parallel markers (including some vocatives)
where the message signalled is diferent from the basic message, and discourse mark-
ers. Basic markers are illustrated by performative expressions (I promise that I will be
there on time) and by pragmatic idioms such as please. Commentary markers can be of
several kinds. For example, frankly comments on the manner in which the message is
conveyed. Fortunately is an assessment marker signalling the speakers evaluation of the
basic message. Evidential markers are illustrated by certainly and conceivably. On the
other hand, afer all would be a discourse marker since it signals the relationship of the
utterance in which it is placed to the preceding utterance:
Mary has gone home. Afer all, she was sick.
Parallel markers can be illustrated by vocatives and by conversational management
markers such as now or okay.
Many pragmatic markers do not ft easily into this framework. Fraser (1996) did
for instance not include well or oh as pragmatic markers and there is no consensus
about what to do with pauses or hesitation and reformulation markers, especially if
they do not seem to qualify as words (uhm,erm).
5. Pragmatic markers and multifunctionality
Pragmatic markers have a number of diferent functions depending on the context.
Tis raises the question whether they have one meaning or many meanings. Te prob-
lem is accentuated by the fact that most items described as pragmatic markers also have
uses which can be described in syntactic terms as clausal constituents, although there
are diferences between individual markers in this respect. Stenstrm (1990b: 162)
illustrates this by means of a comparison between yeah, which occurs as a pragmatic
marker (Stenstrms discourse item) in 100% of the examples in the London-Lund
Corpus, of course which occurs as a pragmatic marker in 11% of the examples and
well as a pragmatic marker in 86% of the instances. Furthermore, some items have
clause-constituent and pragmatic marker functions which are clearly distinguishable
from each other (compare the manner adverb well and the pragmatic marker well),
while for other items the distinction is less clear-cut (compare the epistemic adverb
surely and the pragmatic marker surely). To complicate the picture, even in their use as
pragmatic marker most items are multifunctional.
What frst comes to mind as a solution to the multifunctionality of the pragmatic
markers is homonymy. Homonymy would imply that the number of markers is mul-
tiplied in order to establish a one form one function relationship. In the mono-
semy approach, on the other hand, each pragmatic marker is associated with a single
abstract meaning functioning as the common denominator for the diferent contextu-
ally determined meanings or functions of the marker (meaning minimalism rather
than meaning maximalism).
An alternative to monosemy is to explain multifunctionality as the result of
polysemy: diferent meanings or functions can be related as extensions from a core
(prototype). Kroon (1995) and Hansen (1998) both advocate polysemy. Te poly-
semy approch is ofen combined with or is a prerequisite for the diachronic analysis
of pragmatic markers in the framework of grammaticalization theory (Traugott &
Dasher 2002; Schwenter & Traugott 2000 on the development of the discourse marker
in fact).
A promising approach to describing the multifunctionality of pragmatic markers
is to regard them as constructions. Constructions are pairings of form and meaning,
including the pragmatic constraints associated with the form (stman 2006).
Finally, it should be mentioned that the notion of a unitary meaning of pragmatic
markers is also compatible with relevance theory. In this case relevance-theoretical
principles are used to provide a unifed account of the diferent functions of prag-
matic markers. See e.g. Jucker (1993) on well, which is shown to serve the function of
restoring a breach of common ground.
6. Theoretical approaches to the study of pragmatic markers
Pragmatic markers can have an almost infnite number of functions depending on the
context. Moreover they can overlap with other markers in some of their meanings.
Describing and constraining the multifunctionality of pragmatic markers is therefore
a challenging task. Tis is refected in the richness of theoretical frameworks which
have been suggested. Pragmatic markers can be accommodated in speech act theory,
as shown by Brown and Levinsons (1987) analysis of them as markers of illocution-
ary force (illocutionary force indicating devices ifds) or as speech act adverbials
(Mittwoch 1976; Andersson 1976). In more recent descriptions the context has been
extended beyond the utterance in which the pragmatic marker is situated and also
includes social and contextual factors. stman (1995) has for instance suggested that
pragmatic markers can have meanings with regard to social and cultural parameters
such as politeness, discourse coherence or involvement.
Te importance of a deeper understanding of what goes on in discourse is
illustrated by Schifrin (1987). According to Schifrin, we must go beyond a surface
description of discourse and study the layers of meaning of which it is made up:
exchange (turns, adjacency pairs), information structure (the speakers and hearers
knowledge state), action (speech acts), participation framework (speaker, hearer).
Pragmatic markers (discourse markers) achieve coherence by indexically pointing to
and integrating these domains or meanings in discourse (See Section2 above). Cf. also
Redeker 1990 for a coherence approach to pragmatic markers.
Te development of grammaticalization theory has led to many innovative
studies of pragmatic markers. Diachronic research into the origin and development
of pragmatic markers has tended to explain multifunctionality as the result of gram-
maticalization. In grammaticalization theory the focus is on the systematic semantic
and syntactic developments of lexical elements into elements which have lost most
of their semantic content and gained pragmatic meaning, and have undergone gram-
matical changes at the same time. Traugott (1995) and Traugott and Dasher (2002)
have for instance illustrated the diferent stages in the semantic development of in
fact, besides, indeed into pragmatic markers and the conditions leading to their gram-
maticalization. Te term used by Erman and Kotsinas (1993) and Aijmer (1997) to
describe the development of lexical elements to pragmatic markers is pragmaticaliza-
tion, which focuses on the semantic bleaching and pragmatic enrichment character-
istic of the changes to pragmatic markers. Te term pragmaticalization is also used
in the model proposed by Dostie (2004). Degand and Simon-Vandenbergen (2011)
takes up the issue of the connection between grammaticalization, pragmaticalization
and (inter)subjectifcation.
Relevance theory also provides a useful framework for analysing pragmatic
markers. In particular, this theory (see especially Sperber & Wilson [1986] 1995) draws
attention to the role of pragmatic markers of facilitating the hearers task of decoding
the message. Pragmatic markers are viewed as signals guiding the hearers utterance
interpretation. Tey thus contribute to understanding by reducing the processing efort
needed by the hearer to reach the intended interpretation. Examples of approaches
using a relevance-theoretical framework to analyse pragmatic markers are Blakemore
(1987, 1992), Watts (1988). Blass (1990), Jucker (1993), Ifantidou (2000), Andersen
(2001), Ler (2006).
Conversation analytic approaches (CA) (see e.g. Tsui 1994) capture the fact that
pragmatic markers need to be described in relation to the conversational or discourse
context. In CA the relevant context is formed by the turn and the exchange (adjacency
pair) as well as by larger sequences of action. A well-known study is Heritages analysis
of oh as a change-of-state token marking a change of knowledge-state from uninformed
to informed (1984). Cf. also Hakulinens (1998) analysis of the Finnish nyt (now)
using conversation analytic notions. Pragmatic markers can also mark pre-closing in
a (telephone) conversation. Scheglof and Sacks (1973) for example identifed ways in
which pragmatic markers such as OK and well can signal that the participants want
to move to farewells and close the conversation. Another important function of prag-
matic markers is to mark a response as dispreferred, for instance because it disagrees
with a preceding assessment (Pomerantz 1984).
Te multifunctionality of pragmatic markers has also been analysed in the frame-
work of Natural Semantic Metalanguage (Travis 2006). Te diferent meanings and
functions of pragmatic markers are described by a set of defnitions in terms of semantic
components or features (cf. also Wierzbicka 1976, 1994; Goddard 1994; Fischer 1998).
Many approaches to pragmatic markers have been infuenced by Systemic
Functional Linguistics (SFL), in particular by the distinction of three metafunctions:
experiential, interpersonal and textual (see for instance Halliday 2004). In Hallidays
model pragmatic markers are textual or interpersonal (rather than experiential).
Tey are typically placed in initial position in the clause, which is regarded in SFL as
thematic, so that they form then, together with the experiential Teme, a multiple
Teme. Pragmatic markers with a textual function, which are called continuatives in
SFL, are said to signal a move in the discourse: a response, in dialogue, or a new move
to the next point if the same speaker is continuing (Halliday 2004: 81). Common
continuatives are said to be yes, no, well, oh, now. Interpersonal elements, on the other
hand, are modal comment adjuncts (e.g. certainly, in my opinion, frankly, honestly),
which express the speaker or writers judgement on or attitude to the content of the
message (Halliday 2004: 81). Also vocatives are considered as interpersonal elements,
not contributing to the experiential content of the clause.
Te distinction between interpersonal and textual has been used by Brinton (1996,
2008) to group pragmatic markers into two main classes. For example, discourse mark-
ers have a discourse-marking or textual function which relates to the structuring of dis-
course as text and an interpersonal function which relates to the expression of speaker
attitudes. Among the textual functions are initiating and ending discourse, marking
boundaries in the discourse, signalling topic shif and repairing discourse. Te inter-
personal function describes the use of pragmatic markers to express responses and
attitudes as well as solidarity and face-saving.
7. Methodology
In studies of pragmatic markers the emphasis is on (preferably spoken) corpus
data. Corpora make it possible to investigate the distribution of pragmatic markers
in speech and writing and in diferent registers. English pragmatic markers have
for instance been studied in the London-Lund Corpus (Aijmer 2002; Stenstrm
1990a and b; Svartvik 1980) and in the Bergen Corpus of London Teenager lan-
guage (COLT) (Andersen 2001; Stenstrm & Andersen 1996). Te MICASE corpus
(Michigan Corpus of Academic Spoken English) has been used to study hedges (see
e.g. Mauranen 1994). Van Bogaert (2009) relies on the spoken component of the
International Corpus of English-Great-Britain (ICE-GB).
We are now also able to study pragmatic markers in earlier periods of the language
thanks to the availability of historical corpora such as A corpus of English dialogues
15601760 (Kyt & Walker 2006). Te dialogues come from such text types as trial
proceedings, witness depositions, drama comedy, didactic works and prose fction
over a 200-year period. Te corpus has for instance been used to study hedges in older
English (Culpeper & Kyt 1999).
Pragmatic markers have been studied in a cross-linguistic perspective on the basis of
parallel corpora, which allow one to set up inter-linguistic equivalents, going from source
to target language and back to source. (For English-Swedish and English-Dutch, see e.g.
Aijmer & Simon-Vandenbergen 2003; Simon-Vandenbergen & Aijmer 2002/2003);
further, comparative work is done on the basis of learner corpora (corpora of non-native
speakers English, see below).
Recently the development of multi-modal corpus resources has made it possible
to study the use of gestures co-occurring with linguistic elements such as backchan-
nels which are used to mark active listenership (Knight & Adolphs 2007).
8. Pragmatic markers in the languages of the world
Te largest number of detailed studies deals with pragmatic markers in English. Some
English markers have been researched extensively. For example, well has been examined
by Svartvik (1980), Carlson (1984), Schifrin (1987), Watts (1989), Schourup (1985,
2001), Jucker (1993), Greasley (1994), Norrick (2001), Aijmer & Simon-Vandenbergen
(2003), Aijmer (2009), amongst others.
Regional and social varieties of English are a rich source for the study for prag-
matic markers. Among the non-standard markers like has attracted a lot of interest
(Romaine & Lange 1991; Miller & Weinert 1995; Andersen 2001), especially the socio-
linguistic spread of the quotative be like (Dailey-OCain 2000). Like has been associated
with non-standard varieties as well as with American English. Both like and innit are
frequent in adolescent language (Andersen 2001; Stenstrm & Andersen 1996). In a
number of studies Janet Holmes has investigated the use of several pragmatic markers
by speakers of New Zealand English, with particular reference to gender diferences
(you know (1986), sort of (1988a), of course (1988b)). Vivian de Klerk has studied well
in Xhosa English, a subvariety of Black South African English (2005) and shown that
there are both similarities and diferences with Standard English. Te characteristics
of pragmatic markers in Singlish (Singapore English) have been dealt with by Gupta
(2006). According to Gupta, pragmatic markers can become stereotypes marking the
group and hence be used in fction and for humorous efects. Sebba and Tate (1986)
have studied how agreement markers such as you know what I mean are used in British
Black English.
French linguists have taken an interest in pragmatic markers for a long time. An
early contrastive study involving French is Albrecht (1976), which looks at French
equivalents of the German marker eigentlich. In French linguistics the focus has been
primarily on connectives and reformulation markers, as exemplifed by studies such
as Roulet (1983, 2006) and Rossari (e.g. 1994, 2000, 2006), which deal primarily with
connectives (connecteurs pragmatiques) such as donc, aprs tout. Roulet adopts a
modular methodology with modules restricted to information which can be inter-
related in the discourse organisation (Roulet, Filliettaz & Grobet 2001). Drescher and
Frank-Job (2006) is a theoretically and methodologically oriented study of discursive
markers in the Romance languages.
Important research has been carried out on the so-called ponctuants (Vincent
1993) in Canadian French. In Vincents mainly sociolinguistic study the markers are
treated as discourse variables which are realised diferently depending on the social
context. Vincent (2005) discusses the non-standard par exemple, genre and style in
their function to signal discourse relations. Within the same sociolinguistic tradi-
tion Dines (1980) and Dubois (1993) have written about so-called extension particles
(e.g. tout a and all that) in a corpus of spoken Montral French.
Other pioneering work on pragmatic markers, both synchronic and diachronic,
has been carried out by Hansen. In her 1998 monograph she discusses the markers
bon, ben, eh bien, puis, donc, alors. Pragmatic markers are also characteristic of French
dialects. Pusch (2000) discusses preverbal pragmatic markers in Gascon traditionally
referred to as enunciatives which have long been a puzzle to Romance linguists.
Schwenter (2001) discusses additive particles in Spanish. Moreover, much work
on pragmatic markers in Spanish has been carried out by the research group Val.Es.Co
(Valencia Colloquial Spanish) by Pons Bordera (2006), Briz (1993a and b), Martn
Zorraquino and Portols (1999), Ferrer and Pons Bordera (2001). Italian discourse
markers have been discussed in several works by Bazzanella (e.g. 1990, 2006).
In the Swedish tradition pragmatic markers have ofen been referred to as speech
act adverbials (Andersson 1976). Pragmatic markers are difcult to distinguish
from modal particles and are ofen treated together. Pragmatic markers in Swed-
ish have been discussed by Eriksson (1988), Ottesj (2005), Lehti-Eklund (2003),
Saari (1984). Erman and Kotsinas (1993) and Eriksson (1992) discuss the Swed-
ish particle ba used by adolescents and a close functional correspondence to you
know. Lindstrm and Wide (2005) are interested in pragmatic markers of the you
know type and their derivation as pragmatic markers. Among lesser known varieties
we can mention stmans study of pragmatic markers in Solv, a dialect of Finland
Swedish (e.g. 2006).
Danish discourse particles have been described by Davidsen-Nielsen (1996).
He distinguishes between grammatical discourse particles related to modality
and evidentiality which can be regarded as modal particles and lexical discourse
markers.
In the Finnish tradition the Conversation Analytic approach is strong. Hakulinen
(1998) has studied the particle nyt (now) in diferent sequential contexts and Sorjonen
(1997) has devoted her doctoral dissertation to response markers in Finnish.
Hebrew pragmatic markers have interested a number of scholars, for example
Ariel (1998), Shloush (1998) and Ziv (1998). Yael Maschler (1998) has described
the Hebrew markers occurring at frameshifs in casual Hebrew talk-in-interaction. See
also Maschler (2009).
Among other studies of non-European languages we can mention Nicolles (2000)
study of particles in Amharic and Swahili using the relevance-theoretic notion of
interpretive use.
9. The diachronic study of pragmatic markers
Two tendencies can be noticed in the study of pragmatic markers in historical
pragmatics. On the one hand, there is a lot of interest in pragmatic markers at earlier
stages of the language. Culpeper and Kyt have for example studied hedges in Early
Modern English (1999). Bertin (2002) looks at the emergence of the connector en
efet in medieval French. On the other hand, researchers have taken an interest in
the semantic evolution of pragmatic markers from diverse lexical and grammatical
sources by processes such as grammaticalization (Traugott 1995; Traugott & Dasher
2002) or pragmaticalization (Erman & Kotsinas 1993; Aijmer 1997). Schwenter and
Traugott (2000) have demonstrated how the development of in fact to a pragmatic
marker can be explained by invoking the notion of (pragmatic) scalarity. Brinton
(1996) has studied a number of pragmatic markers diachronically and more recently
(2008) so-called comment clauses (comment clauses with say, I mean, if you will,
as it were, comment clauses with look, whats more and whats else, I gather and I
fnd). Historical analyses of individual pragmatic markers also include well (Finell
1989; Jucker 1997; Defour 2010) and now (Defour 2008). Diachronic analyses are
furthermore found of markers in languages other than English. To give an example,
Hakulinen and Seppnen (1992) discuss the Finnish verb katos path from a verb to
a particle.
10. The contrastive study of pragmatic markers
From an intra-linguistic point-of-view pragmatic markers can be compared on the
basis of similarities and diferences. Smith and Jucker (2000) group together actually,
in fact and well on the basis of their function to mark discrepancy between proposi-
tional attitudes. Te subtle diferences between the functions of actually and in fact are
also the topic of Ohs detailed study of these words in American English (Oh 2000).
Another intra-lingual contrastive study is Jucker and Smiths examination of yeah, like,
you know. Yeah can be analysed from the addressees perspective as a reception marker,
like is a speaker-oriented information-centred presentation marker and you know is an
addressee-oriented presentation marker (Jucker & Smith 1998).
Cross-linguistic comparisons ofen concern cognates. Pragmatic markers may
have cognates in other languages. Te Germanic languages provide many such cases.
An example is Swedish allts (Aijmer 2007), Norwegian alts (Vask & Fretheim 1997),
German also (Fernandez-Villanueva 2007). It is therefore of interest to compare them
semantically and pragmatically. Cognates also exist across language families as a result
of borrowing. Simon-Vandenbergen and Willems (2011) study the semantic-pragmatic
development of English actually ~ French actuellement, and English in fact ~ French en
fait/de fait/au fait. Such diachronic contrastive studies throw light on general principles
furthering or hindering grammaticalization.
Also non-cognate words which are semantic-pragmatic equivalents have been the
object of contrastive studies. Fleischman and Yaguello (2004) were interested in show-
ing that French genre and English like had developed the same pragmatic functions
although their lexical origin was diferent. Tat markers in diferent languages can
develop in similar ways is also shown in a study by Fraser and Malmaud-Makowski
(1996). Tey investigated pragmatic markers in English and Spanish and showed that
the functions of denial and contrast corresponded closely.
Contrastive research in this area has shown that relationships between so-called
equivalents (whether cognates or not) are complex, and that there is most ofen partial
rather than complete overlap in semantic and pragmatic meanings. Tis is for instance
the case with relational markers such as English on the contrary or French au contraire
(Lewis 2006a). An early contrastive study of Italian infatti and English in fact based on
the Italian speakers intuitions showed some similarities of function but mostly difer-
ences (Bruti 1999). Paul Takahara (1998) has studied anyway and its equivalents in
Japanese. As demonstrated by Angela Downing (2006), etymological cognates are not
a guarantee of functional similarities. Te correspondences of English surely are not
Spanish cognates such as seguro, seguramente but we need to look for equivalents which
are not etymologically related (2006). While English actually and French actuellement
have the same origin, the former but not the latter developed into a pragmatic marker
(though there are signs of emergent pragmaticalization in spoken French), as shown
in Simon-Vandenbergen and Willems (forthcoming). Stenstrm (2006) has suggested
English correspondences for the Spanish markers o sea and pues. Willems and Demol
(2006) used a multilingual corpus (English, French, Dutch) to carry out a contras-
tive analysis not only of vraiment and really but also of their Dutch counterparts echt
and werkelijk.
Well has been the object of several contrastive studies: English-Swedish-Dutch,
Aijmer and Simon-Vandenbergen (2003); English-Norwegian, Johansson (2006);
English-Italian, Bazzanella and Morra (2000); English-Spanish, Garca Vizcaino and
Martnez-Cabeza (2005), English-Spanish-Catalan, Cuenca (2008). For the study
of well, Aijmer and Simon-Vandenbergen (2003) used the translations in the other
language as semantic mirrors. Te same methodology was used to investigate the
comparisons of of course in Swedish and Dutch (Simon-Vandenbergen & Aijmer
2002/2003; Aijmer and Simon-Vandenbergen 2004; Aijmer et al. 2006). Comparisons
between German and Dutch are for instance Westheide (1985) on German wohl and
Dutch wel, and Foolen (2006) on German doch and Dutch toch.
Te methodology may involve the use of parallel or contrastive corpora
(Altenberg & Aijmer 2001). On the basis of the English-Norwegian Parallel Corpus
Hasselgrd (2006) compares English now and Norwegian n using the framework of
Systemic Functional Linguistics. Celle (1999), on the other hand, contrasts now with
the German nun and jetzt. Italian allora and French alors are studied in terms of con-
vergence and divergence by Bazzanella et al. (2007). Romero-Trillo (2007) shows that
involvement markers such as English I mean, you know and you see are used diferently
in English and Spanish and Matamala (2007) studies the functions of oh in English sit-
coms and their translations into Catalan. For a contrastive study of pragmatic markers
in English and Catalan oral narratives, see Gonzlez (2004).
11. Pragmatic markers in translation studies
Pragmatic markers have also been of interest to translators. Because of their
multifunctionality and context-boundness they are difcult to translate. Ofen they
are not translated literally but are rendered by words or constructions from many dif-
ferent word classes. Moreover they are ofen omitted from the translation (Aijmer &
Simon-Vandenbergen 2003; Altenberg & Aijmer 2002). Matamala (2007) has stud-
ied the strategies used to translate oh in English sitcoms in the versions dubbed into
Catalan. Compare also Chaume (2004) on discourse markers in audiovisual trans-
lating. Bazzanella and Morra (2000) stress the specifc problems of translating dis-
course markers, illustrating this with the translations of well into Italian. In the study
by Cuenca (2008) the focus is on what we can discover about the multifunctionality of
well on the basis of a contrastive analysis of the flm Four weddings and a funeral and
its translations in Spanish and Catalan.
12. Pragmatic markers in native versus non-native
speaker communication
Te study of pragmatic markers has entered a number of new felds such as second lan-
guage acquisition (interlanguage pragmatics).We can now take advantage of learner
corpora to make comparisons between native and non-native speakers of English. One
of the frst studies was by Nikula (1996), who compared the use of pragmatic markers
with a hedging function in conversations by native (English) speakers with non-native
Finnish speakers. It is especially the existence of spoken learner corpora such as the
LINDSEI Corpus (Louvain International Database of Spoken English Interlanguage)
which invites scholars to make comparisons (de Cock 2004).
Simone Mller (2005; cf. also Mller 2004) has examined a corpus of German
learners use of so, well, you know and like in comparison with native speakers use.
Buysse (2007) examines how Belgian native speakers of Dutch use so in diferent types
of question-answer sequences in an English interview setting. Te results of such
studies show that learners use pragmatic markers diferently from native speakers. A
similar study by Romero-Trillo (2002) described the situation in non-native language
as the pragmatic fossilization of discourse markers. Llinares-Garca and Romero-
Trillo (2006) is a study of discourse markers in the EFL classroom. Another study by
the same authors showed that native and non-native teachers used discourse markers
diferently in CLIL (Content and Language Integrated Learning) contexts and much
more frequently than Spanish teachers in a native context (Llinares-Garca & Romero-
Trillo 2008).
Hasselgren (2002) studied what she referred to as small words as markers of
learner fuency focusing on young Norwegian learners of English. Gilquin (2008)
has shown that the frequency and distribution of hesitation markers (including like,
I mean, you know) were diferent across native/non-native speaker contexts (see also
Fuller 2003).
13. Pragmatic markers and sociolinguistic aspects
Social variation in the use of pragmatic markers is so far a fairly unexplored feld.
One of the earliest studies to draw attention to the social background (geographical
region, education, rurality) of pragmatic markers is Ferraras (1997) study of anyway.
Ferrara suggests that pragmatic markers should be studied from a variationist per-
spective in the social, historical and functional domains. Compare also Ferrara and
Bell (1995) who show that ethnicity is an important factor explaining the spread of
be+like.
Janet Holmes has shown in several studies that pragmatic markers are used dif-
ferently by men and by women (cf. Holmes & Stubbe 1995 and the articles by Holmes
mentioned in Section 5). Erman (1992) has focused on male and female usage of
pragmatic markers in same and mixed-sex conversation.
Age is another important factor. Dailey-OCain (2000) found for instance that
younger people use like more ofen than adults. Compare also Andersen (2001), who
showed on the basis of a comparison of the material in COLT (the Bergen Corpus of
London Teenager Language) with the BNC that adolescents rather than adults are
responsible for the spread of like as a marker, and that like was particularly popu-
lar among teenagers from the highest social class. Te use of be like as a quotative
marker in the speech of British and Canadian youth was examined by Tagliamonte
and Hudson (1999).
Social class can afect the use of pragmatic markers, as shown in Macaulays work
along Bernsteinian lines. Macaulay (2002) found age, gender and social class difer-
ences in two sets of Scottish data that did not appear to be based on the role of shared
information implied by you know. A similar study is Huspek (1989). Huspek discusses
both you know and I think with regard to social class. He shows that you know and I
think have diferent functions for working-class speakers and power-holding speakers.
Simon-Vandenbergen (2002) also verifed Bernsteins hypothesis (1974) that I think is
a middle class expression on the basis of conversation data from the BNC. She found
indications that there are social class diferences in usage. More work on the link
between pragmatic markers and social factors is called for.
Another sociolinguistic line of investigation is the usage of particular pragmatic
markers in specifc registers and genres. One genre which has received attention from
this point-of-view is political discourse. Simon-Vandenbergen (1998, 2000) examines
the use of I think by political speakers as compared with its use in casual conversation
and Fetzer (2008) considers the use of cognitive verbs in general in political dis-
course. Fetzer has also studied politicians uses of hedges such as sort of and kind of
(Fetzer 2009).
14. Pragmatic markers and the future
So far prosodic features of pragmatic markers have been largely neglected and recent
approaches now explore the possibility of integrating prosody into the analysis in a
more systematic way. Prosody plays an important role in distinguishing various uses of
pragmatic markers. For instance, the temporal adverb now and the pragmatic marker
now tend to be realised diferently from a prosodic point-of-view (Aijmer 2002). Dif-
ferent uses of Swedish men (but) have been shown to have diferent prosodic realisa-
tions (Horne et al. 2001). Ferrara (1997) has shown how the intonation pattern difers
for the adverbial and the discourse marker use of anyway. A similar study has been
carried out by Wichmann et al. (2010), which studies the prosody of of course in rela-
tion to its position in the clause and its function. An explanation is provided in terms
of grammaticalization. It is clear that more work in this area is called for and will yield
interesting insights in the connection between grammar and the role of intonation in
expressing stance and structuring information.
From a variationist point-of-view we envisage a further expansion of the feld to
include detailed studies of pragmatic markers in various text types, studies in more
languages as well as more studies comparing native and non-native usage. We also
need more studies of the diachronic developments of pragmatic markers and studies
of pragmatic markers at earlier stages of the language.
Another interesting avenue of further research is the investigation of semantic
felds. Aijmer and Simon-Vandenbergen (2004) made a case for setting up a semantic
feld of expectation, based on cross-linguistic data (English, Swedish, Dutch) gath-
ered from translation corpora. Lewis (2006b), looking at adversative relational mark-
ers (English and French) shows how semantic maps can be set up using comparable
corpora. More work in this area using more languages would deepen our insight into
principles of semantic-pragmatic developments.
Finally, we need more theoretical refection on the category of pragmatic markers
and its place in the grammar. Te concepts of grammaticalization, pragmaticalization,
(inter)subjectifcation need to be used only when the criteria under which they apply
have been strictly defned. Such refection involves defning not only the concepts
themselves but also what we mean by grammar and grammatical categories. In the
wake of increasingly more thorough research into natural spoken language data the
traditional defnitions may be in need of reconsideration.
1
References
Aijmer, K. (1997). I think-an English modal particle. In T. Swan, & O. Jansen Westvik (eds.) Modal-
ity in Germanic lanaguages: Historical and comparative perspectives: 147. Mouton de Gruyter.
(2002). English discourse particles. Evidence from a corpus. John Benjamins.
(2007). Te meaning and functions of the Swedish discourse marker allts-Evidence from
translation corpora. Catalan Journal of Linguistics 6: 3159.
Aijmer, K. (2009). Te pragmatic marker well: a text study. In: O. Navratilova and R. Povoln (eds.)
Coherence and cohesion in spoken and written discourse: 429. Cambridge Scholars Publishing.
Aijmer, K., A. Foolen & A.-M. Simon-Vandenbergen (2006). Pragmatic markers in translation: a
methodological proposal. In K. Fischer (ed.): 101114.
Aijmer, K. & A.-M. Simon-Vandenbergen (2003). Well in English, Swedish and Dutch. Linguistics
41(6): 11231161.
(2004). A model and a methodology for the study of pragmatic markers: Te semantic feld of
expectation. Journal of Pragmatics 36: 17811805.
(eds.) (2006). Pragmatic markers in contrast. Elsevier.
Aikhenvald, A.Y. (2004). Evidentiality. Oxford University Press.
Albrecht, J. (1976). Les quivalents de lallemand eigentlich dans les dictionnaires bilingues et dans la
ralit de lusage. Cahiers de lexicologie 32(1): 6073.
Altenberg, B. & K. Aijmer (2001). Te English-Swedish Parallel Corpus: A resource for contrastive
research and translation studies. In C. Mair & M. Hundt (eds.) Corpus linguistics and linguistic
theory. Papers from the 20th International Conference on English Language Research on Comput-
erized Corpora (ICAME 20) Freiburg im Breisgau 1999: 1533. Rodopi.
(2002). Zero translations and cross-linguistic equivalence: evidence from the English-Swedish
Parallel Corpus. In L.-E. Breivik & A. Hasselgren (eds.) From the COLTs mouth and others.
Language corpora studies in honour of Anna-Brita Stenstrm: 1941. Rodopi.
Andersen, G. (2001). Pragmatic markers and sociolinguistic variation. John Benjamins.
Andersen, G. & T. Fretheim (eds.) (2000). Pragmatic markers and propositional attitude. Benjamins.
:. Anne-Marie Simon-Vandenbergens research for this article was carried out in the framework
of the Interuniversity Attraction Poles (IAP), Phase VI, GRAMIS project, funded by the Belgian
Science Policy.
Andersson, L.-G. (1976). Talaktsadverbial. Nysvenska studier 56: 2546.
Ariel, M. (1998). Discourse markers and form-function correlations. In A. Jucker & Y. Ziv (eds.):
223259.
Arndt, W. (1960). Modal particles in Russian and German. Word 16(3): 323336.
Bazzanella, C. (1990). Phatic connectives as interactional cues in contemporary spoken Italian. Jour-
nal of Pragmatics 14: 629647.
(2006). Discourse markers in Italian: towards a compositional meaning. In K. Fischer (ed.):
449464.
Bazzanella, C. & L. Morra (2000). Discourse markers and the indeterminacy of translation. In
I. Korzen & C. Marello (eds.) Argomenti per una linguistica della traduzione. On linguis-
tic aspects of translation. Notes pour une linguistique de la traduction: 149157. Edizioni
dellOrso.
Bazzanella, C., C. Bosco, A. Garcea, B.G. Fivela, J. Miecznikowski & F.T. Brunozzi (2007). Italian allora,
French alors: Functions, convergences and divergences. Catalan Journal of Linguistics 6: 930.
Bernstein, B. (1974). Classes, codes and control. Routledge and Kegan Paul.
Bertin, A. (2002). Lmergence du connecteur en efet en moyen franais. Linx 46(1): 3750.
Blakemore, D. (1987). Semantic constraints on relevance. Blackwell.
(1992). Understanding utterances. Blackwell.
Blass, R. (1990). Relevance relations in discourse. A study with special reference to Sissala. Cambridge
University Press.
Brinton, L. (1996). Pragmatic markers in English. Grammaticalization and discourse functions.
Mouton de Gruyter.
(2008). Te comment clause in English. Syntactic origins and pragmatic development. Cambridge
University Press.
Briz, A. (1993a). Los conectores pragmticos en espaol coloquial. I: Su papel argumentativo.
Contextos 11: 145188.
(1993b). Los conectores pragmticos en la conversacin coloquial. II. Su papel metadiscur-
sivo. Espaol actual 59: 3956.
Brown, P. & S.C. Levinson (1987). Politeness. Some universals in language usage. Cambridge
University Press.
Bruti, S. (1999). In fact and infatti: Te same, similar or diferent. Pragmatics 9(4): 519533.
Buysse, L. (2007). Discourse marker so in the English of Flemish university students. Belgian Journal
of English Language and Literatures (BELL), New Series 5: 7995.
Carlson, L. (1984). Well in dialogue games: A discourse analysis of the interjection well in idealized
conversation. Benjamins.
Celle, A. (1999). Now, nun, jetzt, marqueurs de qualit ou de quantit, tude contrastive anglais/
allemand. In J. Guillemin-Flescher & A. Deschamps (eds.) Les Oprations de Dtermination,
quantifcation/qualifcation: 131148. Ophrys.
Chafe, W. & J. Nichols (eds.) (1986). Evidentiality: Te linguistic coding of epistemology. Ablex.
Chaume, F. (2004). Discourse markers in audiovisual translating. Meta 49(4): 843855.
Corum, C. (1975). A pragmatic analysis of parenthetic adjuncts. In R.E. Grossman et al. (eds.)
Papers from the Eleventh Regional Meeting of the Chicago Linguistic Society: 131141. Chicago
Linguistic Society.
Cuenca, M.-J. (2008). Pragmatic markers in contrast: Te case of well. Journal of Pragmatics 40(8):
13731391.
Culpeper, J. & M. Kyt (1999). Modifying pragmatic force: Hedges in a corpus of Early Mod-
ern English dialogues. In A. Jucker, G. Fritz & F. Lebsanf (eds.) Historical dialogue analysis:
293312. John Benjamins.
Dailey-Ocain, J. (2000). Te sociolinguistic distribution of and attitudes toward focuser like and
quotative like. Journal of Sociolinguistics 4(1): 6080.
Davidsen-Nielsen, N. (1996). Discourse particles in Danish. In E. Engberg Pedersen, M. Fortescue,
P. Harder, L. Heltof & L. Falster Jacobsen (eds.) Content, expression and structure. Studies in
Danish Functional Grammar: 283314. John Benjamins.
De Cock, S. (2004). Preferred sequences of words in NS and NNS speech. Belgian Journal of English
Language and Literatures (BELL), New Series 2: 225246.
Defour, T. (2008). And so now. Te grammaticalisation and (inter)subjectifcation of now. In
T. Nevalainen, I. Taavitsainen, P. Pahta & M. Korhonen (eds.) Te Dynamics of linguistic varia-
tion. Corpus evidence on English past and present: 18736. John Benjamins.
(2010). Te semantic-pragmatic development of well from the viewpoint of (inter) subjecti-
fcation. In H. Cuyckens, K. Davidse & L. Vandelanotte (eds.) Subjectifcation, intersubjectifca-
tion and grammaticalization (Topics in English Linguistics): 155195. Mouton.
Degand, L. & A.-M. Simon-Vandenbergen (eds.) (2011). Grammaticalization, pragmaticalization
and/or (inter)subjectifcation: Methodological issues for the study of discourse markers. Te-
matic issue of Linguistics 49(2).
De Klerk, V. (2005). Procedural meanings of well in a corpus of Xhosa English. Journal of Pragmat-
ics 37(8): 11831205.
Denniston, J.D. (1934). Te Greek particles. Clarendon.
Dines, E.R. (1980). Variation in discourse and stuf like that. Language in Society 9: 1333.
Dostie, G. (2004). Pragmaticalisation et marqueurs discursifs: analyse smantique et traitement lexi-
cographique. Champs linguistiques. De Boeck Duculot.
Downing, A. (2006). Te English pragmatic marker surely and its functional counterpart in Spanish.
In K. Aijmer & A.-M. Simon-Vandenbergen (eds.): 3958.
Drescher M. & B. Frank-Job (2006). Les marqueurs discursifs dans les langues romanes: approches
thoriques et mthodologiques. Lang.
Dubois, S. (1993). Extension particles, etc. Language Variation and Change 4: 179203.
Eriksson, M. (1988). Ju, vl, d, va, allts. En underskning av talaktsadverbial i stockholmskt
talsprk. In A. Schenstrm & M. Eriksson (eds.) Studier i stockholmssprk: 75120. (MINS).
Institutionen fr nordiska sprk, Stockholms universitet.
(1992). Ett fall av grammatikalisering i modern svenska: ba i ungdomars talsprk. FUMS
rapport 166. Uppsala.
Erman, B. (1987). Pragmatic expressions in English: A study of you know, you see and I mean in
face-to-face conversation. Almqvist & Wiksell.
(1992). Female and male usage of pragmatic expressions in same-sex and mixed-sex interac-
tion. Language Variation and Change 4: 217234.
Erman, B. & U.-B. Kotsinas (1993). Te case of ba and you know. Studier i modern sprkvetenskap
10: 7693.
Fernandez, J.M.M. (1994). Les particules nonciatives dans la construction du discours. Presses
Universitaires de France.
Fernndez-Villanueva, M. (2007). Uses of also in semi-informal German. Catalan Journal of Linguistics
6: 95115.
Ferrara, K. (1997). Form and function of the discourse marker anyway: Implications for discourse
analysis. Linguistics 35: 343378.
Ferrara, K. & B. Bell (1995). Sociolinguistic variation and discourse function of constructed dialogue
introducers: the case of be+like. American Speech 70(3): 265290.
Ferrer, H. & S. Pons Bordera (2001). La pragmtica de los conectorer y las particulas modales.
Universidad de Valencia.
Fetzer, A. (2008). And I think that is a very straight forward way of dealing with it. Te communica-
tive function of cognitive verbs in political discourse. Journal of Language and Social Psychology
27(4): 384396.
(2009). Sort of and kind of in political discourse: hedge, head of NP or contextualization cue?
In A. Jucker, M. Hundt & D. Schreier (eds.) Corpora: Pragmatics and Discourse. Papers from the
29th International Conference on English Language Research on Computerized Corpora: 125147.
Rodopi.
Finell, A. (1989). Well now and then. Journal of Pragmatics 13: 653656.
Fischer, K. (1998). Validating semantic analyses of discourse particles. Journal of Pragmatics 29:
111127.
(2000). From cognitive semantics to lexical pragmatics: the functional polysemy of discourse
particles. Mouton de Gruyter.
(ed.) (2006). Approaches to discourse particles. Elsevier.
Fleischman, S. & M. Yaguello (2004). Discourse markers across languages? Evidence from English
and French. In C.L. Moder & A. Aida Martinovic-Zic (eds.) Discourse across languages and
cultures: 129147. Benjamins.
Foolen, A. (2006). Polysemy patterns in contrast. Te case of Dutch toch and German doch. In
K. Aijmer and A.-M. Simon-Vandenbergen (eds.): 5972.
Fraser, B. (1990). An approach to discourse markers. Journal of Pragmatics 14: 38395.
(1996). Pragmatic markers. Pragmatics 6(2): 167190.
(1999). What are discourse markers? Journal of Pragmatics 31: 931952.
(2006). Towards a theory of discourse markers. In K. Fischer (ed.): 189204.
Fraser, B. & M. Malmaud-Makowski (1996). English and Spanish contrastive markers. Language
Sciences 18: 863881.
Fuller, J.M. (2003). Discourse marker use across speech contexts: a comparison of native and
non-native speaker performance. Multilingua 22: 185208.
Garca Vizcaino, M.J. & M.A. Martinez-Cabeza (2005). Te pragmatics of well and bueno in English
and Spanish. Intercultural Pragmatics 2(1): 6992.
Gilquin, G. (2008). Hesitation markers among EFL learners: Pragmatic defciency or diference?
In J. Romero-Trillo (ed.) Corpus and pragmatics.A mutualistic entente: 119149. Mouton de
Gruyter.
Goddard, C. (1994). Te meaning of lah understanding emphasis in Malay (Bahasa Melayu). Oce-
anic Linguistics 33: 145165.
Gonzlez, M. (2004). Pragmatic markers in oral narrative: Te case of English and Catalan. Benjamins.
Greasley, P. (1994). An investigation into the use of the particle well: commentaries on a game of
snooker. Journal of Pragmatics 22(5): 477494.
Gupta, A.F. (2006). Epistemic modalities and the discourse particles of Singapore. In K. Fischer (ed.):
243263.
Glich, E. (1970). Makrosyntax der Gliederungssignale im gesprochenen Franzsisch. Walter Fink.
Glich, E. & T. Kotschi (1983). Les marqueurs de la rformulation paraphrastique. Connecteurs
pragmatiques et structure du discours. Actes du 2me Colloque de Pragmatique de Genve.
Cahiers de linguistique franaise 5: 305351.
Hakulinen, A. (1998). Te use of Finnish nyt as a discourse particle. In A. Jucker & Y. Ziv (eds.):
8396.
Hakulinen, A. & E.-L. Seppnen (1992). Finnish kato: from verb to particle. Journal of Pragmatics
18: 527549.
Halliday, M.A.K. (2004). An introduction to functional grammar (3rd ed., revised by C.M.M. Matthies-
sen). Arnold.
Hansen, M.-B.M. (1998). Te function of discourse particles. A study with special reference to spoken
standard French. John Benjamins.
Hasselgrd, H. (2006). Not now-On non-correspondence between the cognate adverbs now and n.
In K. Aijmer & A.-M. Simon-Vandenbergen (eds.): 93113.
Hasselgren, A. (2002). Learner corpora and language testing. Smallwords as markers of learner fu-
ency. In S. Granger, J. Hung & S. Petch-Tyson (eds.) Computer learner corpora, second language
acquisition and foreign language teaching: 143173. John Benjamins.
Heritage, J. (1984). A change-of-state token and aspects of its sequential placement. In J.M. Atkinson &
J. Heritage (eds.) Structures of social action. Studies in conversation analysis: 299346. Cambridge
University Press.
Hlker, K. (1988). Zur Analyse von Markern, Korrektur-und Schlussmarker des Franzsichen.
Steiner.
(1991). Franzsisch: Partikelforschung. Lexikon der Romanistischen Linguistik 5(1): 7788.
Niemeyer.
Holmes, J. (1986). Functions of you know in womens and mens speech. Language in Society 15:
122.
(1988a). Sort of in New Zealand womens and mens speech. Studia Linguistica 42(2): 85121.
(1988b). Of course: A pragmatic particle in New Zealand womens and mens speech. Australian
Journal of Linguistics 8(1): 4974.
Holmes, J. & M. Stubbe (1995). You know, eh and other exasperating expressions: analysis of social
and stylistic variation in the use of pragmatic devices in a sample of New Zealand English. Lan-
guage and Communication 15(1): 6388.
Horne, M., P. Hansson, G. Bruce, J. Frid & M. Filipsson (2001). Cue words and the topic structure of
spoken discourse: Te case of Swedish menbut. Journal of Pragmatics 33: 10611081.
Huspek, M. (1989). Linguistic variability and power: An analysis of you know/I think variation in
working-class speech. Journal of Pragmatics 13: 661683.
Ifantidou, E. (2000). Procedural encoding of explicatures by the Modern Greek particle taha. In
G. Andersen & T. Fretheim (eds.): 119143.
Johansson, S. (2006). How well can well be translated? On the English discourse particle well and its
correspondences in English and German. In K. Aijmer & A.-M. Simon-Vandenbergen (eds.):
115137.
Jucker, A. (1993). Te discourse marker well. A relevance-theoretical account. Journal of Pragmatics
19(5): 43553.
(1997). Te discourse marker well in the history of English. English Language and Linguistics
1(1): 91110.
Jucker, A. & S.W. Smith (1998). And people just you know like wow: Discourse markers as negotiating
strategies. In A. Jucker & Y. Ziv (eds.): 171201.
Jucker, A. & Y. Ziv (eds.) (1998). Discourse markers. Description and theory. John Benjamins.
Knight, D. & S. Adolphs (2007). Multi-modal corpus pragmatics: the case of active listenership. In
J. Romero-Trillo (ed.) Corpus and pragmatics. A mutualistic entente: 175190. Mouton de
Gruyter.
Kroon, C. (1995). Discourse particles in Latin: a study of nam, enim, autem, vero, and at. Gieben.
Kyt, M. & T. Walker (2006). Guide to a Corpus of English Dialogues 15601760. (Studia Anglistica
Upsaliensia 130). Acta Universitatis Upsaliensis.
Lehti-Eklund, H. (2003). Te grammaticalization of allts and sledes: Two Swedish conjuncts revis-
ited. In H. Cuyckens, R. Dirven & J. Taylor (eds.) Cognitive approaches to lexical semantics:
123162. Mouton de Gruyter.
Lenk, U. (1998). Marking discourse coherence. Functions of discourse markers in spoken English.
Gunter Narr Verlag.
Ler, S.L.V. (2006). A relevance-theoretic approach to discourse particles in Singapore English. In
K. Fischer (ed.): 149166.
Lewis, D. (2006a). Discourse markers in English: a discourse-pragmatic view. In K. Fischer (ed.):
4359.
(2006b). Contrastive analysis of adversative relational markers, using comparable corpora. In
K. Aijmer & A.-M. Simon-Vandenbergen (eds.): 139153.
Lindstrm, J. & C. Wide (2005). Tracing the origin of a set of discourse particles: Swedish particles of
the type you know. Journal of Historical Pragmatics 6(2): 211236.
Llinares-Garca, A. & J. Romero-Trillo (2006). Discourse markers in the EFL classoom. In
M. Carretero, L. Hidalgo Downing, J. Lavid, E. Martnez-Caro, J. Nef, S. Prez de Ayala &
E. Sanchez-Pardo (eds.) A pleasure of life in words: A festschrif for Angela Downing. Madrid,
Universidad Complutense.
(2008). Discourse markers and the pragmatics of native and non-native teachers in a CLIL
corpus. In J. Romero-Trillo (ed.) Corpus and pragmatics. A mutualistic entente: 191204. Mou-
ton de Gruyter.
Macaulay, R. (2002). You know, it depends-Vous savez, cela dpend. Journal of Pragmatics 6: 746767.
Martn Zorraquino, M.A. & J. Portols (1999). Los marcadoeres del discurso. In I. Bosque &
V. Demonte (eds.) Gramtica descriptiva de la lengua espanola, vol. III: 40514213. Espasa
Calpe.
Maschler, Y. (1998). Rots lishmoa kta? Wanna hear something weird/funny? [lit. a segment]:
Segmenting Israeli Hebrew talk-in-interaction. In A. Jucker & Y. Ziv (eds.): 1359.
(2009). Metalanguage in interaction: Hebrew discourse markers. John Benjamins.
Matamala, A. (2007). Te translation of oh in a corpus of dubbed sitcoms. Catalan Journal of Lin-
guistics (6): 117136.
Mauranen, A. (1994). Teyre a little bit diferentObservations on hedges in academic talk. In
K. Aijmer & A.-B. Stenstrm (eds.) Discourse patterns in spoken and written corpora: 173197.
John Benjamins.
Miller, J. & R. Weinert (1995). Te function of like in dialogue. Journal of Pragmatics 23: 365393.
Mittwoch, A. (1976). Grammar and illocutionary force. Lingua 40: 2142.
Mller, S. (2004). Well you know the type of person: functions of well in the speech of American
and German students. Journal of Pragmatics 36: 11571182.
(2005). Discourse markers in native and non-native English discourse. John Benjamins.
Nicolle, S. (2000). Markers of general interpretive use in Amharic and Swahili. In G. Andersen &
T. Fretheim (eds.): 173188.
Nikula, T. (1996). Pragmatic force modifers: a study in interlanguage pragmatics. University of
Jyvskyl.
Norrick, N. (2001). Discourse markers in oral narrative. Journal of Pragmatics 33: 849878.
Ochs, E. (1996). Linguistic resources for socializing humanity. In J.J. Gumperz & S.C. Levinson (eds.)
Rethinking linguistic relativity: 407437. Cambridge University Press.
Oh, S.-Y. (2000). Actually and in fact in American English: a data-based analysis. English Language
and Linguistics 4(2): 243268.
stman, J.-O. (1981). You know. A discourse-functional approach. John Benjamins.
(1982). Te symbiotic relationship between pragmatic particles and impromptu speech. In
N.E. Enkvist (ed.) Impromptu speech: A symposium: 147177. Publications of the Research Insti-
tute of the bo Akademi Foundation. bo Akademi.
(1995). Pragmatic particles twenty years afer. In B. Wrvik, S.-K. Tanskanen & R. Hiltunen
(eds.) Organization in discourse. Proceedings from the Turku Conference: 95108. University of
Turku.
(2006). Constructions in cross-language research: Verbs as pragmatic particles in Solv. In
Ottesj, C. (2005). Att fortstta och att terg. Institutionen fr lingvistik, Gteborgs universitet.
Pomerantz, A. (1984). Agreeing and disagreeing with assessments: some features of preferred/
dispreferred turn shapes. In J.M. Atkinson & J. Heritage (eds.) Structures of social action. Studies
in conversation analysis: 57101. Cambridge University Press.
Pons Bordera, S. (2006). A functional approach to the study of discourse markers. In K. Fischer
(ed.): 7799.
Pusch, C.D. (2000). The attitudinal meaning of preverbal markers in Gascon: Insights from
the analysis of literary and spoken language data. In G. Andersen & T. Fretheim (eds.):
189206.
Redeker, G. (1990). Ideational and pragmatic markers of discourse structure. Journal of Pragmatics
14: 367381.
Romaine, S. & D. Lange (1991). Te use of like as a marker of reported speech and thought: A case of
grammaticalization in progress. American Speech 66: 22779.
Romero-Trillo, J. (2002). Te pragmatic fossilization of discourse markers in non-native speakers of
English. Journal of Pragmatics 34: 769784.
(2007). Adaptive management in discourse: Te case of involvement markers in English and
Spanish conversations. Catalan Journal of Linguistics 6: 8194.
Rossari, C. (1994). Les oprations de reformulation. Peter Lang.
(2000). Connecteurs et relations de discours: des liens entre cognition et signifcation. Presses
universitaires de Nancy.
(2006). Formal properties of a subset of discourse markers: connectives. In K. Fischer (ed.):
299314.
Roulet, E. (2006). Te description of text relation markers in the Geneva modal of discourse organi-
zation. In K. Fischer (ed.): 115131.
(ed.) (1983). Connecteurs pragmatiques et structure du discours. Actes du 2e Colloque de Prag-
matique de Genve. Cahiers de linguistique franaise 5. Universit de Genve.
Roulet, E., I. Filliettaz & A. Grobet (2001). Un modle et un instrument danalyse de lorganisation du
discours. Peter Lang.
Rhlemann, C. (2007). Conversation in context. A corpus-driven approach. Continuum.
Saari, M. (1984). Ngra pragmatiska partiklar i svenskt talsprk. In L. Huldn (ed.) Festskrif till ke
Granlund 28.4.1984: 213221. Svenska litteratursllskapet i Finland.
Scheglof, E.A. & H. Sacks (1973). Opening up closings. Semiotica 8: 289327.
Schifrin, D. (1987). Discourse markers. Cambridge University Press.
Schourup, L.C. (1985). Common discourse particles in English conversation. Garland.
(2001). Rethinking well. Journal of Pragmatics 33(7): 10251060.
Schwenter, S. (2001). Additive particles and the construction of context. In H. Ferrer & S. Pons Bordera
(eds.): 245262.
Schwenter, S. & E.C. Traugott (2000). Invoking scalarity: the development of in fact. Journal of His-
torical Pragmatics 1(1): 725.
Sebba, M. & S. Tate (1986). You know what I mean? Agreement marking in British black English.
Shloush, S. (1998). A unifed account of Hebrew bekicur in short: Relevance theory and discourse
structure considerations. In A. Jucker & Y. Ziv (eds.): 6182.
Simon-Vandenbergen, A.-M. (1998). I think and its Dutch equivalents in parliamentary debates. In
S. Johansson & S. Oksefell (eds.) Corpora and crosslinguistic research: theory, method and case
studies: 297317. Rodopi.
(2000). Te functions of I think in political discourse. International Journal of Applied Linguis-
tics 10(1): 4163.
(2002). I thinkA marker of middle class discourse? In E. Krkinen, J. Haines, &
T. Lauttamus (eds.) Studia Linguistica et Litteraria Septentrionalia. Studies presented to Heikki
Nyyssnen: 93106. Oulu University Press.
Simon-Vandenbergen, A.-M. & K. Aijmer (2002/2003). Te expectation marker of course. Languages
in Contrast 41(1): 1343.
Simon-Vandenbergen, A.-M. & D. Willems (2011). Cross-linguistic data as evidence for grammati-
calization. In L. Degand and A.-M. Simon-Vandenbergen (eds.): 333364.
Smith, S.W. & A. Jucker (2000). Actually and other markers of an apparent discrepancy between prop-
ositional attitudes of conversational partners. In G. Andersen & T. Fretheim (eds.): 207237.
Sorjonen, M.-L. (1997). Recipient activities. Particles nii(n) and joo as responses in Finnish conversa-
tion. Ph.D. dissertation. University of California.
Sperber, D. & D. Wilson [1986]. (1995). Relevance. Communication and cognition. Blackwell.
Stenstrm, A.-B. (1990a). Pauses in monologue and dialogue. In J. Svartvik (ed.) Te London-Lund
Corpus of Spoken English. Description and research: 211252. Lund University Press.
(1990b). Lexical items peculiar to spoken language. In J. Svartvik (ed.) Te London-Lund Cor-
pus of Spoken English. Description and research: 137175. Lund University Press.
(2006). Te Spanish discourse markers o sea and pues and their English correspondences. In
Stenstrm, A.-B. & G. Andersen (1996). More trends in teenage talk: a corpus-based investigation
of the discourse items cos and innit. In C.E. Percy, C.F. Meyer & I. Lancashire (eds.) Synchronic
corpus linguistics: 177190. Rodopi.
Svartvik, J. (1980). Well in conversation. In S. Greenbaum, G. Leech, J. Svartvik & V. Adams (eds.)
Studies in English linguistics for Randolph Quirk: 167177. Longman.
Tagliamonte, S. & R. Hudson (1999). Be like et al. beyond America: the quotative system in British
and Canadian youth. Journal of Sociolinguistics 3(2): 147172.
Takahara, P.O. (1998). Pragmatic functions of the English discourse marker anyway and its corre-
sponding contrastive Japanese discourse markers. In A. Jucker and Y. Ziv (eds.): 327351.
Traugott, E.C. (1995). Te role of the development of discourse markers in a theory of grammatical-
ization. Paper presented at ICHL XII, Manchester 1995.
Traugott, E.C. & R.B. Dasher (2002). Regularity in semantic change. Cambridge University Press.
Travis, C.E. (2006). Te Natural Semantic Metalanguage approach to discourse markers. In
Tsui, A. (1994). English conversation. Oxford University Press.
Van Bogaert, J. (2009). A reassessment of the syntactic classifcation of pragmatic expressions: Te
positions of you know and I think with special attention to you know as a marker of metalinguis-
tic awareness. In A. Renouf & A. Kehoe (eds.) Corpus linguistics: Refnements and reassessments.
Vask, I. & T. Fretheim (1997). Some central pragmatic functions of the Norwegian particles alts
and nemelig. In Swan, T. & O. Jansen Westvik (eds.) Modality in Germanic lanaguages: Historical
and comparative perspectives: 233292. Mouton de Gruyter.
Vincent, D. (1993). Les ponctuants de la langue et autres mots du discours. Nuit blanche.
(2005). Te journey of non-standard discourse markers in Quebec French: Networks based
on exemplifcation. Journal of Historical Pragmatics 6(2): 188210.
Watts, R. (1988). A relevance-theoretic approach to commentary pragmatic markers: Te case of
actually, really and basically. Acta linguistica Hungarica 38(1/4): 23560.
(1989). Taking the pitcher to the well. Journal of Pragmatics 18(2): 203231.
Westheide, H. (1985). Eine kontrastive Analyse der Partikeln Dt. wohl und Nl. wel. Zeitschrif fr
germanistische Linguistik 13: 186202.
Weydt, H. (1969). Abtnunugspartikel. Die deutschen Modalwrter und ihre Franzsichen Entsprec-
hungen. Gehlen.
White, P. (2003). Beyond modality and hedging: a dialogic view of the language of intersubjective
stance. Text 23(2): 259284.
Wichmann, A., A.-M. Simon-Vandenbergen & K. Aijmer (2010). How prosody refects semantic
change: a synchronic case study of of course. In H. Cuyckens, K. Davidse & L. Vandelanotte (eds.)
Subjectifcation, Intersubjectifcation and Grammaticalization: 103154. Mouton de Gruyter.
Wierzbicka, A. (1976). Particles and linguistic relativity. International Review of Slavic Linguistics
1(2/3): 327367.
(1994). Cultural scripts: A new approach to the study of cross-cultural communication. In
M. Ptz (ed.) Language contact and language confict: 6987. John Benjamins.
Willems, D. & A. Demol (2006). Vraiment and really in contrast. When truth and reality meet. In
Ziv, Y. (1998). Hebrew kaze as discourse marker and lexical hedge: Conceptual and procedural prop-
erties. In A. Jucker & Y. Ziv (eds.): 203221.
Public discourse
Srikant Sarangi
Cardif University
1. Introduction
Te title of this paper public discourse is inherently ambivalent, given the multiple
use of both terms in scholarly work across disciplines. Of the two, the term public
is perhaps more problematic because of its everyday use (e.g. public notice, public
lecture, public interest, going public). In light of this, my discussion here will mainly
centre around the notion of publicness. To begin with, I defne public discourse as
social processes of talk and text in the public domain which have institutionally rati-
fed consequences. For instance, an interrogation of an airline passenger by a cus-
toms ofcer in a secluded space is institutionally grounded and would thus count as
public discourse. Viewed from this angle, public discourse is diferent from vari-
ous versions of the private realm (e.g. family, inter-personal, inner-self etc.). Broadly
speaking, the term discourse can be taken to mean a stretch of talk or text (including
semiotic icons) as well as a form of knowledge and the social processes of produc-
tion and consumption of such knowledge in the Foucauldian sense. Public discourse,
thus defned, includes what goes under the generic rubric of media discourse and
political communication as well as discourse in organisational and professional set-
tings (for a detailed discussion of the scope of public discourse in communicative
terms, see Scollon 1997). However, given the limited scope of this paper, I will link
my arguments to media communication where necessary, and only, in passing, touch
upon the communicative practices in the organisational and professional domain
(see e.g. Alatis & Tucker 1979; Atkinson 1995; Atkinson & Drew 1979; Boden 1994;
Clegg 1990; Labov & Fanshel 1977; Fisher & Todd 1983, 1986; Mishler 1984; OBarr
1982, di Pietro 1982; Silverman 1987).
Te rest of my argument is structured as follows: frst, I spell out in detail various
readings of the term public and the shifing nature of its boundaries. In Section 2,
I draw attention to the situation-talk dialectics in communication-oriented public
discourse studies, followed by an outline of two dominant frameworks Gofman
and Habermas concerned with the notions of public behaviour and public sphere
in diferent ways (Sections 3 and 4). In Section 5, I ofer a brief overview of the shif-
ing nature of boundary marking between private and public domains with relation to
the positioning of the state, pointing out particularly the leakages between these two
types of discourse in the contemporary society (as evidenced in the tradition of media
Public discourse 249
and cultural studies). In Section 6, I focus on the mechanism of information exchange
in the public domain, which is particularly relevant to pragmatics research. I briefy
refer to the dominant models of pragmatics of information exchange, which then
leads me to suggest that a model of social pragmatics (which shares an analytic trend
with critical discourse analysis) can be a useful way of understanding the dynamic
communicative practices in the public domain.
1.1 Multiple readings of publicness
As a way of steering a thematic focus, let me try and characterise the notion of public
by suggesting at least three possible juxtapositions, which are intricately intertwined:
(i) private/public domains; (ii) expert/lay knowledge systems and (iii) micro/macro
social order.
Te private/public dichotomy perhaps better conceptualised as a continuum is
a good starting point. It is worth pointing out that the Latin term publicus derives from
pubes (meaning mature, adult as in puberty) rather than populus (meaning people),
although public as people is reinforced when it is contrasted with the Latin word
privatus which denotes individual afairs separate from the state. Generally speaking,
all human conduct can be located along such a continuum, while recognising the fact
that the boundaries between what constitutes private, public or semi-public (to use
one of Gofmans terms) are being constantly redrawn in any society. We can adopt a
comparative perspective on the changing shape of the public domain vis--vis the
relationship between state and society in diferent periods in history. In the classical
and medieval periods, theoretically speaking, the society (i.e. the public as opposed
to the private) and the state were inseparable. Tis absence of a state-society divide
is also true of the Marxist tradition. State ownership of capital and labour amounts to
public ownership: a case of the state being absorbed into society. In the post-industrial
age, however, the state (as an impersonal locus of authority) is increasingly being seen
as separate from the public.
A second reading of the public (as in public discourse) bears on the juxtaposition
of expert/lay knowledge systems (e.g. public understanding of science, risk etc.).
In his study of environmental discourse in risk societies, Beck (1992) distinguishes
expert perceptions (laboratory knowledge, scientifc rationality) from lay perceptions
(non-knowledge, social rationality). He writes: Te division of the world between
experts and non-experts also contains an image of the public sphere. Te irrational-
ity of deviating public risk perception lies in the fact that, in the eyes of the tech-
nological elite, the majority of the public still behaves like engineering students in
their frst semester (1992: 578). According to Beck, science determines risks
whereas the public perceives risks. In a similar vein, Stehr (1994: 191), among others,
characterises the rise of the technical state in the post-industrial society as the out-
come of the dominance and authority of science and technology and, therefore, the
250 Srikant Sarangi
power of experts silencing all other voices and purposes, other than those given by
technical rationality itself . He refers to this as a process of scintillating, which has
immediate consequences for social action, such as the need for knowledgeable citi-
zens, with a high level of institutional literacy. In his later work, Beck (1995) juxtaposes
expert opinion with counter- opinion. In doing so, he not only moves beyond the para-
digm of public awareness, but in fact challenges the very basis of scientifc credibility
and technological rationality.
A third reading, although diferent in nature, involves the juxtaposition of the
microcosmic and macrocosmic social order within sociology: while the micro deals
with social action and agency in the local, interactional sense, the macro is geared
towards social structure. It may be tempting to equate the micro-macro distinction with
the private-public dichotomy discussed earlier, but such an attempt is unlikely to be
productive (for a discussion of the micro-macro problematic see Alexander et al. 1987;
Knorr-Cetina & Cicourel 1981). Te micro-interactional turn in sociology, beginning
with the ethnomethodology project of Garfnkel (1967) is in some way to be contrasted
with the macro-social turn in linguistics (the critical linguistics tradition) and such
orientations have had a considerable impact on the conceptualisation and analysis of
public discourse.
2. The situation-talk dialectic: public as a feature
of setting vs. public as a feature of talk
It may be useful to impose a distinction between two types of public discourse studies
in communication-oriented research: one dealing with public talk/discourse in terms
of formal linguistic, stylistic and rhetorical features (in felds as diverse as sociolinguis-
tics, anthropology and politics); the other dealing with interactional strategies of com-
munication (in ethnomethodological, pragmatic and discourse analytic traditions).
2.1 (Socio)linguistic markers of public discourse
Over the last two/three decades scholars in sociolinguistics and discourse analysis
have developed a genre-based approach to study language in the public domain. Let us
look closely at one such study. In a rather formalist sense, Labov & Labov (1986) pro-
pose a contrast between public discourse and private/everyday discourse. Tey argue
that public language is characterised by the restriction of the use of the universal
quantifers to strict interpretation, the fxed use of non-restrictive attributes, and the
frequent use of numbers and percentages (p.243). According to them, this particular
style of public discourse is based on logical argument and reasoning, and is governed
by strict rules of evidence and inference (p. 231). Te inconsistent use of linguistic
forms in public discourse, they maintain, is regulated at the broadest level by moral
predication, in this case related to the fundamental principles of social order (p.243)
(there is a confation here however of public/private and spoken/written forms with
regard to the spontaneous and non-spontaneous nature of discourse). Tey conclude
that a great deal of public discourse is oriented towards persuasion, and moving an
audience to action (p.225), a point which I return to in our discussion of the public
sphere in the work of Habermas (Section 4).
2.2 Interaction-based approach
Studies such as Labov & Labov (1986) can be contrasted with interactional stud-
ies of public discourse, especially in the media. A number of scholars in the
conversation-analytic tradition have taken up this line of analysis, focusing on a
wide range of issues: rhetorical strategies such as contrast and listing; elicitation of
applause, specifc turn-taking mechanisms, topic shifs, targeting talk at over-hear-
ing audience etc. (Atkinson 1984; Atkinson & Heritage 1984; Clayman 1988, 1992;
Greatbatch 1988, 1992; Greatbatch & Heritage [in press]; Heritage 1985; Heritage &
Greatbatch 1986, 1991; Hutchby 1996).
It is worth pointing out that media discourse has also attracted scholars from
discourse-analytic and pragmatic traditions. Tere are, for instance, studies into
news discourse dealing with socio-cognitive processes, ideology, power and rela-
tional aspects of identity constitution in the critical linguistic and semiotic tradi-
tions (e.g. Chilton 1985; van Dijk 1988a, 1988b, 1988c; Fowler 1991; Fairclough
1989, 1993, 1995; Hartley 1982; Kress 1983, 1986; Tolson 1996); audience design
and news production in the tradition of sociolinguistics of style (e.g. Bell 1986). A
pragmatic orientation raises, among other things, strategies of evasion (e.g. Harris
1991); the activity structure of news interviews (e.g. Jucker 1986); the use of impli-
cature and presupposition (e.g. Wilson 1990); and the shifing roles of participants
(Tomas forthcoming).
It becomes apparent that the language form/situation distinction suggested
above is not a mutually exclusive one since the kind of public talk Atkinson (1984)
and Labov & Labov (1986) analyse does take place in public settings political or
otherwise. Tis distinction foregrounds the linguistic and situational aspects of pub-
lic discourse: public counts as (i) a feature of talk/discourse (e.g. public documents
which are a genre in their own right); and as (ii) a characterisation of setting/situation
(e.g. a public demonstration by a campaign group or simply a gathering of people). For
our purposes, the discourse which takes place in the public in the Gofmanian sense
(see Section 3) can be distinguished from the discourse of public consequence in
the Habermasian sense (see Section 4). In both senses, public discourse is inherently
constitutive of social identity (Scollon 1997).
252 Srikant Sarangi
3. Gofman and the public order
Following Gofman, it is useful to think of interaction in public settings in relational
terms as a legitimate feld of study. Gofmans interest in social situation or what
he (1964) refers to in his essay titled the neglected situation lies at the root of the
interactional or micro-situational turn in sociology, later developed in the ethnometh-
odological and the conversation analytic tradition. Social situations are defned as an
environment of mutual monitoring possibilities with turns at talking as the unit of
analysis rather than social structure. Gofman points out: rules of conduct in streets,
parks, restaurants, theatres, shops, dance foors, meeting halls, and other gathering
places of any community tell us a great deal about its most difuse forms of social
behaviour (1963: 34).
Tis brings about the issue of gatherings (in the sense of focused encounters)
and aspects of co-mingling, territoriality and orientations in public life along two
dimensions: constraints and social organisation (Gofman 1963, 1971). Gofman is
concerned with the study of social organisation (including total institutions) via the
interactional features within specifc settings. Te interaction order of everyday talk
can form the basis for an understanding of behaviour in institutional settings. In Asy-
lums (Gofman 1961), for instance, rather than studying deviance, he focuses on the
underlying patterns of normal behaviour. Conversation is always a part of a larger
frame of social interaction, and hence talk has to be analysed from the outside in. For
Gofman, social interaction includes all aspects of verbal and non-verbal cues, dress
code, limb discipline, situational improprieties (e.g. in terms of tightness and looseness
of involvement) in behaviour in public and semi-public places.
Te primacy of social interaction is a foregone conclusion in Gofmans
framework: even self-talk (e.g. outcries, mutterings) implies a certain kind of sociali-
sation and a social audience (e.g. overhearers, bystanders). Gofman claims that
when in the presence of others, the individual is guided by a special set of rules,
which have here been called situational proprieties. Upon examination, these rules
prove to govern the allocation of the individuals involvement within the situation,
as expressed through a conventionalized idiom of behavioural cues (1963: 243).
His work builds on etiquette manuals such as Te Laws of Etiquette by A Gentle-
man (1836) (see also Schlesinger 1946). All gatherings have a concern for ftting in
and maintaining interactional tonus or what Bateson (1936) calls tone of appro-
priate behaviour. Gofman draws on game metaphors to characterise the coordinat-
ing aspects of interactants: come into play, stay in play. He (1959) also extends the
dramaturgical metaphors of backstage and frontstage to illustrate the discrepancy
between public and private presentations of self . Rules of social interaction are
thus intimately linked with self-presentation and face-management. For Gofman
(1955: 215), an individuals face is the positive social value which is on loan from
society and it will be withdrawn unless he conducts himself in a way that is worthy of
it. In sum, self-hood and face-management are inevitably intersubjective processes
in a given socio-cultural space (for a reassessment of Brown and Levinsons model of
face-wants in public discourse settings, see Sarangi & Slembrouck 1997).
From a pragmatic angle, the notion of participant roles is useful for analysing inter-
action in public life organisational and bureaucratic settings (Sarangi & Slembrouck
1996; see Tomas [forthcoming] for a distinction between social roles, discourse roles
and activity roles). A role-analytic perspective can easily link up with discourse repre-
sentation, indeterminacy, responsibility and authority in public settings.
At a more specifc level, Gofman indeed talks about participant rights and
obligations which may be diferentially distributed within an encounter (1963: 100).
For example, children at a dinner table, minimal responses restricted to Yes, sir and
No, sir for certain participant categories; denial of speaking rights in public meet-
ings and stage performances. Levinson (1988) recasts the Gofmanian (1979) notion
of footing in terms of a complex set of participant roles at the level of both pro-
duction and reception source, author, relayer, audience, overhearer, bystander etc.
For him, the dyadic bias and the canonical characterisation of speaker and hearer in
pragmatics research does not pay adequate attention to informational and attitudinal
content. Taking utterance-event, as opposed to the speech event, as his unit of analy-
sis, Levinson emphasizes the need for what he calls the incumbancy/role distinction
in talk analysis. He ofers useful insights about linkages between participant roles on
the one hand and grammatical categories, especially personal deixis, social deixis and
reported speech, on the other hand.
4. Habermas and the public sphere
Tere are striking diferences between Gofman and Habermas with regard to the
conceptualisation of publicness. Habermass (1989) account, unlike Gofmans, is
about the constitution and degeneration of the public sphere, as an ideal type. Tere
seems to be a parallel between Habermass notion of the ideal speech situation and
his conceptualisation of changing nature of the public sphere at a certain historical
point. Habermas focuses on the rise of the bourgeois sphere in the western world, in
particular, Germany, France and Britain in the eighteenth and nineteenth centuries.
Originating in literary circles, the public sphere is seen as being increasingly politicised
during the French revolution.
In what is referred to as his linguistic turn, Habermas (1984) proposes a theory
of communicative action as a form of coordinated social interaction oriented toward
understanding (Verstndigung) and consensus (Einverstndnis). Tis theory of com-
municative action is intermeshed with his framework of the lifeworld (e.g. individual,
254 Srikant Sarangi
family, peer group) and the systems (e.g. state, market). For Habermas, communica-
tive action (as distinct from strategic and instrumental action) is paramount to the life-
world and its reproduction. In other words, the notion of the lifeworld (which means
implicit assumptions, intuitive knowledge, moral values etc.) is the horizon-forming
context of communication as it makes understanding happen.
According to Habermas, the private and the public constitute two spheres in the
lifeworld. Te public sphere (in its political and cultural sense) is thus an extension of
the lifeworld, or more appropriately, it functions as mediator between state, market
and individual, that is, systems and lifeworld (Forns 1995: 80). In the Habermasian
sense, the public sphere comes to be seen as distinct from organisational and bureau-
cratic state institutions. Te institutional spheres or felds are, however, dependent
on the ordering principles of both systems, but also on communicative action based
in the lifeworld (Forns 1995: 74). In passing, it can be noted that Gramsci (1971),
however, draws the line between private and public elsewhere: institutions such as
churches, schools, newspapers are seen as part of the private domain and the State
(of the ruling class) is neither public nor private. In his seminal work on ideology
and ideological state apparatuses, Althusser (1994) maintains that it is unimportant
whether the institutions in which ideological state apparatuses are realised are public
or private; what matters is how they function (see Section 5.2 on the colonisation of the
private domain through the adoption of surveillance procedures).
Staying within the political feld, Habermas defnes the public sphere as a body of
private persons assembled to form a public to discuss matters of public concern or
common interest (cf. Sennett 1977). Tere are resonances here with the Gofmanian
notion of gathering, but the Habermasian project is strikingly diferent with its focus
on rational argumentation as opposed to interactional order. Te notion of com-
mon interest allows for the bracketing of participants status, as the public sphere is
regarded as a forum for public opinion formation. For Habermas (1989: 36), public
discourse as a kind of social intercourse, [] far from presupposing the equality
of status, disregarded status altogether. As Calhoun (1992: 13) points out, the idea
that the best rational argument and not the identity of the speaker was supposed to
carry the day was institutionalised as an available claim. Te Gofmanian participant
categories are hardly invoked here, but we can anticipate a link between available
role-categories and production/reception formats in the public domain. According
to Habermas (1989: 27):
Te bourgeois public sphere may be conceived above all as the sphere of private people
come together as a public: they soon claimed the public sphere regulated from above
against the public authorities themselves, to engage them in a debate over the general
rules governing relations in the basically privatized but publicly relevant sphere of
commodity exchange and social labor. Te medium of this political confrontation was
peculiar and without historical precedent: peoples public use of their reason.
Tis sense of publicness brings to the fore an opposition between state and society on
the one hand and the practice of rational critical discourse on political matters, on the
other hand (see Calhoun 1992 for an insightful account).
As we have seen, Habermass framework idealises the dimensions of rationality
and intentionality as forms of human action. One reading of Habermass rational
critical discourse suggests that all sections of the public have equal access to this
form of expression and that it is not afected by the power/status of the speakers.
His subsequent observation on validity and truth claims also builds on the ratio-
nality dimension and has little to say about authority and responsibility claims,
thus keeping knowledge and power separate, unlike in the Foucauldian framework
(cf. Foucault 1980). Note, however, that in his later work on the speech-act model
of pragmatics, Habermas pays attention to the felicity conditions. It is also worth-
while to remind us of the language-ideology dimension since the public domain
is also a perceived phenomenon on which people project a number of ideological
beliefs and ideas. Such perceptions generate speech style choices, including ways
of speaking (e.g. going on record). Te public domain is invested with ideologies
which have to do with textual hierarchies (e.g. written is more formal than spoken),
voices of authority, entextualisation practices and replicability (Silverstein &
Urban 1996).
Bourdieus (1991) notion of legitimate language in relation to position of
speaking takes us to the strategic and symbolic aspects of collusion and hegemony.
Consider, for instance, how certain modes of action and talk (habitus), which are
fostered in the private domain of the middle and upper classes, become recognised
as having symbolic capital in the public domain. Children in school are disposed to
learn the know-how of good behaviour, including speaking properly. As Althusser
(1994: 104) observes: the school (but also other State institutions like the Church,
or other apparatuses like the Army) teaches know-how, but in forms which ensure
subjection to the ruling ideology or the mastery of its practice (cf. Bourdieu 1977).
Te reproduction of norms, inequality and power in and through discursive prac-
tices can be captured in a model of duality (see Giddens [1984] notion of duality
of structure with an emphasis on rules, which is similar to Bourdieus notion of
double structuration but with an emphasis on struggles and strategies). Te ideo-
logical, rather than the intentional, dimension of language use becomes a key aspect
of public discourse.
Notice that the public debates Habermas refers to were carried out in the comfort of
private homes initially, thus marring the distinction between the public and the private
and further, excluding, in the Foucauldian sense, other groups from public participa-
tion. Tis means then that public is no longer an open space accessible to all people, as
there are self-imposed boundaries. According to Fraser (1989, 1992), the Habermasian
notion of the public is narrow as it only accommodates the male-dominated bourgeois,
256 Srikant Sarangi
thus privileging the gendered nature of social structure and social action. She argues for
the notion of the public to be pluralised to include other marginalised groups in terms
of gender, class etc. and lists the various ways in which public and private can be
conceptualised:
Public, for example, can mean (1) state-related, (2) accessible to everyone, (3) of
concern to everyone, and (4) pertaining to a common good or shared interest. Each
of these corresponds to a contrasting sense of the private. In addition there are two
other senses of private hovering just below the surface here: (5) pertaining to private
property in a market economy and (6) pertaining to intimate domestic or personal
life, including sexual life. (Fraser 1992: 128)
Tese terms [private and public], afer all, are not simply straightforward
designations of societal spheres; they are cultural classifcations and rhetorical levels.
In political discourse they are powerful terms frequently deployed to delegitimate
some interests, views, and topics and to valorize others. (Fraser 1992: 131)
In addition to the undiferentiated characterisation of publicness, Habermas public
sphere has several shortcomings (see Tompson 1993 for a critique). As we have seen,
there is a tendency to focus on the political domain, and as Ricur (1986: 298) rightly
points out, there is an overestimation of the power of persuasion by discussion the
assertion that fnally the extension of discussion will be enough to change things. Such
a view pays little attention not only to counter-cultures and institutionalised forms of
resistance, but also to media communication and cultural practices in everyday life.
5. Transformation of the public sphere: Public discourse
as mediated communication
For Habermas (1989: 1712), the world fashioned by the mass media is a public
sphere in appearance only the sphere generated by the mass media has taken on
the traits of a secondary realm of intimacy. A further transformation of the public
sphere coincides with the rise of popular culture through mass media (Williams
1961). Te media is gradually becoming the public sphere, in particular with the
growth of political debates, chat shows (comprising expert and lay perspectives on
issues of public interest).
In a media framework of the public sphere, the critical-rational debate has
been substituted by emotional appeal to bring about changes in government policy.
Negrine (1996), for instance, illustrates this with an extensive analysis of the miners
strike in Britain (the same also holds for the medias portrayal of the human sufering
in the case of Zaire and Rwanda to mobilise international political action and, on a
diferent scale, the Snowdrop campaign for handgun ban following the Dunblane
shooting in Scotland). Social constructivists would argue that social life is constituted
in the public domain via exposure to media and thus media no longer simply repre-
sents social values and norms. With regard to information exchange, the media is
taking on the role of educating the public or what Negrine (1996) calls the creation
of an informed citizenry rather than bringing about a rational debate on issues of
public concern. Tis suggests that the media, with its own socio-political agenda,
is not a substitute for the public sphere in the Habermasian sense. Public rights
have been cast as individual needs and rational-critical debate has been replaced by
consumption of culture.
Te consumption of cultural identities in late modern society remains a key aspect
of the public sphere (Featherstone 1994, Gabriel & Lang 1995, Keat & Abercrombie
1991). Advertisements, handbills etc. continue to refect the cultural values of promo-
tional culture (Wernick 1991; Nava et al. 1997), as simulation rather than representa-
tion is regarded as central to post-modern existence (Baudrillard 1988). Tis can be
seen in chat shows where the emphasis is very much on accounting for experience of
x (e.g. being raped, being terminally ill) rather than on a public debate of what caused
such a state-of-afair. We notice how private emotions and feelings are entering the
public sphere, especially through the media institution. What was once secured in
the private domain is increasingly being staged in the public (cf. various chat shows
and political interviews, see Scannell [1991], in particular Tolson [1991], on broadcast
talk as informal and conversational, which is to be seen as institutionalised variants
of casual conversation). In other words, what we have here is a public celebration
of private experience, in a rather non-political sense, which amounts to denigrating
expertise, critical rationality etc. It should be pointed out however that such public
displays of private experience can be interpreted as an appropriation of expert systems,
thus contributing to new discourse technologies (Fairclough 1992).
Te transformation of the critical-rational debate is partly a consequence
of the democratisation process, ofering people more open access to the pub-
lic space. Underlying this is what Tompson (1990) refers to as mediatization.
Te mediatized public sphere is a convincing illustration of the transformation in
the Habermasian sense: massifcation in favour of popular culture. Te public sphere
has very much become the popular culture (in the Bakhtinian sense of the carni-
valesque), with tendencies of globalisation and regionalisation and calls for cultural
citizenships. In contemporary terms, this has led to the emergence of sub-political cam-
paign groups (Greenpeace, Anti-abortion, Friends of the Earth, Animal Rights etc.).
Tese groups form political opinions through mobilising expert and lay perspectives
on public issues and their main strength comes from globalisation tendencies no
longer constrained by the political climate of any particular nation-state or what
258 Srikant Sarangi
Habermas characterises as the states uncommitted friendly disposition. Increasingly,
we notice a convergence of rhetorical strategies as the state, the politicians, the pres-
sure groups and the media claim to be speaking on behalf of the public.
5.1 The states role in the confation of public and private
discourses in contemporary societies
Te consumption model of public discourse is a consequence of the arrival of the
televisual media, but it is also the result of a shif from a political to an economic model
of states relationship with its citizens. Te culmination of this shif is noticeable in the
privatisation of public utilities in welfare states. Te public goods and public services
are being replaced and transformed by privatisation programmes. Tis transformation
has as much to do with the failing of the public services, as it has to do with the need
for new markets for large multinational companies (Martin 1993). Te state is taking
on a regulatory role to monitor and facilitate the performance of the private sector.
In other words, the state functions as a mediator and this has implications for mecha-
nism of information exchange, accountability and sanctions.
We notice increasing privatisation of public sectors and growth of the enterprise
culture (Keat & Abercrombie 1991), with a high premium placed on accountability.
Te private/public distinction gets blurred in the model of the free market economy,
as can be seen in the competition between state and private sector to intervene with
private lives as well as with the extension of services of institutions such as banks in
areas of savings, loans, insurance packages and pension schemes all rolled into one.
Citizens are being treated as consumers and the state is seeking to fulfl consumer
needs/rights (Keat, Whiteley & Abercrombie 1994; du Gay 1996). Te development
of the contemporary consumer culture involves a distinctive mode of domination:
the new mode of domination distinguishes itself by the substitution of seduction for
repression, public relations for policing, advertising for authority, and needs-creation
for norm-imposition (du Gay 1996: 79; see also Bauman 1987; Baudrillard 1988).
Te creation of what Negrine (1996) calls next step agencies (Benefts Agency,
Child Support Agency etc.), which fall outside the domain of public accountability
as do state government departments, also marks a shif from a model of public sphere
(characterised by collective rational debate) to a model of public relations (character-
ised by the proliferation of the service industry dealing with information exchange
based on individual needs and eligibility criteria). Among other things, this is realised
discoursally in what Fairclough (1992) refers to as conversationalisation the sim-
ulation of equality and informality in the public sector. As Sarangi & Slembrouck
(1996) point out, on the one hand conversationalisation can be seen as a reaction to
bureaucratisation of public life, but on the other hand, we should not overlook how
institutional power in the bureaucratic sphere is being sustained through methods of
surveillance, of which conversationalisation is an intimate part. A further trend here is
the prevalence of promotional discourse in all aspects of social life, including the uni-
versity sector (Fairclough 1993; Sarangi & Slembrouck 1996; Wernick 1991).
Te transformation of citizens into consumers in the bureaucratic feld can be
analysed in terms of what Habermas refers to as the colonisation of the lifeworld.
Te logic of the economic market and administrative power extends to the private
sphere. Communicative action is replaced by strategic action in order to achieve ends,
which undermines rational consensus and validity claims. It is perhaps the case that the
systems (state and market) continue to colonise the lifeworld, but within the lifeworld
there is a confation of private-public spheres (see Crossley 1996 and Forns 1995 for a
useful discussion of the mutual exchange fows between the spheres within a communi-
cative framework). It is important to keep the distinction between, on the one hand, the
colonisation of the lifeworld by the system (which gives rise to distorted communica-
tion) and, on the other hand, the fuzzy border between public and private spheres as they
mutually infltrate each other. At the level of information exchange, public and private
discourses continue to draw upon each other and the hybridity is as much noticeable in
the political domain as in the cultural domain of the public sphere.
5.2 Surveillance and control: Information exchange as a site of struggle
Te state ofen redefnes the public/private domains with the use of surveillance
and discipline as methods of social control. Consider the public/private dimen-
sion of social life in relation to the state, the individual self and the family as an
organisational unit. With regard to the shifing nature of parentchildren relation-
ship vis--vis state intervention, we can examine the setting up of bureaucratic
institutions such as the Child Support Agency in the UK to monitor failed fam-
ily lives and to make absent fathers pay for their childrens well-being. A further
example of state intervention in family life relates to the issue of parents power to
use corporal punishment in order to monitor their childrens behaviour. According
to Weber (1922), the surveillance capacity of the state is such that the distinction
between the public and the private is itself a matter of state defnition (cf. Habermass
[1973] account of how this is accomplished through the extension of the Law to
regulate the private domain).
As Giddens (1985) points out, heightened surveillance is a prerequisite for social
organisation and social control. Following Foucault (1979), it can be argued that the
state and other private organisations use panoptical methods of surveillance to control
information. As Webster (1995: 54) puts it: organisation and observation are Siamese
twins, which have grown together with the development of the modern world (for a
fuller discussion of surveillance as a key technique of modern power both in state and
business organisations, see Dandeker 1990; Giddens 1985). In the transformed public
260 Srikant Sarangi
sphere with its locus on consumer culture, the market adopts surveillance as a mode of
regulation and encourages forms of auto-surveillance and self-discipline.
Tere seems to be a constant tension between the state and the market wanting
to obtain as much information as possible in order to be able to control social
action (see Gandy [1993] for a discussion of the relationship between informa-
tion and power). Consider, for instance, the current trend in the UK of super-
markets obtaining and storing information about households for the purposes of
issuing reward/bonus cards in itself a marketing strategy with some economic
advantage to the loyal customers. It is also possible to argue that such information
exchange takes place in a given socio-political context, and therefore institutionally
literate individuals would deploy fltering strategies when it comes to providing
information about themselves. In other words, social agents as clients may choose
to withhold vital information in pursuit of their own interest. Tis might mean
being economical with truth and constructing a productive other (Barsky 1994). It
would appear therefore that asymmetric information exchange is central to both the
cultural and political domains of the public sphere.
6. Pragmatic theories of information exchange and the public sphere:
Towards a social pragmatics
Given that pragmatics is inevitably concerned with meaning-making against a set of
parameters such as power, status, rights and obligations, various pragmatic theories
of human communication can be reassessed to account for public discourse. In this
section therefore let us briefy examine the descriptive and explanatory adequacy of
various pragmatic theories of information exchange to account for communication in
the public domain (to include state, market and media).
It is safe to argue that information exchange lies at the heart of the three most
dominant models of pragmatic inquiry: the cooperative principle and conventional/
conversational implicatures (Grice 1975), face wants and positive/negative politeness
strategies (Brown & Levinson 1987), and principles of relevance and message process-
ing (Sperber & Wilson 1986). One of the weaknesses of these models is that commu-
nication is viewed as primarily inter-personal and understanding-oriented. It is also
worth pointing out that much of the critique and application of these models of infor-
mation exchange is grounded, for example, in cross-cultural or gender-related difer-
ences. It is important to consider the possibility of extending these pragmatic theories
to the public domain generally, focusing mainly on information exchange vis--vis
issues of role-relationships, identity construction, strategic goal imposition etc. Te
public domain here would include not only face-to-face communicative encounters
between clients/consumers and institutional/organisational representatives of various
kinds, but also the dynamic aspects of inter-institutional communicative networks in
relation to contemporary processes of social change. Such a programme will blend
well with the Habermasian characterisation of public discourse as systematically dis-
torted communication, without subscribing to his universal pragmatics model. In a
further stage, strategic aspects of public discourse can be linked to participant roles,
rights and obligations.
In a series of papers Sarangi & Slembrouck (1992, 1997, 1997, forthcoming)
argue in favour of a social pragmatic model of information exchange in the public
domain. Te point of departure for a model of social pragmatics is that most of
pragmatics theory building is based on an archetypical view of casual conversation
at an interpersonal level. With some exception in terms of illustrative examples, this
is true of Grice (1975), Brown & Levinson (1987) and Sperber & Wilson (1986).
Additionally, a social pragmatic model of information exchange analysis has advan-
tage over a conversational-analytic model of sequential analysis. While the latter
is constrained by practical reasoning, local logic and manifest aspects of account-
ing, the former subscribes to the distinction between strategic and communicative
rationality vis--vis role-categories and authority claims. Note, however, Garfnkels
(1967: 38) disclaimer: a knowledge of how the structures of everyday activities are
routinely produced should permit us to tell how we might proceed for the efective
production of desired disturbances (this is the Habermasian equivalent of distorted
communication).
Future research in information exchange in the public domain, carried out within
a framework of societally-based pragmatics, will supplement the current work within
critical discourse analysis, mainly targeted at power, social struggle, ideological invest-
ments on the one hand and globalisation tendencies on the other hand. Combining
insights from Gofman and Habermas, future pragmatic studies of media can focus on
claims of authority via discourse roles as well as on validity/truth claims (this will be
a strand which is rather diferent from CA-oriented media discourse). Communica-
tion which is not face-to-face and which has multiple audiences requires a pragmatic
framework to account for multiple sources of authorship and audience. A prag-
matic framework grounded in social theory (social structure and agency) will enable
researchers to bring together the linguistic parameters at interactional/relational and
the institutional/rational levels.
References
(Anon.) Te laws of etiquette, by A Gentleman (1836). Philadelphia: Carey, Lee and Blanchard.
Alatis, J. & G. Tucker eds. (1979). Language in public life. Georgetown University Press.
Alexander, J., B. Geisen, R. Munch & N.J. Slemser (eds.) (1987). Te micro-macro link. University of
California Press.
262 Srikant Sarangi
Althusser, L. (1994). Ideology and ideological state apparatuses (notes towards an investigation). In
S. iek (ed.): 100140.
Atkinson, J.M. (1984). Our masters voices: the language and body language of politics. Routledge.
Atkinson, J.M. & P. Drew (1979). Order in court. London: Macmillan.
Atkinson, J.M. & J. Heritage (1984). Structures of social action: studies in conversation analysis. Cam-
bridge University Press.
Atkinson, P. (1995). Medical talk and medical work. Sage.
Barsky, R. (1994). Constructing a productive other: discourse theory and the convention of refugee hear-
ing. John Benjamins.
Bateson, G. (1936). Naven. Cambridge University Press.
Baudrillard, J. (1988). Selected writings. Polity.
Bauman, Z. (1987). Legislators and interpreters. Polity Press.
Beck, U. (1992). Risk society: towards a new modernity. Sage.
(1995). Ecological politics in an age of risk. Polity Press.
Bell, A. (1986). Te language of news media. Blackwell.
Boden, D. (1994). Te business of talk: organisations in action. Polity Press.
Bourdieu, P. (1977). Outline of a theory of practice. Cambridge University Press.
Bourdieu, P. (1991). Language and Symbolic Power. Polity Press.
Brown, P. & S. Levinson (1987). Politeness: some universals in language use. Cambridge University
Press.
Calhoun, C. (ed.) (1992). Habermas and the public sphere. MIT Press.
Chilton, P. (1985). Language and the nuclear arms debate: nukespeak today. Frances Pinter.
Clayman, S. (1988). Displaying neutrality in television news interviews. Social Problems 35/4:
474492.
(1992). Footing in the achievement of neutrality: the case of news interview discourse. In P.
Drew & J. Heritage (eds.): 163198.
Clegg, S. (1990). Organisations: organisation studies in the postmodern world. Sage.
Crossley, N. (1996). Intersubjectivity: the fabric of social becoming. Sage.
Dandeker, C. (1990). Surveillance, power and modernity: bureaucracy and discipline from 1700 to the
present day. Polity.
Drew, P. & J. Heritage (eds.) (1992). Talk at work: interaction in institutional settings. Cambridge
University Press.
Drew, P. & A. Wootton (eds.) (1988). Erving Gofman: exploring the interaction order. Polity Press.
Ensink, T., A. van Essen & T. van der Geest eds. (1986). Discourse analysis and public life. Foris
Publications.
Fairclough, N., N. (1989). Language and power. Longman.
(1992). Discourse and social change. Polity.
(1993). Critical discourse analysis and the marketisation of public discourse: the universities.
Discourse and Society 4/2: 133168.
(1995). Media discourse. Arnold.
Featherstone, M. (1994). Consumer culture and postmodernism. Sage.
Fisher, S., S. & A.D. Todd (eds.) (1983). Te social organisation of doctor-patient communication.
Washington, D.C.: Center for Applied Linguistics.
(eds.) (1986). Discourse and institutional authority: medicine, education and law. Ablex.
Forns, J. (1995). Cultural theory and late modernity. Sage.
Foucault, M. (1979). Discipline and punish: the birth of the prison. Vintage Books.
(1980). Power/knowledge. Harvester.
Fowler, R. (1991). Language in the news: discourse and ideology in the press. Routledge.
Fraser, N. (1989). Unruly practices: power, discourse and gender in contemporary social theory. Polity
Press.
(1992). Rethinking the public sphere: a contribution to the critique of actually existing democ-
racy. In C. Calhoun (ed.): 109142.
Gabriel, Y. & T. Lang (1995). Te unmanageable consumer: contemporary consumption and its frag-
mentation. Sage.
Gandy, O.H. (1993). Te panoptic sort: a political economy of personal information. Westview Press.
Garfnkel, H. (1967). Studies in ethnomethodology. Prentice Hall.
du Gay, P. (1996). Consumption and identity at work. Sage.
Giddens, A. (1984). Te constitution of society: outline of the theory of structuration. Polity.
(1985). Te nation state and violence: volume two of a contemporary critique of historical mate-
rialism. Polity.
Gofman, E. (1955). On facework: an analysis of ritual elements in social interaction. Psychiatry 18,
213231.
(1959). Te presentation of self in everyday life. Doubleday.
(1961). Asylums. Penguin.
(1963). Behaviour in public places. Free Press.
(1964). Te neglected situation. American Anthropologist 66/6: 133136.
(1971). Relations in public: micro studies of the public order. Basic Books.
(1979). Footing. Semiotica 25: 129.
Gramsci, A. (1971). Selections from the prison notebooks: Lawrence & Wishart.
Greatbatch, D. (1988). A turn-taking system for British news interviews. Language in Society 17:
401430.
(1992). On the management of disagreement between news interviewees. In P. Drew &
J. Heritage (eds.): 268301.
Greatbatch, D. & J. Heritage (in press) Te news interview: studies in the history and dynamics of a
social form. Sage.
Grice, H.P. (1975). Logic and conversation. In P. Cole & J.L. Morgan (eds.) Syntax and semantics,
Vol. 3: Speech acts: 4159. Academic Press.
Habermas, J. (1973). Legitimation crisis. Heinemann.
(1984). Teory of communicative action, vol 1: reason and the rationalization of society. Beacon
Press.
(1989). Te structural transformation of the public sphere: an inquiry into a category of
bourgeois society. Polity.
Hartley, J. (1982). Understanding news. Methuen.
Heritage, J. (1985). Analysing news interviews: aspects of the production of talk for an overhearing
audience. In T.A. van Dijk (ed.): 95117.
Heritage, J. & D. Greatbatch (1986). Generating applause: a study of rhetoric and response at party
political conferences. American Journal of Sociology 92: 110157.
(1991). On the institutional character of institutional talk: the case of news interviews. In
D. Boden & D. Zimmerman (eds.) Talk and Social Structure: 123162. Polity Press.
Harris, S. (1991). Evasive action: how politicians respond to questions in political interviews. In
P. Scannell (ed.): 7699.
Hutchby, I. (1996). Confrontation talk: arguments, asymmetries and power on talk radio. Erlbaum.
Jucker, A. (1986). News interviews: a pragmalinguistic analysis. John Benjamins.
Keat, R. & N. Abercrombie (eds.) (1991). Enterprise culture. Routledge.
264 Srikant Sarangi
Keat, R., N. Whiteley & N. Abercrombie (eds.) (1994). Te authority of the consumer. Routledge.
Knorr-Cetina, K. & A.V. Cicourel (1981). Advances in social theory and methodology: toward an inte-
gration of micro- and macro-sociologies. Routledge & Kegan Paul.
Kress, G. (1983). Te politics of newspaper language. International Journal of the Sociology of Lan-
guage 32: 4358.
(1986). Language in the media: the construction of the domains of public and private. Media,
Culture and Society 8: 395419.
Labov, W. & D. Fanshel (1977). Terapeutic discourse: psychotherapy as conversation. Academic Press.
Labov, W. & T. Labov (1986). Public discourse and the problem of social order. In T. Ensink et al.
(eds.): 225243.
Levinson, S.C. (1988). Putting linguistics on a proper footing: explorations in Gofmans concepts of
participation. In P. Drew & A. Wootton (eds.): 161227.
Martin, B. (1993). In the public interest?: privatisation and public sector reform. Zed Books.
Mishler, E. (1984). Te discourse of medicine: dialectics of medical interviews. Ablex.
Nava, M., A. Blake, I. Macrury & B. Richards (eds.) (1997). Buy this book: studies in advertising and
consumption. Routledge.
Negrine, R. (1996). Te communication of politics. Sage.
Obarr, W. (1982). Linguistic evidence: language, power and strategy in the courtroom. Academic Press.
di Pietro, R.J. (ed.) (1982). Linguistics and the professions. Ablex.
Ricur, P. (1986). Lectures on ideology and utopia. Columbia University Press.
Sarangi, S. & S. Slembrouck (1992). Non-cooperation in communication: a reassessment of Gricean
pragmatics. Journal of Pragmatics 17/2: 117154.
(1996). Language, bureaucracy and social control. Longman.
(1997). Cooperative rationality in public discourse: the case of welfare leafets and citizen
charters. In Jean-Pierre van Noppen & Marc Maufort (eds.) Power, cooperation and confict in
English language and literature. Belgian Association of Anglicists in Higher Education.
(1997). Confrontational asymmetries in institutional discourse: a sociopragmatic view of
information exchange and face management. In J. Blommaert & C. Bulcaen (eds.) Political Lin-
guistics. John Benjamins.
(forthcoming) Te discursive construction of the citizen-client as a consumer: welfarisation
in late capitalist society.
Scannell, P. (ed.) (1991). Broadcast Talk. Sage.
Schlesinger, A.M. (1946). Learning how to behave. Te Macmillan Company.
Scollon, R. (1997). Handbills, tissues, and condoms: a site of engagement for the construction of
identity in public discourse. Journal of Sociolinguistics 1/1: 3962.
Sennett, R. (1977). Te fall of public man. Cambridge University Press.
Silverman, D. (1987). Communication and medical practice: social relations in the clinic. Sage.
Silverstein, M. & G. Urban (eds.) (1996). Natural histories of discourse. University of Chicago Press.
Sperber, D. & D. Wilson (1986). Relevance: communication and cognition. Blackwell.
Stehr, N. (1994). Knowledge societies. Sage.
Tomas, J. (forthcoming) Te dynamic pragmatics. Longman.
Tompson, J.B. (1990). Ideology and modern culture: critical social theory in the era of mass commu-
nication. Polity.
(1993). Te theory of the public sphere. Teory, Culture and Society 10/3: 173189.
Tolson, A. (1991). Televised chat and the synthetic personality. In P. Scannell (ed.): 178200.
(1996). Mediations: text and discourse in media studies. Arnold.
van Dijk, T.A. (1988a). News as discourse. Erlbaum.
(1988b). News analysis. Erlbaum.
van Dijk, T.A. (ed.) (1985). Handbook of Discourse Analysis, vol 3. Academic Press.
(ed.) (1988). Discourse and communication: new approaches to the analysis of mass media and
communication. Erlbaum.
Weber, M. (1968 [1922]) Economy and society: an outline of interpretive sociology. Bedminster Press.
Webster, F. (1995). Teories of the information society. Routledge.
Wernick, A. (1991). Promotional culture: advertising, ideology and symbolic expression. Sage.
Williams, R. (1961). Te long revolution. Chatto & Windus.
Wilson, J. (1990). Politically speaking: the pragmatic analysis of political language. Blackwell.
iek, S. (ed.) (1994). Mapping ideology. Verso.
Text and discourse linguistics
Jan-Ola stman & Tuija Virtanen
University of Helsinki & bo Akademi University
1. On terminology
When in the late 1960s syntacticians and semanticists became convinced of the
usefulness of considering stretches longer than individual sentences in their anal-
yses, at least two terms came to be used in parallel fashion: text linguistics and
discourse analysis.
Since many of the European proponents of the drive to go beyond the sentence
also had a keen interest in stylistics, rhetoric, and literature, it was only natural that
their initial focus was indeed directed at concrete texts. In the Anglo-American tradi-
tion, again, the connection of discourse analysis to literature was a loose one if at all
present. As spoken language was very much in focus within this tradition, cognitive
and social aspects of discourse became exceedingly relevant.
In the 1970s, attempts were made to assign slightly diferent meanings to the terms
text and discourse. For instance, discourse was seen as text + situation whence
text = discourse minus situation (cf. e.g. Widdowson 1973).
Present-day text linguistics is just as interested in processual aspects of lan-
guage as discourse analysis. In fact, the interests of the proponents of the two felds
of inquiry overlap greatly. Here we use the phrase text and discourse linguistics
(TDL) as an umbrella-term for all issues that have been dealt with in the linguistic
study of text and discourse. Although our general approach to discourse is to see it
as an all-encompassing (and, admittedly, vague) notion of going beyond the sen-
tence/utterance, for the purpose of delimiting the subject matter of this entry we
have in some respects accepted Levinsons (1983) controversial distinction between
discourse analysis (including text linguistics), and approaches such as conversation
analysis and dialogical analysis, which are closely related to ethnomethodology and
attempt as much as possible to avoid a priori theorizing. Te exposition will, how-
ever, not focus on theoretical models or analytical categories per se, but rather on
important areas of research.
In this overview we focus on the early, formative decades of TDL in order to
provide a background understanding of where it all started and why and how we do
the things we do in TDL research today.
Text and discourse linguistics 267
2. Historical overview
Although discourse studies did not hit mainstream linguistics until fairly late, interest
in longer stretches of verbal material and its relation to particular communicative situ-
ations goes far back. More scholarly interest was typically channeled through stud-
ies in argumentation and rhetoric, so that in the Socratic dialogues, Plato combined
argumentation, dialectics and aesthetics. Oratory and persuasive elements of language
were at the heart of ancient rhetoric. Te interest in rhetoric reached a second peak as
part of the medieval trivium. Although rhetoric has perhaps not been in the center of
academic linguistic attention in modern times, many of its categories and much of its
methodology are still explicitly or implicitly made use of today in, say, newspaper
and propaganda analyses.
Te frst half of the 20th century witnessed the publication of many important
studies which can be seen as laying the foundations for present-day work in text and
discourse linguistics. One very important branch was that of the Russian Formalists.
For instance, Vladimir Propps well-known analyses of Russian wonder tales infu-
enced later views about the structure and main ingredients of any story. Without
this background, it is unlikely that Mikhail Bakhtin could have produced his 1920s
and 1930s studies, which dissociated him from this tradition and which were later
to become so infuential. As early as the mid-19th century, Henri Weil had done his
ground-breaking research on word order, which was much later to become as impor-
tant as the infuence from the Russian Formalists for generations of Prague School
linguists, including Roman Jakobson and Vilm Mathesius.
Bronislaw Malinowskis analyses of the communicative behavior of the Trobriand
Islanders introduced the importance of phatic communion to linguists. J.R. Firth
followed in his teachers footsteps and insisted that the ultimate meaning of a message
only comes about afer it has been placed in its context-of-situation. Firths views were
paralleled in the Ebbinghaus-Bartlett debate in psychology, when F.C. Bartlett showed
how important the knowledge of what today we would call schemata, discourse topics
and cultural background is for accurate memory and understanding of a narrative.
Kenneth Lee Pike and his colleagues at the Summer Institute of Linguistics are
also forerunners in this feld. Tey were simply forced to develop systematic meth-
ods to deal with longer stretches of discourse in order to adequately translate Bible
texts into very diferent kinds of languages. (Cf. Grimes infuential 1975 study, and
Longacre 1976.)
Although Zellig Harris (1952) was among the frst to use the term discourse
analysis, many of his successors within mainstream linguistics in the 1950s and 1960s
continued to focus on formalizations of rules for describing and explaining sentence-
level syntactic phenomena. However, this focus and in particular its implications
created counter-reactions even within transformational-generative grammar itself.
268 Jan-Ola stman & Tuija Virtanen
Especially within generative semantics, scholars started to ask questions like the fol-
lowing: If it takes longer to process passive constructions than active ones, why would
language users need or ever want to use a passive construction? At one stage of theoriz-
ing, an extra box called context was added to the other three (syntax, semantics, and
phonology), with an arrow going from it in the direction towards the syntactic com-
ponent. Sentence-external triggers became part of the jargon.
Te question of how discourse infuences manifestations on the sentential level
(e.g. word order) was one prevalent research perspective. Another problem concerned
the nature of rules for stretches longer than the sentence: some linguists (e.g. van
Dijk 1980) built architectures from below, others (e.g. Labov 1972) from above. (See
also de Beaugrande 1980, and the discussion around de Beaugrande 1982 in Journal
of Pragmatics 6.) Both of these perspectives are still relevant today and continue to
complement each other.
As we get further into the 1970s, text and discourse studies start to catch on all
over the world. Section 3 gives an overview of some of the most important felds of
study from the 1970s to the 1990s. Afer the initial ground-breaking phase in diferent
directions, basically two approaches developed virtually simultaneously: a structural
approach, focusing on texts and discourses (both spoken and written) as products to
be analyzed as such; and a more dynamic approach, mainly interested in discourse as
an interactional process infuenced by various cognitive and social factors. Naturally,
these ways of approaching discourse complement one another, as we can very well
study a product in order to get at the processes assumed to lie behind it.
Te last thirty years constitute a long enough temporal stretch to have witnessed
various approaches at abstract theorizing in the feld. In addition to discussions on
more global perspectives, a host of particular notions and concepts have been defned.
Being an umbrella concept, the notion of TDL also brings with it the risk of becoming
too all-embracing. Tus, diferentiating between discourse analysis, text linguistics,
pragmatics, semiotics, and even linguistics has become a feld of inquiry in itself.
But perhaps more than anything else, TDL has witnessed a staggering amount of
empirical studies during the last decades, including cross-cultural studies, studies of
cognitive processes, of computer-mediated discourse, and of non-verbal communi-
cation, and it has raised important questions about the very ontology of orality and
literacy. Te mid-1980s saw the publication, not only of a number of textbooks on
discourse analysis, but also of the four-volume Handbook of discourse analysis, edited
by Teun van Dijk (1985). New textbooks and monographs in the feld are continu-
ously being published. What follows is a representative list of books dealing explicitly
with diferent aspects of discourse analysis up and until the early 1990s: de Beau-
grande & Dressler (1981), Brown & Yule (1983), Edmondson (1981), Enkvist (1975),
Gumperz (1982), Grimes (1975), Hoey (1983), Longacre (1983), Lundquist (1983),
Sinclair & Coulthard (1975), and Stubbs (1983). In the early 1990s, some of the more
infuential overviews were Schifrin (1994), Kerbrat-Orecchioni (1990/1992/1994),
Renkema (1993), Nunan (1993), Maingueneau (1991), and Cook (1989). Important
journals specifcally devoted to TDL were started at around this time, and they have
continued to be infuential in the feld: Discourse and Society, Discourse Processes, Dis-
course Studies and Text and Talk. New journals that have been started during the last
ffeen years have focussed on specifc perspectives on TDL.
3. Important felds of study
Tis section presents a few important and foundational felds of research within TDL.
Some of these felds have received attention all along, e.g. information structure
and cohesion. Others, like coherence and grounding, have grown in importance as
research has continued to go deeper into the realms of meaning and function. Issues
having to do with discourse types and genres, again, have always been in the back-
ground, but only gradually received systematic attention in linguistics.
3.1 Information structure
One starting point for introducing TDL concepts into linguistics was the need to
explain variations in word order, the use of the active and the passive and other
seemingly synonymous syntactic constructions. One way to approach such variation
is to ask why a text is a better text if it contains a sentence with word order A rather
than word order B. In other words, what factors in the surrounding co-text of the
sentence trigger the use of a particular word order?
Most sentences occur in a larger co-text which infuences their form and,
conversely, the syntactic form of sentences contributes to their textual ft (cf. Enkvist
1981). Syntactic devices which allow the text producer to regulate the fow of given
and new information include passivization, clefing, extraposition and dislocations.
An early interest in this kind of structuring of information (especially in the Prague
School; in the U.S.A. by such prominent linguists as Dwight Bolinger, e.g. 1952, and
gradually also in generative grammars) brought textual aspects into sentence grammar.
Unfortunately, however, the terminology is particularly confusing here. Europeans usu-
ally spoke of themes and rhemes, while, on the other side of the Atlantic, topics and
comments were referred to when dividing a clause or sentence into two information-
structural parts. Another set of terms include, on the one hand, given, old or known
information, and, on the other hand, new or unknown information. Tese may, but
need not, coincide with the notions of theme and rheme. Jan Firbas of the Prague School
suggested (e.g. Firbas 1959) that information structure should not be talked about as
a binary opposition, but as a scale of communicative dynamism. Detailed patterns of
theme progression were developed by Frantiek Dane (e.g. 1974).
One way of distinguishing between the terms was to view theme and rheme
positionally (cf. Halliday 1967, 1985), topic in terms of aboutness, i.e. what a proposition
is about (cf. van Dijk 1977, 1980), and given and new information cognitively, in terms
of the stage of activation assumed by the text producer (cf. Chafe 1976, 1994; Prince
1981; on accessibility, activiation and topicality, see also Lambrecht 1994).
It has been proposed that the basic principle of information structure is that
themes, topics, and given information precede rhemes, comments, and new informa-
tion. (Cf. Section4 on iconicity.) Since this sequence is seen as the default order by
text receivers, it becomes necessary for the text producer to either choose verbs that
allow her/him to use the prototypical word order of her/his language, or to signal
deviation from the basic pattern, for instance through the use of a clef sentence, or
marked focus in speech. In spoken discourse, given information is frequently recov-
erable from the situation, which may allow the speaker to either start with, or even
neglect, all other except the most crucial information (a typical case would be sports
commentaries given under time pressure, cf. Enkvist 1989). Starting with crucial
information frst in an utterance is also an efective way of getting an interlocutors
attention.
Research in this feld has taken two complementary directions. First, there have
been attempts at fnding a close correspondence between the cognitive and the linguis-
tic structures (an early suggestion is Chafe 1980). Secondly, inherent givenness values
have been ascribed to diferent kinds of linguistic elements, e.g. setting up a givenness
hierarchy among NPs (cf. e.g. Givn 19841990; Ariel 1990).
3.2 Cohesion
Many aspects of what is dealt with in the previous section take us into the area
of cohesion. Hence another early interest in TDL concerns the question of how
sentences and in later studies, propositions are explicitly linked together in a text
by diferent kinds of overt ties. Such cohesion devices include repetition of items,
coreference, synonymy, antonymy, hyponymy, comparison, indexical relationships,
ellipsis and substitution, conjunction, and structural iconicity.
Problems of cohesion and textuality were tackled in France by Michel Charolles,
in Germany by Jnos Petf, in Austria by Wolfgang Dressler, in Finland by Nils Erik
Enkvist, and in Japan by Yoshihiko Ikegami. An infuential work in the area of sur-
face cohesion is the sentence-based model of Halliday & Hasan (1976), in which
cohesion devices are grouped into fve diferent categories: reference (endophoric and
exophoric), substitution, ellipsis, conjunction, and lexical cohesion. Hoey (1991) is an
early, detailed study of lexical cohesion.
Elements in a text which have the same referent or application form a cohesive
chain (Halliday & Hasan 1976), which ties together parts in the text, and thus the
text itself into a whole. Such cohesive chains were also viewed in a processual light, as
signals that a particular text/discourse strategy has been chosen by the text producer
to structure the text in view of her/his communicative goal(s). Tus, chains also
facilitate the text receivers task of discourse processing since they help to segment the
text (see Enkvist 1987; Givn 1983; Virtanen 1992a). Te most common discourse
strategies involve temporal, locative, and/or participant/topic-oriented continuity.
With the advance of studies in conversation analysis and ethnomethodology,
cohesion in spoken interaction also received considerable attention early on. In par-
ticular, studies of the behavior of diferent kinds of adjacency pairs (such as a ques-
tion-answer sequence, a greeting followed by another greeting) proved very fruitful.
Another vast feld of research deals with the functions and use of special signals to
segment the discourse fow and to indicate that a change of topic is taking place. Such
discourse markers or pragmatic particles have been found to be the organizers of
discourse par excellence. In English, such typical markers would include anyway, so,
and well then. In addition to their discourse-segmenting function, these markers also
indicate speaker hearer relations and attitudes (early studies include Fernandez 1994;
stman 1981; Schourup 1985; Schifrin 1987).
3.3 Coherence
Not all relations among the various parts of a discourse are explicitly marked with
the kinds of cohesive devices talked about in the preceding section. Still, as speakers
of a language and as members of a discourse community and culture, we have built
up expectations and background knowledge that we can rely on to understand our
interlocutor even if s/he does not state everything overtly and explicitly. Such implicit
textuality is usually referred to as coherence. While surface cohesion may be seen as
a characteristic of discourse as a product, coherence is rather connected with inter-
pretability, the text world which the text receiver is building around the text (Enkvist
1989), on the basis of context, background knowledge, and the text producers choice
of discourse strategy (cf. 3.2).
Te distinction between surface cohesion and underlying coherence was early on
extensively addressed by Widdowson (1973), stman (1978), de Beaugrande & Dressler
(1981), Charolles (1981), and Petf (1989). Enkvist (1978) uses the term pseudo-
coherent for a string of sentences with overt surface cohesion, which is nevertheless
not a coherent text. For instance, if in a text about a whistling boy we suddenly fnd the
sentence Whistling is a nine-letter word, we might say that whistling in this sentence
creates cohesion with previous mentionings of the activity through repetition. But the
joint text that these two parts form, might rather be characterized as pseudo-coherent.
In France, the concept of enunciation (nonciation) was suggested as a domain
of mediation by which discourse is produced. Te concept is based on Benvenistes
(19661974) semiotic approach to language and was further developed by Culioli
(1990), and Anscombre & Ducrot (1983), cf. also Ducrot (1984). In a sense, the
distinction between coherence and cohesion corresponds to their distinction between
enunciation (proper) and uttered enunciation, respectively.
Te distinction between cohesion and coherence can of course only be upheld
in theory the concepts are convenient tools for the discourse analyst. Textual ele-
ments cannot discretely be categorized as either cohesion or coherence markers. A
question-answer sequence might be said to have a cohesive tie due to, say, the particu-
lar prosody, word order, or graphically the occurrence of a question mark in the
frst part. But it is clear that it will not be easy to fnd unambiguous surface indicators
for, e.g. an acceptance of an apology. All elements in a sentence or utterance ultimately
contribute in some way overtly or covertly to the textuality of a discourse, to making
it look a coherent whole. Even the pragmatic particles we mentioned in the preceding
section, which might seem to be explicit indicators of coherence, are really very implicit
markers. Interlocutors hardly ever notice that they use them. Speakers can deny that
they have used a pragmatic particle in a way that they cannot deny having transmit-
ted the propositional content of their utterance. Admittedly, this is a very monological
view of how communication works, and in the last couple of decades more dialogical
approaches have been advanced, where Malinowskis distinction between communica-
tion and communion is blurred, and the latter arguably gets to be the main impetus for
the former.
Tere are at least six major approaches that early on addressed the question of
how coherence could be explained; these are still relevant today. Some address the
issue with the help of micro-level analyses of what happens when two sentences are
combined. One early representative of this approach was work done within Sperber &
Wilsons relevance theory (cf. Blakemore 1992).
Another tradition is van Dijks (1977, 1980) model of macrostructures. A mac-
rostructure is based on a summary of the content of a discourse, and it can be recur-
sively reduced to fnally yield only one macroproposition. A very similar approach
is in terms of the notion of discourse topic (see Ochs-Keenan & Schiefelin 1976;
for a broader interpretation, see Brown & Yule 1983). Te idea here is that in the
same way as sentences have topics/themes as pegs to hang the rest on (cf. 3.1.), so
too, discourses have topics, usually expressed in the title or frst sentence/utterance
of a unit of discourse.
A third approach to coherence is to systematize and further develop the ideas
brought forth in the traditions of argumentation and rhetoric. Perhaps the best
known approach in this line of thinking is rhetorical structure theory, as put forth
by William Mann and Sandra Tompson (cf. Mann et al. 1992). Tis model further
develops insights in Grimes (1975) and is set up to reveal underlying rhetorical rela-
tions between units of text, e.g. causal and circumstantial relations, and the discourse
functions such relations have.
A fourth set of approaches tackle coherence by focusing on the interactive prop-
erties of discourse. Tough methodologically very diferent, speech act theory, the
Birmingham School of discourse analysis, and conversation analysis all treat coher-
ence in a very similar manner, through adjacency pairs, turn-taking, and similar
principles. Communicative acts similar to those developed in speech act theory are
being made use of to show the intricacies of the underlying coherence of a text.
A ffh approach follows in the footsteps of Bartlett. In order to understand and
adequately interpret a text, the receiver has to be able to build up a particular frame of
reference for it, a schema that organizes the ensuing discourse. Ideally, a text should
clearly indicate what its discourse topic is. It should be able to call forth a cognitive
scenario to which all the rest of the discourse can easily be related, and in terms of
which it can be adequately understood. For instance, the occurrence of Once upon a
time raises certain story-schema expectations in a reader.
Te sixth attempt at coming to terms with the concept of coherence is to use
notions like implicitness, negotiation and adaptability as explanatory tools in addi-
tion to the ever-important concepts of inference and context. A discourse is a process
whose purpose and efect changes as it unfolds. Producers and receivers of discourse
partake in a constant act of adaptation, changing their predispositions as they go along,
and negotiating the unfolding coherence. Verschueren (1987) showed the importance
of adaptability, Gumperz (1982) stressed aspects of negotiation, and developed coher-
ence on an interactive basis, and stman (1986) suggested three basic parameters of
implicitness which infuence the coherence of a discourse: politeness, involvement,
and what he calls coherence. Te society and culture at large impose requirements
and norms on its members; through the use of implicit reference to aspects of polite-
ness, we take our interlocutor into account and adapt our message to her/his world of
experience; and by the manner in which we express afect and involvement, we embed
feelings and prejudices in our discourse.
Coherence and implicitness are like context and function notions which, as
core phenomena, constantly need to be addressed in pragmatics and demand atten-
tion and adequate explanations. Te goal, however, should be to discuss and explicate
them rather than to defne them.
3.4 Grounding
Getting a deeper understanding of what makes discourse coherent takes us a long way,
but in order to adequately process a piece of discourse, we have to be able to pay atten-
tion to its salient parts more than to its digressions. Certain parts of a discourse are
bound to be more important than others.
Some elements, especially in spoken interaction, are used to check that the listener
understands or is paying attention, other elements evaluations, expressions of feeling
and camaraderie, jokes are added to make discourse more interesting, and others
are of course directly concerned with the propositional content of what one wants to
communicate. Languages use various devices to indicate that some parts are in some
way more foregrounded, and others more backgrounded, or digressive. Speakers have
diferent strategies available to them to indicate what point of view they take on what
is being said. Grounding has usually been studied in narratives (see, e.g. Hopper &
Tompson 1980). Tis has lead some linguists to equate foreground in discourse with
the temporally sequential main thread of actions and events in a prototypical story.
But temporal sequentiality is only one factor that infuences what is taken to be fore-
grounded information. For instance, the main participants in a story are ofen referred
to by using pronouns and zero anaphora, while reference to minor participants would
usually involve full noun phrases of various kinds. Tus, reduced noun phrases come
to be another foregrounding device. A fair number of studies have been devoted to the
search for an inventory of such grounding factors. Some languages have, thus, been
shown to have specifc foregrounding particles (for instance, this was the function of
Old English a, see 5.2.). Te following are some further factors that have been found
to infuence what is foregrounded in narrative text: a high degree of individuation and
agentivity of the subject, afectedness of the most afected participant, a dynamic, telic,
and punctual situation time, perfective aspect, and volitional verbs. For a survey of the
feld of grounding, see Wrvik (1990, 2002).
Te foreground-background distinction derives from the concepts of fgure and
ground in Gestalt psychology: the foregrounded part of the text the fgure is ofen
also viewed as its most salient portion (cf. e.g. Chvany 1985; for salience, see Osgood
1980). One direction in research on grounding has retained this original cognitive ori-
entation. Since events or sequences of events cannot be talked about without relating
them to some observer, i.e. to a human being who interprets them as such in the frst
place, one focus of grounding studies has moved to encompass issues of point of view.
Te language users point of view which in turn is infuenced by her/his interest,
empathy and involvement guides all observation and hence afects the grounding
values of linguistic elements (see e.g. Bjrklund 1993; Ehrlich 1990).
One area of research that awaits detailed study is how grounding works and how
it is marked in non-narrative text.
3.5 Discourse types and genres
When discourse linguists became interested in phenomena such as grounding and
point of view, they approached an area that has of tradition been part of the study of
literature. Another area that has constantly received attention in literary studies is that
of genre.
In linguistics, the last thirty years have seen interesting attempts at systemati-
cally discussing genre, or more generally, the diferent kinds and types of text that we
encounter. To a large extent, the impetus for categorizing texts and discourses came
from the practical needs of corpus linguists. For instance, texts in the Brown and LOB
corpora of written English made a frst distinction between fction and non-fction,
and had further sub-divisions for each category. Such intuitive classifcations have
continued to be used for other corpora, e.g. in the Helsinki Corpus of diachronic texts.
Since such classifcations are primarily situation-based i.e. they are made on the basis
of the purpose for which a text is produced, rather than on purely linguistic criteria
they correspond to our intuitive feelings about whether texts/discourses are of the
same type or not; in processual terms, whether the use of a particular kind of text
achieves the communicative purpose at hand. A type of text understood intuitively as
being of a certain kind can then be analyzed in terms of what its prototypical linguistic
characteristics are. However, Biber (1989) using the same method as Carroll (1960)
had employed in stylistics approached the problem from a very diferent point of
view. He used factor analysis on a large set of linguistic features in a variety of texts,
and interpreted the resulting clusterings of these features as functional types of text,
such as imaginative narrative, and learned exposition.
Issues relating to genre diferences have been dealt with in ancient rhetoric ever
since the Ad Herennium. In modern times, Crystal & Davys 1969 book on stylistics
was one of the frst in linguistics to systematically account for diferences between
genres. Halliday (1978) uses the term register. In his model, registers can be defned
in terms of variables such as feld, tenor, and mode variables which are used in
systemic functional linguistics in general. Within this approach to TDL, the notion of
genre has particularly been developed by Martin (e.g. 2006).
A complementary approach to the primarily situational ones mentioned above,
is one that is based on discourse-functional criteria. Text diferences in this sense are
talked about as text types, or discourse types. Whereas genre and register refer to
text-externally defnable classes of text (such as fairy-tale, news report, diary), used for
a specifc purpose, in a specifc communication situation, with ofen particular inter-
locutor roles (e.g. story-teller and audience), classifcations into text/discourse types
1

are made on a text-internal basis, to indicate variation according to the organization
and content of texts. Discourse types are prototypical categories, and text-internally
characterizable discourse types may be found in several text-externally defned genres
or registers. Tus, the narrative discourse type is representative not only of fairy-tales,
but also of genres like news stories, diaries, and reports of scientifc experiments. Tese
and many other genres may manifest coherence in terms of such prototypical charac-
teristics of narratives as temporal succession and/or agent orientation. And, the other
way round, a diary, for instance, would not necessarily have to be a narrative discourse
type, although typically, of course, it is.
:. In this overview we use the terms discourse type and text type synonymously; see, however,
Virtanen (1992b).
Traditional text-type categories such as narrative, descriptive and argumentative
discourse have been widely discussed in the literature. Werlich (1976) distinguishes fve
idealized types of text: narrative, descriptive, instructive, expository, and argumenta-
tive text. Tese abstract categories are further discussed in terms of fve parallel text
forms, which are divided into a number of text form variants (i.e. everyday text catego-
ries such as anecdotes, advertisements, and sermons). In addition to Werlichs studies,
the most infuential works in this feld are Longacre (1983) and Kinneavy (1971).
Te most infuential approach to genre in the 1990s came from applied linguis-
tics and in particular the seminal work by Swales (1990). It is proftable in genre
analyses not to work with too broad categories like journalese, but to distinguish
between genres like news report, editorial, background article, letter to the editor,
sports commentary, profle article, and advertisement; or even 'subgenres' like the
headline and the lead. Such hierarchies or categories must not, however, become
a straitjacket for the analyst; Bibers work clearly shows how much such categories
overlap. Genres are continuously on the move: emergent and established genres are
maintained or altered to ft the communicative needs of communities of people.
Still, in the analysis of authentic discourse, it is crucial to pay attention to varia-
tion in terms of discourse types, text types, genres, and registers. Otherwise there
is a risk of uncontrolled comparisons, where for instance the type of discourse afects
the results of an investigation but these are interpreted as evidence for some other, e.g.
dialectal or sociolectal, phenomena. Tere is still a tendency for this variation not to
be systematically addressed, for stories to be taken as discourse par excellence, and for
generalizations about language to be made on the basis of the linguistic characteristics
of stories.
4. Other trends
In addition to a continuing interest in the traditional felds discussed above, certain
trends particularly stood out at the turn of the millenium. Tis section will very briefy
survey some of these areas. As has continuously been stressed in this overview, TDL
covers a great number of sub-areas, and the overlap with other felds especially con-
versation analysis is immense. It is thus no wonder that research in TDL touches on
almost every feld in linguistics.
One issue that has received extensive attention in TDL studies the last twenty years
is the diference between speech and writing. Although there is still a general bias in
linguistics in favor of written discourse, discourse analysts have just as ofen studied
spoken interaction so much so that for some linguists, the term discourse itself refers
to spoken language. Tere is, however, no either-or divide between spoken and writ-
ten language. Bibers (1988) extensive corpus study confrms the existence of variation
within and across speech and writing. A related dichotomy or rather, scale ofen
discussed in TDL is that of orality vs. literacy. Cultures can be discussed in relation to a
prototypical set of characteristics associated with orality and literacy. According to Ong
(1982), an oral culture is basically formulaic, participatory, situational, and a literal cul-
ture analytic, distanced and abstract. In this sense, texts can be regarded as more or less
oral, independent of whether they de facto are spoken or written. As research within
computer-mediated discourse/communication (CMD/CMC) has advanced, the ques-
tion of how e.g. chat discourse should be classifed in relation to speech and writing has
attracted a lot of interest. Herring (e.g. 1996) has convincingly argued that CMC is not
reducible to either, but a third mode of expression, showing systematic variation across
languages, contexts, users and technological modes.
Chafe (1982) very early on drew attention to a parameter of involvement vs. detach-
ment: a speaker/writers involvement in a text manifests itself in the frequent occurrence
of particular linguistic features, for instance, frst person references and verbs like think,
indicating cognitive processes. Detachment, again, may be manifested through the use
of the passive. According to Chafe, involvement is also ofen coupled with fragmenta-
tion (cf. contractions, pragmatic particles), and detachment with integration (cf. attrib-
utive adjectives and nominalizations). Prior to Chafe, Labov (1972) had carried out
ground-breaking research into the use of evaluative elements in language, showing how
evaluative elements may be found dispersed throughout any story. Fleischman (1990)
found a clustering of evaluative elements around the most salient peak of narratives.
Two early collections of articles on involvement, or another term ofen used in prag-
matic research on afect are Ochs (1989) and Caf & Janney (1994).
Isomorphism and the discoursal impact of the iconicity of structures also need
to be given attention in any analysis of discourse. On all levels of language, we can
fnd tendencies toward iconic relationships. Te obvious case in a prototypical narra-
tive is that events are displayed in a sequential order mirroring the order in which
these events might have been experienced or interpreted as having taken place in real
time. Fairy tales, for instance, are very ofen presented in this manner. Other examples
include the following: horoscopes are ordered in a temporal order according to the
sequence in which the signs have been assigned stretches of time; recipes frst give the
ingredients that are causally primary for the food production to take place; and travel
guides tend to take the reader by his/her hand and lead him/her through the scen-
ery or place under description. Important early references include Haiman (1985),
Enkvist (1981), Givn (1989), and stman (1989).
Although Critical discourse analysis (CDA) has developed into a method in its own
right, the attribute critical is indicative of the similarity in its goals with those of Criti-
cal linguistics. Both attempt to describe the persuasive and manipulative functions of
language, and the linguistic manifestations of these, e.g. in news texts and advertise-
ments. Tus, TDL is used as a tool to unravel hidden prejudices and insincerities in
the speeches and documents produced by politicians and other wo/men in power. Te
original impetus for critical linguists was to raise the consciousness of the members of a
particular society, to go behind the superfcial meanings and look for the attitudes and
ideologies that messages implicitly carry with them. A stereotypical issue would be the
use of agentless passives to avoid mentioning the actor or controller behind an event.
Te journal Discourse and Society was specifcally established to deal with such issues;
a more recent infuential journal in this feld is the Journal of Language and Politics.
Te sphere of interest of CDA is close to that of sociolinguistics and social psychology,
where the impact of factors like age, ethnicity, sex, class, and the attitudes of speakers/
writers is in focus. However, critical approaches are not just concerned with the impact
of social factors on language, but also with the impact of language on society.
Computational applications have also had a major impact on TDL in recent
decades, notably through work associated with corpus studies. Some corpora have also
been planned in view of the study of discourse phenomena, both for spoken discourse
like the London-Lund Corpus of British English and the Santa Barbara Corpus of Amer-
ican English, and for specifc felds of research, like the Helsinki Corpus of diachronic
texts, the CHILDES corpus of acquisitional data, and ICLE, the International Corpus
of Learner English. Examples of two early but very diferent kinds of computational
approaches to discourse are Hovy (1988) and Hoey (1995). Eduard Hovy designed a
program, PAULINE, with a large set of parameters related e.g. to the formality and
addressee of texts, and showed how PAULINE can produce diferent texts with the
same propositional content for diferent audiences. Michael Hoey was interested in
what he calls lexical bonding within and across texts: on the basis of the recurrences
of the same or similar lexical elements in diferent texts on the same topic, one could
ultimately establish the pool of knowledge that a culture has about that topic.
One of the spheres of research in the 1960s that sparked the then growing inter-
est in large chunks of text was literary criticism and stylistics. Recently, TDL and nar-
ratology have established a fruitful give-and-take relationship. Propp, the Russian
Formalists, and Bakhtin all directly or indirectly inspired the work of narratologists like
Roland Barthes, Tzvetan Todorov, A.J. Greimas, and Grard Genette. Bakhtins studies
of voices, polyphony, and heteroglossia have been particularly infuential: a piece of
discourse typically manifests several perspectives, several voices, simultaneously, and
the words we use evoke other discourses.
2
A similar notion was developed as intertex-
tuality by Julia Kristeva (1969). Te point of focus here is precisely how we match a new
text with texts and situations we have encountered before. Discourse types and genres,
as well as Hoeys notion of cross-textual bonding, and the whole feld of allusions are
also closely related to intertextuality. Other important notions that have been devel-
oped in parallel in TDL and in narratology include point of view (cf. 3.4), and Genettes
i. Te Geneva School approach to discourse cf. Roulet et al. (1985) makes extensive use of the
concept of polyphony, as does Oswald Ducrot.
(1972) distinction between histoire and rcit, going back to the Russian Formalists
distinction between fabula (the activity as it is experienced to have taken place) and
sjuet (the way the activity is presented in a text). More recently, narrative studies have
tended to become more dialogically oriented; cf. e.g. Bamberg (2007).
Contrastive studies have a long history in linguistics both for language teach-
ing purposes, and for the purpose of setting up language universals and typologies.
In pragmatics, as such studies have received a more contextual and functional ori-
entation, we speak about cross-cultural studies. One of the frst infuential studies in
this area, with a direct infuence on research in TDL, is Kaplan (1966), attempting to
show how people from a variety of cultures construct paragraphs diferently. Galtungs
(1979) discussion of the characteristics of academic writing in the Teutonic, Saxonic,
Romance and Nipponic styles is another early study in this feld, showing how com-
munication difers across cultures in accordance with diferent patterns of thought and
of argumentation. Te most persistent challenge for cross-cultural discourse analysts
is no doubt to fnd a workable metalanguage. Today, the feld of cross-cultural studies
is variably referred to as contrastive text linguistics, contrastive pragmatics, or contras-
tive rhetoric, depending on the particular perspective of the analyst. Early important
studies in this rapidly growing feld include Blum-Kulka et al. (1989), Kaplan (1987),
Polanyi (1981), Purves (1988), Clyne (1987), Rgent (1985), and the journals Applied
Linguistics and Languages in Contrast.
A recent trend in pragmatics and TDL addresses the ontology of culture as such,
arguing that culture does not exist without discourse. Discourse gives structure and con-
tents to what we understand by culture. TDL in this view thus becomes an umbrella
concept not only for text studies, but for language and interaction studies in general.
Te impact of addressing questions of ontology has far-reaching consequences. For
instance, everybody will agree that context is one of the most crucial concepts in
pragmatics; and intuitively it may seem straightforward to relate a message to a set of
extralinguistic parameters. However, if contextualization itself is to be seen in terms of
discourse, the status of any presumed common denominator for understanding will be
questionable. Further research in this feld will no doubt give us a better understand-
ing of notions like variability, indeterminacy and ambivalence. Relevant early refer-
ences include Sherzer (1987), Duranti & Goodwin (1992), Auer & Di Luzio (1992),
Bauman & Briggs (1990).
5. Applications
So far our overview has dealt with aspects of TDL from a rather theory-centered point
of view. Tis brief section will draw together and mention the most important areas of
applicational work within TDL.
5.1 Practical applications
An early interest within TDL was classroom interaction, and studies of the character-
istics of teacher-student interaction by Sinclair & Coulthard (1975), Coulthard (1977),
Brazil et al. (1980), and other Birmingham-based discourse analysts brought this
feld to the very center of TDL (cf. also Edmondson 1981). Whereas the Birmingham
school addressed both theoretical and practical questions, others focused primarily
on the renewed interest in texts as an inspiration for developing new kinds of teaching
methods and materials. Widdowson (1979) is one of the forerunners here.
Te analysis of cross-cultural communication (cf. 4) is also ofen connected with
pedagogical aims, and it has proved particularly useful for those engaged in the teach-
ing of writing, for students of business negotiations across cultures, and for translators.
Reading and writing ofen go hand in hand, both benefting from various applications
of TDL (cf. e.g. Hoey 1991). On the basis of reading experiments, van Dijk & Kintsch
(1983) showed that diferent readings result in diferent macrostructures. Te technique
of process writing makes TDL even more useful to the teaching of academic and other
writing, since the focus is not solely on the resulting text, but on the process as such,
and thus on the various available strategies at the disposal of the student/writer. (On
writing, see, e.g. Connor & Kaplan 1987; Connor & Johns 1990.) Recently, Bakhtins
concept of diferent voices and ways of how to make use of this concept have started to
infuence developments even in this feld.
Tere are also specifc journals devoted to applications of TDL methods, such
as Written Communication and College Composition and Communication. Other
professional felds to which TDL has proftably been applied include psychotherapy,
law, business communication, advertising, academic writing and studies of difer-
ent languages for specifc purposes, e.g. technical or engineering English. For early
applications of TDL to various professions, see Di Pietro (1982), Bhatia (1993), and
Drew & Heritage (1992).
5.2 Acquisitional and diachronic studies
Work on the acquisition of discourse has been fairly abundant during the last decades.
Two of the frst systematic attempts at tying down TDL notions at large were the proj-
ects conducted in the late 1970s and early 1980s by Lily Wong Fillmore on second-
language acquisition and Susan Ervin-Tripps study of frst-language acquisition.
Te frst book-length studies on developmental pragmatics (Bates 1976; Ochs
& Schiefelin 1979) also appeared very early on. Within frst language acquisition,
early work concentrated on the order and manner in which children acquire difer-
ent communicative acts (cf. McTear 1979), and how they learn to understand them;
for instance, how they learn to interpret directives in the speech of their peers and
caregivers (cf. Ervin-Tripp et al. 1990). Another feld of interest was the acquisition of
modality and of discourse-organizing markers and hedges in a variety of languages (cf.
Gerhardt 1991; Clancy 1980).
In work on second-language acquisition and foreign-language learning, issues
like the difculty in teaching and learning the acceptable use of pragmatic particles
and discourse markers have been under constant debate. Tis work has resulted in
a number of contrastive studies on such particles in diferent languages, pointing to
functional correspondences between particles, prosody, and gestures.
TDL interest in diachronic aspects of language was primarily due to the establish-
ment of corpora such as the Helsinki Corpus (cf. Kyt 1991), which allow easy access to
a representative collection of texts from diferent genres. Moreover, Traugotts (1982)
early suggestion of a general principle for semantic change in words also contributed
considerably to this interest. Te former has seen large-scale attempts at tracking
down the histories of diferent genres Bibers multidimensional approach is a case in
point. Te gist of the latter is that Traugott suggested a general principle of subjectifca-
tion according to which a word will change from having a propositional meaning to
becoming more textual and ultimately more interactional and attitudinal.
In addition to these two developments, diachronic applications of the issues dealt
with in Sections 3 and 4 are also gaining in importance. Tus, e.g. the phenomenon
that discourse-organizing particles and adverbials foreground certain elements, show
the peak-salience of others, and the backgrounding, digressive function of still oth-
ers have received extensive attention due to the fndings that English seems to have
changed from marking grounding distinctions with particles in the Old English period,
to relying more on tense and aspect distinctions in Modern English (cf. Enkvist 1972;
Wrvik 1990). Te establishment of the Journal of Historical Pragmatics in 2000 greatly
enhanced research in this feld.
6. Final remarks
In this overview we have not attempted to give hard-and-fast defnitions of text and
discourse. At times, there has been competition between discourse analysts and prag-
maticists about which one is more general. For instance, Schifrin (1994) includes her
view of what pragmatics is as one subfeld of her Approaches to discourse.
3
Although we
have indicated the possibility of seeing discourse as a very general term, on the same
level of abstraction as, or even more general than, pragmatics, we have here retained
Morris (1938) view of pragmatics as the most general term.
To a certain extent, of course, this is a question of the hierarchization of tradi-
tional labels of academic subjects. A slightly more interesting response to this debate is
. Van Dijk (1985: Vol. 2) has pragmatics as one dimension of discourse.
that we are faced with diferent perspectives. Discourse and pragmatics have the same
felds of interest, but diferent aspects in focus. Tus, discourse will typically require
larger stretches of text or conversation, whereas for pragmatics this is not necessarily
the case. Te function of, for example, sound symbolism, or the specifcation of poten-
tial functional set-ups for pragmatic particles can be approached linguistically without
explicit reference to a particular discourse. Furthermore, pragmatics will be interested
in, say, contact phenomena, irrespective of whether these are defnable in relation to
discourse. Finally, if we see pragmatics as the study of functional aspects of language,
and semiotics as that of the communication of symbolic values with the help of sym-
bols, there is also a close relationship between semiotics and TDL.
Te variety of methods and theoretical approaches used within TDL might seem
fragmentative to non-practitioners, but discourse analysts would no doubt consider it
a sign of fexibility and open-mindedness.
References
Anscombre, J.-C. & O. Ducrot (1983). L argumentation dans la langue. Mardaga.
Ariel, M. (1990). Accessing noun-phrase antecedents. Routledge.
Auer, P. & A. di Luzio (eds.) (1992). Te contextualization of language. John Benjamins.
Bates, E. (1976). Language and context. Academic Press.
Bamberg, M. (ed.) (2007) Narrative - State of the art. John Benjamins.
Bauman, R. & C.L. Briggs (1990). Poetics and performance as critical perspectives on language and
social life. Annual review of Anthropology 19: 5988.
de Beaugrande, R. (1980). Text, discourse, and process. Ablex.
(1982). Te story of grammars and the grammar of stories. Journal of Pragmatics 6: 383422.
de Beaugrande, R. & W. Dressler (1981). Introduction to text linguistics. Longman.
Benveniste, E. (1966/1974). Problmes de linguistique gnrale. Gallimard.
Bhatia, V.K. (1993). Analysing genre. Longman.
Biber, D. (1988). Variation across speech and writing. Cambridge University Press.
(1989). A typology of English texts. Linguistics 27: 343.
Bjrklund, M. (1993). Narrative strategies in echovs Te steppe. bo Akademi University Press.
Blakemore, D. (1992). Understanding utterances. Blackwell.
Blum-Kulka, S. et al. (eds.) (1989). Cross-cultural pragmatics. Ablex.
Bolinger, D. (1952). Linear modifcation. Publications of the Modern Language Association 67:
11171144.
Brazil, D. et al. (1980). Discourse intonation and language teaching. Longman.
Brown, G. & G. Yule (1983). Discourse analysis. Cambridge University Press.
Caf, C. & R.W. Janney (eds.) (1994). Involvement in language. Special issue of Journal of Pragmatics
22(3/4).
Carroll, J.B. (1960). Vectors of prose style. In T.A. Sebeok (ed.) Style in language: 283292. MIT Press.
Chafe, W.L. (1976). Givenness, contrastiveness, defniteness, subjects, topics, and point of view. In
C.N. Li (ed.) Subject and topic: 2555. Academic Press.
(1980). Te deployment of consciousness in the production of a narrative. In W.L. Chafe(ed.)
Te pear stories: 950. Ablex.
(1982). Integration and involvement in speaking, writing, and oral literature. In D. Tannen
(ed.) Spoken and written language: 3553. Ablex.
(1994). Discourse, consciousness and time. University of Chicago Press.
Charolles, M. (1981). Coherence as a principle in the interpretation of discourse. Text 3: 7199.
Chvany, C.V. (1985). Foregrounding, transitivity, saliency (in sequential and non-sequential prose).
Essays in Poetics 10: 127.
Clancy, P.M. (1980). Te acquisition of narrative discourse. Ph.D. Diss., University of California at
Berkeley.
Clyne, M. (1987). Cultural diferences in the organization of academic texts: English and German.
Connor, U. & A.M. Johns (eds.) (1990). Coherence in writing. TESOL.
Connor, U. & R.B. Kaplan (eds.) (1987). Writing across languages. Addison-Wesley.
Cook, G. (1989). Discourse. Oxford University Press.
Coulthard, M. (1977). An introduction to discourse analysis. Longman.
Crystal, D. & D. Davy (1969). Investigating English style. Longman.
Culioli, A. (1990). Pour une linguistique de l nonciation. Ophrys.
Dane, F. (1974). Functional sentence perspective and the organization of the text. In F. Dane (ed.)
Papers on functional sentence perspective: 106128. Academia.
van Dijk, T.A. (1977). Text and context. Longman.
(1980). Macrostructures. Erlbaum.
(ed.) (1985). Handbook of discourse analysis, 4 vols. Academic Press.
van Dijk, T.A. & W. Kintsch (1983). Strategies of discourse comprehension. Academic Press.
Drew, P. & J. Heritage (eds.) (1992). Talk at work. Cambridge University Press.
Ducrot, O. (1984). Le dire et le dit. Minuit.
Duranti, A. & C. Goodwin (1992). Rethinking context. Cambridge University Press.
Edmondson, W. (1981). Spoken discourse. Longman.
Ehrlich, S. (1990). Point of view. Routledge.
Enkvist, N.E. (1972). Old English a an action marker? Neuphilologische Mitteilungen 73: 9093.
(1975). Tekstilingvistiikan perusksitteit. Gaudeamus.
(1978). Coherence, pseudo-coherence, and non-coherence. In J.-O. stman (ed.): 109128.
Research Institute of the bo Akademi Foundation.
(1981). Experiential iconicism in text strategy. Text 1: 97111.
(1987). A note towards the defnition of text strategy. Zeitschrif fr Phonetik, Sprachwissen-
schaf und Kommunikationsforschung 40: 1927.
(1989). Connexity, interpretability, universes of discourse, and text worlds. In S. Alln (ed.)
Possible worlds in humanities, arts and sciences: 162186. Walter de Gruyter.
Ervin-Tripp, S. et al. (1990). Politeness and persuasion in childrens control acts. Journal of Pragmat-
ics 14: 307331.
Fernandez, M.M.J. (1994). Les particules nonciatives dans la construction du discours. Presses
University de France.
Firbas, J. (1959). Toughts on the communicative function of the verb in English, German and
Czech. Brno Studies in English 1: 3968.
Fleischman, S. (1990). Tense and narrativity. Routledge.
Galtung, J. (1979). Essays in methodology 2. Ejlers.
Genette, G. (1972). Figures III. Seuil.
Gerhardt, J. (1991). Te meaning and use of the modals hafa, needto and wanna in childrens speech.
Givn, T. (ed.) (1983). Topic continuity in discourse. John Benjamins.
(1989). Mind, code and context. Erlbaum.
(1984/1990). Syntax, 2 vols. John Benjamins.
Grimes, J.E. (1975). Te thread of discourse. Mouton.
Gumperz, J.J. (1982). Discourse strategies. Cambridge University Press.
Haiman, J. (ed.) (1985). Iconicity in syntax. John Benjamins.
Halliday, M.A.K. (1967). Notes on transitivity and theme in English, Part2. Journal of Linguistics 3:
199244.
(1978). Language as social semiotic. Edward Arnold.
(1985). An introduction to functional grammar. Edward Arnold.
Halliday, M.A.K. & R. Hasan (1976). Cohesion in English. Longman.
Harris, Z. (1952). Discourse analysis. Language 28: 130.
Herring, S.C. (ed.). (1996). Computer-mediated communication. John Benjamins.
Hoey, M. (1983). On the surface of discourse. Allen & Unwin.
(1991). Patterns of lexis in text. Oxford University Press.
(1995). Patterns of lexis beyond text. In B. Wrvik et al. (eds.) Organization in discourse.
University of Turku.
Hopper, P.J. & S.A. Tompson (1980). Transitivity in grammar and discourse. Language 56: 251299.
Hovy, E. (1988). Generating natural language under pragmatic constraints. Erlbaum.
Kaplan, R.B. (1966). Cultural thought patterns in intercultural education. Language Learning 16: 120.
(1987). Cultural thought patterns revisited. In U. Connor & R.B. Kaplan (eds.): 921.
Kerbrat-Orecchioni, C. (1990/1992/1994) Les interactions verbales, 3 vols. Colin.
Kinneavy, J.L. (1971). A theory of discourse. Prentice-Hall.
Kristeva, J. (1969). Semeiotik. Seuil.
Kyt, M. (1991). Manual to the diachronic part of the Helsinki Corpus of English texts. University of
Helsinki.
Labov, W. (1972). Language in the inner city. University of Pennsylvania Press.
Lambrecht, K. (1994). Information structure and sentence form. Cambridge University Press.
Levinson, S.C. (1983). Pragmatics. Cambridge University Press.
Longacre, R.E. (1976). An anatomy of speech notions. Peter de Ridder Press.
(1983). Te grammar of discourse. Plenum Press.
Lundquist, L. (1983). Lanalyse textuelle. CEDIC.
Maingueneau, D. (1991). Lanalyse du discours. Hachette.
Mann, W.C. et al. (1992). Rhetorical structure theory and text analysis. In W.C. Mann &
S.A. Tompson (eds.) Discourse description: 3978. John Benjamins.
Martin, J.R. (2006) Genre, ideology and intertextuality: A systemic functional perspective. Linguis-
tics and the Human Sciences 2: 275298.
McTear, M. (1979). Hey! Ive got something to tell you: a study of the initiation of conversational
exchanges by preschool children. Journal of Pragmatics 3: 321336.
Morris, C.W. (1938). Foundations of the theory of signs. In O. Neurath et al. (eds.) International
encyclopedia of unifed science 1: 78137. University of Chicago Press.
Nunan, D. (1993). Introducing discourse analysis. Penguin.
Ochs, E. (ed.) (1989). Te pragmatics of afect. Text 9.
Ochs Keenan, E. & B.B. Schiefelin (1976). Topic as a discourse notion. In C.N. Li (ed.) Subject and
topic: 335384. Academic Press.
Ochs, E. & B.B. Schiefelin (eds.) (1979). Developmental pragmatics. Academic Press.
Ong, W.J. (1982). Orality and literacy. Methuen.
Osgood, C.E. (1980). Lectures on language performance. Springer.
stman, J.-O. (1981). You know. John Benjamins.
(1986). Pragmatics as implicitness. Ph.D. Diss., University of California at Berkeley.
(1989). Testing iconicity. Belgian Journal of Linguistics 4: 145163.
(ed.) (1978). Semantics and cohesion. Research Institute of the bo Akademi Foundation.
Petf, J.S. (1989). Constitution and meaning. In M.-E. Conte et al. (eds.) Text and discourse connect-
edness. John Benjamins.
di Pietro, R.J. (ed.) (1982). Linguistics and the professions. Ablex.
Polanyi, L. (1981). What stories can tell us about their tellers world. Poetics Today 2: 97112.
Prince, E. (1981). Toward a taxonomy of given-new information. In P. Cole (ed.) Radical pragmatics:
223255. Academic Press.
Purves, A.C. (ed.) (1988). Writing across languages and cultures. Sage.
Rgent, O. (1985). A comparative approach to the learning of specialized written discourse. In
P. Riley (ed.) Discourse and learning: 105120. Longman.
Renkema, J. (1993). Discourse studies. John Benjamins.
Roulet, E. et al. (1985). L articulation du discours en franais contemporain. Peter Lang.
Schifrin, D. (1987). Discourse markers. Cambridge University Press.
(1994). Approaches to discourse. Blackwell.
Schourup, L.C. (1985). Common discourse particles in English conversation. Garland.
Sherzer, J. (1987). A discourse-centered approach to language and culture. American Anthropologist
89(2): 295309.
Sinclair, J.M. & R.M. Coulthard (1975). Towards an analysis of discourse. Oxford University Press.
Stubbs, M. (1983). Discourse analysis. Blackwell.
Swales, J. (1990). Genre analysis: English in academic and research settings. Cambridge.
Traugott, E.C. (1982). From propositional to textual and expressive meanings. In W.P. Lehmann &
Y. Malkiel (eds.) Perspectives on historical linguistics: 245271. John Benjamins.
Verschueren, J. (1987). Pragmatics as a theory of linguistic adaptation. IPrA Working Document 1.
Virtanen, T. (1992a). Discourse functions of adverbial placement in English. bo Akademi University
Press.
(1992b). Issues of text typology. Text 12: 293310.
Wrvik, B. (1990). On the history of grounding markers in English narrative. In H. Anderson &
K. Koerner (eds.) Historical linguistics 1987.531542. John Benjamins.
(2002). On grounding in narrative: A survey of criteria and models. bo Akademi University.
Werlich, E. (1976). A text grammar of English. Quelle & Meyer.
Widdowson, H.G. (1973). Directions in the teaching of discourse. In S.P. Corder & E. Roulet (eds.)
Teoretical linguistic models in applied linguistics: 6576. AIMAV & Didier.
(1979). Explorations in applied linguistics. Oxford University Press.
Text linguistics
Robert de Beaugrande
University of Vienna
1. The rise of text linguistics
A decisive step in the rise of modern Western Continental linguistics was to instate
the central tenet that language should be considered a uniform, stable and abstract
system apart from variations due to individual or social contexts and activities. Te
description of the language should be couched in highly general statements about the
language as a whole or even in universal ones about all languages. Tis (Saussurean)
programme inaugurated the search for the complete set of constraints upon language
in and for itself (langue) without reference to the use of language (parole). Te con-
straints should cover all valid instances or sets of instances of the language.
Instead of accepting the central tenet as an incontestable a priori principle, as
mainstream linguistics has widely done, we might view it as a working hypothesis that
should be tested. Te crucial tests would surely be the convergence among the data
discovered and described, and the consensus among linguists about the constraints
that govern the data. Beyond the scope of phonemics and morphemics, however, these
tests have not been met. Instead, we can observe linguists choosing between two com-
pensatory strategies for tapping constraints that had not yet been properly discovered
and stated. Either they have sought to describe corpuses of recorded real data that are
naturally constrained by their contexts of occurrence, or they have used their own
intuition and introspection as a source of constraints while inventing language data,
mostly isolated sentences. Te second strategy looks easier than the frst but turns out
not to be because it obliges linguists to devise artifcial constraints to describe some
well-formed sentences and to exclude some ill-formed ones that would be unlikely
to appear in real data. Not surprisingly, the switch from real corpuses over to invented
sentences led to an increasing breakdown in convergence and consensus.
Te situation was obscured somewhat by the versatile role of the sentence. For
Saussure (1966 [1916]: 124), the sentence was the ideal type of syntagma, but it
belongs to speaking [parole], not to language [langue]. For Bloomfeld (1933: 170f),
the sentence was an independent form not included in any larger (complex) linguistic
form; perhaps all languages distinguish sentence types. For Chomsky (1957: 55 and
1965: 34), the sentence was a main notion that must be defned by general linguistic
theory, and the unit to which a descriptively adequate grammar assigns structural
descriptions in accordance with the linguistic competence of the native speaker.
Text linguistics 287
For Halliday & Hasan (1976: 8, 244, 235), a sentence was the highest unit of gram-
matical structure and was defned as any set of clauses that are hypotactically and/or
paratactically related sentences cannot be rearranged, as a coordinate structure can,
in diferent sequences. Tis versatility suggests that the sentence provided a strategic
compromise between language versus use by being both (1) a formal or grammatical
pattern whose obligatory or optional constituents are specifed by the general language
system, and (2) a fairly self-contained utterance pattern. Every reasonable sentence,
real or invented, entails some constraints refecting its possible uses. So the sentence
has readily been accepted as the standard realisation of semantic and pragmatic enti-
ties like the proposition, predication, utterance, assertion, or speech act.
Yet even so versatile a notion of the sentence has not supplied linguistic theories
and models with any defnitive set of constraints, either for corpuses of real data or for
accumulations of invented data. Naturally, some linguists looked beyond the sentence
boundary, giving a vital impulse to text linguistics in the late 1960s and early 1970s.
For the sake of continuity, early text linguists based the text in more abstract or con-
crete ways on the sentence, witness the tactic of studying the constraints only within
sentence pairs, especially question and answer, the grammatical structure of the pre-
ceding sentence being treated as a minimal context (e.g. Isaenko 1965).
To the degree that the sentence was ofcially a grammatical or syntactic unit,
research was dominated by notions like text grammar (e.g. van Dijk 1972), text
syntax (e.g. Dressler 1970), hyper-syntax (Palek 1969), and macro-syntax (Glich
1970), which allowed some double-tracking by referring to the sentence while moving
beyond it. Research sought to project the prevailing concepts and methods from sen-
tences over to texts with relatively modest revisions, yet justifed text studies by raising
grammatical issues not well accounted for within the single sentence. For instance, the
ellipsis found in faithful recordings of spontaneous conversation could be grasped
as a grammatical connection among sentences rather than a phenomenon inside the
single sentence, yet can be constrained by deriving each instance from an underlying
sentence [that] native speakers reconstruct (Gunter 1963: 137, 140).
Such double-tracking at the boundaries of the sentence eventually led to a
reas sessment of its role in grammar. In a genuine text grammar, the sentence could be
neither the largest linguistic form to be studied (Bloomfeld) nor the only unit to which
structural descriptions are assigned (Chomsky). Such a grammar was expected also
to specify the formal features that make a sequence of sentences into a text (Isaenko
1965: 164); or to distinguish whether two successive sentences belong to the same text
(Dressler 1970: 66), or to specify the formal properties determining a typology of texts
(van Dijk 1972: 5); or to explain how speakers of a language [] are able to distinguish
between grammatical and ungrammatical texts (van Dijk 1972: 4); and so on.
But no such text grammar managed to supply adequate constraints. Non-
grammatical factors were required even to account for manifestations that are undeniably
288 Robert de Beaugrande
grammatical, such as pronouns (e.g. Steinitz 1971), articles (e.g. Weinrich 1969)
and tenses (e.g. Weinrich 1964); and the same holds far more plainly textual factors,
such as cohesion, coherence, topic progression, narrativity, and situationality, which
also afect grammar. So text linguistics energetically explored semantic and prag-
matic issues, including those previously discounted for describing abstract, isolated
sentences.
Tis exploration engendered a wide spectrum of defnitions for the text. For
Isenberg (1970: 1), a text was a coherent sequence of sentences. For Harweg
(1968a: 148), a text was a series [Nacheinander] of linguistic units constituted by
uninterrupted pronominal chaining. For van Dijk, a text was the abstract theo-
retical construct underlying discourse (1976: 3); or a sentence sequence with a
macrostructure, i.e. a global structure of meaning (1978: 41; cf. van Dijk 1979).
For Hartmann (1968: 220), the text was a fnite ordered set of textually complicated
partial signs of various sorts and functions and hence the linguistically signifcant
originary sign of language. For Weinrich (1976: 186), a text was an ordered sequence
of language signs between two noticeable discontinuations [Unterbrechungen] of
communication. For Halliday & Hasan (1976: 1, 4, 23) a text was any passage, spoken
or written, of whatever length, that does form a unifed whole; it was variously said
to be defned by relations of meaning or by cohesion and register. For Schmidt
(1973: 144f), a text was a linguistic realisation of textuality as an organisational form
for the formation of pluralistic complexes, hence a matrix for the mutually relevant
and coherent occurrence of elements in the language system.
Tis brief gallery of text defnitions in text studies was emblematic for the lively
discussions of fundamental concepts and problems merely implicit in sentence lin-
guistics, as earlier approaches now came to be called. If the text is so diverse an entity
and yet consists of sentences, then the sentence should also have a high diversity that
has hitherto not been acknowledged or strictly formalised, but has remained inherent
in the intuitions of linguists or in the conditions of social contexts. Some of this diver-
sity was ambivalently acknowledged at times, e.g. when sentence linguists sought
richer constraints by adding a semantic component but balked at advancing it from
an interpretative to a properly generative role, doubtless fearing to lose the formality;
the syntactic features and logical form proved even less conducive to convergence
and consensus than did purely syntactic notions.
When the text became an object of syntactic inquiry, the pressure redoubled
to reintroduce the semantic and pragmatic constraints that had previously been
diluted by discounting real corpuses of empirical data and centring linguistic theory
on the syntax of the single sentence. Tus, Weinrich (1967: 112) redefned syntax
as a dialectic that investigates all semantic elements of language in their relation
to the speech situation. Dresslers (1970: 68) text-syntax favoured hypotheses
that take account of the actualities of the speech process. Glichs (1970: 46, 297)
macro-syntax eschewed the concept of sentence as highly questionable how-
ever you defne it, and looked for the arrangement signals (Gliederungssignale)
marking the beginning or end of spoken texts or text segments. Danes (1970)
model of thematic progression in texts refected his well-known three-level
approach to syntax, wherein the grammatical structure of sentence was paralleled
by its semantic structure and its pragmatic organisation of utterance (1964: 226f).
And Petfs (1978: 40) ambitious text-structure/world structure theory () aims
at describing the syntactic, semantic, and pragmatic structures of natural language
text through a formal semiotics.
It was high time to consider whether the text would belong on the side of
language or of language use. Te simplest mainstream recourse of putting it among
language use had the drawback that linguistic theory had not just neglected use,
but had described it in negative terms to rationalise its neglect. For Saussure (1966
[1916]: 13, 15, 19, 11), parole was individual, heterogeneous, and momentary;
speech cannot be studied for we cannot discover its unity. For Chomsky (1965: 4, 44,
58, 201), actual speech in performance was scattered and defcient and had a
degenerate quality, replete with false starts fragments, and deviant expressions.
Text linguists in turn remained cautious about putting the text in the domain of
actual use even though they viewed use much more positively. Van Dijk (1972: 3f) rea-
soned: although theories, and especially grammars, do not directly describe empirical
phenomena, but only formally reconstruct the entities, relations and systems underly-
ing these phenomena, it is necessary to postulate entities or regularities which are as
close as possible to the empirical phenomena without losing their general character.
So he proposed to make the text a theoretical concept and an abstract unit [that]
underlies a meaningful utterance [of a] discourse; any utterance is the token of a text
(1978: 41, 1972: 317f, 1976: 3). Similar reasoning motivated such conceptions as Bellerts
(1970: 366) idealised text with no digressions expressing one uniform reasoning,
continuous plot, etc.; Dresslers (1970: 65) potential text in the sense of a pre-linguis-
tic text; Schmidts (1973: 150) meta-linguistic text formulary, being a set of lin-
guistic constituents ordered by a thematic deep structure; and Harwegs (1968a: 152,
1971: 179) emic text defned by text-immanent criteria as opposed to etic text deter-
mined by extra-linguistic criteria not based on language structure. Like Saussures
speech, Harwegs etic side lacked unifying delimiting criteria, witness the etic text
beginnings [of] novels, short stories, poems, scientifc treatises; still, he allowed that
emic and etic texts might coincide to varying degrees (1968a: 152f, 1968b: 348).
For Hartmann, in contrast, sentence formation proceeds through rules of
individual languages, whereas text formation proceeds through principles apply-
ing to all languages and through extralinguistic stylistic and individual requirements,
e.g. motives, intentions, efects (1968: 218f). Tis formulation justifed treating
the sentence within a grammar in a langue perspective yet suggested that speakers
who form a text are transcending the boundary between langue and parole plus the
boundary between the particular language and all languages. Te text would be both
more general and more specifc than the sentence, and text theory would need to span
the whole range from abstract universals over to concrete situational factors.
Hartmanns vision signalled his remarkable agility as a theorist, which pow-
erfully infuenced his students (e.g. Harweg, Koch, Schmidt, Rieser, Wienold, de
Beaugrande). Like the British functional school that did not call itself text linguis-
tics (e.g. Firth, Halliday, Sinclair, and their students), he realised that separating the
text from the communicative context and situation obscures rather than clarifes the
question of what a theory should account for. So we should not confne the term
text to a theoretical abstraction as proposed by van Dijk: beside the huge task of
analysing and describing actual texts, we incur the task of translating between the
actual text and the hypothetical underlying text, typically a formal representation,
the interpretation of which is vastly less conductive to convergence and consensus
that is that of the text, as van Dijk himself and his co-workers in the Konstanz project
soon discovered (cf. van Dijk et al. 1972).
Today, the text is widely defned as an empirical communicative event given
through human communication rather than specifed by a formal theory. Each such
event rides on a dynamic dialectic between the virtual system of language (the rep-
ertory of possibilities) and the actual system constituted by the choices of the text
producer; the text is thus on neither side of language versus use, but integrates and
reconciles the two. Our task is to describe, as empirically and realistically as we can,
the processes whereby communicative participants can and do produce and receive
texts. Te task plainly demands interdisciplinary research between text linguistics and
psychology, sociology, ethnography, and so on, all of whom work with real data from
the standpoint of human activities.
Defning the text and our main task this way brings text linguistics into line
with the research tradition of discourse analysis, which has operated bottom up with
realistic data obtained from feldwork. Since its earliest formulations (e.g. Harris 1952),
investigators have looked a real data beyond the sentence and developed a grammar
incorporating semantic and pragmatic issues, such as how diferent languages deploy
grammatical resources for expressing events and participants in narratives (see now
Longacre 1990). Te ethnographic emphasis obviated the need for the abstract theo-
retical formalisations cultivated in mainstream linguistics, including the isolation of
language in and for itself from the use of language (cf. Stubbs 1993; Sinclair 1994).
Today, the same emphasis has brought discourse analysis into the mainstream, while
formalist linguistics drifed into a scientifc crisis by failing to meet its announced
goals, notably to formulate a complete generative grammar for any natural language.
Te constraints once excluded on theoretical grounds now had to be reintroduced,
and discourse analysis steadily merged with text linguistics.
Te latest and most decisive step has been the compilation of the Bank of English,
a vast computerised corpus of contemporary spoken and written English discourse,
which enables reliable statements about the typical patterns and collections of English
(Sinclair 1992a, 1992b; Baker et al. eds. 1993). Some of the many regularities that are
not open to mere intuition or introspection emerged at 20 million words (1992 fgure);
others at 200 million words (1994 fgure). For example, the verb brook was found to
have a clearly defned set of constraints on its usage (Sinclair 1994). Grammatically, it
takes negation (usually not just before it or, no just afer), and does not appear with the
frst or second person singular as the subject. Semantically, its direct object is a con-
cept associated with opposition, interference, or delay. Pragmatically, the subject must
be in a position of sufcient authority to carry the performative perlocutionary force
entailed in declaring what it will not brook.
Te implications Sinclair (1994) draws from such data are likely to be felt through
all of linguistics. (He provocatively compares the large corpus to the telescope in
astronomy and the microscope in biology instruments which vastly enhanced the
scientifc status of inquiries). First, the rif between langue versus parole is fnally
healed, because we can derive the regularities in the language fairly directly from real
data. Second, the pendulum swings decisively back toward convergence and consensus
because such constraints as those upon the verb brook are strongly implicated by the
totality of data in the display,
1
whereas the constraints upon a single invented sentence
that might hardly ever be uttered can be endlessly disputed. Tird, the semantic and
pragmatic constraints set aside by formalist models are efectively restored and placed
on a more realistic basis than ever before. Fourth, we see that the appropriate unit for
analysis is not the morpheme, word, or phrase but the co-text, the utterance environ-
ment, of variable lengths, in which words of phrases tend to appear together. And ffh,
we no longer need to cut up linguistics into tidy levels or components for separate
analysis and description; even the staid division between grammar and lexicon proves
to be merely two complementary points of view upon the same phenomena.
Text linguistics too will feel the impact. First, the dialectic between virtual system
and actual system becomes far clearer and more direct when the function of a word
of phrase in the representative corpus can be explicitly correlated with the function it
has in a given text. Second, the text as whole does not need a formal circumscribed
defnition; for most purposes, it can be perfectly well grasped as one actualization
within a set of attested cross-sections we can pull up for display as we choose. Tird,
:. Sinclair (1994) suggests that the displays shows competence if read vertically and performance
if read horizontally; but this competence would at most be probabilistic because speakers of
English know how to produce data that have not yet occured in the corpus, although they may
well not be likely to.
the sentence boundary loses its theoretical signifcance and becomes an empirical phe-
nomenon which, like the other boundary signals of punctuation or intonation, occurs
with higher probability afer some word combination (Firths collocations) and gram-
matical patterns (Firths colligations) than others. Fourth, the production of a text is
viewed as a set of interactive local choices, some of which are made well before the
key word appears in the sequence. And ffh, we can use subcorpora as a data-driven
bottom-up method for characterising text types and discourse domains, a task for
which speculative schemes have remained inefectual.
We can now reach a consensus about realistic defnitions of our two main terms:
text being the individual communicative event within the ongoing interaction of dis-
course. Our shared research interests should decide how abstract or concrete, how
theoretical or empirical, these terms should be in specifc projects with explicit goals,
e.g. studies of discourse for special purposes (DSP) in technical felds (an analogue to
register studies) with the goal of developing strategies to make specialised knowledge
more generally accessible. Te same interests could resolve other thematic questions in
text linguistics that are not resolvable on general theoretical grounds, e.g. how far we
would analyse a text in an ascendant mode from smaller units to the text or in a descen-
dant mode from text to its constituent units (Kallmeyer et al. eds. 1971: 57).
Text research is also swinging into line with discourse processing, an interdisciplin-
ary exploration of the processes that make discourse both feasible useful in so many set-
tings (e.g. Freedle ed. 1977, 1979; Beaugrande 1980, 1984; van Dijk & Kintsch 1983) and
a major sector in cognitive science, itself a convergence among psychology, computer
science, sociology, and anthropology, along with their ofshoots and intermediaries
like psycholinguistics, cognitive psychology, artifcial intelligence, sociolinguistics, and
social psychology. Te question of ascendant versus descendant, for instance, could be
assessed in light of fnding that text receivers combine both tactics, doing bottom-up
scans of the local text units while forming and testing top-down hypotheses about the
global organisation, e.g. the macro-structures of overall coherence.
Further, text linguistics has become strongly engaged with pragmatics (e.g.
Schmidt 1973, 1976). Instead of giving a set of rules that turn these objects into
acts and place these acts in a situation (van Dijk 1976: 190), pragmatics investi-
gates how texts and co-texts (not sentences) are always already acts in situations, even
while the syntactic or semantic operations are being carried out. Hence, pragmatics
forms the outermost framework for approaching syntax and semantics in terms of
communicative activities.
And fnally, text linguistics has assumed a cordial relation to stylistics (e.g. Enkvist
1977), which mainstream linguistics tended to marginalise by situating it in the use of
language rather than in language itself. For Weinrich (1969: 69), the border consists
of text linguistics exploring the syntactic function of an element such as the indefnite
article in the language, i.e. in every real or possible text, whereas stylistics explores
the particular use in a certain text by one author; in between lies text typology,
stating for which text types certain afnities can be determined. Here, we may feel
reminded of Weinrichs (1967: 112) already cited defnition of syntax as a dialectic
that investigates all semantic elements of language in their relation to the [pragmatic]
speech situation; the relation between the function of an element in the language and
in a given text is surely dialectical too. As with Hartmann and Sinclair, the old static
opposition between language and parole was here given a dynamic reinterpretation
making the text the entity that interfaces the two sides.
Te increasing engagement of text linguistics with the four domains just cited
discourse analysis, discourse processing, pragmatics, and stylistics has both increased
our fexibility and guided us in resetting our priorities. We have grown more aware of
how to coordinate theoretical with empirical, and categories with data, when we set up
and implement a research project. Instead of following one set of orthodox all-purpose
procedures, we seek theoretical principles for telling which to address. Moreover, we
acknowledge that, as a communicative event, the text does not simply hold still like an
inert object while we analyse it. Instead, we become engaged in retextualising it, which
can infuence the conditions of its textuality. We must consider how far our intentions
and interpretations as investigators parallel, or difer from, those of the discourse par-
ticipants whose activities are now expressly included in the scope of our studies, e.g.
how far we may develop more elaborate representations to register and resolve multiple
meanings, or single out deviations or disturbances, well beyond what would reach the
consciousness of a speaker or hearer. Such issues too can be empirically approached.
In retrospect, then, we might distinguish three stages in the rise of text linguistics.
Te earliest stage (late 1960s, early 1970s) was dominated by text grammar and
remained fairly circumscribed and uniform by working to extend prevailing theories
from sentence to text. Te next stage (late 1970s, early 1980s) was more dominated
by textuality as a structure with both linguistic and social aspects. Te principles
of textuality apply whenever an artifact is treated as a text; so textuality is not a lin-
guistic property or feature (or set of these) which some samples possess and other
do not, but an empirical human predisposition and activity wherever communica-
tive events occur. Te mutual relevance assigned to linguistic forms is cohesion and
that assigned to the meanings is coherence; intentionality covers what speakers intend,
and acceptability what hearers engage to do; situationality concerns ongoing circum-
stances; and intertextuality covers relations with other text, notably those of the same
or a similar text type.
Te third and most recent stage has been dominated by textualisation, the social
and cognitive processes entailed in the actual production and reception of texts. Tis
trend too was driven by the search for richer and more global constraints, but now
assuming that process models can provide more than can product models starting
from a text-artifact.
Te 1990s look toward a general science of text and discourse. Its syntax could
be a cognitive functional grammar wherein world-knowledge constrains the
organisation of phrases, clauses, and clause complexes (not just sentences). Its seman-
tics could be a post-classical epistemology that brackets knowledge and content by
examining them not via reference, reality, or truth value but via the applied model-
ling styles such as realism with parameters of design such as complexity, novelty, and
determinacy. Its pragmatics could be a critical view of communication as an ongoing
interaction whereby the signifcance of a situation (real or hypothetical) is being nego-
tiated, speaking turns are assigned, and relations of power or solidarity are enacted.
Tese three domains are organised within a text agenda refecting the ongoing docket
of actions, plans, and goals, and within a text economy of focuses, values, and priori-
ties. Te three thus merge in a framework for exploring the relation of texts to such
factors as text types, style, register, language for special purposes (LSP) and terminol-
ogy, but also socialisation, education, translation, ideology, gender, and emotion.
2. Some central issues
Te following central issues in text studies have crystallized in the work of numerous
researchers:
1. Te text is not merely a linguistic unit, but an event of human action, interaction,
communication, and cognition. Te notion of the text being merely an instance
of language has a long tradition both in folk-wisdom and in the various spe-
cialised disciplines that have studied texts from ancient grammar and rhetoric
up through modern linguistics. Te notion was hardly challenged as long as the
text did not have the status of a central or explicitly defned term. When language
studies reached the point in the search for constraints where the text did attain
this status, its multi-media qualities as a social and interactional event came into
focus. Te perennial question, Is it still linguistics? generated during half a cen-
tury of self-conscious questing for language by itself , seems quaint or irrelevant
now. Te constraints bearing upon texts must be comprehensively explored, what-
ever their sources.
2. Te main source of data is naturally occurring texts and discourses. Trough
history, most studies of language have expropriated language data without
expressly justifying its authenticity. Most samples in earlier studies were in prac-
tice taken from prestigious sources such as literary or sacred texts, whose spe-
cialised qualities were not considered problematic. Modern linguistics recognised
the problems and emphatically turned to ordinary speech but, somewhat contra-
dictorily, proposed to describe it apart from texts. Te proposal eventually led to
descriptions based on invented data, which was artifcial in many respects including
even its supposed ordinariness. Today, large corpuses of real data reveal the argu-
ments in favour of this tactic to be either obsolete, e.g. that we cannot get our hands
on enough data; or spurious from the start, e.g. that speakers have some idealised
competence whose relation to what they actually say is deeply uncertain. Te cor-
puses also show that the supposed uniqueness of every utterance impels us not
to retreat from surface structure to deep structure but rather to exploit the com-
monalities of collocation and colligation that we can derive from the data.
3. Text analysis is rich and expansive. Te amount of potentially relevant data for
describing a text as a communicative event is in principle open. We can be content
when we have uncovered some non-obvious and useful insights about the human
consequences of the text, e.g. how and how well it makes knowledge accessible to
people who need it.
4. In modern-day text research, the investigator has to engage and re-engage with
the texts rather than strive for an idealised separation of subject from object or
scientist from data.
5. Te motives for doing text research in terms of epistemological interests bearing on
the relations between texts and society have become more and more important for
researchers in the feld. Tus, e.g. projects addressing issues of inequality in the
access to knowledge through discourse are socially relevant and thus enhanced.
Critical text research should ultimately try to establish freedom of access to
knowledge and to reveal and redistribute communicative power structures.
6. Text linguistics has an interdisciplinary perspective; linguistics cannot be the sole
basis for a science of text and discourse. To understand a phenomenon as com-
plex as language, we need to regard it from continually alternating and evolving
viewpoints.
Tese issues have increased in importance within text linguistics. Tey all go in the
direction of Coserius (1971: 189) vision of a new dimension wherein the text is
a perspective concerning all of linguistics, and every linguistic discipline would be
reconceived from this standpoint.
References
Baker, M., G. Francis & E. Tognini-Bonelli (eds.) (1993). Text and technology. John Benjamins.
de Beaugrande, R. (1980). Text, discourse, and process. Ablex.
(1984). Text production. Ablex.
de Beaugrande, R. & W. Dressler (1981). Introduction to text linguistics. Longman.
Bellert, I. (1970). On a condition of the coherence of texts. Semiotica 2: 335363.
Bloomfeld, L. (1933). Language. University of Chicago Press.
Chomsky, N. (1957). Syntactic structures. Mouton.
(1965). Aspects of the theory of syntax. MIT Press.
Coseriu, E. (1971). Intervention in the discussion Textlinguistik. In W.-D. Stempel (ed.) Beitrge zur
Textlinguistik: 189216. Fink.
Dane, F. (1964). A three-level approach to syntax. Travaux Linguistiques de Prague 1: 225240.
(1970). Zur linguistischen Analyse der Textstruktur. Folia Linguistica 4: 7278.
van Dijk, T. (1972). Some aspects of text grammars. Mouton.
(1976). Text and context. Longman.
(1978). Textwissenschaf. Niemeyer.
(1979). Macro-structures. Erlbaum.
van Dijk, T. et al. (1972). Zur Bestimmung narrativer Strukturen auf der Grundlage von Textgram-
matiken. Buske.
van Dijk, T. & W. Kintsch (1983). Strategies of discourse comprehension. Academic Press.
Dressler, W. (1970). Modelle und Methoden der Textsyntax. Folia Linguistica 4: 6471.
Dressler, W. (ed.) (1977). Current trends in text linguistics. De Gruyter.
Enkvist, N.-E. (1977). Stylistics and text linguistics. In W. Dressler (ed.): 174190.
Freedle, R. (ed.) (1977). Discourse production and comprehension. Ablex.
(1979). New directions in discourse processing. Ablex.
Glich, E. (1970). Makrosyntax der Gliederungssignale im gesprochenen Franzsisch. Fink.
Gunter, R. (1963). Elliptical sentences in American English. Lingua 12: 137150.
Halliday, M. & R. Hasan (1976). Cohesion in English. Longman.
Harris, Z. (1952). Discourse analysis. Language 28: 130/474494.
Hartmann, P. (1968). Zum Begrif des sprachlichen Zeichens. Zeitschrif fr Phonetik, Sprachwissen-
schaf und Kommunikationsforschung 21: 205222.
Harweg, R. (1968a). Pronomina und Textkonstitution. Fink.
(1968b). Textanfnge in geschriebener und gesprochener Sprache. Orbis 17: 343388.
Isaenko, A. (1965). Kontextbedingte Ellipse und Pronominalisierung im Deutschen. Beitrge zur
Sprachwissenschaf, Volkskunde und Literaturforschung: 163174. Akademie.
Isenberg, H. (1970). Der Begrif Text in der Sprachtheorie. Akademie der Wissenschafen der DDR.
Kallmeyer, W., W. Klein, R. Meyer-Hermann, K. Netzer & H.-J. Siebert (eds.) (1971). Lektrenkolleg:
Einfhrung in die Textlinguistik. Universitt Bielefeld.
Longacre, R. (1990). Storyline concerns and word order typology in East and West Africa. UCLA Dept.
of Linguistics.
Palek, B. (1969). Cross-reference. Charles University Press.
Petf, J. (1978). A formal semiotic text theory as an integrated theory of natural language. In
W. Dressler (ed.): 3446.
de saussure, F. (1966). [1916]. Course in general linguistics. McGraw-Hill.
Schmidt, S.J. (1973). Texttheorie. Fink.
(ed.) (1976). Pragmatik 2. Fink.
Sinclair, J.M. (1992a). Priorities in discourse analysis. In M. Coulthard (ed.) Advances in spoken dis-
course analysis: 7988. Routledge.
(1992b). Te automatic analysis of corpora. In J. Svartlik (ed.) Directions in corpus linguistics:
379397. Mouton de Gruyter.
(1994). Large-corpus linguistics at the crossroads (video). University of Vienna.
Steinitz, R. (1971). Nominale Proformen. In W. Kallmeyer et al. (eds.): 375397.
Stubbs, M. (1993). British traditions in text analysis. In M. Baker et al. (eds.): 133.
Weinrich, H. (1964). Tempus. Kohlhammer.
(1967). Syntax als Dialektik. Poetica 1: 109126.
(1969). Textlinguistik. Jahrbuch fr Internationale Germanistik 1: 6174.
(1976). Sprachen in Texten. Klett.
Index
A
A corpus of English dialogues
15601760, :,I
Abercrombie, N., :,,, :,8
Abrams, K., :o,
Abtnungspartikel (1969), ::,
Achard, P., o,
acknowledge, ,o
acquisition, :8o
acquisitional and diachronic
studies, :8o:8I
act of uttering, 8,, ,:
acte de discours, 8:
activation modes, I,I8
actualization, ,,
Adam, J.-M., 8
adaptability, :,,
addressivity, I:I
adjectives, ,,
Adolphs, S., :,:
afect, Io
Aijmer, K., ::,
Aikhenvald, A.Y., ::
Aitchison, J., Io,, II,
Alatis, J., :8
Albrecht, J., :,:
Aldrich, V.C., II,
alehtic modalities, ,,
Alexander, J., Io,, :,o
Altenberg, B., :,,, :,o
Althusser, L., o,, :,, :,,
ancient rhetoric, FSP, Io:Io
Andersen, G., ::, ::,,
:,o:,:, :,,
Andersson, L.-G., ::,, :,,
anecdote, I
Anscombre, J.-C., 88, :,I
Anthonissen, C., ,o,:
Antonopoulou, E., Io
Applied Linguistics, :,,
appraisal theory, ,
appraisal, I,
appraisal, defned, I
appreciation, I,
appreciative modalities, ,,
Apte, M., I,,, I,I8
argumentation analysis, o
argumentation theory, II,
Ariel, M., :,,, :,o
Aristotle., Io,, Io,, II:,
I,,, I,o, I,o
arms race, I8:, I8
Arnauld, A., ,,
Arndt, W., ::,
Aronsson, K., :o:
art of rhetoric, I,oI,,
Askehave, I., I:8
Atkinson, J.M., :8, :,I
Atkinson, P., :,I
Attardo, S., Io, I,,I,o
attitude, Io:o
attitudinal assessment, :I:,
attitudinal modes, interplay
between, I,:o
attribute, :8
au contraire (on the
contrary), :I,:I
Aubouin, E., I,o
Auer, P., :,,
Authier-Revuz, J., ,, 8,, ,o,:
axiologic adjectives, ,,
B
Bacry, P., Io, Io8
Baker, M., :,I
Bakhtin, M.M., 8o, II,, I:o,
I:, I:,, :o8:I:
Bally, C., ,,8I
Ballys enunciation, ,,,
Ballys work on stylistics, ,
Bamberg, M., I,I, I,,, :oI, :o,
Banfeld, A., :Io, :I8
Bank of English, :,I
Banks, D., ,I
Barsky, R., :oo
Barthes, R., I,:, :,8
Bartlett, F.C., :o,, :,,
basic pragmatic markers, ::,
Basso, K.H., I,
Bates, E., :8o
Bateson, G., :,:
Baudrillard, J., :,,, :,8
Bauman, R., ,, I:I, I,o, I,,
Bauman, Z., I,o
Bazerman, C., I:
Bazzanella, C., :,,, :,,, :,o
Beaufort, A., I:,
Beck, U., :,, :,o
Becker, J., II,
Beeman, W.O., I,, I,o
Behrens, L. II,
Bell, A., :,I
Bell, B., :,,
Bellert, I., :8,
Benke, G., ,I
Benveniste, E., 8I8o, ,,,,
Berge, K-L., I:I
Bergson, H., I,,, I,
Berkenkotter, C., I:
Bertin, A., :,
Bex, T., I:8
Bhatia, V.K., I:o, I,o, :8o
Biber, D., I:,, :,,
Billig, M., ,o
Binsted, K., I,
Bjrklund, M., I:o, I,,, :,
Black, M., II:
Blakemore, D., :,o, :,:
Blass, R., :,o
Blommaert, J., ,o, ,:, ,,, o:, o,
Bloomfeld, L. :8o, :8,
Blum-Kulka, S., I,,, :o:, :,,
Boden, D., :8
Bolinger, D., :o,
Bonnafous, S., oo
Bourdieu, P., ,:, Io,Io8, I,o
Bradford, A.N., Io
Branner, R., I,, I,
Brazil, D., :8o
Bral, M., ,,
Brals pragmatic and historical
perspective, ,,
Briggs, C.L., I:I, :,,
Brinton, L., ::,, ::o, :,I, :,
Briz, A., :,,
298 Index
Brockmeier, J., :o,
Brne, G., I,
Brown, G., ,, ,
Brown, P., II,, ::,, :,,, :oo, :oI
Bruner, J. I,o, I,,
Brunet, E., 8I
Brnner, G., ,I
Brunot, F., ,,, ,,
Bruti, S., ,,
Bublitz, W., ,,8
Butler, J., II
Buttny, R., I,
Buysse, L., :,o
Byrne, R.W., I8,
C
Caf, C., I,o, :,,
Cahn, M., II
Caldas-Coulthard, C.R., :,, ,o
Calhoun, C., :,, :,,
canned jokes, I,
capital, Ioo
Capps, L., :oo, :o,
Carbaugh, D., :o,
Carbonell, J.G., II,
Carlson, L., :,:
Carr, D., I,,
Carroll, J.B., :,,
CARS model, I:o
CDA. see critical discourse
analysis (CDA)
Cedeborg, A., :o:
Celle, A., :,o
Cervoni, J., ,,, ,,
chains of speech
communion, I:o
Channell, J., :o
Character-Did-It
hypothesis, Io
Charaudeau, P., ,:
Charolles, M., :,o, :,I
Chatman, S., I,:
Chaume, F., :,o
cheat-checker, overriding,
I8I8,
cheater detection device, I8,
CHILDES corpus, :,8
Chilton, P., ,, o,, ,:, I,oI88
Chiss, J.-L., ,, ,,
Chlopicki, W., I,8, Io
Chomsky, N. :8o, :8,, :8,
Chouliaraki, L., I:, ,,
Christie, F., I, ,, I::
chrono-logic, I,:
Chvany, C.V., :,
Cicero, M.T., Io:, Io,, III, I,,,
I,o, I,o, I,,
Cicourel, A.V., :,o
CL principles, ,,,,
CL. see critical linguistics (CL)
Clancy, P.M., :8I
Clark, D.D., ,:, ,,
Clayman, S., :,I
Clegg, S. :8
CLIL (Content and Language
Integrated Learning), :,,
Clyne, M., :,,
Coe, R., I:,
co-nonciation, ,
co-enunciation, ,:
co-enunciator, 8,
Cofn, C., I8
cognitive and communicative
role of FSP, III
cognitive functional
grammar, :,,
cognitive linguistics and
humor, I,
coherence and implicitness, :,,
coherence, 8, ,, :,I:,,
coherence, as default
assumption, ,
coherence, defned, ,8
coherence, perspectives, ,8
cohesion as condition for
coherence, o:
Cohesion in English, ,8
cohesion vs. coherence, :,:
cohesion, 8, ,,, ,8, :,o:,I
cohesion, defned, ,,
cohesive chain, :,o
cohesive, I
collective symbolism, o,
College Composition and
Communication, :8o
Collins, J., I,:, I,,
collocations, ,,, 8
Colston H.R., II, I,I
comment clauses, :,
commentary markers, ::,
commodifcation, IoI
common interest, :,
communication, I8,
communicative situation, ,,
computational humor, I8I,
computer-mediated discourse/
communication
(CMD/CMC), :,,
concur, ,I
conjunction types, ,,
connectives, :,:
connectivity, :,
Connerton, P., ,I
Connor, U., :8o
Connors, R.J., Io, Io,
consistency-checking
device, I8,I8,
constitutive heterogeneity, ,o
constitutive intertextuality, I,,
constructions, ::,
context, :o8, :,,, :,,
contextual narrative
analyses, :oo:o
contextual, ,,
contextualization, ,
continuatives, :,I
contrastive pragmatics, :,,
contrastive rhetoric, :,,
contrastive study of pragmatic
markers, :,:,o
contrastive text linguistics, :,,
conventionalized FSP, Io,
conversation analysis
(CA), I,Io, Io:Io,,
:,o, :,,, :,o
conversational humor, IIo
conversational jokes, I,
conversational storytelling, I,,
conversationalization, IoI, :,8
Cook, G., :o,
cooperative principle and
humor, I,8I:
Cope, B. I::, I:,
Coquet, J.-C., ,o
Corbett, E.P.J., Io, Io,
corpus data, :,I
Corpus of London Teenager
language (COLT), :,I
Corum, C., ::,
Coseriu, E., Io,Io,, II
Coulson, S., I,
Coulthard, M., :,, :o8, :8o
Couper-Kuhlen, E., I
Cours de linguistique
gnrale, ,
Courtine, J.-J., o,
CP violation, I,,
CP. see cooperative principle
Crawford, M. I,, I8
Index 299
Crismore, A., :o
Critchley, S., II
critical discourse analysis
(CDA), :, ,oo,, I8o, I88,
:I:, :I,, ::o, :,,
critical linguistics (CL), :, ,oo,
critical linguists, I,,
critical theories, ,I,:
cross-cultural
communication, :8o
Crossley, N., :,,
cross-textual bonding, :,8
Crystal, D., :,,
crystallized dialogue, :II
Cuenca, M.-J., :,,, :,o
Culioli, A.74, 8,88, ,,, ,o
Culioli, enunciation, 8o
Culioli, language use, 88
Culioli, theory of enunciative
operations, 8o
Culpeper, J., :,I, :,
Curc, C., Io, II
D
daemons, I,,
Dailey-Ocain, J. :,:, :,,
Damrad-Frye, D., I,,
Dandeker, C., :,,
Dane, F., :o,
Danon-Boileau, L., ,,, ,
Darwinian metaphor, I8,
Dascal, M., I:
Dasher, R.B., ::,, :,o, :,
Davidsen-Nielsen, N., :,,
Davies, B., :oI
Davies, C., I8
Dawkins, R., I8:I8
de Beaugrande, R., :, I,,, I,8,
:o8, :,o, :,I, :8o:,,
de Cillia, R., o,
De Cock, S., :,o
(de)contextualization, I,:
De Klerk, V., :,:
De Man, P., II
De Palma, P., II
de Saussure, F., Io,
Debatin, B., Io,
Defour, T., :,
Degand, L., :,o
deixis, ,,,
delayed laughter, I
Demol, A., :,,
deontic modalities, ,,
Derks, P., I,
Desmet, P., ,,
developmental pragmatics, :8o
Devitt, A., II,, I:,, I:, I:8
di Luzio, A., :,,
di Pietro, R.J., :8, :8o
diachronic study of pragmatic
markers, :,
dialogic contraction, :8:,
dialogic expansion, :8:,
dialogism and
polyphony, :o,:I:
dialogism, I,8, I,,, :o8:o,
dialogized heteroglossia, :o,
dialogue analysis, :o,, :I:
dichotomies, o
dictum, ,o, ,o
Dines, E.R., :,,
disclaim, :,, ,I,,
discourse analysis, I, I:, o,
Discourse and Society, :o,, :,8
discourse communities, I:,, I:o
discourse formations, o,, I8o
discourse genres, :,:,o
Discourse markers (1987), ::
discourse markers, ::, ::o::,
discourse markers, diferent
usesof, :I:I,
discourse particles, ::,, ::, ::o
Discourse Processes, :o,
Discourse Studies and Text and
Talk, :o,
discourse theory, o,
discourse topic, :,:
discourse types, :,:,o. see
also text type
discourse, I, ,I, o,, o,, 8, 8,,
I8o, :,,, :,o
discourse-historical
approach, oIo:
Discourses in Place: Language in the
Material World (2002), ,8
discursive pragmatics, I, I:
discursive situation, ,,
diskursives Gewimmel
( discursive swarming), o,
distancing, :,
distribution, ,,
Dittmar, N., I:o
Dolan, R. J., I,
Dolitsky, M., I:
Dostie, G., :,o
double enunciation, ,8
double structuration, :,,
double-voicedness, I,8, I,,
Downing, A., :,,
Drescher M., :,:
Dressler, W., I,,, :o8, :,o, :,I, :8,
Drew, P., Io, I,,, ::o, :8, :8o
du Gay, P. :,8
duality of structure, :,,
Dubois, J., ,I, ,
Dubois, S., Io,, Ioo, Io8, IIo,
III, II,
Ducard, D., 8o
Ducrot, enunciation theory, 8,
Ducrot, O., ,,8, 8I, 8,,
8,,o, ,:
Dumarsais, C., Io
Duranti, A., I,:, I,,, :,,
Durrer, S., ,o, ,8
E
Eco, U., IIIII,, II,
Edmondson, W., o, :o8, :8o
Eggins, S., ,, II:, I,
Eggs, E., III
Ehrlich, S., :,
eigentlich, :,:
ellipsis, ,,
embrayeurs, ,,
enacted heteroglossia, Ioo
Encyclopedia Universalis, 8I
endorsing, :,
engagement resources, ,:
engagement, :,,o
English for Specifc Purposes
(ESP), I:,I:,
Enkvist, N.E., Io, :o8:,I,
:,,, :8I
nonc, ,:, ,,, ,:
nonciateur, ,:
nonciation, ,, o,, ,,:
Ensink, T., ,,
entertain, :8, ,o
entextualisation practices, :,,
enunciation linguistics, ,I, ,,
enunciation theory, ,I, 8I
enunciation, ,:, 8I, 8:, ,:, :,I
enunciative (nonciatif, -ive), ,:
enunciative dedoubling, ,:
enunciative heterogeneity, ,o
enunciative polyphony, 8,
enunciative pragmatics, ,I, ,
enunciative, 8,, ,:
enunciatives, :,,
300 Index
enunciator, ,:, ,:, :I:
epidemiology, I8:
episodic structure, I,
epistemic modalities, ,,
epistolary communication, ::o
equivalents, :,,
Eriksson, M., :,,
Erman, B., ::, :,o, :,,, :,, :,,
Ervin-Tripp, S., I,, :8o
ESP. see English for Specifc
Purposes (ESP)
Essai de la smantique, ,,
ethnic group, I8
ethnicity and humor, I8
evaluative adjectives, ,,
evaluative language, I,I8
Evaluative Semantics, I,
Evaluative Units, I8
Everts, E., I,
evidential markers, ::8
exemplum, I,
expert/lay knowledge
systems, :,:,o
exposition, I:,
extended enunciation, ,:
extension particles, :,,
F
family humor style, I,
Fanshel, D., :8
Fass, D., II,
Fay, B., ,I
Feak, C., I:,
Featherstone, M., :,,
Feez, S., I
Fein, O., I:
Fernandez, J.M.M., ::, :,, :,I
Fernndez-Villanueva, M., :,
Ferrara, K., :,,, :,8
Ferrer, H., :,,
Ferro-Luzzi, G.E., I,,
Fetzer, A., :,8
Feyaerts, K., I,
fgure and ground, :,
fgure, :,
fgures of diction, Io,
fgures of rhetoric, Io:
fgures of speech (FSP), 8, Io:.
see also fgures of rhetoric;
rhetorical fgures
fgures of thought, Io,, III
fller, ::o
Filliettaz, I., :I,, :,:
Fillmore, C.J., I,,, I,
Finell, A., :,
Firbas, J., o, :o,
Fischer, K., ::,, ::, :,o
Fisher, S., :8
Fleischman, S., I,o, :oo, :,,, :,,
Flowerdew, J., ,,, I:o
Fludernik, M., I,:
Fogelin, R.J., II,
Fontaine, J., 8I
Fontanier, P., Io, Ioo
Foolen, A., :,,
footing, :oI, :,,
Forabosco, G., I,o, I,8
Forns, J., :,, :,,
Foucauldian
poststructuralism, ,8,
Foucault, M., Io, II, o,, o, 8o,
8I, I,8, :,,, :,,
Foucaults archaeological
approach, Io
Fowler, R., ,,, I,,, :o,, :I:, :,I
frames, ,, I,,
Francis, G., ,,, I,:
Frandsen, F., I:,
Frank-Job, B., :,:
Fraser, B., ::, ::,, ::8, :,,
Fraser, N., :,,, :,o
Freedle, R., :,:
Freedman, A., I:,, I:,
Freeman, D.C., Io, II
French enunciation theory, ,, ,I
French pragmatics, ,I
Fretheim, T., ::, ::,, :,
Freud, S., II,, I,,, I,o
Fritz, G., ,, 8
Frow, J., II,, I:o, I:8
frozen stylistics, Io,
FSP, classes of, Io,, Io, Io8III
FSP, cognitive and
communicative role of, III
FSP, defnition of, IoIo,
FSP, deviation theory, Io,
FSP, linguistic operations, Io,
FSP. see fgures of speech (FSP)
Fuchs, C., ,I, ,:
Fuller, G., I,
Fuller, J.M., :,,
functional conversational
analyses, II,
functional sentence
perspective, ,,
fundamental kinesis
hypothesis, ,,
G
Gabriel, Y., :,,
Gandy, O.H., :oo
Garca Vizcaino, M.J., :,,
Garfnkel, H., ,, :,o
Gasquet-Cyrus, M., I,
Gee, J.P., ,:, I,,
Geluykens, R., ,
gender diferences and
humor, I,I8
general theory of
enunciation, ,,
General Teory of Verbal
Humor (GTVH), I,,
generative grammar, I,,
Genette, G., I,:, :,8
Genre Analysis, I:,
genre, o, II,I,I, :,o
Georgakopoulou, A., ,, I,o:o
geosemiotics, ,8
Gerhardt, J., :8I
Gernsbacher, M.A., ,
Gibbs, R.W.J., Io,, II:, II,,
II, I:
Giddens, A., ,, :,,, :,,
Gilquin, G., :,,
Giora, R., II, I:
Givn, T., ,, :,o, :,I, :,,
Glenn, P., I,, I
Gliederungssignale, ::,
Glynos, J., II, I:
Goatly, A., II,, II
Goddard, C., :,o
Goel, V., I,
Gofman and public order,
:,::,,
Gofman, E., Io, :oI, :8, :,:,
:,,, :oI
Gonzlez, M., :,o
Goodwin, C., :,,
Goodwin, M.H. :oI
Goutsos, D.191, :oI
Gouvard, J.-M., ,o
Graefen, G., ,I
Graesser, A. C., I,
Graham, E., I,
Graham, P., ,:, ,,
Grammaire gnrale et raisonne
(1660), ,,
grammaticalization theory, ::,
grammaticalization, :,,
Index 301
Gramsci, A., :,
Greasley, P., :,:
Greatbatch, D., :,I
Greimas, A.J., I,o
Grice and Gricean analyses, I,8
Grice, H.P., II,, I,8, I,,
Grimes, J.E., :o,, :o8, :,:
Groarke, L., II,
Grobet, A., II,, :,:
ground, :,
grounding, :,,:,
Groupe Relpred, ,:, ,,
Groussier, M.-L., 8o, 8,
Gruber, H., I8, I:I
Guillaume, G., ,,, ,, ,88o
Guillaumes psychomechanic
model, 8o
Glich, E., ::,, ::,, :8,
Gumperz, J.J., ,, ,, :o8, :,,
Gunter, R., ,,, :8,
Gupta, A.F., :,:
H
Habermas and public
sphere, :,,:,o
habitus, Ioo, Io,
Haiman, J., :,,
Hakulinen, A., :,,, :,
Handbook of discourse
analysis, :o8
Hanks, W., I,:
Hansard report, I,I
Hansen, M.-B.M., ::, ::o,
::8, :,,
Harr, R., :oI
Harris, S., :,I,
Harris, Z., :o,, :,o
Hartley, J., :,I
Hartmann, P., :88, :8,, :,,
Harweg, R., :88, :,o
Hasan, R., ,8, ,,, , ,, I::,
I:8, :,o, :8,
Hasselgrd, H., :,,
Hasselgren, A., :,,
Hatim, B., I,:
Hausendof, H., :oo
Hausmann, F.J., I8
Hay, J., I,, Io
Heath, S., I,,
Hebrew pragmatic markers, :,,
Heer , ,,, oI
hegemony theory, IoI
Heller, M., o,
Hellman, C., ,
Helsinki Corpus, :,8, :8I
Heritage, J., ,I, 8,, Io,, :,I, :8o
Herman, D., I,
Herring, S.C., :,,
hesitation marker, ::o
heteroglossia, I,8, Ioo
Hill, J., :oI
Hirst, G., ,,
Hodge, R., ,,, I,,
Hoey, M., :o8, :,o, :,8, :8o
Hlker, K., ::,, ::o
Holmes, J., I, Io, I8, I,,
:,:, :,,
homonymy, ::8
Hopper, P.J., :,
Hopper, R., II,
Horne, M., :,8
Hovy, E., :,8
Howarth, D., II, I:
Hudson, R., :,,
Hulstijn J., I,
human communication, I8,
humor as non-bona-fde
communication, I,8I,,
humor, I,,I,
Hunston, S., I,, :,
Hunter, K.M., :o:
Huspek, M., :,,
Hutchby, I., :,I
hybrid construction, :Io
hybridization, IoI
hybrids, I,o
Hyland, K.20, 129
Hymes, D., ,, II,, I:o, I,,
Hyon, S., I:I
hyper-syntax, :8,
hysteronproteron, Io,
I
ICLE (International Corpus of
Learner English), :,8
iconicity, ,,, o, :,,
ideal discourse
structuralism, :I:
ideal speech situation, :,,
idologme, Ioo
ideology, o,, I,,
Iedema, R., I, I,, :,, :, ,
Ifantidou, E., :,o
Ilie, C., III
implicative modalities, ,8
implicitness, :,,
incongruity, I,o
incongruity-resolution (IR)
model, I,o
indexicality, 8, II, ,8
inference, :,,
information assessment, ,,
information structure, :o,:,o
informativeness approach,
III:
Ingsoc, I,8
instance d vnement
nonciatif, 8I
instance du discours, 8,
integrated pragmatics, 88
interaction-based approach of
public discourse, :,I
interdiscourse, o,
inter-discursivity, I,,
interlanguage pragmatics, :,o
internal dialogization of
discourse, :II
internally dialogized hybrid, :I,
International Corpus of
English-Great-Britain
(ICE-GB), :,I
interpretative repertoires, Io
interpretive use, :,,
(inter)subjectifcation, :,o, :,,
inter-subjective modalities, ,8
intersubjective stance, :o:I
intertextual analysis, I,8I,,
intertextualities as
contextualities, I,:
intertextuality and
hegemony, IoI
intertextuality, ,, I,oI,,
:o,, :,8
intimacy (re)afrmation, I,,
involvement vs.
detachment, :,,
irony, :I,
Isaenko, A., :8,
Isenberg, H., :88
isotopy, I,o, I,,
isotopy-disjunction model
(IDM), I,oI,,
J
Jafe, S., II,
Jger, S., ,I, o,
Jakobson, R., 8o, ,:, ,,, Io,
Io,, IIo, II
Janney, R.W., I,o, :,,
302 Index
Januschek, F., o
JAPE, I,
Jaspers, J., Io
Jeferson, G., I, I,,
Jenkins, M.M., I,
Jessop, B., ,,
Jodlowiec, M., Io
Johansson, M., ,, ,, ,I ,8
Johansson, S., :,,
Johns, A.M., :8o
Johnson, M. Ioo, II:, II,
Johnson-Laird, P., I,
Johnstone, B., :oo
Joly, A., ,,, ,8, 8o
Josselson, R., I,:
Journal of Historical
Pragmatics, :8I
Journal of Language and
Politics, :,8
Journal of Pragmatics, I,
Jucker, A., ::, ::o, ::,, :,o,
:,:, :,
judgement & appreciation as
institutionalized afect, :o
judgement, Io
justifcation, I,o
K
Kalantzis, M., I::, I:,
Kallmeyer, W., :,:
Kaplan, R.B., :,,, :8o
Keat, R., :,,, :,8
Keith-Spiegel, P., I,,
Kennedy, J., II,
Kennedy, V., II,
Kerbrat-Orecchioni, C.,
,:, ,,, ,,
Kerby, A., I,o
Kienpointner, M., Io:II,
Kinneavy, J.L., :,o
Kintsch, W., II,, :8o, :,:
Kittay, E.F., Io,, II:
Kleiber, G., ,,
Klemperer, V., o
Kloepfer, R., Io,
Knape, J., Io:, Io,, III
Knight, C., I8o, I8
Knight, D., :,:
Knorr-Cetina, K., :,o
Koller, V., ,:
Kopperschmidt, J., II
Krner, H., I,
Kotschi, T., ::,
Kotsinas, U.-B., :,o, :,,, :,
Kotthof, H., I,, I,
Kvecses, Z., Ioo, Io,, II:II
Krebs, J.R., I8:, I8
Kress, G., I:, ,,, ,,, ,o, ,8, I::,
I:,, I,o
Krings, H., ,I
Kristeva, J., I,o, I,,, I,8, Ioo,
:o,, ::I
Kroon, C., ::8
Krzyanowski, M., oI, o:
Kutas, M., I,
Kyt, M., :,I, :,, :8I
L
L appareil formel de
l nonciation, 8:
La nature des pronoms, 8:
La pense et la langue, ,,
Labov, W., :o, I,,, I,, :8, :,o,
:,I, :o8, :,,
Labovs study of narrative,
I,I,,
Lacan, J., ,o, II,
Laclau, E., II, I:
Lakof, G., Ioo, II:, II,
Lakof, R., I,I
Lala, M.-C., ,
Lambrecht, K., :,o
Lampert, M.D., I,
Lamy, B., Io
Lancelot, C., ,,
Lang, T., :,,
Lange, D., :,:
Language and Control (1979), I,,
language and thought, I,8
Language as Ideology (1979,
1993), 179
Language in New
Capitalism, ,,
Language of New Labour, ,,
Languages in Contrast, :,,
langue, ,, Ioo, :8o
langue/parole, 8I, :,I
Lausberg, H., Io
Le marxisme et la philosophie
du language (1977), 8o
Le Querler, N., ,,, ,8
learning, ,,
Ledin, P., I:I
Lee, D., I8o
Leech, G.N., ,:, Io,, Io8, II,
legitimate language, :,,
Lehti-Eklund, H., :,,
Lemke, J., ,I, ,:, ,,, Io,
Lenk, U., ,,, ::
Leontjev, A.N., o,
Ler, S.L.V., :,o
Les feurs du mal, I,I
Lesarten approach, o
Lesweisenanalyse, o
Levelt, W.J.M., II,
Levinson, S.C., I:, II,, I:I, :,,,
:oo, :oI
Lvi-Strauss, C., II
Levy, E.T., ,
Lewandowska-Tomaszczyk, II,
Lewis, D., ::o, :,,, :,8
lexical cohesion, ,,
lexical discourse
markers, :,,
lexico-grammatical
description, I::
lexicometry, o:o,
Lhomme dans la langue (Man in
language), 8I
Liao, Chao-Chih, I,
Liddle, M., 8o
Linde, C., ,, I,:
LINDSEI Corpus, :,o
Lindstrm, J., :,,
Linell, P., I:I
linguiste smioticien, 8I
linguistic analysis, I,8
linguistic communication, I8,
linguistic determinism, I,8
linguistic habitus, Io8
linguistic manipulation, I,8
linguistic market, Ioo
linguistic signs, 8,
linguistic structures, ,, I8o
linguistic/discursive turn, ,
Linguistique gnrale et
linguistique franaise
(1932), ,,
Link, J., o,
Linn of Rhetoric, Io
literary narrative, I,:
Little Dorrit, :II
Llinares-Garca, A., :,,
Locke, J., III, II:, I,,
Loehr, D., I,
logico-rhetorical module,
I8,, I8,
London-Lund Corpus, :,I, :,8
Long, D. L., I,
Index 303
Longacre, R.E., I,,, :o,, :o8, :,o
longer texts, I,8
Louhiala-Salminen, L., I:o
Louw, B., I,
Luke, A., I:,
Lundquist, L., :o8
Lyons, J., :I
M
Maas, U., o
Macaulay, R., :,,
machiavellian
manipulation, I8o
Macintyre, A., I,,
Macken, M., ,, I:,
Macken-Horarik, M.
macroproposition, :,:
macrostructure, :,:
macro-syntax, :8,, :8,
Maingueneau, D., o,, ,,, :o,
mais (but), :I
Malmaud-Makowski, M., :,,
Malrieu, J.P., I,, I8
manifest intertextuality, I,,
manipulation, I,oI88
Mann, W.C., Io,, :,:
Maranho, T., I,o
Marchart, O., II
marked informativeness, II
market, Ioo
Markkanen, R., :o, :I
Marksizm i flosofja
jazyka, :o,
Marnette, S., ,
Marouzeau, J., Ioo
Marra, M., Io
Martn Zorraquino, M.A., :,,
Martin, B., :,,, :,8
Martin, J.R., I, I,, Io, :o, ::, :,,
,, ,,, ,:, ,,, I::, I:,, I:,,
I:,, I,o
Martinez-Cabeza, M.A., :,,
Maschler, Y., :,,
mass media, I,,
Matamala, A., :,o
Matouschek, B., o:
Matthiessen, C.M., I
Mauranen, A., :,I
Maybin, J., Io, Io,
McCloskey, D.N., Io,
McGhee, P., I,o
McNeill, D., ,
McTear, M., :8o
meaning, I,,, :I,
meaning-making, o, :oo
mconnaissance, ,:
Mediated Discourse: Te Nexus
of Practice (2001), ,8
mediatization, :,,
Medway, P., I:,, I,o
Meeuwis, M., o,
meme, I8:, I8,
memetics, I8,
Mertz, E., I,,
metaphor, IIo
metarepresentations, I8,
Meunier, A., ,o
Mey, J., I,, ,:, o,
Meyer, M., ,o
Meyer, P.G., :o
MICASE corpus, :,I
Michaels, S., I,,
micro/macro social order, :,o
Miller, C.R., I:
Miller, J., :,:
Mishler, E., I,I, :o:, :8
Mittwoch, A., ::,
modal particles, :,,
modal subject, ,o
modalities of enunciation, ,o
modality, ,,o, ,,, ,o
modalization, ,,o, ,o
modus, ,o
Moerman, M., I,I
Moeschler J., 8,
Monk, L., I:o, I:,
monosemy, ::8
Morain, G.G., I8
Morel, M.-A., ,,
Morin, V., I,o
Morra, L., :,,, :,o
Morreall, J., I,,
Morris, C.W., :8I
Morris, J., ,,
Mulkay, M., I,, ::o
Mller, S., :,o
multimodality, ,o
Mumby, D.K., :o:
Muntigl, P., ,:, o:, I:I
Murphy, J.J., Io,
Myers, G., :o, I:,
N
Nardini, G., I,
narrative analysis, I,I
narrative and reality, I,,
narrative as mode of
communication, I,oI,:
narrative canon, :o,
narrative evaluation, I,o:oo
narrative function, I,o, I,,
narrative mode, I,o
narrative organization, I,,I,o
narrative orthodoxy, :o,
narrative performance, I,o
narrative texts, I,o
narrative, ,, I:,, I,o:o
Nash, W., Io
National Socialist language, o
natural histories of discourse, I,:
Natural Semantic
Metalanguage, :,o
Nava, M., :,,
Navon, D., I:
NBF modes, I,,
NBF. see non-bona-fde (NBF)
negation and negative
polarity, :I,:I
negotiation, :,,
Negrine, R., :,o:,8
Nerlich, B., ,:, ,,
new rhetoric, ,, I:,I:,
Newspeak, I,8
nexus analysis, ,8
Nichols, J., :I, :I
Nietzsche, F., II,
Nijholt, A., I,
Nikula, T., :,o
Nilsen, D.L.F., I8
Nineteen eighty-four, I,8, I8,
Nlke, H., ,o
nominalization, I,,
non-bona-fde (NBF), I,8
Normand, C., 8o, 8I, 8,, 8,
Norrick, N.R., I,, I, I,,
Io, I,
nouns, ,,
Nunan, D., :o,
O
OBarr, W., :8
OHalloran, K.L., I:
OKeefe, D.J., I8,
Oakley, T., I,
Oberhuber, F., o:
objective adjectives, ,,
objective modalities, ,8. see also
implicative modalities
Ochs, E., Io, I,I, :o:, :o,, :,:,
304 Index
:,,, :,,, :8o
Olbrechts-Tyteca, L., II:
Ong, W.J., II,, :,,
operative time, ,8
orders of discourse, ,8,, IoI
Orlikowski, W.J., I:
Orwell, G., I,8, I,,
Osgood, C.E., :,
stman, J.-O., I,, ::,::o, ::,,
:oo:8:
Ottesj, C., :,,
oxymoron/paradoxon, Io,
P
Pagano, A., :o, ,:
Palek, B., :8,
Palmer, F.R., :I
Paltridge, B., II,, I:,
paradigmatic (logicosemantic)
mode, I,o
parallel markers, ::,, ::8
parole, ,, 8I, ,o, I,o, Io:, :8o
paronymy, I,o
participant roles, :,,
passivization, I,,
PAULINE, :,8
Pear Stories, I,,
Pcheux, M., ,:, ,,, o,
Pelinka, A., oI
Perelman, C., II:
persuasion, I,o
Petf, J.S., :,o, :,I
phrase, 8,
Pielenz, M., Io,, II:
Pilegaard, M., I:,
place semiotics, ,8
Plato, I,,
Plett, H.F., Io,, Io8III, II
Plum, G.A., I
Polanyi, L., I,,, :oo, :,,
political lexicometry, o,
polyphonic organization of
discourse, :I,:I8
polyphonic utterances, :Io
polyphony, , I,8, I,,, :o8::I,
:o,, :Io, :I,
polysemy, ::8
Pomerantz, A., :,o
ponctuants, :,,
Pons Bordera, S., :,,
Portols, J., :,,
post-structuralist discourse
theory, Io
Potter, J., Io
pour autant, :I
power behind discourse, I8o
power in discourse, I8o, I8,
power, ,:
pragmatic markers, 8, ::,:,,
see also ller; hesitation
marker; discourse markers;
discourse particle;
pragmatic particle
pragmatic markers and
future, :,8:,,
pragmatic markers and
sociolinguistic
aspects, :,,:,8
pragmatic markers in languages
of world, :,::,,
pragmatic markers in native
vs. non-native speaker
communication, :,o:,,
pragmatic markers in translation
studies, :,o
pragmatic markers,
classifcation, ::,::8
pragmatic particle, ::o
Pragmatic particles twenty years
afer, ::,
pragmatic theories of information
exchange, :oo:oI
pragmaticalization, :,o, :,,
preface, I,
presupposition, :I,
Priego Valverde, B., I,, I,
prima facie, I,,
Prince, E., :,o
Prince, G., I,:
principles of textuality, :,,
private/public dichotomy, :,
Problmes de linguistique
gnrale I-II (1966, 1970),
,, 8I
Problemy tvorchestva
Dostoevskogo, :o,
process models, :,,
process writing, :8o
proclaim, :8
pronounce, ,o,I
pronouncements, ,o
Propp, V., :,8
prosody, ,,,8, :,8
prospective devices, ,,
pseudo-coherent, :,I
public discourse, o, :8
public sphere, :,,, :,
publicness, o, :8, :,:,o, :,,
Puech, C., ,,
Purves, A.C., :,,
Pusch, C.D., :,,
Q
Quaderni di Semantica, I,,
quantitative conversational
analyses, I,Io
quantity, ,,
Quasthof, U., :oo
Quintilian, M.F. Io:, Io,, Io,,
I,,, I,o
R
Rampton, B., Io,
Raskin, V., I,,I,,, I8, I,
reading analysis, o
(re)contextualization/
entextualization, I,oI,
Redeker, G., ::,
reference, ,8
referential humor, I,,I,o
referential narrative
analyses, I,:I,,
Rgent, O., :,,
registers, II, :,,
Reid, I., I::
Reinhart, T., :oo
Reinsdorf, W., II,
Reisigl, M., ,,, oI
Relations de temps dans le verbe
franais, 8
relevance theory, II, I,,II,
I8:, I8, :,o
relevance-theoretic approaches
to humor, I,,II
relevance-theoretical
principles, ::,
Renkema, J., ,o, :o,
repertoires, I:
replicability, :,,
reported speech, ,
representational resources, ,,
resolution phase, I,o
response, I
restricted enunciation, ,:
retrospective cohesive, ,,
Rhetorica ad Herennium, Io,
rhetorical fgures, Io:
Rhetorical Genre Studies, I:,
rhetorical manuals, I,o
rhetorical structure theory, :,:
Richardson, J., I:,
Rickford, J., o,
Index 305
Ricoeur, P., Io,Io,, III, II:
Rimmon-Kenan, S., I,:
Ritchie, G., I,
Romaine, S., :,:
Romero-Trillo, J., :,o, :,,
Rosch, E.H., II
Rose, D., I::, I:,, I:,, I,o
Rosier, L., ,
Rossari, C., ::,, :,:
Rothery, J., I,, I::
Roulet, E., :o8::o, :,:
Roulland, D., ,8, 8o
Ruch, W., I,8
Rhlemann, C., ::,
rule-strengthening
deviation, Io8
rule-violating deviation, Io8
Rutter, J., I,
S
Saari, M., :,,
Sacks, H., :,
Samet, J., :, ,
Samraj, B., I:o, I:,
Sanders, T., ,
Sandig, B., III
Santa Barbara Corpus, :,8
Sapir, E., I,8
Sapir-Whorf hypothesis, I,8
Sarangi, S., :8:oo
Sauer, C., ,,, o
Sayer, A., ,,
Scannell, P., :,,
Schaefer, J.-M., ,8
Schfner, C., I8I
Schank, R., :, ,
Scheglof, E., 8, I,,
Scheglof, E.A., I,, I, :,o
schemata, ,, I,,
Schiefelin, B.B., :,:, :8o
Schifrin, D., ,I, :oo, :oI, ::,
::o, ::,, :,:
Schlesinger, A.M., :,:
Schmidt, S.J., :88, :,o, :,:
Schmitz, A., II,
Schourup, L.C., ::, :,:
Schrder, H., :o, :I
Schryer, C., I:
Schwenter, S., ::,, :,,, :,
scintillating, :,o
Scollon, R., I:, ,,, ,8
Scollon, S., ,,
script-based semantic theory of
humor, I,,I,8
scripts, ,, I,,
Searle, J., II,
Sebba, M., :,:
self-talk, :,:
semantic connectivity ,
semantic mirrors, :,,
semantic prosodies, I,, I8
Semantic Script Teory of
Humor (SSTH), I,,
semantics, I,oI,8
semiotic principles, Io,
Senf, G, Io
Sennett, R., :,
sentence linguistics, :88
sentence, :8,, :8,
Seppnen, E.-L., :,
sequential implicativeness,
Ioo, Io,
Sexl, M., II
SFL principles, ,
Sherzer, J., ,, I,, I, I8,
I,,, :,,
shifers, 8o, ,,
Shloush, S., :,,
Short, M., I,:
shown heterogeneity, ,o
Shuman, A., :oI
Sidnell, J., I:
Sifanou, M., Io
SIGET, I:I
signifcation, Ioo
Silverman, D., :8
Silverstein, M., I,:, I,,, :,,
Simon-Vandenbergen, A.-M.,
8, ::,:,,
Sinclair, J.M., I,, I,, , :o8,
:8o, :,o, :,I, :,,
situation/context, ,:,,
situation-talk dialectic, :,o:,I
Slade, D., ,, I,
Slembrouck, S., I,oI,
Slovo v romane, :Io
Smart, G., I:,
Smith, S.W., :,
Sobkowiak, W., I,o, I8
sociability, I,,
social functions, I
social interaction, :,:
social semiotics, ,,,8
social situations, :,:
socio-cognitive model, ooI
sociolinguistic contact, Io,
sociolinguistic legitimacy,
Io,I,o
(socio)linguistic markers of
public discourse, :,o:,I
sociolinguistics of humor,
IoI8
socio-political theory, Io
Sorjonen, M.-L., :,,
speaker, ,:, ,:, :I:
speaker-hearer
communication, I,8
speaking subject, :I:
speech act adverbials, :,,
speech act theory, ,,, ::,
speech and writing, :,o
speech genres, I:o
speech-act model, :,,
Spencer, J.R., II,
Sperber, D., II,, II, I,,, Io,
I8:I8, :,o, :oo, :oI, :,:
Sperry, L., I,
Spillner, B., Io,, IIo, II
Spitzer, L., Io,
split personality, ,,
split subject, 8,
Spooren, W., ,
Springfellow, F., II,
stacked humor, Io
standards of textuality, I,,
Stavrakakis, Y., II
Stehr, N., :,
Stein, N., I,
Steinbrink, B., Io
Steinitz, R., :88
Stenglin, M., I,
Stenstrm, A.-B., ::,, ::8, :,I,
:,:, :,,
stereotype, I8
Stock, O., ,, I,
story schema, I,
Strapparava, C., I,
strategic communicative
action, I8I
Stubbe, M., Io, :,,
Stubbs, M., :o8, :,o
stylistics, ,
subjective adjectives, ,,
subjective modalities, ,8
subjectivity, ,,,
substitution, ,8
sujet de l nonciation, 8,
Suls, J., I,o, I:
surprise efect, II
Svartvik, J., :,I, :,:
Swales, J. I:,I:8
Swiggers, P., ,,
306 Index
Sydney genre theory,
I::, I:,, I:o
Sydney School, I:II:,
syntax, :88, :,,
systemic functional linguistics
(SFL), ,, I:I, :,o
systemic functional
linguistics, I
T
Tagliamonte, S., :,,
Takahara, P.O., :,,
Tannen, D., I,I,
Tardy, C.M., I:
Tate, S., I:
tautology, Io,
Taylor, C., I,I, :o,
TDL and narratology, :,8
TDL applications, :,,:8I
TDL. see text and discourse
linguistics (TDL)
tellers positioning, :oI:o:
teller-tale-telling, I,,
telling, I,I
text and discourse linguistics
(TDL), :oo:8:
text form variants, :,o
text grammar, :8,, :,,
text linguistics, :, I,,I,8, :oo,
:8o:,,
text types, :,,, :,o
text, ,, ,I, I,o, :88, :,o
text-syntax, :8,, :88
textual islet (lot textuel), ,I
textual narrative analyses,
I,,:oo
textual prosody, ,8
textualisation, :,,
the carnivalesque, I,8, Io,
Te Greek particles (1934), ::,
Te Laws of Etiquette, :,:
the neglected situation, :,:
Te Selfsh Gene, I8,
thematic progression, :8,
theme and rheme, :o,:,o
theory of argumentation, 88
theory of argumentative
scales, 88
theory of communicative
action, :,,
theory of enunciation, ,,, 8I
theory of linguistic practice, Ioo
theory of semantic blocks, 88
Tiesmeyer, L., ,,
Tomas, J., :,I, :,,
Tompson, J.B., ,I, :,o, :,,
Tompson, S.A., :,:, :,
thought and social action,
I8II8
thought control, I,,I,8
Titscher, S., o
Todd, A.D., :8
Todorov, T., Io,, Io8, III,
II,, :o8
Toolan, M., :oo
Torfng, J., I:
Tournier, M., o,
Trabasso, I,
transcendental
subject, 8o
transformation of public
sphere, :,o:oo
Traugott, E.C., ::,, :,o,
:,, :8I
Travis, C.E., :,o
Trew, T., I,,
tropes, Io,
Tsakona, V., I,8
Tsui, A., :,o
Tucker, G., :8
two-stage processing of
humor, I:
typological criteria,
I8I,
U
Ueding, G., Io
unexpectedness, I,,
unintentional humor, I:
unitary meaning of pragmatic
markers, ::,
Urban, G., I,, I,:, I,,
utterance modalities, ,o,8
utterance situation, ,,
utterance, ,,, 8,, ,:, Io,
uttered enunciation, :,:
utterer, ,:
V
Vaid, J., I:
Valette, M., ,, ,8, ,,, 8o
values of appreciation, :o
Van Bogaert, J., :,I
van der Geest, T., :,I
van Dijk, T.A., :,I
van Eemeren, F.H., II:
van Hecke, T., 8I
van Leeuwen, T., I:, ,,, ,o, ,8, I,o
variedness-in-language, Io,
Vask, I., :,
Ventola, E., I::
verbal humor, I,,I,o
verbs, ,,
verisimilitude
( vraisemblance), I,o
Verschueren, J., o:, Ioo, :,,
Viana, A., I,
Vincent, D., ,, :,,
violence symbolique, ,:
Vion, R., ,,, ,,, ,o, ,8
Virtanen, T., :, I,I, :oo:8:
Voloinov, V.I., ,:, ,,, I,8,
Io:Io,, I,:
Voloinov, V.N., :I, :o,
Vuchinich, S.,
W
Walker, T., :,I
Wrvik, B., :,, :8I
Watts, R., :,o, :,:
Waugh, L.R., ,, II
Weber, M., :,,
Webster, F., :,,
Weiner, E.J., I,, I:
Weinert, R., :,:
Weinrich, H., II,, :88, :,:
Weiss, G., o:
Wenzel, P., o:
Werlich, E., I:,, :,o
Wernick, A., :,,, :,,
Westheide, H., :,,
Wetherell, M., Io
Weydt, H., Io,
White, P.R.R., I,, oo
Whiteley, N., :,8
Whiten, A., I8,
Whorf, B.L., I,8
Wichmann, A., :,8
Widdowson, H.G., :oo,
:,I, :8o
Wide, C., :,,
Wierzbicka, A., :,o
Willems, D., :,, :,,
Williams, G., o,
Williams, R., :,o
Wilson, D., o, II,, II, I,,, Io,
I8:, :,I, :oo
Wilson, J., :,I
Wolfson, N., I,o
Index 307
Woodbury, A., I,,
Woolard, K., I,,
world-knowledge, :,,
Written Communication, :8o
Yaguello, M., :,,
Yamaguchi, H., I,,, Io
Yates, J.A., I:
Young, K.G., I8,, I,o
Yu, N., II
Yule, G., ,, ,, :o8, :,:
Yus, F., Io
Zac, L., II
Zajdman, A., I,
Ziv, A., I,o, I8
Ziv, Y., ::o, :,,

Discursive Pragmatics Handbook of Pragmatics - Highlights - 2011

Uploaded by

Copyright:

Available Formats

Discursive Pragmatics Handbook of Pragmatics - Highlights - 2011

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Discursive Pragmatics Handbook of Pragmatics - Highlights - 2011

Uploaded by

Copyright:

Available Formats

Discursive Pragmatics

You might also like