You are on page 1of 1450

Foundations of Pragmatics

HoPs 1

Handbooks of Pragmatics

Editors

Wolfram Bublitz
Andreas H. Jucker
Klaus P. Schneider
Volume 1

De Gruyter Mouton

Foundations of Pragmatics

Edited by

Wolfram Bublitz
Neal R. Norrick

De Gruyter Mouton

ISBN 978-3-11-021425-3
e-ISBN 978-3-11-021426-0
Library of Congress Cataloging-in-Publication Data
Foundations of pragmatics / edited by Wolfram Bublitz, Neal R.
Norrick.
p. cm. (Handbook of pragmatics; 1)
Includes bibliographical references and index.
ISBN 978-3-11-021425-3 (alk. paper)
1. Pragmatics. I. Bublitz, Wolfram. II. Norrick, Neal R.
P99.4.P72F68 2011
4011.45dc22
2011013980

Bibliographic information published by the Deutsche Nationalbibliothek


The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie;
detailed bibliographic data are available in the Internet at http://dnb.d-nb.de.
2011 Walter de Gruyter GmbH & Co. KG, Berlin/Boston
Cover image: Vetta Collection/iStockphoto
Typesetting: Drlemann Satz GmbH & Co. KG, Lemfrde
Printing: Hubert & Co. GmbH & Co. KG, Gttingen
Printed on acid-free paper
Printed in Germany
www.degruyter.com

Preface to the handbook series


Wolfram Bublitz, Andreas H. Jucker and Klaus P. Schneider
The series Handbooks of Pragmatics, which comprises nine self-contained volumes, provides a comprehensive overview of the entire field of pragmatics. It is
meant to reflect the substantial and wide-ranging significance of pragmatics as a
genuinely multi- and transdisciplinary field for nearly all areas of language description, and also to account for its remarkable and continuously rising popularity
in linguistics and adjoining disciplines.
All nine handbooks share the same wide understanding of pragmatics as the
scientific study of all aspects of linguistic behaviour. Its purview includes patterns
of linguistic actions, language functions, types of inferences, principles of communication, frames of knowledge, attitude and belief, as well as organisational
principles of text and discourse. Pragmatics deals with meaning-in-context, which
for analytical purposes can be viewed from different perspectives (that of the
speaker, the recipient, the analyst, etc.). It bridges the gap between the system side
of language and the use side, and relates both of them at the same time. Unlike syntax, semantics, sociolinguistics and other linguistic disciplines, pragmatics is defined by its point of view more than by its objects of investigation. The former precedes (actually creates) the latter. Researchers in pragmatics work in all areas of
linguistics (and beyond), but from a distinctive perspective that makes their work
pragmatic and leads to new findings and to reinterpretations of old findings. The
focal point of pragmatics (from the Greek prgma ,act) is linguistic action (and
inter-action): it is the hub around which all accounts in these handbooks revolve.
Despite its roots in philosophy, classical rhetorical tradition and stylistics, pragmatics is a relatively recent discipline within linguistics. C.S. Peirce and C. Morris
introduced pragmatics into semiotics early in the twentieth century. But it was not
until the late 1960s and early 1970s that linguists took note of the term and began
referring to performance phenomena and, subsequently, to ideas developed and advanced by Wittgenstein, Ryle, Austin and other ordinary language philosophers.
Since the ensuing pragmatic turn, pragmatics has developed more rapidly and diversely than any other linguistic discipline.
The series is characertised by two general objectives. Firstly, it sets out to reflect the field by presenting in-depth articles covering the central and multifarious
theories and methodological approaches as well as core concepts and topics characteristic of pragmatics as the analysis of language use in social contexts. All articles are both state of the art reviews and critical evaluations of their topic in the
light of recent developments. Secondly, while we accept its extraordinary complexity and diversity (which we consider a decided asset), we suggest a definite
structure, which gives coherence to the entire field of pragmatics and provides

vi

Wolfram Bublitz, Andreas H. Jucker and Klaus P. Schneider

orientation to the user of these handbooks. The series specifically pursues the following aims:
it operates with a wide conception of pragmatics, dealing with approaches that
are traditional and contemporary, linguistic and philosophical, social and cultural, text- and context-based, as well as diachronic and synchronic;
it views pragmatics from both theoretical and applied perspectives;
it reflects the state of the art in a comprehensive and coherent way, providing a
systematic overview of past, present and possible future developments;
it describes theoretical paradigms, methodological accounts and a large
number and variety of topical areas comprehensively yet concisely;
it is organised in a principled fashion reflecting our understanding of the structure of the field, with entries appearing in conceptually related groups;
it serves as a comprehensive, reliable, authoritative guide to the central issues
in pragmatics;
it is internationally oriented, meeting the needs of the international pragmatic
community;
it is interdisciplinary, including pragmatically relevant entries from adjacent
fields such as philosophy, anthropology and sociology, neuroscience and psychology, semantics, grammar and discourse analysis;
it provides reliable orientational overviews useful both to students and more
advanced scholars and teachers.
The nine volumes are arranged according to the following principles. The first
three volumes are dedicated to the foundations of pragmatics with a focus on micro
and macro units: Foundations must be at the beginning (volume 1), followed by
the core concepts in pragmatics, speech actions (micro level in volume 2) and discourse (macro level in volume 3). The following three volumes provide cognitive
(volume 4), societal (volume 5) and interactional (volume 6) perspectives. The
remaining three volumes discuss variability from a cultural and contrastive (volume 7), a diachronic (volume 8) and a medial perspective (volume 9):
1. Foundations of pragmatics
Wolfram Bublitz and Neal R. Norrick
2. Pragmatics of speech actions
Marina Sbis and Ken Turner
3. Pragmatics of discourse
Klaus P. Schneider and Anne Barron
4. Cognitive pragmatics
Hans-Jrg Schmid
5. Pragmatics of society
Gisle Andersen and Karin Aijmer

Preface to the handbook series

6. Interpersonal pragmatics
Miriam A. Locher and Sage L. Graham
7. Pragmatics across languages and cultures
Anna Trosborg
8. Historical pragmatics
Andreas H. Jucker and Irma Taavitsainen
9. Pragmatics of computer-mediated communication
Susan Herring, Dieter Stein and Tuija Virtanen

vii

Acknowledgements

This initial, foundational volume in this series of handbooks represents the work of
a large number of individuals over a fairly long period of time. We must, of course,
thank our contributors for producing high quality articles reflecting scholarly rigor
and a strong sense of what our readers seek in a handbook of this scope. Without
conscientious, willing reviewers, projects like this are simply impossible. Our sincere thanks go out to all our reviewers, though we cannot identify them by name.
We editors are surrounded by competent, trustworthy staff in Augsburg and Saarbrcken, Germany. We own a debt of gratitude to Claudia Enzweiler for proofreading and Sylvia Monzon for her clerical services, to Katharina Rters, Katrin Stuis
and Claudia Rieger for their help with the desk editing and compilation of the indexes. Finally, we are grateful to the De Gruyter team: Barbara Karlson, Wolfgang
Konwitschny and Anke Beck.

Table of contents

Preface to the handbook series . . . . . . . . . . . . . . . . . . . . . . . .


Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v
ix

Introduction: the burgeoning field of pragmatics


Wolfram Bublitz and Neal R. Norrick . . . . . . . . . . . . . . . . . . . . .

Part I Conceptual foundations


1.

2.

3.

4.

Pragmatics as a linguistic concept


Anita Fetzer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

23

Micropragmatics and macropragmatics


Piotr Cap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

51

Pragmalinguistics and sociopragmatics


Sophia Marmaridou . . . . . . . . . . . . . . . . . . . . . . . . . .

77

Metapragmatics
Axel Hbler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

Part II Theoretical foundations


5.

The rise of pragmatics: a historiographic overview


Wataru Koyama . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

6.

Semiotic foundations of pragmatics


Winfried Nth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167

7.

Pragmatics in modern philosophy of language


Nikola Kompa and Georg Meggle . . . . . . . . . . . . . . . . . . . 203

8.

Foundations of pragmatics in functional linguistics


Saskia Daalder and Andreas Musolff . . . . . . . . . . . . . . . . . . 229

9.

Foundations: ethnomethodology and Erving Goffman


Christine Domke and Werner Holly . . . . . . . . . . . . . . . . . . 261

xii

Table of contents

10.

Pragmatics in Habermas Critical Social Theory


Maeve Cooke . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

289

Part III Key topics in pragmatic description


11.

12.

13.

14.

Deixis and indexicality


William F. Hanks . . . . . . . . . . . . . . . . . . . . . . . . . . . .

315

Reference and anaphora


Monika Schwarz-Friesel and Manfred Consten . . . . . . . . . . . .

347

Speech acts
Elena Collavin . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

373

Types of inference: entailment, presupposition, and implicature


Yan Huang . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

397

Part IV The place of pragmatics in the description of discourse


15.

16.

17.

18.

Pragmatics and grammar


Arnulf Deppermann . . . . . . . . . . . . . . . . . . . . . . . . . .

425

Pragmatics and semantics


John Saeed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

461

Pragmatics and prosody: prosody as social action


Elizabeth Couper-Kuhlen . . . . . . . . . . . . . . . . . . . . . . .

491

Pragmatics and literature


Jacob L. Mey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

511

Part V Methods and tools


19.

20.

Approaching the data of pragmatics


Monika Bednarek . . . . . . . . . . . . . . . . . . . . . . . . . . .

537

Experimental pragmatics
Richard Breheny . . . . . . . . . . . . . . . . . . . . . . . . . . . .

561

Table of contents

xiii

21.

Corpus-based pragmatics I: qualitative studies


Gisle Andersen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 587

22.

Corpus-based pragmatics II: quantitative studies


Christoph Rhlemann . . . . . . . . . . . . . . . . . . . . . . . . . . 629

23.

The transcription of face-to-face interaction


Roger J. Kreuz and Monica A. Riordan . . . . . . . . . . . . . . . . 657

About the authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 681


Name index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 691
Subject index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 706

xiv

Table of contents

Introduction: the burgeoning field of pragmatics


Wolfram Bublitz and Neal R. Norrick
1.

The aim of the handbook

The present volume, as the initial handbook in the nine volume series Handbooks
of Pragmatics, provides a comprehensive overview of the foundations of pragmatics. Its in-depth articles cover the roots and evolution of those central theories
and approaches as well as key concepts and topics that are characteristic of twenty-first century pragmatics as an approach to the means and ways of using language
in authentic social contexts. The articles provide reliable orientational overviews
useful to researchers, students, and teachers. They offer both state-of-the-art reviews of their topics and critical evaluations in the light of on-going developments.
Thus, topics are considered not only within their contemporary scholarly context
but are also critically evaluated from one or more current perspectives. As the
opening volume in the series, Foundations of Pragmatics provides historical, conceptual, theoretical and methodological vantage points from which the articles in
the following eight volumes can be related to each other and to the development of
the entire field.
The series as a whole seeks to be reasonably comprehensive to account for the
exceptionally vast, unusually heterogeneous and still rapidly expanding field of
pragmatics. At the same time, it sets out to give structure and coherence to the field
through its organization and choice of topics as well as by revealing traits and contours that have evolved during the past half-century. The nine volumes are meant to
reflect the substantial and wide-ranging significance of this transdisciplinary subject for nearly all areas of language description, to trace its origins and influences,
as well as to account for its remarkable and continuously rising popularity in linguistics and adjoining disciplines.
2.

The history of pragmatics

Even though its roots can be traced back to early classical traditions of rhetoric and
stylistics, to Immanuel Kants conception of pragmatics as empirical and purposive and to William James, who pointed out its practical nature, modern pragmatics
is a fairly recent discipline. Its inauguration as an independent field of study within
semiotics took place early in the 20th century by C. Morris, R. Carnap and ultimately C.S. Peirce. The classic division between syntax, semantics, and pragmatics goes back to Morris, who distinguished three separate dimensions of semiosis within his science of signs:

Wolfram Bublitz and Neal R. Norrick


One may study the relations of signs to the objects to which the signs are applicable.
This relation will be called the semantical dimension of semiosis, symbolized by the
sign DSEM; the study of this dimension will be called semantics. Or the subject of study
may be the relation of signs to interpreters. This relation will be called the pragmatical
dimension of semiosis, symbolized by the sign DP; the study of this dimension will be
named pragmatics. One important relation of signs has not yet been introduced: the formal relations of signs to one another. [] This third dimension will be called the syntactical dimension of semiosis, symbolized by the sign DSYN, and the study of this dimension will be named syntactics. (Morris 1938: 2122)

Syntax studies the relations signs bear to other signs, semantics the relation between signs and objects, and pragmatics the relation between signs and their interpreters. Of course, there were and are differences of opinion on where exactly to
draw the line between semantics and pragmatics, as many of the chapters in this
volume will show.
Some thirty years elapsed before pragmatics finally made its way into modern
linguistics in the late 1960s, when linguists began to explore so-called performance phenomena. To this end, they adopted ideas developed and advanced by
L. Wittgenstein, G. Ryle, P. Strawson, J.L. Austin and other eminent (ordinary or
natural) language philosophers. It seems safe to claim that the ensuing pragmatic
turn was most notably induced by J.L. Austin, J.R. Searle and H.P. Grice, who
were interested in utterance meaning rather than sentence or word meaning, i.e. in
studying unique historical events created by actual speakers to perform linguistic
acts in actual situational contexts in order to accomplish specific goals. Other
scientific movements that nourished pragmatics include anthropology (B. Malinowski, P. Wegener, A. Gardiner), contextualism (J.R. Firth) and functionalism
(K. Bhler, R. Jakobson, D. Hymes), ethnomethodology (H. Garfinkel, E. Goffman, H. Sacks) and European sociology (J. Habermas). This volume addresses all
of these influences.
Since the pragmatic turn, pragmatics has arguably developed more rapidly and
diversely than any other linguistic discipline. Since the 1970s, the early AngloAmerican framework of pragmatic linguistic study has been immensely expanded
and enhanced by research in Continental Europe and elsewhere. With historiographic hindsight, it can be seen that the broadening, i.e. the interdisciplinary expansion, of the field of pragmatics has been a cumulative process; the broader conception of pragmatics chronologically (and causally) followed the narrower one.
We can easily detect a first shift from the binarity of early structuralist concepts
(such as lexical/word meaning versus causal/sentence meaning) to the multiplicity
of speech act related concepts. A further step towards a (conceptual as well as methodological) widening of the field took place in the 1970s and 1980s, when structure and action oriented pragmatics linked up with the emerging interactional paradigms in sociology (in general) and ethnomethodology (in particular). The ensuing
developments have seen inter alia a realignment of context (from a static and au-

Introduction: the burgeoning field of pragmatics

tonomous to a dynamic and collaborative concept, which is cognitively, situationally and socio-culturally much more refined), and a shift from the concept of the
unilateral, i.e. the individual speakers (speech) act (as advocated by pioneers like
Austin) to the interactionally expanded concept of the bi- or (in some types of interactive computer-mediated forms of communication) multi-lateral inter-act. As
such, contemporary broad pragmatics takes account of the interactional turn
that can be observed in the most recent development of interactive (Web 2.0-based)
media formats.
Taken all in all, the establishment and development of linguistic pragmatics has
been an authentic success story.

3.

The definition of pragmatics

Despite its scientific acclaim, the notion of pragmatics remains somewhat enigmatic and is still difficult to define. This holds for its readings in everyday discourse as well as in scholarly contexts. Nonetheless, when we refer to attitudes and
modes of behavior as pragmatic, we mean that they have a factual kind of orientation in common. People who act pragmatically or take a pragmatic perspective
generally have a preference for a practical, matter of fact and realistic rather than a
theoretical, speculative and idealistic way of approaching imminent problems and
handling everyday affairs. To put it differently, they share a concrete, situation-dependent approach geared to action and usage rather than an abstract, situation-independent and system-related point of view. To assume a pragmatic stance in
everyday social encounters as well as in political, historical and related kinds of
discourse, means to handle the related affairs in a goal-directed and object-directed, common-sense and down to earth kind of way. Such an understanding of
pragmatics as an attitude in non-scientific discourse has obviously left its traces in
scientific definitions of the term. By and large we can say that in semiotics and
philosophy, pragmatic characterizes those theoretical and methodological approaches that are oriented toward use and context rather than toward some system,
and that they regard use and context as creating a high degree of analytical surplus.
While essentially the same is true for linguistics in general, there is no commonly accepted definition of pragmatics in linguistics which would refer to a
single, unified and homogeneous field of study. In contemporary linguistics, we
can identify a narrow and a broad way of delineating pragmatics (of which the
former is sometimes allocated to an Anglo-American and the latter to a Continental [European] tradition of pragmatics, cf. Huang 2007: xi). According to the
narrow view, pragmatics is understood as the systematic investigation of what and
how people mean when they use language as a vehicle of action in a particular context and with a particular goal in mind. Thus, the context-dependency of utterance
meaning is the central component of more narrowly defined accounts of prag-

Wolfram Bublitz and Neal R. Norrick

matics, which focus on a few key issues that can be juxtaposed with related issues
in other modules of language theory such as grammar and semantics. Those issues
include indexicality/deixis (versus anaphora), presuppositions, implicatures (versus entailments) and speech acts (versus types of sentences), to name only the most
conspicuous topics.
In this volume, as indeed in the whole Handbooks of Pragmatics series, we
adopt a much broader point of view and understand pragmatics as the scientific
study of all aspects of linguistic behavior. In particular,
pragmatics includes patterns of linguistic actions, language functions, types of inferences, principles of communication, frames of knowledge, attitude and belief, as well as
organisational principles of text and discourse. Pragmatics deals with meaning-in-context, which for analytical purposes can be viewed from different perspectives (that of
the speaker, the recipient, the analyst, etc.). It bridges the gap between the system side of
language and the use side, and relates both of them at the same time. Unlike syntax, semantics, sociolinguistics and other linguistic disciplines, pragmatics is defined by its
point of view more than by its objects of investigation. The former precedes (actually
creates) the latter. Researchers in pragmatics work in all areas of linguistics (and
beyond), but from a distinctive perspective that makes their work pragmatic and leads
to new findings and to reinterpretations of old findings. The focal point of pragmatics
(from the Greek prgma act) is linguistic action (and inter-action); it is the hub around
which all accounts in these handbooks revolve. (Preface to the handbook series)

Pragmatics is fundamentally concerned with communicative action in any kind of


context. The multifaceted research paradigm of pragmatics has provided new directions and perspectives in the arts and humanities, philosophy, cognitive science,
computer science and the social sciences. Pragmatic perspectives have been employed in information technology and in the social sciences, particularly in economics, politics and education.
In the pragmatic perspective, language use and language users in interaction
are primary, as opposed to language as a system of signs or a set of rules. The pragmatic perspective scrutinizes neither just individual words nor sentences nor even
isolated texts, but rather whole speech events or language games in real social contexts, considering both the present state of affairs and its connectedness with prior
and succeeding actions. It rejects a localization of language in a limited segment of
the acts of speaking, understanding and responding or within the conscious of the
individual. It supplants a view of language as an abstraction without variation by
speaker, region or time, of language as a non-cultural, non-social, static, depersonalized fact independent of context and discourse. Pragmatics goes beyond the perspective of written texts with their carefully marshalled grammatical sentences to
embrace everyday talk and the messiness of language in real embodied human
contexts, where participants with personalities, feelings and goals interact in complex ways with physical objects and other participants within institutions and communities.

Introduction: the burgeoning field of pragmatics

That pragmatics overcomes any narrow focus on language as a system of signs


or a set of rules does not mean, of course, that pragmatics (in contradistinction to
other scientific disciplines) attempts to describe language without recourse to the
systematic level of theoretical abstraction. Ultimately and in order to create and secure a sound scientific foundation, pragmatics, like any other scientific theory, has
to accomplish the transfer from the level of token to the level of type. For the pragmatic scholar, specific concrete linguistic events (and their contexts) are only relevant in as far as their properties and constraints can be integrated under analogous
conditions into a general concept of language and communication. While pragmatics is on a par with other scientific theories in this respect, it transcends them in
various ways, in particular by adopting a constructivist point of view which allows
for the inclusion of new contextual, situational and cognitive variables. Describing
language pragmatically thus means going beyond the description of language as an
autonomous, type-restricted principle and taking into account extra-linguistic phenomena and conditions emanating from the context and concrete situation of language use. Turning contextual and related conditions into prototypical conditions
(and thus theorizing them) constitutes the pragmatic surplus.

4.

The purview of pragmatics

If pragmatics is more a perspective or orientation toward language use than a particular theory or specific research object, then the data practitioners choose to analyze assume great importance. In linguistic research, as Saussure realized, it is the
viewpoint that determines the object rather than the other way around. In the tradition growing out of Natural Language Philosophy the intuitions and casual observations of the individual writer provide the stuff for analysis. Introspection counts
as data. All the early work on speech acts, presupposition and inference was based
on introspection and invented examples in imagined contexts (Austin, Searle,
Strawson and Grice).
Other traditional areas of pragmatic research were based on authentic data from
the start. Indexicality/deixis, discourse markers, particles and the like are ubiquitous in texts, and examples for study are thus easy to collect. As corpora of spoken
language became increasingly available and reliable, the problem of evidence versus introspection began to take care of itself. Scholars no longer needed to guess
about distributions and frequencies of linguistic items. One might begin with a
small pre-selected corpus representing a specific set of items or types of interaction
to identify a range of functions for a particular linguistic feature before embarking
on a general search in a large corpus to determine distributions and frequencies in
the corpus at large only to return to a small corpus representing specific contexts
for careful qualitative analysis. By contrast, investigations of large corpora might
provide the impetus for research, illustrating a range of items or contexts not found

Wolfram Bublitz and Neal R. Norrick

in a small corpus and suggesting hypotheses to be tested. For statistical purposes,


larger was presumably always better, because scholars tended to correlate frequency with typicality (the more frequent the more typical). On the other hand, increasing interest in multi-modal data and description in research on spoken interaction brought researchers back to small, carefully annotated corpora. By its very
nature, research on prosody, gaze, gesture and the details of interaction with objects in a specific physical setting must proceed from a narrow recording and transcript of a particular event: Large corpora are extremely difficult to manage in
these circumstances.
Many pragmaticians maintain that language data must be as authentic as possible, recorded in real life contexts where something is at stake for the participants.
But the whole matter of recording, often with camcorders as well as microphones,
raises the spectre of the Observers Paradox. How natural is the language interaction, when we require technical apparatus to record it? Some pragmatic scholars
ascribe to the notion that language use must be understood in terms of embedded
practices, as part of the social world in which listening and speaking are modalities
of action, in line with the treatment of speech production and reception as embodied social facts (Hanks 1996), and practices as shared ways of doing things, ways
of talking, beliefs and values as a function of engagement in an activity (Wenger
1998; Eckert 2000). Conceptualizing speaking and listening in terms of practices
affords a natural purchase on nonverbal features of participant behavior, including
gaze, posture and gesture in face-to-face interaction (Goodwin 2003). This has
consequences for what counts as data, how much context must be considered, what
must be transcribed and in how much detail.
At the same time, certain questions will always go unanswered on the basis of
corpus data, no matter how careful the transcription and no matter how large the
corpus, since every individual exchange is participant designed and context dependent. With some kinds of structures like prosody, discourse markers and interjections, there is the problem of how to get enough examples to make sensible generalizations. Experimental procedures could complement corpus-based research in
order to generate parallel structures and contexts to test for recurrent speaker and
listener practices. Indeed, Experimental Pragmatics represents a new area of research at the intersection of traditional linguistic pragmatics and psycholinguistics. It seeks to apply experimental techniques from psycholinguistics to solve
problems defined in linguistic pragmatics (see Breheny this volume). So far such
experiments have focused on issues surrounding the conversational maxims and
implicatures of Grice (1975) and his followers (both Neo-Griceans like Horn 1984
and Levinson 2000, and Post-Griceans like Sperber and Wilson 1986 and Carston
2002), but creative experiment design could extend this new research paradigm to
all sorts of questions about spoken language. All the same, investigations of large
corpora provide a body of natural data against which experimental results can be
evaluated.

Introduction: the burgeoning field of pragmatics

The discussions between scholars of different persuasions about what constitutes appropriate data are ongoing. The historical development of different data
sources and areas of inquiry as well as the continuing debates about their usability
and validity resonate in this handbook. Indeed, the authors contributing to this
handbook were chosen to represent various positions in these debates.

5.

Organization of this Handbook

To distinguish pragmatics as a linguistic discipline from pragmatics as an attitude or


a mode of behavior (a practical rather than an idealistic approach to problems and
dealings) as used in everyday, in political, in economic and in other kinds of discourse, the present handbook Foundations of Pragmatics opens with a section on
Conceptual Foundations consisting of the description of pragmatics as a linguistic
concept. Fundamental concepts covered are: micropragmatics, i.e. the pragmatics of
utterance based concepts such as speech acts etc. (which are covered in more detail
in vol. 2 of the series) as opposed to macropragmatics, i.e. the pragmatics of discourse or text based concepts such as topics, discourse markers etc. (cf. a more detailed overview in vol. 3 of the series), pragmalinguistics (dealing with forms) as opposed to sociopragmatics (dealing with pragmatic strategies), and metapragmatics.
The second major section is dedicated to the Theoretical Foundations of pragmatics. It begins with a historiographic overview of how pragmatics was established as a linguistic field of study, how it developed and spread, and how it eventually became an international success story. After all, since the so-called
pragmatic turn, linguistic pragmatics has, arguably, developed more rapidly and
diversely than any other linguistic field. The scientific disciplines that nourished
pragmatics in the first place range from semiotics (Peirce, Morris, Carnap) and
(Natural Language) philosophy (Wittgenstein, Strawson, Austin, Searle, Grice) to
anthropology (Malinowski, Wegener, Gardiner), contextualism (Firth) and functionalism (Bhler, Jakobson, Hymes), from ethnomethodology (Garfinkel, Goffman, Sacks) to European sociology (Habermas).
The third major section, Key Topics in Pragmatic Description, investigates
those central concepts of pragmatic description which were the first to make their
way from grammar into pragmatics (indexicality and deixis, followed by reference
and anaphora), on the one hand, and those notions that rapidly became the focal descriptive construct in mainstream pragmatics (speech acts and types of inference
such as entailment, presupposition, implicature), on the other.
In the fourth major section, the Place of Pragmatics in the Description of Discourse, the scope of pragmatics within linguistic description is delimited vis--vis
grammar, semantics, prosody. A final chapter on pragmatics and literature underscores the contention that the social aspect of language use is as prominent in literary contexts as in any non-literary context.

Wolfram Bublitz and Neal R. Norrick

Methods and Tools constitutes the fifth and final major section. It contains an
initial chapter comparing various methodological approaches to pragmatics such
as intuition, introspection, ethnographic field work, elicitation and corpus analysis,
before it then proceeds with articles on experimental pragmatics, corpus-based
pragmatics and the transcription of face-to-face interaction.
No handbook can include every individual perspective and specific topic even
regarding the foundations of an area of study. Our decisions regarding which fields
and issues to include and how they should be addressed reflect both our own views
of pragmatics and the expertise and interests of our authors. As authors were recruited for the handbook project, joined us, and began to think and write about their
areas, topics were redefined, expanded and narrowed, titles were adjusted, some
boundaries were sharpened, others redrawn. Recent developments in pragmatics,
the general availability of new kinds of multi-media data and analytic procedures
along with dynamically evolving attitudes toward topics and research methods
helped shape the contents of this volume. Inevitably, the exigencies of scheduling
and the special interests of the contributing authors conspire to yield a final set of
articles representing individual views at a specific point in time. We embrace this
particular character of the snapshot offered by the articles in this handbook.

6.

Part I Conceptual foundations

In the lead article, Pragmatics as a linguistic concept, Anita Fetzer seeks to limit
the concept of linguistic pragmatics. The chapter compares pragmatics as a linguistic concept with pragmatics as used in everyday, in political, in economic and
other kinds of discourse. It discusses what is special about linguistic pragmatics,
covering General Pragmatics as well. Pragmatics as a common-sense notion is
functionally synonymous with practical or just right at that stage, describing a
language-bound and/or action-based phenomenon within a particular situation, as
is reflected in its usage in everyday, in political, in economic and other kinds of discourse. In linguistics, semiotics or philosophy, pragmatics refers to a theoretical
concept, comprising contextual, generally usage-related constraints and requirements, for instance presuppositions, indexical expressions, felicity conditions,
common ground and background assumptions. These are necessary conditions for
communication to be felicitous in context. Thus, the perspective is shifted from the
analysis of the language system and its constitutive parts to that of its rule-governed instantiation and embeddedness in context. In General Pragmatics the focus
of investigation is on universals of communicative action, such as directness versus indirectness. These are refined in Linguistic Pragmatics, where they undergo
language- and culture-specific modification. That change in perspective has important consequences for methodology. To account for the parts-whole connectedness, a relational frame of reference is required which accommodates the relation

Introduction: the burgeoning field of pragmatics

between the language system and language use, and interlocutors and what they do
with language in context, comprising linguistic context (or co-text), social context
and cognitive context.
In the second article, Micropragmatics and macropragmatics, Piotr Cap addresses the central distinction between micropragmatics and macropragmatics.
This chapter distinguishes micropragmatics as referring to speech acts/actions etc.
from macropragmatics as directed at the discourse/text level. Defining pragmatics
by its general perspective, rather than by its specific object of investigation, entails
looking at the analytic scope of pragmatics in terms of micro- and macro-level concepts. These have been traditionally (cf. Mey 1993) associated with the opposition
between the analysis of speech act force at an utterance level and the analysis of
global intentionality at the level of a discourse/text. Following this division, as
well as its later refinements, the article, first, reviews the locutionary, illocutionary
and perlocutionary constituents of force of an utterance, relating them to pragmatic
concepts which characterize the process of encoding and decoding its message
(deixis, presupposition, implicature etc.). Second, it demonstrates how individual
utterances comprising different sets of constituents and markers of force, including
syntactic markers of explicitness and inferable carriers of implicitness, can form
sequences to shape the global intentionality of a discourse/text. Invoking such notions as speech event and macro speech act, the article shows which pragmatic concepts utilized in (micro-) analysis of individual utterances are essentially complementary, i.e. which markers of force denoted by these concepts can collectively
generate complex macro illocutions responsible for accomplishing global
discourse/text goals.
The important distinction between Pragmalinguistics and sociopragmatics
forms the topic of chapter 3 by Sophia Marmaridou. In Principles of Pragmatics
(1983), Leech introduced a distinction between pragmalinguistics (roughly, what
form to use to achieve an intended pragmatic effect) and sociopragmatics (roughly,
when to use a particular pragmatic strategy). This chapter describes past and current research on this distinction in pragmatics. Within a broad definition of pragmatics as the scientific study of all aspects of linguistic behavior, the distinction
between pragmalinguistics and sociopragmatics is intended to focus on two methodological approaches. Pragmalinguistics concerns the study of the particular resources that a given language provides for conveying pragmatic meaning (illocutionary and interpersonal), whereas sociopragmatics, as the sociological interface
of pragmatics (Leech 1983: 10), relates pragmatic meaning to an assessment of
participants social distance, the language communitys social rules and appropriateness norms, discourse practices, and accepted behaviors (Thomas 1995). The
above methodological distinction to some extent relates to earlier work in Marxist
pragmatics (Mey 1979; Prucha 1983), foreshadows work in Critical Discourse
Analysis (Fairclough 1995), and intersects with studies in sociolinguistics. A critical assessment of the two concepts and the respective methodologies emerges from

10

Wolfram Bublitz and Neal R. Norrick

current research practices. Even though the borderline between the two concepts
often seems fuzzy, their relevance to language learning, teaching and testing has
been enhanced (McNamara and Roever 2006) and so has its significance in historical corpus linguistics (Archer and Culpeper 2003).
In the fourth chapter, Axel Hbler addresses the subject of Metapragmatics.
He focuses on metapragmatics concerning the actual management of discourse.
Key tasks in metapragmatics are the investigation of that area of speakers competence which reflects the judgments of appropriateness of their own and other
peoples communicative behavior and the exploration of the know how regarding the control and planning of, as well as feedback on, the ongoing interaction.
The human capacity for reflecting about language and language use/communication is a feature of language-related competence that no animal is endowed with.
Part of this capacity extends to areas which linguists have become accustomed to
referring to as metapragmatics. The term is fairly new, and its usage not yet fully
regulated; it accommodates various systematic perspectives and pertains to a wide
range of language-reflexive phenomena. The article provides a comprehensive survey along two lines: (1) what linguists (and experts from other disciplines such as
philosophy or social-psychology) have observed about how people act and interact
and what they express about their practice; and (2) what these experts have observed about their own descriptive, analytical and theorizing practice. Line (1) includes theories of pragmatics as prefiguring observation and resulting from it; and
it takes account of what could be categorized as (implicit or explicit) meta-practice
by members of a speech community (e.g., monitoring, remedying, evaluating by,
e.g., quoting, topicalizing, metaphorizing). Line (2) comprises meta-theoretical
considerations, which (critically) reflect upon and compare varying theoretical
stances or elaborate on the difference to be drawn between a scientific view on
pragmatics and folk theories.

7.

Part II Theoretical foundations

The opening chapter 5 in this section, The rise of pragmatics: a historiographic


overview, by Wataru Koyama traces the history of the still young field of linguistic pragmatics. This chapter stresses the success story as well as the newness of
the field of pragmatics. Especially after the artificial impoverishment of linguistics
due to structuralism and generative grammar, there was clearly a need for pragmatics as a field to amalgamate insights from rhetoric, anthropology, functionalism
etc. When and by whom was linguistic pragmatics invented? Who were its
(extra-linguistic) predecessors? What was the nature of the scientific and, particularly, the linguistic environment, in which pragmatics could arise (the 1950s; logical positivism etc.). Then the author considers further developments: how and why
pragmatics developed, slowly at first, abruptly later in the 1980s (Levinson 1983,

Introduction: the burgeoning field of pragmatics

11

International Pragmatics Association, Journal of Pragmatics etc.). Finally, he


looks at how todays pragmatics differs from earlier stages of pragmatics.
In the sixth article, Semiotic foundations of pragmatics, Winfried Nth relates pragmatics to its foundations in semiotics. The chapter shows how semiotics
influenced the development of the theoretical foundations of pragmatics, focusing
on the contributions of Peirce, Morris and Carnap. Semiotics, the general study of
signs and sign processes, is fundamental to pragmatics, the study of language use,
speech acts, and communicative processes, and it extends the framework of language use to include nonverbal, visual, and other signs in the context of verbal behavior. According to the general framework outlined by Carnap and Morris, pragmatics is one of the three branches of semiotics besides syntax and semantics, but
twentieth century structuralist semiotics tended to neglect the pragmatic dimension
in its study of language and discourse. Central issues of pragmatics, such as the
questions concerning intention and purpose, meaning and reference, word and object, interpretation and communication, or the relationship between language and
action, find fundamental answers in the writings of the founder of general semiotics and philosophical pragmatism (or pragmaticism), Charles S. Peirce.
Chapter 7, Pragmatics in modern philosophy of language by Nikola Kompa
and Georg Meggle, traces the origins of linguistic pragmatics in the Philosophy of
Language, particularly in Natural Language Philosophy in the work of Wittgenstein, Strawson, Austin, Searle, Grice. There are two paradigms of Pragmatics in
modern studies in the Philosophy of Language. In the first, semantics is taken as
given and the aim is to incorporate as many context-sensitive aspects of language
as possible into that semantics; in the second, research is trying to spell out the semantics on a pragmatical (action-theoretical) basis itself. Accordingly, the difference between the two paradigms could be considered that between semantic pragmatics on the one side and pragmatic semantics on the other side. Characteristic
contributions to the first approach are those by Frege, Carnap, Kaplan and Stalnaker; characteristic contributions to the second are by Wittgenstein, Austin,
Grice, Lewis and Bennett. This article considers the basic problems as well as the
present state of the art in both paradigms of pragmatics and attempts to achieve
some clarity about their deeper (dis-) connection in particular
In chapter 8, Foundations of pragmatics in functional linguistics, Saskia
Daalder and Andreas Musolff trace the roots of pragmatics in functional approaches to linguistics. Pragmatics addresses the functions of language as found in
the work of Bhler, Jakobson, Hymes. This chapter traces the historical connections from these sources to contemporary thinking in pragmatics. The study of linguistic pragmatics received a major impetus from the systematic analysis of language functions as started by Karl Bhler, who distinguished three fundamental
dimensions of the linguistic sign: it functions at the same time as symbol, symptom
and signal. The model has proved its enduring value in the development of functionalist language theories, via its modification and augmentation into a six-func-

12

Wolfram Bublitz and Neal R. Norrick

tions-model by Roman Jakobson. His model in turn informed subsequent functionally oriented theories of language and communication, e.g. Dell Hymes
Ethnography of Speaking (1962) and Michael A.K. Hallidays concept of Language as Social Semiotic (1978). In retrospect, Bhlers model (and, to a lesser extent, Jakobsons modified model) can be judged to suffer from its forced integration with the structuralist concept of the linguistic sign, which initially
hampered a fully fledged pragmatic interpretation.
In their chapter 9, Foundations: Ethnomethodology and Erving Goffman,
Christine Domke and Werner Holly discuss two related questions: What is Ethnomethodology or rather what are the crucial questions and assumptions of this
qualitative research approach? And how did Ethnomethodology impact linguistic
pragmatics? At first they overview the studies of Harold Garfinkel which contain
the groundbreaking thoughts about the most significant question: How is social
reality constructed and organized by activities of the societys members? Accordingly, it is shown how conversation analysts (e.g. Sacks, Schegloff) are applying
Garfinkels fundamental assumption of the social reality accomplished through
peoples everyday life activities in analyzing talk-in-interaction. Subsequently, the
main focus lies on the interface between social order, which Ethnomethodology
considers as produced by people interacting in their daily life, and the fundamental
pragmatic question what people are doing using speech and talking. How the idea
of (re-)produced social order is in line with and can be linked to the pragmatic
focus on the role of language/verbal interaction in different forms of daily interaction.
The work of Habermas has been central in the development of pragmatics in
Europe. In her contribution in chapter 10, Pragmatics in Habermas Critical Social Theory, Maeve Cooke considers the influence of Habermas on linguistic
pragmatics. She deals with the pragmatic turn that Jrgen Habermas introduced to
critical social theory with his program of universal pragmatics in the 1970s. With
this program, subsequently referred to as formal pragmatics, Habermas investigated everyday linguistic interaction from the point of view of its orientation towards validity. His investigations were guided by the critical intentions of his social theory. His principal intentions were to provide a normative basis for a conception of practical rationality without reference to otherworldly authorities or
entities and to develop a corresponding normative account of social action. However, his formal pragmatics has profound implications for social theory in general
and, indeed, for anyone concerned with questions of meaning and validity. This article sets out the key elements of Habermas formal pragmatics, shows its implications for social theory and beyond, and discusses some important objections that
can be raised against him.

Introduction: the burgeoning field of pragmatics

8.

13

Part III Key topics in pragmatic description

Pragmatics first made inroads into linguistics through the pioneering work of Yehoshua Bar-Hillel. In this chapter 11, William Hanks describes past and current research on Deixis and indexicality from a pragmatic perspective. He starts from a
historical sketch of indexicality in the semiotics of Peirce and Morris, and traces
the development of the concept by philosophers, linguists, linguistic anthropologists, sociolinguists (via style, register, contextualization) and conversation
analysts. Called shifters by Jespersen and Jakobson, deictics encode a relational
structure projecting from an indexical ground (a context of utterance) to a referential figure (the object denoted) via a relation (proximity, perceptibility, anaphora,
etc.). Deictic systems differ cross-linguistically in ways just beginning to emerge.
Pronouns are a related type of shifter and are closely tied into participation frameworks as treated by Goffman, conversation analysts and linguistic anthropologists. Recent debate is summarized regarding the role of spatial proximity vs.
cognitive-social accessibility as a basis for deixis. The final section outlines empirical approaches to pragmatic research on deixis.
Reference and anaphora are inherently pragmatic in nature and their treatment
in linguistic description helped bring pragmatic notions into prominence.
Chapter 12 on Reference and anaphora by Monika Schwarz-Friesel and Manfred
Consten concerns reference, the area in scientific research which deals with the
ancient problem of how language relates to the world. In the tradition of philosophical logics and formal semantics, reference has been defined as a static relationship between expressions and the things they denote. In the course of pragmatic change in linguistics, models of reference have arisen that (to different
extents) take cognitive properties of language users into account. In the framework
of the latter notion, anaphora is a specific kind of reference that has to be resolved
mainly through the activation of discourse based knowledge established by the
preceding text. Thus, anaphora resolution is analysed as a process of reactivating
pre-mentioned referents. The traditional view of anaphora as a means of referential
continuity and the availability of given information has to be revised. From a procedural perspective, anaphora may add new information to the mental files of the
referents in the text-world model.
Elena Collavin addresses the key topic of Speech acts in chapter 13. The
chapter provides an overview of the notions of act/action (versus behavior) and
speech act: how they developed, how they made their way into linguistics, how
they became the core notions of various (speech act) theories, how the latter developed etc. Collavin presents an account of the notion of speech act, which includes
an illustration of the main formulations of the theory, its limitations, and its current
relevance and application in the broad field of pragmatic research. She traces the
origins of this notion and describes the main formulations of Speech Act Theory,
focusing on conventionality, intentionality, and action as the key aspects of speech

14

Wolfram Bublitz and Neal R. Norrick

acts around which similarities and incompatibilities in different versions of the theory are established. The discussion then turns to the criticisms moved to various
notions of speech act. The relevance of speech acts for the empirical analysis of
language use in interaction has been strongly debated. Collavin discusses how
some notion of speech act can relate to conceptualizations nowadays fundamental
in the study of language usage such as frame, voicing, indexicality, multimodality,
shared cognition and embodiment.
Chapter 14 provides an overview of the varieties of inference recognized in
pragmatics, how they developed, how they made their way into linguistics, how
they became core notions of various theories, how the latter developed etc. In this
chapter, Types of inference: entailment, presupposition, and implicature, Yan
Huang provides a critical survey of three fundamental types of inference recognized in semantics and pragmatics, namely entailment, presupposition and implicature, covering their origin, development, and the roles they play in various semantic/pragmatic theories in linguistics and the philosophy of language. Of the three
categories of inference, entailment is essentially semantic in nature, though the notion of pragmatic entailment is discussed. Presupposition straddles the semanticspragmatics boundary, resulting in semantic and pragmatic presupposition. Finally,
with regard to the two subtypes of implicature, while conversational implicature is
pragmatic in nature, conventional implicature can be considered either as semantic
or pragmatic, depending on how the distinction between semantics and pragmatics
is drawn. Notions parallel to conversational implicature (such as explicature) and
the argument that conversational implicature should not be treated as inference are
also considered. Finally, comparisons between entailment and presupposition on
the one hand, and between presupposition and implicature on the other, are presented.

9.

Part IV The place of pragmatics in the description of discourse

Part IV opens with chapter 15 Pragmatics and grammar by Arnulf Deppermann.


This chapter details the delineation of pragmatics and grammar and the integration
of pragmatics and grammar into linguistic theory in relation to each other, focusing
on issues, methods and areas covered. Deppermann discusses theoretical approaches, linguistic phenomena and empirical methods concerning the relationship
between grammar and pragmatics. He sketches functional approaches to grammar,
which claim that grammar is shaped and constrained by cognition, text and/or social interaction (functional grammar, systemic-functional linguistics, construction
grammar, cognitive grammar, interactional linguistics). Historical studies on the
paths of grammaticalization, psycholinguistic insights from usage-based approaches, and conversation analysis are introduced as methods for the study of
the relationship between grammar and pragmatics. The relevance of grammar for

Introduction: the burgeoning field of pragmatics

15

pragmatics is discussed concerning a broad range of issues, such as cognitive processes (implicature, inference, information structure), the expression of cognitive
states (epistemic stance), conversational organization (repair, turn-construction
and turn-taking, collaborative action, intersubjective grounding), action structure
(speech acts, genres), bodily action (gesture) and socio-structural concerns (politeness, power, social styles).
In chapter 16 on Pragmatics and semantics, John Saeed discusses some contrasting views of the relationship between semantics and pragmatics. His chapter
details the delineation of pragmatics and semantics in linguistic theory, focusing
on issues, methods, areas covered. Within the philosophical and formal traditions,
he considers the growth of contextualism, that is the increasing emphasis on the
underspecification of meaning in linguistic form, illustrated by phenomena beyond
traditional deixis (indexicals) such as gradable and comparative adjectives, quantifiers, possessives etc. The article also looks at the defense against this from semantic minimalism (e.g. Cappelen and Lepore 2005). As an example of contextualism, Saeed discusses aspects of Relevance Theory (Sperber and Wilson 1986,
Carston 2002), with its identification of pragmatic principles reaching deep into
what earlier accounts viewed as semantic content. In particular, he considers the
theorys notions of explicatures and higher level explicatures, and looks at its view
of lexical pragmatics. Finally, he brings into the discussion the quite different view
from cognitive linguistics, where the very distinction between semantics and pragmatics is questioned. So in theories like Fauconniers, mechanisms of mental
spaces and conceptual blending seek to model phenomena from across the traditional divide, including e.g. tense, modality, presupposition and metaphor. An interesting question is whether a pragmatic theory like Relevance Theory and the
cognitive semantics of writers like Lakoff, Fauconnier and Langacker are in fact
compatible.
Elisabeth Couper-Kuhlen describes the role of prosody in pragmatics, and the
integration of prosody into linguistic theory. In chapter 17 Pragmatics and prosody, she contemplates what has been achieved so far in the field of prosody and
interaction and what still remains to be done. Instead of terms such as suprasegmentals, sentence types and illocutions, she speaks of prosodic resources for
single- and multi-turn construction, action formation and floor management. Comparing the use of prosodic resources across turns-at-talk, she acknowledges the
contribution of prosody to marking the continuation of same-speaker turns, to organizing sequences of turns, to cueing problems, negotiating agreement and signaling affiliation in talk. But recent developments also present challenges to researchers in the field of prosody and interaction: (i) What counts as prosodic? and
(ii) What is prosodys place in the analysis of multimodal interaction? The chapter
discusses these issues and ways of dealing with them in the future.
In chapter 18, Jacob Mey discusses Pragmatics and literature. According to
Mey (2000), pragmatics is concerned with the way humans use language in a social

16

Wolfram Bublitz and Neal R. Norrick

context: When people use language in a literary context, the social aspect is no less
prominent. Literary pragmatics, therefore, concerns itself with the relation between the language users in the context of literature, that is, the producers (the
authors) and the consumers (the readers). In the context of literary production,
the processes of producing and consuming are paramount. The product (the written
text) can only be evaluated properly in terms of its social conditions and the concomitant roles of (authorial) production and (readerly) consumption. Here, the key
concept of voice (originally due to Bakhtin) plays a major role. The author assumes a narrative voice, while retaining a distinct personal voice. Conversely, the
readers have a voice, inasmuch as they identify with, and co-construct, the narrative personae; in this way, they take part in the building of the literary universe
and making it come alive. Some implications of this pragmatic perspective are followed up through analysis of snippets of literary texts.

10.

Part V Methods and tools

The initial chapter 19 in this final part by Monika Bednarek, Approaching the data
of pragmatics, provides an overview of how various aspects of language use
are studied using different methodological approaches: intuition, introspection,
ethnographic field work, elicitation (and experimentation), and corpus analysis.
It compares critically the various methodologies that are employed in areas such
as conversation analysis, politeness theory, and speech act theory as well as in related functional approaches. The chapter covers both quantitative and qualitative
methodologies ranging from intuition and introspection to interviews, discourse
completion tasks, ethnographic field work, elicited data and corpus analysis. It
takes a broad approach to the data of pragmatics in that it not only includes research in key pragmatic areas but also disciplines related to pragmatics such as ethnography and sociolinguistics. It describes different methodologies and compares
the insights gained from them as well as the different perspectives they provide on
linguistic data.
In chapter 20, Experimental pragmatics, Richard Breheny surveys recent research in this emerging field. Experimental pragmatics aims to use the methods of
experimental psychology to test claims found in the more theoretical linguistics literature. The strategy of testing theoretical ideas experimentally has been around
for a while and is notably exemplified in the work of Clark, Sanford, Gibbs and
Glucksberg. More recently, there has been an upturn in the interest in experimental
methods, coinciding with renewed interest in pragmatic questions such as the nature of generalized implicatures, presupposition, metaphor and the like. Breheny
reviews studies involving children, adolescents and adults from clinical populations as well as typically developing groups using the full range of methods to be
found in psycholinguistics and developmental psychology (including both beha-

Introduction: the burgeoning field of pragmatics

17

vioral and neuro-scientific methodologies). The contribution also contains a discussion of the difficulty of drawing empirically testable predictions from claims
made in the theoretical literature.
The final three chapters in the volume all address the role and forms of corpus
data in linguistic pragmatics. The first two consider corpus-based research in
pragmatics, first qualitative then quantitative studies, and the third discusses
methods of transcription. In chapter 21, Corpus-based pragmatics I: qualitative
studies, Gisle Andersen describes the practice and potentials of corpus-based research in pragmatics, focusing on qualitative approaches to language data. He
considers the development of spoken and written corpora like the Brown Corpus,
the London-Lund Corpus, the Helsinki Corpus, the British National Corpus, the
Santa Barbara Corpus of Spoken American English etc., and their growing influence on research in pragmatics over the last thirty years, paying attention to
special corpora, large general corpora, continually growing monitor corpora and
web-based corpora. Corpora allow for in-depth studies of patterns of frequency,
distribution and collocation, thus facilitating statistically based observations of
innovative language use and variation. Dynamic and ever-growing corpora allow
for the observation of emerging repetitive patterns such as the routinization of
forms with discourse marking functions, ongoing grammaticalization and the development of new abstract and interactional meanings from old forms. In contrast
to quantitative, statistical approaches to corpus data, qualitative studies see corpora primarily as a source of natural data and a way of overcoming the fallible introspection of armchair linguistics. In the area of qualitative pragmatic studies,
corpora have provided a constantly growing pool of data, allowing researchers to
test their hypotheses about individual items and constructions, as well as practices
and usage within particular groups, but also suggesting new avenues of research.
Diachronically oriented corpora are responsible for the rapid progress made
in historical pragmatics. At the same time, corpora representing specific groups,
genres, registers or stages of a language have led to research of various kinds, including the language of adolescents, foreign language learners and other user
groups.
In the companion chapter 22, Corpus-based pragmatics II: quantitative
studies, Christoph Rhlemann takes up quantitative corpus-based research. Due
to the massive dependence of pragmatic phenomena on context, corpora, as a
relatively decontextualized medium, have long been seen by some researchers as
unfit for use in pragmatic research. Nonetheless, corpus linguistic analyses, both
qualitative and quantitative in orientation, have produced a wealth of new insights
into key pragmatic phenomena. The aim of this article is to illustrate key quantitative corpus studies into phenomena of pragmatic interest. The article is divided
into four sections. Following a brief introduction that addresses general issues in
corpus linguistic research on pragmatic phenomena, such as the question of how
much or little context-sensitive corpora are, the second section introduces quanti-

18

Wolfram Bublitz and Neal R. Norrick

tative studies carried out on the small number of corpora that are pragmatically annotated. The third, and largest, section discusses key corpus linguistic research into
pragmatic phenomena such as speech acts, deixis, evaluation, discourse marking,
backchannelling and vagueness. Further, as a borderline case between pragmatics
and semantics, research into semantic prosody will be reviewed. The concluding
section looks to the future, outlining inter alia recent attempts at building multimodal corpora that will potentially shed light on the interplay between utterance
and kinesics.
The final contribution 23 deals with various methods of transcribing face-toface interaction and their significance for pragmatic theorizing. In this chapter,
The transcription of face-to-face interaction, Roger J. Kreuz and Monica A.
Riordan consider issues in the representation of linguistic data. Researchers in
pragmatics frequently transcribe audio- or videotaped recordings of conversational
participants. A comprehensive transcription coding system would help to unify diverse lines of research and make collaboration and interpretation of results easier.
However, a comprehensive system currently does not exist. Kreuz and Riordan
review coding schemes that have been previously proposed and assess their
strengths and weaknesses. Most systems have focused solely on the acoustic (i.e.,
lexical and intonational) aspects of discourse between adult speakers, but they also
review systems that have been developed to transcribe the speech of other populations (e.g., children, aphasics and cognitively impaired individuals). In addition,
they make recommendations about how non-acoustic information can be coded.
Specifically, the challenges and utility of coding for facial and gestural information
are addressed. It is hoped that a more comprehensive coding system will make it
easier for researchers to describe and study the multimodal aspects of conversational interaction.
Taken together, these chapters constitute an outline of the fundamental issues
and trends in the study of pragmatics at the beginning of the twenty-first century.
They can be taken as a kind of extended definition of the term pragmatics as
practiced today, along with critical attempts to set down the history of the discipline and identify the sorts of questions taken to be important and the sorts of
answers being proposed. We hope in this way to contribute to the definition and coherence of the field, and thereby to foster research in the generation to come.

References
Archer, Dawn and Jonathan Culpeper
2003
Sociopragmatic annotation: New directions and possibilities in historical corpus linguistics. In: Andrew Wilson, Paul Rayson, and Tony McEnery (eds.),
Corpus Linguistics by the Lune: A Festschrift for Geoffrey Leech, 3758.
Frankfurt am Main: Peter Lang.

Introduction: the burgeoning field of pragmatics

19

Cappelen, Herman and Ernie Lepore


2005
Insensitive Semantics. Oxford: Blackwell.
Carston, Robyn
2002
Thoughts and Utterances: The Pragmatics of Explicit Communication. Oxford: Blackwell.
Eckert, Penelope
2000
Linguistic Variation as Social Practice. Malden, MA: Blackwell.
Fairclough, Norman
1995
Critical Discourse Analysis. London: Longman.
Goodwin, Charles
2003
Pointing as situated practice. In: Sotaro Kita (ed.), Pointing: Where Language,
Culture, and Cognition Meet, 217242. Mahwah, NJ: Erlbaum.
Grice, Herbert Paul
1975
Logic and Conversation. In: Peter Cole and Jerry L. Morgan (eds.). Syntax and
Semantics. Vol. 3: Speech Acts, 4158. New York: Academic Press.
Halliday, Michael A.K.
1978
Language as Social Semiotic. London: Arnold.
Hanks, William
1996
Language and Communicative Practices. Boulder, CO: Westview Press.
Horn, Lawrence R.
1984
Toward a new taxonomy for pragmatic inference: Q-based and R-based implicature. In: Deborah Schiffrin (ed.), Meaning, Form, and Use in Context: Linguistic Applications, 1142. Washington DC.: Georgetown University Press.
Huang, Yan
2007
Pragmatics. Oxford: University Press.
Hymes, Dell
1962
The ethnography of speaking. In: Thomas Gladwin and William C. Sturtevant
(eds.), Anthropology and Human Behavior, 1353. Washington: Anthropological Society.
Leech, Geoffrey
1983
Principles of Pragmatics. London: Longman.
Levinson, Stephen
1983
Pragmatics. Cambridge: University Press.
Levinson, Stephen
2000
Presumptive Meanings. The Theory of Generalized Conversational Implicature. Cambridge, MA: MIT Press.
McNamara, Tim and Carsten Roever
2006
The social dimension of proficiency: How testable is it? Language Learning
56/2: 4379.
Mey, Jacob L.
1979
Introduction. In: Jacob L. Mey (ed.), Pragmalinguistics. Theory and Practice,
1017. The Hague: Mouton.
Mey, Jacob L.
1993
Pragmatics. An Introduction. Oxford: Blackwell.
Mey, Jacob L.
2000
When Voices Clash: A Study in Literary Pragmatics. Berlin: Mouton de
Gruyter.

20

Wolfram Bublitz and Neal R. Norrick

Morris, Charles
1938
Foundations of the Theory of Signs. In: Otto Neurath, Rudolf Carnap and
Charles Morris (eds.). International Encyclopaedia of Unified Science,
77138. Chicago: University of Chicago Press.
Prucha, Jan
1983
Pragmalinguistics: East European Approaches. Amsterdam: Benjamins.
Sperber, Dan and Deirdre Wilson
1986. Relevance: Communication and Cognition. Oxford: Blackwell.
Thomas, Jenny
1995
Meaning in Interaction: An Introduction to Pragmatics. London: Longman.
Wenger, Etienne
1998
Communities of Practice: Learning, Meaning, and Identity. Cambridge: Cambridge University Press.

Part I
Conceptual foundations

1.

Pragmatics as a linguistic concept1


Anita Fetzer

1.

Introduction

Pragmatics is fundamentally concerned with communicative action and its felicity


in context, investigating action with respect to the questions of what action is, what
may count as action, what action is composed of, what conditions need to be satisfied for action to be felicitous, and how action is related to context. These research questions and the object of research require action in general and communicative action in particular to be conceived of as relational concepts, relating action
and context, relating action and communicative action, relating communicative action and interlocutors, and relating interlocutors with the things they do with words
in context. The inherent parts-whole perspective requires the explicit accommodation of a theory of context, capturing the dynamic processes from the connectedness between parts and wholes thus transcending clearly delimited frames of investigation.
The multifaceted research paradigm of pragmatics has provided new directions
and perspectives in arts and humanities, philosophy, cognitive science, computer
science and social science. Pragmatic perspectives have been employed in information technology and in the social sciences, particularly in economics, politics
and education. In line with Wittgensteins conception of language games and Levinsons conception of activity types (Levinson 1979; Wittgenstein 1958), game
theory (Lewis 1979) has been adapted to their particular contextual constraints and
requirements. The change of perspective from objects to interlocutors (or users),
and the accommodation of their wants and needs is reflected in the paradigm-anchored collocations of pragmatic thinking and learning, pragmatic software, pragmatic design, pragmatic modelling and pragmatic technology on the one hand, and
pragmatic foreign policy, pragmatic politicians, pragmatic sanction and pragmatic
nationalism on the other. Hence, it is no longer solely individual actions and their
perlocutionary effects, which are at the heart of analysis, but rather whole games,
considering both the present state of affairs and its connectedness with prior and
succeeding actions. In philosophical contexts, pragmatics is related closely to William James, who pointed out its practical nature, and to Charles Peirce, who
referred back to Immanuel Kants conception of pragmatics as experimental, empirical and purposive thought (e.g., The New Encyclopaedia Britannica 1976; cf.
also Koyama this volume). Pragmatics is also frequently connected with Greek
pragma, meaning action and instructive or useful (e.g., Bublitz 2009; Cummings 2005).

24

Anita Fetzer

The connectedness between action and purpose is also reflected in its ordinary-language meaning learning by doing (cf. Longman DCE 2003; Collins
Cobuild 1990). As a non-scholarly, ordinary-language notion, pragmatics is functionally synonymous with practical or just right at that stage, describing a surface-bound and action-based phenomenon within a particular situation. It occurs
primarily in its adjectival form, foregrounding practical and unexpected, rather
than theoretical and dogmatic considerations. For instance, politicians and economists may argue for a case on wholly pragmatic grounds, brushing aside traditionally established borders, which tend to result from presupposed and taken-forgranted ideology. Whenever an action, decision or solution is assigned the status
of a pragmatic action, a pragmatic decision or a pragmatic solution, it is contrasted implicitly with its theory-based and abstract-principles-driven counterparts. Against this background, the adjective pragmatic in its function as a modifier anchors the head of a noun phrase to the here and now of discourse. To employ
Gricean terminology from his cooperative principle, it signals that the ongoing discourse is such as is required, at the current stage at which it occurs, by the accepted purpose or direction of the talk exchange (Grice 1975: 45). Pragmatics in
its non-scholarly ordinary-language meaning thus meets with one of the most fundamental premises of scholarly pragmatics, the Gricean cooperative principle.
There is a further bridging point between non-scholarly pragmatics and scholarly
pragmatics: the differentiation between brute facts and institutional facts. According to Searle (1995), facts, that is brute facts, mental facts, social facts and institutional facts, constitute and regulate the world. To have a good knowledge of the
present situation, interlocutors need to have a good knowledge of what regulates
the world and what constitutes the world. Only then may they construct alternative
visions of a present situation or alternative social realities.

2.

Pragmatics in context

Pragmatics is frequently conceptualized as the science of language use, the study


of context-dependent meaning and the study of speaker-intended meaning, presupposing the existence of language, language user and context on the one hand, and
context-independent meaning on the other. To capture its multifaceted nature, definitions tend not to concentrate on the questions of what pragmatics is and what it
does, but rather on what it is not and what it does not do. The majority of conceptualizations pay tribute to Charles Morriss definition as the study of the relation
of signs to interpreters (Morris 1971: 6) and to Austins differentiation between
constative and performative (Austin 1971), foregrounding the performance of
communicative action and the effects the uttering of words may have. Against this
background, pragmatics is considered to be the study of invisible meaning (Yule
1996: 127), the science of the unsaid (Mey 2001: 194), the study of meaning as it

Pragmatics as a linguistic concept

25

emerges in language use (Marmaridou 2000: 1), and the study of linguistic acts
and communicative action and their appropriateness (Bublitz 2009; van Dijk 2008;
Fetzer 2004, 2007).
There have been various attempts to systematize the multifaceted and heterogeneous field of pragmatics, or to bring some order into the pragmatic wastebasket
(Bar-Hillel 1971), as is reflected in the differentiation between the broad fields of
scholarly pragmatics: general (or non-linguistic) pragmatics and linguistic pragmatics. Implicit in the philosophy-anchored general-action and linguistic-actionbased paradigms is the premise that pragmatics is conceptualized best as a perspective, which comprises a general pragmatic perspective, a social perspective, a compositional perspective and a relational perspective. These are introduced below.
2.1.

The pragmatic perspective

The pragmatic-perspective paradigm does not represent a clearly delimited field of


research but rather offers a general perspective towards the object to be examined.
It provides a point of view for analysis, in which not the object in isolation is examined but rather the nature of its connectedness with users and other objects, or with
the theoretical framework to which it is anchored. The perspective is thus shifted
from an analysis of the language system and its constitutive parts to that of its rulegoverned instantiation and embeddedness in context, considering its generalized
and particularized conditions of use. That change in perspective has important consequences for methodology. To account for the parts-whole connectedness, a relational frame of reference is adopted, which accommodates the relation between
the language system, language use and language users, and what they do with
words in context, comprising linguistic context (or co-text), social context and
cognitive context.
The pragmatic-perspective paradigm provides a general cognitive, social and
cultural perspective on linguistic phenomena in relation to their usage in forms of
behaviour, accounting for the dynamics of language and language use, as is reflected in the premises that meaning is not a product and given but rather dynamic,
multifaceted and negotiated in context. Language use is dependent on variability,
and language users adapt to contextual constraints accommodating them not
only in the formulation of utterances but also in their interpretation (e.g. Mey 2001;
Verschueren 1999). As in functional grammar and systemic functional grammar
(Givn 1993; Halliday 1994), the dynamic construct of meaning is a necessary
condition for making linguistic choices. Following Verschueren (1999: 5558), the
choices are made at every possible level of structure as regards form and strategy.
In line with the sociolinguistic premise of linguistic variation, which claims that
speaker-intended meaning can be formulated with different linguistic surfaces
(Brown 1995), these choices are not equivalent in status. They carry along their alternatives and are always seen and interpreted against the background of other,

26

Anita Fetzer

more or less marked possibilities. While unmarked variants are seen as more frequent in distribution, more conventional in formulation and easier to process,
marked variants are less frequent in distribution, less conventional in formulation
and harder to process (Fetzer 2002, 2007).
The pragmatic perspective is not only reflected in the more general approach
examined above, considering linguistic communication as well as linguistic behaviour. It is also found in the particularized, hyphenated fields of pragmatics, as in
cognitive pragmatics (cf. vol. 4 Cognitive pragmatics of this series Handbooks of
Pragmatics), empirical pragmatics (cf. Bednarek this volume, Andersen this volume, Rhlemann this volume), experimental pragmatics (cf. Breheny this volume), formal pragmatics, historical pragmatics (cf. vol. 8 Historical pragmatics of
this series Handbooks of Pragmatics), intercultural pragmatics (cf. vol. 7 Pragmatics across languages and cultures of this series Handbooks of Pragmatics), interlanguage pragmatics, sociopragmatics (cf. vol. 5 Pragmatics of society of this
series Handbooks of Pragmatics) and theoretical pragmatics.
2.2.

The social perspective

An analysis of the rule-governed and strategic use of language in context anchored


in a social-perspective goes beyond traditional linguistic analyses. It also considers
social and sociocultural aspects of communication in an explicit manner, assigning
them the status of a constitutive part of communicative action, thus overlapping
with the research paradigms of anthropological linguistics (Duranti 1997; Hanks
1996), interactional sociolinguistics (Auer and Di Luzio 1992; Gumperz 1992,
1996, 2003), (critical) discourse analysis (Fairclough 1992; van Dijk 2008) and
ethnomethodological conversation analysis (Garfinkel 1994; Heritage 1984).
Societal pragmatics examines social parameters which affect the production
and interpretation of utterances, placing language use in an external relation to language users. It is based on the premise that language use and social structure are
connected dialectically, (re)constructing social and sociocultural context by confirming (or disconfirming) social values in interaction, for instance gender, power
and social status. Consequently, the language user and her/his social roles and
identities are at the heart of societal pragmatics, analysing social and sociocultural
context on the micro level of face-to-face interaction as well as on the macro level
of institution. In interaction, language users interactionally organize their social
roles and identities while at the same time constituting social context. To use Heritages (1984: 242) terminology, the production of talk is doubly contextual. An utterance relies upon the existing context for its production and interpretation, and it
is, in its own right, an event that shapes a new context for the action that will follow.
The social perspective to pragmatics has shifted the focus of investigating context-dependent meaning from semantics-based methodologies and their premise of

Pragmatics as a linguistic concept

27

semantic well-formedness and propositional truth to social context-based approaches and their premise of negotiated communicative meaning and sociocultural appropriateness (Fetzer 2007), thus adding a further layer of meaning: social
and sociocultural meaning. In this perspective, meaning is not given per se and carried as such into the utterance, metaphorically speaking. Rather, meaning, viz.
lexical meaning and utterance meaning, is brought into the utterance and its diverse layers of meaning are brought out in the utterance, foregrounding salient
meanings while backgrounding non-salient ones (to employ interactional-sociolinguistic terminology, Gumperz 1992, 1996). Language is assigned the status of a
sociocultural construct which is used strategically by rational language users in
context, considering possible perlocutionary effects their utterances may trigger as
regards negative and positive politeness (Brown and Levinson 1987), for instance.
The social-meaning perspective also prevails in ethnographic studies, sociolinguistics and interactional sociolinguistics, as well as in ordinary language philosophy (Brandom 1994; Recanati 2004; Searle 1969, 1975, 1995). It is in disaccordance with frameworks conceptualizing language and language use as mutually
exclusive concepts, as is the case with the dichotomies of linguistic competence
and communicative performance, I(nternalized)-language and E(xternalized)-language, or langue and parole.
In the Anglo-American research tradition, the social perspective and the pragmatic perspective are referred to as European-Continental pragmatics. This field
defines pragmatics in a far broader way, encompassing much that goes under the
rubric of sociolinguistics, psycholinguistics, and discourse analysis (Huang 2007:
4), thus not falling strictly within formal linguistic theory (Horn and Ward 2004:
xi). The Anglo-American tradition tends to favour the compositional perspective.
2.3.

The compositional perspective

In the pragmatics-as-component perspective pragmatics is assigned the status of


a necessary component (or module) of grammar, thus anchoring pragmatics firmly
to linguistics (Birner and Ward 2006; Carston 2002; Huang 2007; Horn 1984;
Jaszczolt 2005; Levinson 2000; Sperber and Wilson 1995). Pragmatics is conceived of as an individual component of grammar, interacting with the components
of semantics, morphology, syntax and phonology, as is the case in its most prominent representatives: theoretical pragmatics and formal pragmatics. Radical pragmatics departs from the premise of strict modularity or discrete compositionality
by merging semantics and pragmatics. It is based on the premise of sense generality and contextualism, presuming that the output of semantic meaning is underdetermined and assigned determinate meaning in context through inference and pragmatic enrichment.
The compositional perspective is based on the premise that language is a theoretical construct focussing on language-internal rules and regularities, and on in-

28

Anita Fetzer

ferencing processes, thus considering language use and context in an explicit


manner. In contrast to societal pragmatics, however, its conception of context tends
to be restricted to linguistic context. In the case of deixis and its realization as deictic expressions, e.g., temporal and local deixis, participant deixis and discourse
deixis (Hanks 1996 and this volume; Levinson 1983), the compositional perspective needs to accommodate physical components anchored in the social context,
e.g., speaker and hearer, and time and place. Participants tend to be conceptualized
as the prototypical constructs of Model Speaker, Model Addressee or Model Person, viz. ideally competent interlocutors in the domains of linguistic competence,
communicative competence and rationality (Brown and Levinson 1987; Jaszczolt
2005), abstracting from their social, discursive and interactional roles. The same
holds for the conceptualization of time and place.
At the heart of compositional-perspective investigations are pragmatic universals, viz. deixis, reference, non-literal meaning, indefiniteness, presupposition, information structure, discourse marker, speech act and implicature. To account for
their function, pragmatics differentiates between an abstract construct and its linguistic realization (or representation). For instance, the pragmatic universal of deixis is examined with respect to deictic expression, and the universal of speech
act is analysed with respect to its constitutive acts and their direct, indirect, and
conventionalized indirect realizations (Sbis 2002; Searle 1995; Brown and Levinson 1987). Closely connected with indirect speech acts are speaker- and heareroriented cognitive operations: the former are conventional implicature, generalized and particularized conversational implicature (Levinson 2000; Horn 1989;
Huang this volume), and impliciture (Bach 2006), and the latter are inference and
abduction (Cohen, Morgan and Pollack 1992; Givn 1989, 2005). Inferencing is
also indispensable for the definition of intentionality of communicative action and
non-literal meaning, for indefiniteness and discourse markers, and for presupposition and information structure. Another pragmatic universal is face (Brown and
Levinson 1987; Goffman 1981, 1986), which is also of key importance to the social perspective.
More recently, the compositional perspective has undergone a shift in focus
and methodology, calling for the accommodation of methodological compositionality to bridge the gap between two competing paradigms (Jaszczolt 2005).
Only then is it possible to account for speech act and uptake, speech act type and
cognitive reality, and speech act type and situation (mis)match. The change in direction from monological, speaker-centred methodologies to dialogical, co-participant-centred interaction requires the explicit consideration of the connectedness
amongst world knowledge, linguistic knowledge, discourse knowledge, the properties of the human inferential system and stereotypes and presumptions about society and culture, as is accommodated in the relational perspective.

Pragmatics as a linguistic concept

2.4

29

The relational perspective

The three perspectives adopted towards the heterogeneous field of pragmatics all
share a particular perspective towards their object of investigation, accounting for
its connectedness with context as regards utterance meaning and conditions of use.
For this reason, they adopt a more or less explicit relational frame of reference.
They go separate ways, however, in their accommodation of social context. While
the social perspective considers social context to be a key concept, the compositional perspective tends to disregard it, concentrating on linguistic context and
cognitive context. All of them are based on the premise that language is as a rulegoverned system, and all of them intend to account for form, function and meaning. The types of meaning under examination and their conceptualizations are different, though.
In a relational frame of reference, language may be examined from a parts perspective, analysing the form, function and meaning of individual parts, viz.
sounds, morphemes and constituents. It may be investigated from a holistic viewpoint, viz. phonological words and lexical words, and propositions and sentences,
and it may be analysed from a parts-whole perspective considering the nature of
their connectedness. The latter perspective is intrinsically dynamic, focussing not
on the object as such, but rather on its embeddedness in context, its conditions of
use and its appropriateness. The examination of linguistic form does not generally
cause delimitation problems, while the analyses of linguistic meaning and function
have become increasingly messy, necessitating the explicit accommodation of a
pragmatic wastebasket for non-fits and other-fits.
Pragmatics is generally delimited from semantics by shifting the focus of investigation from what is said, viz. semantic, literal or truth-conditional meaning, to
what is meant, viz. what is communicated implicitly, what is implicated, utterance
meaning or meaningnn (e.g., Grice 1975; Jaszczolt 1999, 2005; Levinson 1983,
2000; Recanati 2004; Saeed this volume). While the object of research is meaning
in semantics as well as in pragmatics, accounting for world-to-language, languageto-world and language-to-world-to-mind relations, the delimitation of the object,
the focus of investigation and the methodologies employed differ. In semantics the
meaning of a lexeme, proposition or discourse unit is defined from a unit-of-investigation-internal perspective, and in pragmatics, the meaning of an utterance,
speech act or conversational contribution is defined from unit-of-investigation-external and unit-of-investigation-internal perspectives. Metaphorically speaking,
semantics examines what is in the container, and pragmatics examines what is
brought out of the container (Reddy 1979).
The following figure attempts to bring some order in the pragmatic wastebasket
by systematising the two closely related but yet distinct fields of research, which
intersect and overlap in their analysis of meaning in context.2 While meaning is
conceptualised as conventional, truth-conditional meaning with discrete bound-

30

Anita Fetzer

Figure 1. Meaning in context

Pragmatics as a linguistic concept

31

aries in semantics, pragmatic meaning is seen as speaker-intended, non-truthconditional meaning with fuzzy boundaries, as is indicated by the dotted lines. Furthermore, meaning is autonomous and represented by well-formed context-independent sentences in semantics, and it is inferred and represented by appropriate
context-dependent utterances in pragmatics. Semantics and pragmatics meet at the
interpretation interface, where the calculation of utterance meaning is informed by
the linguistic code and at the same time by pragmatic inferencing. While the
former is a unidirectional operation as is signified by single arrows, the latter is dialectical and interactive in nature as is indicated by double arrows, capturing the dynamics of pragmatic meaning:
The performative and its constitutive parts, viz. meaningnn, non-truth-conditional meaning and presupposition, are among the most important pillars of pragmatics. They are also relevant to the definition of speech acts and their realization
in context as utterance meaning, which is a linguistic-pragmatics phenomenon par
excellence. It is at the interpretation interface where context-dependent implicatures are calculated and where pragmatics interacts with semantics, syntax, morphology and phonology.
In the following, the research domain of general pragmatics is introduced, concentrating on its bridging points with linguistic pragmatics.

3.

General pragmatics

General pragmatics is set firmly in the research paradigm of philosophy, considering cooperation, action theory, intentionality, rationality and context, thus providing relevant bridging points for language-anchored linguistic pragmatics. Both research paradigms analyse pragmatic universals and pragmatic mechanisms, such
as speech act and implicature (e.g. Austin 1971; Bach 1992, 2006; Grice 1975;
Sbis 2002; Searle 1975; Vanderveken and Kubo 2002), indexical and deixis (Recanati 2004; Levinson 1983), common ground and presupposition (Akman et al.
2001; Bouquet et al. 1999), focussing on the connectedness between parts and
wholes, and between content and context (Stalnaker 1999). While general pragmatics adopts an action-based frame of reference, linguistic pragmatics examines
the form and function of speech acts in various languages.
In general pragmatics, action and cooperation are analysed in the framework of
an ideal pragmatic situation in which communication is neither distorted by social
factors, e.g. institutional constraints, unequal social hierarchies and other forms of
oppression (e.g., Habermas 1998, cf. Cooke this volume), nor by interpersonal
considerations, such as face-wants and face needs (Brown and Levinson 1987),
politeness or etiquette (Watts, Ide and Ehlich 1992). The high degree of idealization inherent in this research paradigm is also reflected in its conception of language and language use, which does not refer to natural language as spoken by or-

32

Anita Fetzer

dinary language users in context but rather an ideal language built in accordance
with the language-system specific constraints and requirements. In an ideal pragmatic situation, all necessary felicity conditions obtain, such as normal input and
output conditions, propositional content and preparatory conditions, sincerity and
essential conditions (Searle 1969). Consequently, rational agents have ideal linguistic, communicative and cognitive competence as well as an ideal world knowledge (cf. Cohen, Morgan and Pollack 1992).
Pragmatic universals can be assigned the status of basic building blocks of
communication. Among the most prominent ones are the speech act and its constitutive acts, viz. propositional act composed of further constitutive acts and illocutionary act (Searle 1969; Vanderveken and Kubo 2002), indexicals, for instance
pure and impure indexicals, intentionals and perspectivals (Recanati 2004), and
presuppositions and presupposition triggers (Huang 2007; Levinson 1983; Stalnaker 1999). Another basic building block is the rational agent or model user,
who employs building blocks in a strategic manner to speaker-intend communicative meaning, thus achieving particular goals in context. The model user can be a
natural user, that is speaker and hearer of a natural language, or an artificial user,
such as a robot or a dialogue system. In accordance with the fundamental pragmatic premises of rationality and intentionality, the model user is equipped with a
reasoning device, accommodating pragmatic inferencing and default inferencing,
top-down reasoning and bottom-up reasoning, and abduction. This is a necessary
condition for a model user to calculate and formulate speaker-intended implicatures and to perform the hearer-intended logical operations of inference and abduction (Cohen, Morgan and Pollack 1992; Givn 1989, 2005; Huang this volume;
Sperber and Wilson 1995).
3.1.

Presupposition and common ground

Presupposition refers to a proposition or an inference whose validity is taken for


granted for a sentence to be true or for a speech act to be felicitous. Philosophy and
linguistics distinguish between pragmatic presupposition and semantic presupposition. The satisfaction of semantic presupposition is a necessary condition for
the truth value of a sentence, and the satisfaction of pragmatic presupposition is
necessary for a speech act to be appropriate in context. Pragmatic presuppositions
are accommodated in speech act theorys felicity conditions, which are considered
as linguistic and social context categories and their satisfaction is assigned the
status of a default configuration (Sbis 2002). Presuppositions are generated by
presupposition triggers, for instance definite descriptions, factive verbs, aspectual
verbs, change of state verbs, iteratives and clefts (cf. Huang 2007, this volume; Levinson 1983).
Some contexts allow presupposition neutralisation, and more recently presupposition triggers have been classified as hard presupposition triggers and soft pre-

Pragmatics as a linguistic concept

33

supposition triggers. Presuppositions are connected intrinsically with propositions


and assumptions which are taken for granted and therefore do not need to be made
explicit. They are organized and administered in the framework of context sets
(Stalnaker 1999), which serve as common ground in communication. Consequently, pragmatic presupposition may be seen as a restriction on common ground.
Common ground is indispensable for philosophical and cognitive conceptions
of knowledge where it serves as background for reasoning and for retrieving
speaker-intended meaning and other types of implicit meaning, such as indexical
expression or implicature. In the field of computer science, common ground and
world knowledge are frequently conceptualized as a database. It is seen as comprising a set of propositions, which serve as a resource for the understanding of utterances. Common ground is implicit but can be made explicit via propositions,
and model users presuppose its validity and fall back on it when they retrieve implicatures (cf. Bublitz 2006).
Common ground in the sense of background also plays a fundamental role in
dialogue system modelling. According to Vanderveken and Kubo (2002), model
users negotiate the compatibility of background with utterances and their felicity
and satisfaction conditions in and through the process of communication. However, background and context are not identical because possible contexts of utterance can have different backgrounds. As a consequence of this, background contains not only mutual knowledge of facts about the conversational background but
also knowledge about the world and of the world, such as ethical norms and sociocultural values, transcending the common sense notion of context. Searle (1995)
considers background to be a necessary condition both for literal and non-literal
meaning, thus assigning it the status of a basic premise for felicitous communication. He defines background as an open-ended set of skills, pre-intentional assumptions and practices, which are not representational but rather enable intentional acts and states to be made manifest.
In Relevance Theory (Sperber and Wilson 1995), common ground is conceived
of as a common set of premises for inference rules. The relevance-theoretic conception differs from the traditional notion of mutual knowledge by its attempt to
avoid the logical consequence of infinite regress, which follows from the code
model of communication. To avoid infinite regress, Sperber and Wilson base their
theoretical framework on an approximation of mutual knowledge, namely cognitive environments and mutual manifestness.
Common ground is a presupposed common knowledge base, which is a necessary condition for felicitous communication. To that base, model users anchor their
communicative action and they fall back on it, should they require further information, which may not be encoded explicitly. The overall encompassing concept of common ground has been given a context-dependent interpretation as conversational record (Thomason 1992) based on Lewiss concept of accommodation
(Lewis 1979). Conversational record contains public information, presumptions,

34

Anita Fetzer

and an update operator. Other context-dependent subsets of common ground are


Clarks notions of personal common ground, which contains a model users subjective experience, and communal common ground, which stores social experience
(Clark 1996), and Fetzers dialogue common ground, which is differentiated in individual dialogue common ground, containing a model users representation of
her/his dialogue common ground, and collective dialogue common ground, viz. the
model users representation of the set of model users representation of dialogue
common ground (Fetzer 2004). All of the subsets need to be interconnected, constantly updated and, if necessary, revised.
3.2.

Context

The analysis of context-dependent meaning is at the heart of pragmatics, and for


this reason context is one of its key objects of investigation. The theory, practice
and implementation of context are also of relevance to diverse fields of investigation, ranging from philosophy and computer-mediated communication to cognitive science, in particular dialogue management, artificial speech production,
artificial intelligence, distributed knowledge representation, robotics and information technology.
The heterogeneous nature of context and the context-dependence of the concept itself have made it almost impossible for the scientific community to agree
upon one commonly shared definition or theoretical perspective, and frequently,
only a minute aspect of context is described, modelled or formalized (cf. Akman et
al. 2001; Blackburn et al. 2003; Bouquet et al. 1999). Because of its multifaceted
nature and inherent complexity, context is no longer considered an analytic prime
but rather seen from a parts-whole perspective as an entity containing sub-entities
(or sub-contexts).
The multilayered outlook on context contains a number of different perspectives. First, context is conceived as a frame of reference whose job it is to frame
content by delimiting the content while at the same time being framed and delimited by less immediate adjacent frames. The nature of the connectedness between the different frames is a structured whole composed of interconnected
frames (Goffman 1986). The gestalt-psychological figure-ground scenario prevails
in psychological and psycholinguistic perspectives on context. It is also adopted in
cognitive pragmatics as is reflected in the relevance-theoretic conception of context as an onion, metaphorically speaking. The individual layers are interconnected
and their order of inclusion corresponds to their order of accessibility (Sperber and
Wilson 1995), as is reflected in inferencing and other kinds of reasoning.
Second, context is seen as a dynamic construct, which is interactionally organized in and through the process of communication. This view prevails in ethnomethodology (Garfinkel 1994; Goodwin and Duranti 1992; Heritage 1984), interactional sociolinguistics (Gumperz 1996, 2003) and sociopragmatics (Bublitz 2009;

Pragmatics as a linguistic concept

35

Fetzer 2007, Schmid 2003), where context is assigned the dual status of process
and product. The dynamic outlook is based on the premise of indexicality of social
action, and the joint construction of context. In the primarily qualitatively oriented
paradigms, context is connected intrinsically with adjacency pair, conditional relevance and the turn-taking system on the micro level, and with institution on the
macro level, whose order is captured through context-independent and contextsensitive constraints and requirements. Closely related to the conception of context
as a dynamic construct is its relational conception, which conceives it as relating
communicative actions and their surroundings, relating communicative actions,
relating individual participants and their individual surroundings, and relating the
set of individual participants and their communicative actions to their surroundings (Fetzer and Akman 2002).
Third, context is seen as given as is reflected in the presuppositional approach
to context, which is also referred to as common ground or background information
(Stalnaker 1999). Here, context is seen as a set of propositions, which participants
take for granted in interaction. This allows for two different conceptions of context: a static conception in which context is external to the utterance, and an interactive one, in which context is imported into the utterance while at the same time
invoking and reconstructing context.
The context-dependence of context is reflected in its status as given and external to the utterance, reconstructed and negotiated in communication, indexical, and
never saturated. A further classification of context is anchored to a holistic conception of context embedding its constitutive parts of model user, conversational contribution, surroundings and their presuppositions, viz. cognitive context, linguistic
context and social context (Fetzer 2002, 2004).
Linguistic context comprises language use and is delimited by the constraints
and requirements of genre. Language is composed of linguistic constructions (or
parts) embedded in adjacent linguistic constructions (further parts), composing a
whole clause, sentence, utterance, turn or text. Linguistic context is functionally
synonymous to co-text (de Beaugrande and Dressler 1981; Widdowson 2004), denoting a relational construct composed of local and global adjacency relations.
Analogously to speech-act-theoretic constitutive rules and regulative rules (Searle
1969), linguistic constructions constitute text. The production and interpretation of
linguistic constructions is constrained by the rules of grammar, and the production
and interpretation of speech acts are constrained by felicity conditions.
Cognitive context is the foundation on which inference and other forms of reasoning are based. Constitutive elements of cognitive context are mental representations, propositions, contextual assumptions and factual assumptions. Since cognitive contexts are anchored to an individual but are also required for a cognitively
based outlook on communication, they need to contain assumptions about mutual
cognitive environments. Thus, cognitive context is not only defined by representations but also by meta-representations. In the social-psychological paradigm,

36

Anita Fetzer

cognitive context is conceptualized along the lines of the gestalt-psychological distinction between figure and ground and the related metacommunicative concepts
of frame and framing (Bateson 1972; Goffman 1986). Frame is seen as a delimiting
device, which is (or delimits) a class or set of messages (or meaningful actions)
(Bateson 1972: 187). Because of its delimiting function, psychological frames are
exclusive, i.e. by including certain messages (or meaningful actions) within a
frame, certain other messages are excluded and they are inclusive, i.e. by excluding certain messages certain others are included (ibid.). This also holds for context which, analogously to frame is also structured and metacommunicative, or to
use Batesons words: the hypothesis depends upon the idea that this structured
context also occurs within a wider context a metacontext if you will and that
this sequence of contexts is an open, and conceivably infinite, series (Bateson
1972: 245).
Social context comprises the context of a communicative exchange and is defined by deducting linguistic context and cognitive context from a holistic conception of context. Constituents of social context are, for instance, model users, the
immediate concrete, physical surroundings including time and location, and the
macro contextual institutional and non-institutional domains. The connectedness
between language and language use on the one hand, and linguistic context, social
context and cognitive context on the other is reflected in deixis, viz. temporal deixis, local deixis, participant deixis, discourse deixis and social deixis (Hanks
1996). Furthermore, the category of model user (as speaker and hearer) can no
longer be conceived of as an analytic prime but needs to be refined by the accommodation of footing (Goffman 1981; Levinson 1988). The importance of social
context to communication is spelled out by Hanks as follows: Hence it is not that
people must share a grammar, but that they must share, to a degree, ways of orienting themselves in social context. This kind of sharing partial, orientational and
socially distributed may be attributed to the habitus, or relatively stable schemes
of perception to which actors are inculcated (Hanks 1996: 235).
3.3.

Cooperation

The pragmatic principle of cooperation is connected intrinsically with coordination and collaboration on the one hand (Clark 1996; Cummings 2005; Grosz and
Sidner 1992), and with the pragmatic principles of rationality and intentionality
(Cohen, Morgan and Pollack 1992; Searle 1983) on the other. The principles are
examined in philosophy as regards their contents, function and status (Brandom
1994; Penco 1999), and they are analysed in psychology as mental representations,
mental states and beliefs (Bouquet et al. 1999; Sperber and Wilson 1995). Against
this background, language and language use are seen as fundamentally cooperative, and subsequently as intrinsically discursive and social (e.g., Grice 1975;
Lewis 1979; Clark 1996).

Pragmatics as a linguistic concept

37

The connectedness between cooperation and intentionality, and the differentiation between I-intentions and we-intentions are at the heart of Searles conception
of speech act theory. To use his own words: Collective intentionality presupposes
a Background sense of the other as a candidate for cooperative agency; that is, it
presupposes a sense of the others as more than just conscious agents, indeed as actual or potential members of a cooperative activity (Searle 1991: 414). In his
work, Searle is very explicit about the nature of the parts-whole connectedness of
individual speech acts and I-intentions on the one hand, and of a conversation or
discourse and we-intentions on the other: The reason that we-intentions cannot be
reduced to I-intentions, even I-intentions supplemented with beliefs and beliefs
about mutual beliefs, can be stated quite generally. The notion of a we-intention of
collective intentionality implies the notion of cooperation (Searle 1991: 406).
In the Gricean approach to natural-language communication, in which language is seen as both purposive and social, and rational and cognitive, cooperation
counts as the fundamental premise of communication. As a philosophical concept,
cooperation is anchored to the premise of rationality and intentionality. Consequently, conversations are rational endeavours and cooperative efforts, in which
model users realize and recognize a common purpose or set of purposes, or, at
least, some mutually accepted direction. This is reflected in the Gricean cooperative principle which reads as follows: Make your conversational contribution
such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged (Grice 1975: 45). The cooperative principle is a necessary requirement for the calculation of pragmatic
meaning or meaningnn, and is for this reason not open for negotiation. Model users
may, however, employ the pragmatic strategy of opting out, which is generally accompanied by explicit references to some higher-order moral principle, such as
confidentiality or trust. When they opt out, they generally spell out the reasons for
their non-compliance, thus acting in accordance with the cooperative principle on a
deeper level. The cooperative principle subsumes four conversational maxims, the
maxim of quality, the maxim of quantity, the maxim of relation, and the maxim of
manner (Grice 1975: 4546). Further constitutive parts of the cooperative principle are the conventional and conversational implicatures, which are dealt with
below in the section 4.2 on linguistic pragmatics.
A cooperation-anchored outlook on language and language use not only is
manifest in a conversation as a whole, but also in its building blocks of speech act
and in the act of referring and predicating. Reference operates on the phrasal level
and involves a speakers use of a linguistic expression, which tends to be a noun
phrase, to induce a hearer to access or create some entity in her/his mental model of
the discourse. Reference may be performed directly with a noun phrase and indirectly as deferred reference in the contexts of pars-pro-toto configurations, meaning transfer and figurative meaning. Referring seen from a cooperation viewpoint
is not restricted to the semantic domain of a proposition but also applies to the in-

38

Anita Fetzer

terpersonal and interactional domains, viz. negotiation of meaning and common


ground.
The status of cooperation as a universal pragmatic principle has been examined
extensively in general pragmatics, and there seems to be a general agreement on its
validity. However, the claim that the Gricean cooperative principle is universal has
been under severe attack. There have been modifications as regards the adoption of
a further constitutive pragmatic principle on the same level, the politeness principle (Leech 1983), and there have been various modifications of the number of
conversational maxims required for the calculation of pragmatic meaning as well
as their speaker- and hearer-oriented conceptualizations and interpretations (Horn
1984; Lakoff 1973; Levinson 2000).
Pragmatics is a field of research which intersects with various paradigms. As
regards meaning, it cannot but consider semantics. As regards rationality, it cannot
but consider cognitive science, and as regards cooperation, it cannot but consider
philosophy. Thus, pragmatics needs linguistics and linguistics needs pragmatics.
For a thorough investigation of meaning, we need to examine how things are done
with words, and how conversational contributions are structured to achieve particularized goals in context. This is done in linguistic pragmatics, where generalpragmatic principles and pragmatic universals undergo language- and culture-specific modification.

4.

Linguistic pragmatics

Linguistic pragmatics is defined as the science of language use, and [i]n the same
way as human actions change existing reality, linguistic actions also change the
world (Marmaridou 2000: 22). Linguistic pragmatics and general pragmatics
share almost identical goals: general pragmatics examines pragmatic principles,
mechanisms and universals in the context of action theory, rationality and intentionality, while linguistic pragmatics focuses on their instantiation in language and
language use. Hence linguistic pragmatics overlaps with general pragmatics, sharing its generalized principles, mechanisms and universals, and departs from the
generalized framework by concentrating on language as a general construct and on
languages as particularized instantiations.
In linguistics, language tends to be examined in its own right, accounting for
language use, but not for the model user. The language system comprises the constitutive subsystems of phonology, morphology, syntax and semantics. These may
be assigned the status of autonomous modules, as is the case in the formal paradigms, they may be conceived of as connected subsystems, as is the case in systemic functional grammar (Halliday 1994), or as dialectically connected subsystems with fuzzy boundaries, as for instance the cognitive paradigm (Givn 1993,
2005), where language use is connected dialectically with the language system and

Pragmatics as a linguistic concept

39

the model user. Language is seen as a dynamic construct changing through patterned use in context, in which grammatical and pragmatic aspects of interpretation
interface (Ariel 2008: 257259).
The dialectical connectedness between languages constitutive subsystems
requires the explicit accommodation of pragmatic interfaces. The pragmatic perspective adopted towards the object to be examined does not consider the object in
isolation but rather its connectedness with its conditions of use and with the user.
This is reflected in grammaticalization or grammaticization (Traugott 1988) and in
pragmaticalization (Aijmer 1997). Pragmatic principles shape the lexicon, as is reflected in Horns division of pragmatic labour, viz. his hearer-oriented Q-principle
and speaker-oriented R-principle (Horn 1984), and their impact on the structure
and use of lexical items, e.g. pragmatic strengthening, scalar implicature or polarity items. It is also manifest in deictic expressions and tense and aspect markers,
whose meanings are constant but whose referents vary with the speaker and the
hearer, and with time and place of the utterance, and on the style, register or purpose of a speech act (Ariel 2008; Huang 2007). Furthermore, pragmatic principles
have a strong influence on the information status of constituents within and across
sentences regarding focus, topic (or theme) and comment (or rheme). Deviations
from the default generate implicatures, such as emphasis implicating a set of alternatives (Birner and Ward 1998; Lambrecht 1994).
4.1.

Speech act

Speech act theory is connected intrinsically with J.L. Austins groundbreaking


lectures on ordinary language philosophy, edited posthumously (Austin 1971).
Austin scrutinized the then prevailing view that language describes the world in
terms of true and false, pointing out its shortcomings if applied to natural (or ordinary) language. He compared and contrasted the constative view, according to
which sentences are true or false, with the performative approach, according to
which sentences are used to perform speech acts, and came to the conclusion that
model users do not just describe the world, but rather do things with words, even
when describing the world. This is further refined by J.R. Searles modifications
and refinement in his research on speech acts (Searle 1969, 1991), systematizing
the cognitive, linguistic and social contexts, in which speech acts are performed
by differentiating between constitutive rules, which are a necessary part of the
language game and make speech acts count as particularized speech acts, e.g.
promise, threat or rejection, and regulative rules, which are normative in nature.
Considering language use from an action-theoretic frame of reference has paved
the way for investigating linguistic meaning as well as speaker-intended meaning. It has shifted the focus of investigation from the rigid framework of formal
semantics to the action-theoretic premises of rationality, intentionality and communication.

40

Anita Fetzer

Speech act theorys basic unit of investigation is the speech act, and depending
on the frameworks employed, speech acts are divided into locutionary, illocutionary and perlocutionary acts (Austin 1971), or propositional and illocutionary acts
(Searle 1969), and their respective sub-acts, viz. reference and predication in the
Searlean paradigm, and the phonetic act, phatic act and rhetic act in Austins frame
of reference. Speech acts and their constitutive acts are performed simultaneously,
and a speech act may be realized as a direct speech act, as an indirect speech act
and as a conventionally indirect speech act. To account for a speech acts perlocutionary effects as regards the social and interpersonal planes of discourse, speech
acts have been further refined and redefined as face-threatening acts, in which the
impact of a speech acts perlocutionary effects on the interpersonal plane is calculated with respect to the model persons, to employ Brown and Levinsons terminology, negative and positive face wants (Brown and Levinson 1987).
Speech act theory is anchored in action theory on the one hand, and in language
and language use on the other. Speech acts are described in various ways: they are
classified as explicit performative utterances, viz. utterances in which the speech
act contains a performative formula realized by a first-person pronoun and a
speech-act verb, e.g. I/we hereby close the meeting or we hereby request you to pay
your dues; they are classified as first-order speech act and second-order speech act,
accounting for the differentiation between direct and indirect speech acts; they are
classified as micro speech act and macro speech act, accounting for the larger unit
of genre (van Dijk 1981; Fetzer 2002), and as nth order speech acts, capturing the
multi-layered nature of performing speech acts in context. Furthermore, speech
acts are categorized with respect to possible directions of fit between language and
the world, that is the words-to-world direction of fit for assertion, the world-towords direction of fit for directive and commissive, the double direction of fit for
declaration and the empty direction of fit for expressive (Searle 1979). In addition
to the illocutionary-point anchored taxonomies, speech act verbs are classified and
systematized according to the way they are used in a particular language (cf. Collavin this volume).
Speech act theory combines language-internal and language-external factors
by systematizing them in the framework of felicity conditions, which are classified
with regard to propositional content conditions, preparatory conditions, essential
conditions and sincerity conditions. While propositional content conditions generally specify restrictions on the content of the proposition, preparatory, essential
and sincerity conditions explicate language-external conditions concerning the
language game as well as speaker- and hearer-specific requirements.
Speech acts are neither true nor false but rather felicitous or infelicitous. To be
felicitous, they need to be produced (and interpreted) in accordance with generalized felicity conditions, viz. normal input and output conditions, the essential and
sincerity conditions, and with particularized felicity conditions, viz. proposition
content condition.

Pragmatics as a linguistic concept

41

Speaking a language is seen as engaging in a (highly complex) rule-governed form of behavior (Searle 1969: 12). The rule-governed behaviour is specified in speech act theorys felicity conditions and in the requirement that the model
user in her/his role as speaker intends that her/his utterance will produce in a model
user in her/his role as hearer a belief that the sincerity and essential conditions obtain by means of the recognition of the intention to produce this belief, and that the
speaker intends that recognition to be achieved by means of the recognition of the
utterance as one conventionally used to produce such beliefs. Secondly, that the
semantic rules of the language spoken by the model users are such that the utterance is uttered sincerely and such as is required just in case all of the previous conditions obtain.
4.2.

Implicature

The cooperative principle counts as a universal principle in pragmatics, where it


represents the solid base to which communication in general and the formulation
and interpretation of communicative action in particular are anchored. This holds
especially for the calculation of context-dependent communicative meaning and
the necessary processes of inferencing required for the contextualization and enrichment of underspecified conversional contributions. To use Ariels words,
[u]nderdeterminacy is an inherent characteristic of human language, since no
natural language sentence can encode interlocutors intended statements fully
(Ariel 2008: 265). In a similar vein, but more explicitly, Levinson (1995) argues
that intentionality is a fundamental premise of natural-language communication.
In his words, human interaction, and thus communication depends on intentionascription. Achieving this is a computational miracle: inferences must be made
way beyond the available data. It is an abductive process of hypothesis formation,
yet it appears subjectively as fast and certain the inferences seem determinate,
though we are happy to revise them when forced to do so (Levinson 1995: 241).
Underdeterminacy and the necessary processes of inferencing are connected
with the Gricean paradigm and its differentiation between what is said and what is
meant, which both refer to utterances produced in context (Grice 1975); and they
are also connected with Relevance Theory (Sperber and Wilson 1995). In both
frames of reference, communication is seen as a context-dependent endeavour, in
which communicative meaning may go beyond the level of what has been said.
Hence, what is said cannot be equated with pure linguistic meaning but rather is
closely related to the conventional meaning of the words (the sentence) [] uttered (Grice 1975: 44). Unlike the rather controversial status of what is said in
semantics and pragmatics, what is meant has always been equated with what is
implicated.
Relevance Theory differentiates between a code model of communication and
an inference model. While the former is conventional, the latter is inferential and

42

Anita Fetzer

hence pragmatic by definition. Sperber and Wilson define inference as follows:


Inference is the process by which an assumption is accepted as true or probably
true on the strength of the truth or probable truth or other assumptions. It is thus a
form of fixation of belief (Sperber and Wilson 1995: 68). Their theory uses demonstrative inference as is employed in various types of logic, and non-demonstrative inference, which is employed in the process of inferential comprehension.
Unlike local deductive reasoning, inferential comprehension is global and based
on inductive rules.
Pragmatic inferencing occurs at various levels of comprehension. It is connected with the differentiation between explicature, which is a combination of linguistically encoded and contextually inferred conceptual features (Sperber and
Wilson 1995: 182), and different kinds of implication, viz. trivial implication, analytic implication, synthetic implication and contextual implication. Against this
background, implicature is a contextual assumption or implication which a
speaker, intending her utterance to be manifestly relevant, manifestly intended to
make manifest to the hearer. We will distinguish two kinds of implicatures: implicated premises and implicated conclusions (Sperber and Wilson 1995: 194195).
Implicated premises must be supplied by the hearer, who must either retrieve them
from memory or construct them by developing assumption schemas retrieved from
memory, and implicated conclusions are deduced from the explicatures of the
utterance and the context. What makes it possible to identify such conclusions as
implicatures is that the speaker must have expected the hearer to derive them, or
some of them, given that she intended her utterance to be manifestly relevant to the
hearer (1995: 195). With these tools, Relevance Theory explicates the relevant inferencing processes involved in communication. They delimit themselves from the
Gricean paradigm where, from their perspective, the successes of human non-demonstrative inference must be explained by appealing not to logical processes of
assumption confirmation, but to constraints on the formation and exploitation of assumptions (1995: 81), viz. the cooperative principle and the maxims.
Grice (1975: 4344) differentiates between implicate and the related nouns implicature (cf. implying) and implicatum (cf. what is implied). He distinguishes between two basic types of implicature: conventional implicature and conversational
implicature. The latter is subdivided into generalized conversational implicature
and particularized conversational implicature. Generalized conversational implicature is also referred to as default implicature or pragmatic regularities (Bach
2006). This is in line with Levinsons claim that utterance-types carry generalized
implicatures []: rational speakers meannn both what they say (except in non-literal uses of language) and what that saying implicates; different layers of meaning
all come under the umbrella of meaningnn (Levinson 2000: 373). While conventional implicature is connected closely with linguistic form, for instance with connectives (e.g., but), implicative verbs (e.g., manage, forget to), honorifics or nonrestrictive relative clauses, conversational implicatures are essentially connected

Pragmatics as a linguistic concept

43

with certain general features of discourse, viz. the Gricean conversational maxims.
Both conversational and conventional implicatures can be suspended: conventional implicatures are detachable, but not cancellable and conversational implicatures are cancellable but not detachable (for a thorough analysis of implicature, cf.
Huang this volume).
Conversational implicatures are generated or triggered by model users exploiting a maxim. That is to say, a model user gets in a conversational implicature if s/he
flouts a maxim, blatantly failing to fulfil it. Grice (1975: 4950) characterizes the
notion of conversational implicature as follows:
A man [or person, A.F.] who, by (in, when) saying (or making as if to say) that p has implicated that q, may be said to have conversationally implicated that q, PROVIDED THAT
(1) he is to be presumed to be observing the conversational maxims, or at least the cooperative principle; (2) the supposition that he is aware that, or thinks that, q is required
in order to make his saying or making as if to say p (or doing so in THOSE terms) consistent with this presumption; and (3) the speaker thinks (and would expect the hearer to
think that the speaker thinks) that it is within the competence of the hearer to work out,
or grasp intuitively, that the supposition in (2) IS required.

The conversational implicature is characterized by some degree of indeterminateness because it is, by definition, defeasible, non-detachable, calculable and nonconventional (cf. Bach 2006; e.g., Levinson 1979, 1983). Moreover, conversational implicatures are universal and reinforceable (Ariel 2008: 13). That is to say,
they can be asserted explicitly without bringing about a redundancy effect. It needs
to be pointed out, however, that the universality claim only holds for the inferential
mechanism and not for the relevant background assumptions.
Implicature is a cognitive mechanism anchored to model users in their roles as
speakers or producers of conversational contributions, and inference is a cognitive
mechanism anchored to model users in their roles as hearers. Cognitive processes
can be local, they can be global or they can be both. Not only natural model users,
but also non-natural intelligent agents may perform the cognitive operations of implicating and inferring, thus drawing invited inferences. For the operations to be
felicitous, compatible contexts for reasoning and drawing analogies are required
(cf. e.g., Bouquet et al. 1999; Cohen, Morgan and Pollack 1992).
In Gricean pragmatics, the conversational maxims are seen as specifications of
some unmarked communicative context representing the ground on which the conversational implicature is calculated. For this reason, deviations, however common, are seen as special or marked and signify that the speaker intends to communicate conversationally implicated meaning. The signalling is only possible
against the background of the Gricean premise of a reflexive intention, which is
succinctly explicated by Levinson: [w]hat distinguishes a Gricean reflexive intention from other kinds of complex reflexive intention is that the communicators
goal or intention is achieved simply by being perceived: recognition exhausts or
realizes the intention (Levinson 1995: 228).

44

Anita Fetzer

5.

Outlook

Neither general pragmatics nor linguistic pragmatics examine its objects of investigation in isolation but rather focus on their conditions of use, the connectedness
with their surroundings, and the necessary and sufficient conditions which assign
the object, e.g. intentionality, rationality, model user or action, the status of a particular object and make it count as that object. While general pragmatics concentrates on the analysis of these fundamental premises of practical action, identifying
their necessary and sufficient conditions, linguistic pragmatics establishes the explicit connection between those foundations and their language-specific and language-use specific constraints and requirements.
Pragmatics is more of a perspective towards an object under investigation than
the examination of the object as such. For this reason, it needs to touch on and interface with neighbouring disciplines, in particular philosophy, cognitive science,
neuroscience, linguistics and the social sciences. In spite of the diversity of the
field, the key research question of pragmatics is concerned with communicative
action, particularly with the expression and interpretation of meaning in context.
To tackle that question felicitously, pragmatics needs to accommodate extra-linguistic world knowledge, cultural and social stereotypes, and situation of discourse on the one hand, and word meaning, sentence structure and the cognitive
system, especially inference and abduction on the other. The generalized knowledge is stored as typed world knowledge and typed discursive knowledge. These
defaults represent pillars, against the background of which the argument from ignorance holds:
The argument from ignorance, it thus emerges, is an integral part of the inferential
mechanism by means of which implicatures are recovered in conversation. This argument permits us to conclude that conversational principles and maxims are being observed on the grounds that there is no reason to believe that they are not being observed.
Such a conclusion is at all possible because these principles and maxims have the status
of presumptions in communication principles that stand in the absence of counterindications. (Cummings 2005:109)

Presumptions or defaults concern both linguistic and extra-linguistic knowledge,


which need to be stored as social and cultural world-knowledge defaults on the one
hand, and as linguistic and discursive defaults on the other. Only then is it possible
to account for a pragmatic perspective which accommodates the inherent dialogic
and pragmatic principles of speech act and uptake, speech act type and cognitive
reality, and speech act type and situation (mis)match, which is calculated against
the background of the argument of ignorance. If a conversational contribution is in
accordance with the presumptions, no counterindications are found. Consequently,
the contribution is assigned the status of a match and interpreted accordingly. If a
conversational contribution is in disaccordance, there are counterindications and a
particularized process of inferencing is generated to calculate nonce meaning.

Pragmatics as a linguistic concept

45

The dialogue-oriented pragmatic mechanisms of inferencing and abduction


require not only a dynamic framework accounting for the parts-whole connectedness, but also a conscious model user who is accountable for her/his actions (Fetzer
2004). At the same time, s/he makes use of automatic processing and default interpretations which figure as salient and strong interpretative probabilities, unless
counterindications in the context signify a non-default match. Only an integrated
framework, making allowances for methodological compositionality and synthesizing relevant findings (Fetzer 2004; Jaszcolt 2005; Marmaridou 2000) may capture the multifaceted nature of pragmatics.

Notes
1. I am grateful to the anonymous reviewer and the editors of this volume for helpful comments on the first version of this article, and I would like to thank Varol Akman and Teun
van Dijk for discussing previous versions of this chapter.
2. The seemingly clear-cut distinction between semantics and pragmatics, though useful for
analytic reasons, is highly idealized and has been strongly contested (cf. Bublitz 2009).

References
Aijmer, Karin
1997
I think an English modal particle. In: Toril Swan and Olaf Jansen (eds.), Modality in Germanic Languages. Historical and Comparative Perspectives,
147. Berlin: de Gruyter.
Akman, Varol, Paolo Bouquet, Richmond Thomason and Roger A. Young (eds.)
2001
Modeling and Using Context. Third International and Interdisciplinary Conference, CONTEXT 2001, Dundee, UK, July 2001, Proceedings. Springer: Heidelberg.
Ariel, Mira
2008
Pragmatics and Grammar. Cambridge: Cambridge University Press.
Auer, Peter and Aldo di Luzio (eds.)
1992
The Contextualization of Language. Amsterdam: Benjamins.
Austin, John L.
1971
How to Do Things with Words. Cambridge: Cambridge University Press.
Bach, Kent
1992
Communicative intentions, plan recognition, and pragmatics: comments on
Thomason and Littman and Allen. In: Cohen, Philip R., Jerry Morgan and
Martha E. Pollack (eds.), Intentions in Communication, 389400. Cambridge,
MA: MIT Press.
Bach, Kent
2006
The top 10 misconceptions about implicature. In: Betty Birner and Gregory
Ward (eds.), Drawing the Boundaries of Meaning. Neo-Gricean Studies in
Pragmatics and Semantics in Honour of Laurence R. Horn, 2130. Amsterdam: Benjamins.

46

Anita Fetzer

Bar-Hillel, Yehoshua
1971
Out of the pragmatic wastebasket. Linguistic Inquiry 2: 401407.
Bateson, Gregory
1972
Steps to an Ecology of Mind. Toronto: Chandler Publishing Company.
Beaugrande, Robert de and Wolfgang Dressler
1981
Einfhrung in die Textlinguistik. Tbingen: Niemeyer.
Birner, Betty and Gregory Ward
1998
Information Status and Noncanonical Word Order in English. Amsterdam:
Benjamins.
Birner, Betty and Gregory Ward (eds.)
2006
Drawing the Boundaries of Meaning. Neo-Gricean Studies in Pragmatics and
Semantics in Honour of Laurence R. Horn. Amsterdam: Benjamins.
Blackburn, Patrick, Chiara Ghidini, Roy M. Turner, and Fausto Giunchiglia (eds.)
2003
Modeling and Using Context. 4th International and Interdisciplinary Conference, Context 2003, Proceedings. Heidelberg: Springer.
Bouquet, Paolo, Patrick Brezillon, Massimo Benerecetti, Francesca Castellani and Luciano
Serafini (eds.)
1999
Modeling and Using Context. Second International and Interdisciplinary Conference on Modeling and Using Context, Context99, Trento, Italy, September
1999, Proceedings. Heidelberg: Springer.
Brandom, Robert B.
1994
Making it Explicit: Reasoning, Representing, and Discursive Commitment.
Cambridge, MA: Harvard University Press.
Brown, Penelope
1995
Politeness strategies and the attribution of intentions: the case of Tzeltal irony.
In: Esther Goody (ed.), Social Intelligence and Interaction, 153174. Cambridge: Cambridge University Press.
Brown, Penelope and Stephen C. Levinson
1987
Politeness. Some Universals in Language Usage. Cambridge: Cambridge University Press.
Bublitz, Wolfram
2006
It utterly boggles the mind: knowledge, common ground and coherence. In:
Hanna Pishwa (ed.), Language and Memory: Aspects of Knowledge Representations, 359386. Berlin: de Gruyter.
Bublitz, Wolfram
2009
Englische Pragmatik. (2nd rev. ed.). Berlin: Erich Schmidt.
Carston, Robyn
2002
Thoughts and Utterances. The Pragmatics of Explicit Communication. Oxford: Blackwell.
Clark, Herbert H.
1996
Using Language. Cambridge: Cambridge University Press.
Cohen, Philip R., Jerry Morgan and Martha E. Pollack (eds.)
1992
Intentions in Communication. Cambridge, MA: MIT Press
Collins Cobuild Essential English Dictionary
1990
Glasgow and London: Collins.
Cummings, Louise
2005
Pragmatics: A Multidisciplinary Perspective. Edinburgh: Edinburgh University Press.

Pragmatics as a linguistic concept

47

Duranti, Alessandro
1997
Linguistic Anthropology. Cambridge: Cambridge University Press.
Fairclough, Norman
1992
Discourse and Social Change. Cambridge: Polity Press.
Fetzer, Anita and Varol Akman
2002
Contexts of social action: guest editors introduction. Language and Communication 22(4): 391402.
Fetzer, Anita
2002
Communicative intentions in context. In: Anita Fetzer and Christiane Meierkord (eds.), Rethinking Sequentiality: Linguistics Meets Conversational Interaction, 3769. Amsterdam: Benjamins.
Fetzer, Anita
2004
Recontextualizing Context: Grammaticality Meets Appropriateness. Amsterdam: Benjamins
Fetzer, Anita
2007
Non-acceptances in context. Intercultural Pragmatics 4(4): 493520.
Garfinkel, Harold
1994
Studies in Ethnomethodology. Cambridge: Polity Press.
Givn, Talmy
1989
Mind, Code and Context. Hillsdale: Laurence Erlbaum.
Givn, Talmy
1993
English Grammar: A Function-Based Introduction. Amsterdam: Benjamins.
Givn, Talmy
2005
Context as Other Minds. Amsterdam: Benjamins.
Goffman, Erving
1981
Forms of Talk. Oxford: Blackwell.
Goffman, Erving
1986
Frame Analysis. Boston: Northeastern University Press.
Goodwin, Charles and Alessandro Duranti
1992
Rethinking context: an introduction. In: Alessandro Duranti and Charles
Goodwin (eds.), Rethinking Context. Language as an Interactive Phenomenon,
142. Cambridge: Cambridge University Press.
Grice, Herbert Paul
1975
Logic and conversation. In: Peter Cole and Jerry L. Morgan (eds.), Syntax and
Semantics, 4158. New York: Academic Press.
Grosz, Barbara J. and Candace L. Sidner
1992
Plans for discourse. In: Philip R. Cohen, Jerry Morgan and Martha E. Pollack
(eds.), Intentions in Communication, 417444. Cambridge, MA: MIT Press.
Gumperz, John J.
1992
Contextualization and understanding. In: Alessandro Duranti and Charles
Goodwin (eds.), Rethinking Context: Language as an Interactive Phenomenon, 229252. Cambridge: Cambridge University Press.
Gumperz, John J.
1996
The linguistic and cultural relativity of inference. In: John J. Gumperz and
Stephen C. Levinson (eds.), Rethinking Linguistic Relativity, 374406. Cambridge: Cambridge University Press.

48

Anita Fetzer

Gumperz, John J.
2003
Response essay. In: Susan L. Eerdmans, Carlo L. Prevignano and Paul J. Thibault (eds.), Language and Interaction. Discussions with John J. Gumperz,
105126. Amsterdam: Benjamins.
Habermas, Jrgen
1998
On the Pragmatics of Communication. Cambridge, MA: MIT Press.
Halliday, Michael A.K.
1994
Introduction to Functional Grammar. London: Arnold.
Hanks, William F.
1996
Language form and communicative practices. In: John J. Gumperz and
Stephen C. Levinson (eds.), Rethinking Linguistic Relativity, 232270. Cambridge: Cambridge University Press.
Heritage, John
1984
Garfinkel and Ethnomethodology. Cambridge: Polity Press.
Horn, Laurence R.
1984
Toward a new taxonomy for pragmatic inference: Q-based and R-based Implicature. In: Deborah Schiffrin (ed.), Meaning, Form and Use in Context: Linguistic Applications, 1142. Washington: Washington University Press.
Horn, Laurence R.
1989
A Natural History of Negation. Chicago: Chicago University Press.
Horn, Laurence R. and Gregory Ward (eds.)
2004
The Handbook of Pragmatics. Oxford: Blackwell.
Huang, Yan
2007
Pragmatics. Oxford: Oxford University Press.
Jaszczolt, Kasia
1999
Discourse, Beliefs, and Intentions: Semantic Defaults and Propositional
Attitude Ascription. Oxford: Elsevier Science.
Jaszczolt, Kasia
2005
Default Semantics: Foundations of a Compositional Theory of Acts of Communication. Oxford: Oxford University Press.
Lakoff, Robin
1973
The logic of politeness: or minding you ps and qs. Papers from the 9th Regional Meeting of the Linguistic Society of Chicago: 292305.
Lambrecht, Knud
1994
Information Structure and Sentence Form. Topic, Focus and the Mental Representation of Discourse Referents. Cambridge: Cambridge University Press.
Leech, Geoffrey
1983
Principles of Pragmatics. London: Longman.
Levinson, Stephen C.
1979
Activity types and language. Linguistics 17: 365399.
Levinson, Stephen C.
1983
Pragmatics. Cambridge: Cambridge University Press.
Levinson, Stephen C.
1988
Putting linguistics on a proper footing: explorations in Goffmans concepts of
participation. In: Paul Drew and Anthony Wootton (eds.), Erving Goffman. Exploring the Interaction Order, 161227. Cambridge: Cambridge University
Press.

Pragmatics as a linguistic concept

49

Levinson, Stephen C.
1995
Interactional bias in human thinking. In: Esther Goody (ed.), Social Intelligence and Interaction, 221260. Cambridge: Cambridge University Press.
Levinson, Stephen C.
2000
Presumptive Meanings: The Theory of Generalized Conversational Implicature. Cambridge, MA: MIT Press.
Lewis, David
1979
Scorekeeping in a language game. Journal of Philosophical Logic 8: 339359.
Longman Dictionary of Contemporary English
2003
Harlow: Pearson Education Limited.
Marmaridou, Sophia
2000
Pragmatic Meaning and Cognition. Amsterdam: Benjamins.
Mey, Jacob
2001
Pragmatics. An Introduction. Oxford: Blackwell.
Morris, Charles
1971
Writings on the General Theory of Signs. The Hague: Mouton.
Penco, Carlo
1999
Objective and cognitive context. In: Paolo Bouquet et al. (eds.), 2nd International and Interdisciplinary Conference on Modeling and Using Context
(Context99), 270283. Heidelberg: Springer.
Recanati, Franois
2004
Literal Meaning. Cambridge: Cambridge University Press.
Reddy, Michael
1979
The conduit metaphor. In: Andrew Ortony (ed.), Metaphor and Thought,
284324. Cambridge: Cambridge University Press.
Sbis, Marina
2002
Speech acts in context. Language and Communication 22(4): 421436.
Schmid, Hans-Jrg
2003
Introduction: context and cognition a concern shared by cognitive sciences.
In: Ewald Mengel, Hans-Jrg Schmid and Michael Steppat (eds.), Anglistentag
2002 Bayreuth. Proceedings, 363367. Trier: Wissenschaftlicher Verlag Trier.
Searle, John R.
1969
Speech Acts. Cambridge: Cambridge University Press.
Searle, John R.
1975
Indirect speech acts. In: Peter Cole and Jerry L. Morgan (eds.), Syntax and
Semantics, 5982. New York: Academic Press.
Searle, John R.
1979
Expression and Meaning: Studies in the Theory of Speech Acts. Cambridge:
Cambridge University Press.
Searle, John R.
1983
Intentionality. Cambridge: Cambridge University Press.
Searle, John R.
1991
Collective Intentions and actions. In: John R. Searle, Herman Parret and Jef
Verschueren (eds.), (On) Searle on Conversation, 401415. Amsterdam: Benjamins.
Searle, John R.
1995
The Construction of Social Reality. New York: The Free Press.

50

Anita Fetzer

Sperber, Dan and Deidre Wilson


1995
Relevance: Communication and Cognition. Oxford: Blackwell.
Stalnaker, Robert
1999
Context and Content. Oxford: Oxford University Press.
The New Encyclopaedia Britannica.
1976
The University of Chicago, Vol 14. Chicago: Helen Hemmingway Benton.
Thomason, Richmond H.
1992
Accommodation, meaning, and implicature: interdisciplinary foundations for
pragmatics. In: Philip R. Cohen, Jerry Morgan and Martha E. Pollack (eds.),
Intentions in Communication, 325363. Cambridge: MIT Press.
Traugott, Elizabeth Closs
1988
Approaches to Grammaticalization. Amsterdam: Benjamins.
Vanderveken, Daniel and Susumo Kubo (eds.)
2002
Essays in Speech Act Theory. Amsterdam: Benjamins.
Vanderveken, Daniel and Susumo Kubo
2002
Introduction. In: Daniel Vanderveken and Susumo Kubo (eds.), Essays in
Speech Act Theory, 121. Amsterdam: Benjamins.
Van Dijk, Teun
1981
Studies in the Pragmatics of Discourse. The Hague: Mouton.
Van Dijk, Teun
2008
Discourse and Context. A Sociocognitive Approach. Cambridge: Cambridge
University Press.
Verschueren, Jef
1999
Understanding Pragmatics. London: Arnold.
Watts, Richard, Sachiko Ide and Konrad Ehlich (eds.)
1992
Politeness in Language. Berlin: de Gruyter.
Widdowson, Henry
2004
Text, Context, and Pretext. Critical Issues in Discourse Analysis. Oxford:
Blackwell.
Wittgenstein, Ludwig
1958
Philosophical Investigations. Oxford: Blackwell.
Yule, George
1996
The Study of Language. Cambridge: Cambridge University Press.

2.

Micropragmatics and macropragmatics


Piotr Cap

1.

Introduction

The present chapter discusses micropragmatics (the pragmatics of utterance based


concepts such as deixis, anaphora, presupposition, etc.) and macropragmatics (the
pragmatics of discourse based concepts such as speech events, global intentionality or macro speech acts). Its aim is to show how micropragmatics and macropragmatics (have come to) complement each other, both conceptually and in terms of
the division of labor across the field of contemporary linguistic pragmatics (henceforth: pragmatics).1
After a brief overview of the perspectivist approach to pragmatics in section 2.,
I turn to the discussion of context (section 3.) and discuss how its potential and
actual manifestations affect pragmatic inferences at different levels of the utterance and discourse structures. In 4., the perspectivist approach and the considerations of context are bridged to constitute two crucial factors defining micro- and
macropragmatics, in addition to several other factors and developments that have
led to such a distinction. In section 5., I deal with the four concepts central to the
methodology of pragmatics in general: deixis, presupposition, implicature and
speech acts. I discuss them from a micropragmatic perspective, showing their contribution to the proposition of an utterance, its illocutionary force, and the range
and kinds of effects the utterance is able to bring about. At the same time, however,
I introduce the macropragmatic approach, by raising three observations. The first is
the different potential each of the four concepts possesses to contribute to the description of function and effects of an utterance. The second is the fact that, in the
process of describing an utterance, some of the concepts go together and form
more sophisticated methodological tools, better than others. The third is the different range of utterance contexts each of the concepts is able to account for, which
opens up prospects for their hierarchical application in analysis. In 6., I take these
observations as prompts for an integrated study of intentionality at the macro level
of discourse. Addressing such notions as speech event and macro speech act, I
revisit the potential of individual speech acts for combining interacting discourse
topics into larger functional units. I discuss the conceptual infiniteness pertaining
to speech events in terms of their composition and hierarchy, and show the resulting relativity of the concept of the macro speech act.

52
2.

Piotr Cap

The pragmatic perspective

Pragmatics offers a unique, function-based perspective on all aspects of human


(linguistic) communication (cf. Verschueren 1999; Fetzer 2002). It is concerned
with all facets of a communicative act or a series of acts, such as the speaker, his/
her background knowledge and contextual assumptions, the lexical and grammatical constituents of an utterance, the hearers interpretations and patterns of inferencing, etc. All these are studied against a network of social factors, preconditions,
norms and expectations that govern communication, both within a culture and
across cultures. Since communicative acts involve linguistic units, whose choice is
dictated by language-internal rules, as well as their interpersonal, social and cultural embedding, pragmatic studies bridge the system and the use side of language.
They examine what is lexically and grammatically available for a speaker to accomplish a communicative goal, and at the same time explore the ways in which
the linguistic potential is realized in a specific social context.
Since the arsenal of language forms and the catalogue of social embeddings in
which these forms are used in communication are virtually infinite and heterogeneous, pragmatic scholarship naturally draws upon a number of diverse disciplines
in linguistics (and beyond). It is generated collectively by phonologists, morphologists, syntacticians, discourse analysts but also psychologists, sociologists and
other scientists in the Humanities as long as their work has a functional-communicative bent. Pragmatics is thus defined by its perspective more than by a set
number of its objects of investigation.2 It offers explanations which apply to different lexical and structural levels of language word, phrase, sentence, discourse.

3.

Context

The conception of pragmatics as a (functional) perspective foregrounds and endorses its preoccupation with context. Different types of context are naturally salient at different levels of communication (utterance, series of utterances, discourse, genre) but there are also manifestations of different aspects of context at
any specific level. Hence, context is a fundamental concept to account for in describing the micro-macro dialogue in pragmatic studies.
In a nutshell (cf. e.g. Bazzanella 2002), one can speak of a static and a dynamic conception of context. The static view (cf. e.g. Halliday and Hasan 2000;
Cummings 2005) sees context as a set of variables (linguistic, cognitive, social)
that surround strips of text. They constitute an a priori, static infrastructure of resources which are there for interlocutors to determine the meaning of utterances at
hand. On the dynamic view (cf. e.g. Duranti and Goodwin 1992; Bublitz 2003,
2006), context is never given and interlocutors keep creating current contexts for
current utterances the moment a sentence is uttered it becomes part of the envi-

Micropragmatics and macropragmatics

53

ronmental resources on which the contextual interpretation of the following as well


as the preceding utterances has to draw (Bublitz 2003: 383).
As the next section will show in detail, the dynamic view of context is especially
important in elucidating the relationship between micropragmatics and macropragmatics the prerequisite being, of course, the move from a discourse-participant to
a discourse-analytic perspective. The complementarity of micropragmatics (pragmatics of the utterance) and macropragmatics (pragmatics of the discourse) consists in the micro-level analysis getting verified from the macro standpoint, a procedure often leading to the redefinition of the original analytic track. Text and context are, not only in interlocution but also in analysis, constantly matched with and
against each other, both prospectively and retrospectively. Any micropragmatic
analysis of an utterance, reshaped as a result of macropragmatic considerations of
the discourse to which the utterance contributes, is thus an example of reinterpretation from hindsight endorsed by the dynamic view of context. Such a reinterpretation is often indicative of the change in the number and kind of textual cues
which are used before and after the reshaping takes place. It is likely that in the
original micro-pragmatic analysis the reliance on the most prominent textual cues
will be significant, since the macro factors and constraints are, as yet, disregarded.
However, the reshaped micro-analysis may well see some of the initial cues
brushed aside, often for the benefit of the cues originally neglected (see section 4.).
All in all, the dynamic notion of context is a vital concept for micro- and macropragmatics since it allows speculations over, a) the extent to which the interpretation of meaning is guided by text; b) what factors may cause such an interpretation
to be insufficient; c) at which level of analysis such a deficit becomes visible.
The dynamic conception of context recognizes one more important premise for
the existence of the micro-macro dialogue, i.e. the different degrees of accessibility of contextual cues, at different stages of interpretation.3 The linguistic cues
(often referred to as co-text) are immediately accessible, but some of the cognitive and social cues might not be. This, altogether, makes discourse participants abstract from the holistic context (the context which is often defined at the macro
level, according to the characteristics of an attitudinal frame, genre, speech event,
discourse type) the cues that seem actually relevant to a given communicative situation (micro-level). The above process has been extensively described in terms of a
number of methodologies, which account for the relation between context and subcontexts (cf. Goffman 1986), figure and ground (cf. Langacker 1987, 2001), frame
and framing (cf. Bateson 1972; Fetzer 2007), generalized (default, unmarked) context and particularized (marked) context, global context and local context, etc.
Their common assumption is that the concept of context is best investigated from a
parts-whole perspective, which elucidates its function of delimiting content. Such
a perspective is naturally in line with the perspectivist view of pragmatics in general and, at the same time, congruent with the scope of micropragmatics and macropragmatics.

54
4.

Piotr Cap

Micropragmatics and macropragmatics: conceptual origins


and scope

As has been indicated before, micropragmatics can be defined as the study of illocutionary force at the utterance level. In contrast, the focus of macropragmatics is
not on the utterance, but on a series or sequence of utterances which form discourses/texts, seen as bearers of global intentionality of the speaker (i.e. the intentionality resulting from different speech act configurations, often referred to as
speech events) and as initiators of complex effects.
The distinction between micro- and macropragmatics follows from several factors as well as research developments. First, it has to do with flexible definitional
boundaries of pragmatics. If pragmatics involves the study of meaning in context,
then context is not limited in formal ways, such as the length of sentence/utterance
or the amount of its discourse history. Contextual considerations which are used for
determining the function and effect of an utterance may or may not stop within the
boundaries of the language form used. Thus, we need a conceptual handle on the interface between smaller and larger functional units of discourse (e.g. speech act
versus speech event), coded in smaller or larger forms (e.g. utterance versus text).
Second, since context is complex, dynamic and multilayered (cf. section 3.),
the accessibility of contextual cues is a matter of degree (cf. Marmaridou 2000).
The more access to contextual (including co-textual) embedding of an utterance or
a series of utterances, the better profiled the meaning. Hence a tendency on the part
of the analyst to pile up contextual cues and to add to the picture as much of the
prehistory of the discourse situation (cf. Mey 1993) as possible (and feasible).
But since access to contextual cues is each time different and invariably limited by
a dynamic network of constraints, the analyst can hardly ever claim to have exhausted all context to determine the meaning. In this sense, the micro-macro dichotomy involves a relation between the minimal and the maximal contextual
input into analysis, irrespective of its scope, i.e. the length of the language form
under scrutiny.
The domains of micro- and macropragmatics are complementary in terms of
their contribution to analytic labor. There is no micropragmatic analysis that would
not provoke a macropragmatic extension of scope; similarly, there is no macropragmatic study that would not question, retrospectively, its micropragmatic components, thus prompting revision or modification of the original analytic track. Altogether, we arrive at a bottom-top-bottom cycle of upgrades on the explanatory
power of both micropragmatic and macropragmatic concepts. This cycle involves
looking at different aspects of context (linguistic, cognitive, social) from two different perspectives, a more linguistic bottom-up perspective and a more social topdown perspective.
By way of illustration, consider a sample analytic procedure whose objective is
to account for the pragmatics of speeches of the American cold war presidents (say,

Micropragmatics and macropragmatics

55

Harry S. Truman or Dwight D. Eisenhower). Imagine that a micro-level analysis of


utterances making up these speeches points to a large number of deictic, referential
and anaphoric markers, embedded in a multitude of direct speech acts. In contrast,
some other units or phenomena explored in such an analysis, for instance implicatures and presuppositions, turn out to be underrepresented. The next step is to classify these findings under controlling categories, such as speech events or macro
speech acts, to establish the performative denominators and thus postulate global
function(s) of the discourse genre (i.e. the presidential cold war speeches) as a
whole. This macro-analytic task requires inspecting the extralinguistic context,
which informs and complements the linguistic observations. The likely outcome of
the extralinguistic considerations is that the cold war presidents address a world
that is ideologically divided and respond to a need to clearly define to their audience the essence of this divide, which eventually leads them to construct the wellknown us and them opposition. Hence the rhetorical urge for directness and
clarity, rationalizing the opulent use of indexicals or anaphoric expressions and
banning most forms of implicit communication such as implicatures or presuppositions. Reaching this conclusion means that the analyst was able to use micropragmatic considerations as building blocks of a functional hypothesis, which he or she
then tested against contextual factors (including his or her expert knowledge), in
order to generate a macro-function of the genre. This is, roughly, how the analytic
curiosity about regularities governing the salience of some phenomena observed at
an utterance level can provoke the macropragmatic extension of scope. However,
the upgrade cycle does not stop here, there is still room for a reverse procedure,
which has the macropragmatic findings relate to the original component premises.
Accordingly, the analyst may want to return to the bottom level of an utterance to
look for more data, to further strengthen the macro conclusion(s). This may mean
leaving out some of the now useless forms (e.g. markers of implicitness) and
delving more deeply into those which have passed the extralinguistic context
verification. As a result, further micropragmatic activity can follow, with a view to,
either narrowing down the study of the original parameters, e.g. deixis or direct
speech acts, to only those aspects which are in line with the macropragmatic conclusion, or, possibly, identifying more bottom-level forms which are relevant, for
instance more cohesive devices (cf. Halliday and Hasan 1976), relational propositions (cf. Mann and Thompson 1983), etc. Which is where the first round of the
bottom-top-bottom cycle ends, but, of course, more rounds can still take place,
with the micro-macro conclusions refining each other virtually ad infinitum. The
existence of such a dialogue is probably among the most significant methodological arguments for having the two concepts, micropragmatics and macropragmatics,
in the theoretical framework of the field.
The distinction between micro- and macropragmatics has been strongly influenced by work in disciplines which pragmatics addressed while defining its
scope, and upon which it imposed its functional perspective. The ethnography of

56

Piotr Cap

communication (cf. Hymes 1974; Saville-Troike 1989; Gumperz 1992) has provided insights into communicative acts seen as socially situated cultural forms, and
micro-specifications of the more general social and linguistic contexts. Interactional socioliguistics (cf. Gumperz 2003) distinguishes between two levels of inference in analyses of interpretation processes. The higher level accommodates
global inferences of what the exchange is about and what mutual rights and obligations apply, what topics can be brought up and what reply is desired by the cultural macro-convention, and the lower level contains local inferences concerning
what is intended with the speakers own move and what is required by way of a response. Research in corpus linguistics (cf. Biber 1988) has adapted social and institutional factors to the examination of variation in language, both spoken and
written. It is argued (cf. Biber 1988) that conversational and discourse goals comprise both conventional social and institutional goals, as well as personal microgoals. Furthermore, the micro-macro interface has been at the heart of studies in
activity types (cf. Levinson 1979) and communicative projects (cf. Linell 1998).
These studies have paved the way for bridging the micro with the macro on the
plane of the explicit or implicit accommodation of context in interpretative processes. As it seems, the only way in which a distinction between the macro and the
micro domains can be attempted is indeed by looking into the different degrees of
lexical and grammatical coding of context in an utterance or a discourse vis a vis its
extralinguistic coding, i.e. the import of mostly non-indexical cues. Such analytic
endeavors are central to the most recent approaches to the analysis of communicative genres (cf. Bauman 1992; Luckmann 1995; Thibault 2003; Fetzer 2007).
Genre (recall our sample analysis above) is a classificatory concept; it specifies
typical ways in which lexicogrammatical resources are deployed to enact a communicative goal, by matching them against its own delimiting frame. Thus, micro
analysis is an analysis of individualistic phenomena (tokens), which macro analysis verifies against the generic requirements (type). The token contributions are not
merely cumulative, they can get encapsulated in one another or form a hierarchy
as will be shown in the next section.

5.

A micropragmatic perspective on deixis, presupposition, implicature


and speech acts with implications for macropragmatics

While the basic scope of micropragmatics is an utterance, any micropragmatic


analysis is, ultimately, a journey into the macro domain. This follows from the conceptual and methodological characteristics of the most common parameters utilized in description: deixis, presupposition, implicature, speech act. They are never
static components of the analytic enterprise; instead, they prompt an analysis
which is essentially cumulative, interactional and hierarchical. At the bottom of the
hierarchy is the input obtained from the interpretation of the lexical and grammati-

Micropragmatics and macropragmatics

57

cal coding of the utterance (immediate) context. This input (from deixis, presupposition lexical triggers, etc.) is usually insufficient for the utterance interpretation
since it does not include non-textual cues. Accordingly, the analysis proceeds to
the study of inferencing (pragmatic presuppositions, implicatures), which bridges
the linguistic cues with the extralinguistic ones. Finally, the speech act level sees
the function of the utterance emerge from the premises and interpretations (speaker-, hearer-, content-oriented) that have accumulated. Such a function remains,
however, tentative, until a wider discourse context has been addressed for verification (viz. the sample analysis in 4.). Thus, the category of the speech act sits at
the methodological borderline between the micro and the macro domain, which the
latter involves generic, institutional, etc., characteristics and constraints. The
speech act is at the same time an umbrella category for deixis, presupposition, implicature, and, potentially, a component category in the macro considerations. Altogether, micropragmatics feeds sequentially into macropragmatics because the
consecutive-collective use of micropragmatic tools in analysis entails a continual
extension of scope, which eventually straddles the original utterance boundaries.
The four subsections below unfold this sequence, showing, ultimately, the inherent
relativity of the micro-macro distinction.
5.1.

Deixis

The outset of micropragmatic analysis is marked by establishing a tangible, lexicogrammatical architecture of the utterance, and especially the explicit manifestations of the relation between the words used and the context. The analysts task is
to identify the lexicalized pointers which indicate who utters the words to whom
to accomplish what, when and where. This task endorses the status of deixis as the
initiator of the (micro)pragmatic analysis since deixis is the concept that captures
the relationship between the language form and the context in the most evident and
direct manner. It can be viewed as the main phenomenon whereby features of context are encoded in utterances by primarily lexical (e.g. demonstratives) but also
grammatical (e.g. tense) means.4
While the analysis of deixis sets the stage for a more complex scrutiny of the
utterance, its potential to interact with the other parameters of description (presupposition, implicature) is limited. As a concept involving lexicalized or grammaticalized forms of expression, it makes a relatively small contribution to the illocutionary force of the utterance, especially in implicit communication. Naturally,
there are exceptions. For instance, deictic projections (the cues are processed not
from the speakers own perspective, but from the perspective projected on the
hearer, cf. Levinson 1983; Fillmore 1997; Marmaridou 2000; Manning 2001) enable deictic expressions to generate implicatures. If, at 4:01 p.m. on Friday, I send
an e-mail to my publisher which says:

58

Piotr Cap

(1) I will submit the manuscript on Saturday.


I have good reason to expect the publisher (who works until 4 p.m. and only on
weekdays) to read the e-mail no earlier than Monday and, seeing no attachment
containing the ms. in the mailbox, to be puzzled did I mean the Saturday that has
just passed or the one to come? Of course, I retain the rhetorical comfort of denying
whichever interpretation. In fact the comfort starts the moment the Friday e-mail
is sent out I can feel free to submit the manuscript on whichever of the two Saturdays I prefer. There is no way I can be blamed for not submitting the ms. on the
first of the two Saturdays, as I can always respond that I was taking my addressees
perspective (i.e. receiving time), not mine.
Most types of deixis (person, time, place, for an overview cf. Hanks this volume) are analyzed at the utterance level. However, a notable exception is discourse
deixis, which crosses the utterance-discourse boundary, thus becoming a macropragmatic worktool. Discourse deixis involves the use of a lexical item within an
utterance to point to the preceding or following utterances in the same discourse
situation (speech event). The backward and forward reference can be illustrated by
(2) and (3), respectively, where already refers to the earlier stretch of discourse and
here anticipates the upcoming stretch:
(2) As already indicated, all languages possess deictics.
(3) Here goes my argument.
There is no rule to how much textual distance should hold between the deictic expression and its referent. While it is reasonable to expect that the referent of here
will be the immediately following utterance(s), the referent of already is surely not
the closest, preceding utterance, but rather an utterance made much earlier in the
unfolding discourse. This brings us to an important conclusion regarding the pragmatics of discourse deixis. By using a specific number of deictic expressions, the
speaker is able to control the overt connectedness of discourse and, in consequence, its comprehension by the hearer. The presence of deictic markers in utterances which make up a discourse situation where a specific topic is pursued
usually contributes to explicitness and clarity. On the other hand, the speaker may
choose to withhold the use of deixis to purposefully obscure the message.
These exceptions (deictic projections contributing to implicatures; discourse
deixis as controllers of discourse comprehension), however important, do not detract from the core function of deixis in pragmatic studies. Deixis is an essentially
micropragmatic phenomenon which sets analysis of the utterance in motion, but,
by itself, adds little to the characteristics of the speech act(s) residing in it. The
deictic framework of the utterance is sketched to establish the basic, lexically encoded relations between the referent(s) of the utterance and the common ground
of knowledge the utterance assumes to exist between the speaker and the hearer.
Yet, a full account of these relations (and of the utterance function(s) the relations

Micropragmatics and macropragmatics

59

contribute to) needs complementation from concepts that go beyond lexically encoded relations (e.g. presupposition, implicature).
5.2.

Presupposition

Presupposition can be defined as a mechanism whereby the speaker addresses a


body of knowledge and experience, involving both linguistic and non-linguistic
contexts, which he or she assumes to be common to him-/herself and the hearer.
The assumption of the existence of the shared knowledge may cause the speaker
not to grammaticalize (or lexicalize) it in the utterance. This characterization takes
presupposition to be a phenomenon lying at several intersections: the encoded and
the assumed, the semantic and the pragmatic, (or even) the linguistic and the nonlinguistic.5
Presupposition comes in contact with deixis on the plane of its partial anchoring in lexical and structural forms. However, since many instances of presupposition can only be approached with reference to (non-linguistic) context, presupposition also reaches out in the direction of the implicit, constituting, in a sense, a
shared knowledge prerequisite for communicating messages whose final destination is their inference by the hearer. Hence its feasible combination with the apparatus of implicature and, altogether, its relevance to the hierarchy of micropragmatic analysis, which derives its output from both accumulation and interaction of
descriptions offered by the individual conceptual tools. As one of the latter, presupposition targets the communicative act at the stage where it develops upwards
from the lexicogrammatical coding of context to its further abstraction and elaboration by the speaker, with a view to producing a speech act. Throughout this
stage, the speaker decorates the deictic framework of the utterance with instantiations of knowledge shared by the speaker and the hearer with regard to all entities
indexed, referred to, or implied in the utterance.
Traditionally, the more a presupposition was linked with a lexical item or a linguistic construction generating it, the more it was treated as a semantic phenomenon, the other cases deemed pragmatic and worth less attention precisely because of the absence of fixed language forms responsible for enacting particular
presuppositions.6 This view has produced multiple typologies of presupposition,
based on its embedding in lexicogrammatical forms called presupposition triggers.
Furthermore, a number of properties have been assigned to presuppositions, including cancellability (a possibility of denying a presupposition, usually by adding
more content to the utterance where it occurs) and constancy under negation (negating the predication carrying a presupposition does not detach or change the presupposition) (cf. e.g. Beaver 2001).
From the perspective of micropragmatic analysis oriented toward the speech
act characterization of the function of the utterance, as well as the macropragmatic
perspective of the discourse, a rigid distinction between semantic and pragmatic

60

Piotr Cap

presupposition seems far from necessary. It offers little explanatory power compared to an integrated, global view of presupposition as a concept which should be
studied against the utterance and discourse goals it serves. Such a view is quite naturally pragmatic, because even though there are specific lexical items (in fact,
many) associated with specific assumptions (for instance, the assumptions of existence), their descriptive capacity does not expire within the structural boundaries of
the linguistic expression. On the contrary, their significance goes much beyond as
they are able to combine with primarily experiential premises and thus successfully contribute to a network of contextual, often non-linguistic beliefs making up
the entire load of knowledge shared by the speaker and the hearer (see fn. 5).
Most communicative goals served by presupposition (relevant to both the
micro- and macropragmatic analysis) have to do, in one way or another, with economy of expression, though the latter is rarely the only goal sought. If a speaker
could not rely on shared assumptions, the lexical and grammatical load of his/her
utterance would grow in size, potentially obstructing its comprehension. Although
economizing on the linguistic form for the benefit of unobstructed communication
could be an utterance goal in itself, it is often a contribution to a larger utterance or
discourse function. This becomes clear when analyzing, at the macro level, a series
of utterances containing consecutive acts of adjustment to shifting discourse expectations. If I am overweight and say I started jogging after visiting my doctor
and my confession meets with a blatant attack like So it took you going to the doctors to work out like you should, I can always follow a defense line in continuing
Well, to be honest, I tried to do some jogging a few times before, but now I do it
regularly. Since the short form of the initial utterance makes it underdetermined
in terms of meaning, the denial of the only then and never before presupposition
comes rather easy, contributing to the overall explanation and justification (see
Huang 2007 on cancellability of presuppositions).
5.3.

Implicature

The relevance of the analysis of implicature7 to micropragmatics (and, potentially,


macropragmatics) is twofold. First, implicature studies account for the further processing of information that has been encoded by the speaker based on his/her presuppositions. Second, in doing the latter, they eventually recognize the contribution implicature makes to the update of the utterance-discourse context.
From the analytic standpoint, the explanatory powers of presupposition and
implicature are inherently complementary, shedding light on both the speaker and
the hearer side of the speech act formation. They are suited to cover, in combination, the whole process of encoding messages by speakers and decoding them by
their hearers. This process is essentially a continuum, where making a presupposition paves the way for the utterance before it takes on a linguistic form, in which
the presupposition is lexically or non-lexically salient. From that point on, i.e. the

Micropragmatics and macropragmatics

61

point of making the utterance by the speaker, the recovery of the implicature(s) by
the hearer may begin of course, if the hearer senses a prompt to search for it/
them. The inference of the implicature(s), whether in accordance with the speakers expectations or not, finalizes the entire process, thus updating the status of the
interaction and creating a new contextual basis on which to build presuppositions
for further utterances in the exchange. The cycle in question corroborates the dynamic view of context (see section 3.) and endorses the intrinsic relativity of the
micro-macro dichotomy. While the update takes place, technically speaking,
within the utterance, its effect is on the prospective discourse.
Delving deeper, implicatures created within the boundaries of the utterance
(the micro level) are often returned to or readdressed purposefully later on in
the unfolding discourse (the macro level). Since implicature is rarely encoded in
language form,8 it involves a virtually indeterminable number of more or less complex contextual inferences. As such, it constitutes a valuable rhetorical tool
whereby the speaker can control the flow of discourse, adopting his or her consecutive utterances to the current goals. This is due to a central property of implicature,
cancellability (cf. Sadock 1978; Levinson 2000, etc.), which makes it possible for
the speaker to deny, at any moment of speech situation, any implicature he or she
apparently created. We have seen this property as partly relevant to presuppositions, but it is implicature that permits its broadest manifestation. Indeed, a great
many implicatures are cancelled to re-establish adherence to the conversational
norms (see the Cooperative Principle and the Maxims of Conversation by Grice
(1975, 1989) as well as their numerous reformulations and supplements9) but also
to play with the addressee, pull a trick on him/her, or simply annoy him/her. Following on this note, many implicatures are cancelled for ironic or sarcastic effects,
which is well documented in humor studies (cf. Attardo 1990, 1993). Another area
where implicatures and their cancellations surface, is public (especially political)
discourse. Implicatures contribute to the rhetorical safety of public speakers, who,
on the one hand, wish to make statements that are universally acceptable, but on
the other, want to retain the possibility of refining or fine-tuning these statements for hearers who hold different expectations of the meanings conveyed in
them (cf. Cap 2008, 2010). Altogether, the phenomenon of the cancellability of implicature belongs to macropragmatics, since, first, the context that determines the
cancellations is made up of a heterogeneous number of social and institutional factors, second, the distance between implicature and its cancellation is a matter of
discourse, rather than utterance.
5.4.

Speech acts towards macropragmatics

The three brief subsections above have shown that deixis, presupposition and implicature make their distinctive micropragmatic contributions to understanding
how an utterance is built, what its referents are and how they are encoded, what as-

62

Piotr Cap

sumptions are made before the utterance is produced, what effects can be expected
after it has been produced and what inferential processes determine these effects.
They partake in the process of enacting goal(s) of the utterance, from the speakers
intention to realize its envisaged function via application of specific indicators of
force, to the hearers successful (or not) recognition of this function and its results.
A procedure this complex needs a controlling, umbrella parameter of description.
It needs a conceptual tool that is able to cover both speaker and hearer related aspects of the utterance function, and, while doing so, draw upon and thus systematize the particular contributions from deixis, presupposition, and implicature in
order to make them fit for macropragmatic work at the discourse level. The concept
of the speech act seems an excellent theoretical candidate to take up this task.10
The orientation of speech acts to both parties of a verbal exchange, as well as to
its linguistic matter, is visible at a glance from the traditional distinction between
the locutionary, the illocutionary, and the perlocutionary aspects of a speech act.11
While the locutionary aspect is the most objective since it concerns the stable language form of the utterance, the illocutionary and perlocutionary aspects involve a
dynamic negotiation of meaning between the speaker and the hearer. In saying Its
hot in here a speaker may be producing an (implicit) illocutionary act requesting
the hearer to open the window, and the perlocutionary act (effect) might be that the
hearer indeed opens it, but it might also be that he or she turns on the air-conditioning instead. Thus, the illocutionary-perlocutionary relation not only mirrors the
complex process of meaning evolution as sketched at the beginning of this subsection; it also inscribes in the distinction between explicit (direct) and implicit (indirect) ways of communicating a speech act. Consequently, it invokes the notions of
deixis, presupposition, and implicature, since they all situate themselves at some
specific yet different points of the conceptual axis which links what is said with
what is effected.
The classificatory, controlling power of the speech act is further reflected in its
network of felicity conditions, i.e. the conditions that underlie a successful, logical,
felicitous production of different acts.12 For example, a speaker cannot make a
successful order if he or she does not sincerely want the order to be followed, or if
he or she deems the hearer incapable of following it. These two felicity conditions
are excellent illustrations of the connection that holds between the concept of the
speech act and the other micropragmatic concepts a relation we have postulated
at the beginning of this subsection. The speakers awareness of cognitive and social context obtaining at the moment of producing a speech act gives rise to pragmatic presuppositions underlying the utterance that contains this act. Then, once
the act is accomplished, the speakers presuppositions can be assessed against the
effectiveness of implicatures they helped to create. While this part is naturally the
domain of implicature studies, we do get an extra insight from the cover framework of the speech act theory and, specifically, from its contribution to the research
in direct versus indirect acts.

Micropragmatics and macropragmatics

63

Ironically, the greatest advantage of this research is that it almost removes one
of the two groups of acts from our scope of interest, at both the micro and the
macro level of description. As pointed out in several studies, most speech act
usages are indirect (cf. Bertolet 1994; Holdcroft 1994, etc.) and indeed, even the
apparently straightforward act of asking for salt in Can you pass the salt? is, formally speaking, indirect. This, however, should not be discouraging. The orientation of the speech act theory to indirect usages is only reflective of the visible emphasis on indirectness that is salient in the other (micro-)pragmatic concepts; after
all, our discussion of deixis, presupposition, and implicature has been mostly preoccupied with hidden meanings. Thus, speech act theory turns out to be nothing
but congruent with the other apparatuses. Moreover, it extends over all of their
fields of application, from the stage where knowledge is assumed to get encoded in
the utterance, to the stage where the utterance is interpreted. Conceivably, a single
speech act can be realized through recourse to an x number of presuppositions,
an x number of deictic markers in the utterance, which the latter could produce an
x number of implicatures. We have thus arrived at a micropragmatic hierarchy of
analysis, with the speech act constituting a category superordinate over the other
micropragmatic categories. At the same time, however, the speech act should not
be considered the top-most variable of description; in order to account for discourses, rather than individual utterances, we are in need of yet higher-rank concepts.

6.

Macropragmatics

Most speech acts residing in individual utterances tend to combine into larger functional units, thus paralleling the combination of utterances into texts and discourses. This process can be seen from the analysis of (4). Assume that instead of
reading out students names from the attendance list at the beginning of a seminar,
I simply ask:
(4) Hello, are we all here?
In uttering (4), I perform two direct speech acts (greeting and asking), as well as an
indirect act of requesting my students to reveal the names of the absentees. The
speech acts involve a deictic anchoring (e.g., here recognized as classroom), presuppositions (e.g., of some students still missing), and the indirect act involves a
process of inference. This is, with some simplification, where a micropragmatic
analysis of (4) can get us.
However, being myself the producer of (4), I can enjoy the privilege of stating
with absolute certainty that, most of the time, I do not open the class with this utterance for the mere sake of greeting, asking, and requesting, as described above.
That would be unlikely considering that (4) happens regularly and has thus become

64

Piotr Cap

(a part of) expectable routine, a pattern that arises, usually, from some kind of a
larger intention.13 Mark an analogy: If I follow the routine of buying each morning
a newspaper (an action which involves micro-actions analogical to the acts defined
in (4)), the larger intention can be described as a general aspiration to stay updated with the current news. The larger intention in (4) is to begin the seminar,
making sure all things are in place for a productive meeting. This finding does not
sound terribly original in itself, but we will take it as a starting point to show,
eventually, that the nature of intentionality is such that it cannot be a matter of the
individual utterance and thus the scope of inquiry must be extended accordingly.
More often than not, I may need to say more than just (4), to successfully begin
the seminar. I may choose from an apparently infinite catalogue of other utterances
which are able to serve the same intention. Some typical cases may be: Its getting
late, Would you close the doors please?, Now, Adam, listen up, will you?,
Right, shall we begin?, And the chalk is missing as usual!, I cant remember
when we last started on time, etc. All these seemingly diverse utterances contribute, in one way or the other, to realization of the principal intention. Interestingly, they do so while performing different speech acts (assertives, directives, expressives) whose force may also be different, direct or indirect. Consequently, each
of these acts may accommodate a different input from deixis, presupposition, and
implicature.
The routine of performing individual speech acts (which may in turn involve
the other micropragmatic categories) to realize a more general intention, can be defined as a speech event.14 Speech events may contain, as we have seen, a potentially
infinite number of utterances, depending on how many are actually necessary to
carry out the intention. If, for instance, my seminar group enters the classroom discussing a just-finished exam in which they participated, I may be in for a longer
stretch of talk to set up the stage for my own class. Otherwise, it may take a few
short remarks to cover the technicalities and initiate the topic proper. Whatever
happens in actuality is thus dictated, as anything in pragmatics, by context. A
speech event can comprise just as many utterances (and speech acts) as needed to
match the contextual preconditions. These utterances do not have to come in a
monologic pattern. I may keep performing the principal speech event by producing
a few utterances in a row, then pausing, then, possibly, reacting to a question that
comes in the meantime, then resuming the monologue, etc. Thus, my performance
is a genuine discourse performance, which exists in and responds to a social setting, though, of course, some parts of it can mirror textual units and their connectedness patterns.
The concept of infiniteness resurfaces once we go on to discuss speech events
as building blocks for realization of a still larger intention. To readdress the
case initiated by (4), the speech event which realizes the intention to successfully
begin the seminar is not only an umbrella category for a series of individual acts,
but may itself be subordinate to a larger discourse goal be it, for instance, con-

Micropragmatics and macropragmatics

65

ducting a productive seminar meeting as a whole. This goal entails that an apparently infinite number of speech events are carried out on the way, from a speech
event of providing a complex, multi-act explanation to a problem that has arisen
during the session, to a speech event of assigning homework, which may again involve a number of component speech acts. The latter (event) may include: an expressive (e.g., reprimanding students for not completing the previous assignment),
a commissive (e.g., threatening to fail students at the end of the course, if they keep
neglecting their assignments), a directive (e.g., telling students to do a particular
task for the next meeting), an assertive (e.g., describing a rationale for the task),
etc. The variety and diversity of the acts involved is, here, no smaller than in the
case of the speech event performed to begin the seminar.
In this clearly bottom-up fashion, we have approached the problem of the uppermost or global category of intentionality enactment, one whose promise
would be to encompass all the subordinate intentions realized in speech events and
their attendant acts. Such a promise is partly fulfilled by the classical concept of the
macro speech act, which is, in van Dijks words, a global speech act performed by
the utterance of a whole discourse, and executed by a sequence of possibly different speech acts.15 Still, while doing some useful job in the way of systematizing
speech events, the macro speech act suffers from a problem of an inherent relativity as regards the range of its own operation. The definition of the macro act only
corroborates the problem; we do not get to know how much is a whole discourse.
Is it, to return to our example above, the whole body of discourse produced within
the duration of the seminar, controlled by the global intention to make it a productive meeting? Then the relationship between the macro speech act and the component speech events (and their individual acts) seems analytically elegant. But, does
this account exhaust the potential of the macro speech act to combine with further
macro speech acts, to serve a yet-more-global intention? Apparently not. The intention to carry out an academically rewarding seminar can be considered subordinate to the intention to conduct the entire course as planned, which in turn partakes in the intention to perform my (academic) duties properly as a whole, etc.,
etc., which, naturally, makes the consecutive macro acts accumulate accordingly.
To envisage the highest-rank intention, pursued in the highest-rank speech act,
turns, then, a philosophical undertaking.
One of the undisputed methodological values of the search for the clear-cut categories signposting the particular levels of intentionality and its enactment lies,
paradoxically, in the recognition of limits to which intentionality could be accounted for in larger stretches of discourse. Thus, endeavors like the macro speech
act theory should not be carelessly brushed off, as they eventually motivate research, however minimalist, in better demarcated and better empirically equipped
areas. Apart from the aforementioned research in genres, a prominent example of
such an area is Conversation Analysis.16 Conversation analysts have elaborated an
impressive arsenal of techniques for the description of speech act deployment,

66

Piotr Cap

though in necessarily limited contexts, often just co-texts. As a result, the apparatus of macropragmatics has been endowed with a number of relevant concepts,
such as floor, topic, turn, turn-taking, transition relevance place, etc., each
of which contributes its share of insight in how people manage their intentions and
goals in particularized (conversational) settings.
A related and very important significance is that a macro perspective on the
pragmatics of discourse makes the analyst approach intentionality as continually
re-shaped and updated by extralinguistic context and thus work out methodological tools to handle the variability of enactment of the speakers intention(s) over an
extensive period of time. Let us illustrate this phenomenon with an example from
political discourse, a domain rich in speakers attempts to continue with an enactment of a global intention, though often in consecutively modified patterns, responding to the changing extralinguistic reality. Consider an excerpt from one of
G.W. Bushs speeches legitimizing US involvement in the recent Iraq war:
(5) By advancing freedom in the greater Middle East, we help end a cycle of dictatorship and radicalism that brings millions of people to misery and brings
danger to our own people. By struggling for justice in Iraq, Burma, in Sudan,
and in Zimbabwe, we give hope to suffering people and improve the chances
for stability and progress. [] Had we failed to act, the dictators programs for
weapons of mass destruction would continue to this day. For all who love freedom and peace, the world without Saddam Husseins regime is a better and
safer place.17
This speech comes eight months into the war, which started on the assumption that
the Iraqi regime (and its alleged terrorist allies) had developed access to weapons
of mass destruction (WMD), thus becoming a world threat and unwilling to disarm
unless forced to. In his address, Bush attempts to maintain the aura of legitimization of the US intervention, against the increasingly evident collapse of the original premise no weapons of mass destruction were found in Iraq ever since the
US and the coalition troops entered it in March 2003. Since he can no longer employ the single WMD argument to execute his global intention (to keep the legitimization process intact), he deftly switches to a more universal, ideological
rationale. Instead of invoking a direct threat from a destructive impact of WMD, he
concentrates on building up an ideological framework for a potential growth of
such a threat in the future and presents the US strike as part of the necessary means
to ensure that the antagonistic ideologies (dictatorship, radicalism, regime)
do not materialize in the form of physical impact. The localization of these ideologies in more countries than Iraq itself (viz. Burma, Sudan, Zimbabwe) is
in fact a clever rhetorical ploy. By extending the spectrum of the ideological conflict and, thus, the spectrum of US foreign involvement, he encourages the construal of the WMD intelligence failure in terms of an isolated incident, fully justifiable given the range of the American mission as a whole. Altogether, Bush

Micropragmatics and macropragmatics

67

keeps up with the global intention to continue with legitimization of the Iraq war,
now on ideological grounds, but the pragmatic patterns (and, consequently, lexical
choices) responsible for realizing this intention within a macro-temporal reality
need to be updated to meet the extralinguistic developments.
In section 4., I argued that macropragmatic analyses often lead to redefinitions
of the micropragmatic input that has been utilized to build up their tentative versions. They also tend to review the micropragmatic concepts and point to those
whose descriptive potential has not yet been fully exhausted. The analysis of
Bushs speech is a good case in point. At its macro-level, we consider the general
patterns of adaptation of the political speaker to contextual conditions. This leads
us to the more focused question which of the micro-level concepts are able to account for such processes in the possibly richest lexical way, i.e. which of them are
most frequently reflected in lexical items and structures that eventually make up
the matter of the macroanalysis. The analysis of (5) reveals, for instance, that much
explanatory power is yet to be drawn from a careful scrutiny of implicature forms
in the text. If we gather from macro-contextual considerations that Bush is forced
to switch to a new argumentative strategy, yet, as seems logical, without discrediting the previous argumentation, what better way to accomplish the goal than
through implied meanings, which are always subject to cancellation as the speaker
sees fit? Recall the phrase programs for weapons of mass destruction. It is flexible enough to concede that Saddam did not indeed possess WMD understood as a
product ready for use, at the outset of the war, but it does not completely detract
from the original assumption that he did. It is quite likely that a microanalysis of
this implicature (as well as of many other bottom-level forms in (5)) could be overlooked but for the prompt from the macropragmatic approach, involving a vast
range of contextual factors affecting the entire discourse of the Iraq war.
In this way we have returned to the question of the micro-macro dialogue,
which I defined in the fourth section of this chapter as one of the most significant
methodological arguments for having the [distinction between the] two concepts,
micropragmatics and macropragmatics, in the theoretical framework of the field.
It should be said in closing that prompts for such a dialogue need not come from
pragmatic analyses alone; in fact, the dialogic relationship between micropragmatics and macropragmatics is kept alive by insights from approaches which, at
least definitionally, go much beyond linguistics-based disciplines or perspectives.
An example of such an approach is Critical Discourse Analysis (CDA; cf. e.g. Fairclough 1995; Wodak and Chilton 2005), which extends onto domains traditionally
associated with various branches of sociology and social psychology. Rooted in the
conception that discourse is just as much reflective of the existing reality as it is potentially constitutive of a new reality, CDA offers a number of useful ways for the
analysis of the reality (which pragmatics would rather term extralinguistic aspects of context). In so doing, it provides (macro-)pragmatic considerations with
socio-cognitive grounding of discourse (cf. van Dijk 1995, 2002), thus signposting

68

Piotr Cap

the downward, top-down analytic procedure. Two examples of how such a procedure could operate have been, in this article, the suggested refinements and extensions of the analysis of deixis and direct speech acts in the genre of the cold war
presidents speeches, as well as the just discussed hint for a fully fledged study of
implicature in (5). Of course, CDA has much more to offer, as does any approach
or discipline that recognizes the paramount importance of (holistic, dynamic) context in human (linguistic) behavior. In whatever kind of acting, including pragmatic acting, we adapt ourselves to a context as is, but at the same time we
change it with a view to future effects. As Mey (1993: 263) writes, all this is done
through the intermediate use of language as the tool helping us to select the relevant features of any situation in the total context. Micropragmatics and macropragmatics are two perspectives which, on the one hand, statically account for the
different size manifestations of this total, or holistic, context. On the other
hand, they dynamically inform each other (and thus the pragmatic analysis as a
whole) of, first, which instances of the utterance-level use of language are worth a
broader contextual verification to determine their function, second, which parts of
the broader contextual check are applicable as determiners of further analytic activity at the utterance level.

Notes
1. See Fetzer (this volume) for the distinction between general (non-linguistic) pragmatics
and linguistic pragmatics.
2. See Fetzer (this volume) or Huang (2007) for a discussion of perspectivist versus componential views of pragmatics.
3. Cognitive pragmatics (e.g. Sperber and Wilson 1986) illustrates this by the onion metaphor. The pattern of inclusion of the individual layers, which are interconnected, reflects
their order of accessibility in the process of inferencing.
4. Foundations of research in deixis are presented in Bhler (1934/1982). Among the most
useful modern overviews are Nunberg (1993), Fillmore (1997), and Levinson (2004); cf.
also Hanks (this volume).
5. The standard view of presupposition (see Huang this volume) is that (i) it involves a proposition whose truth is taken for granted in the utterance of a sentence and remains in force
even if the sentence is negated, and that (ii) it is engendered by certain lexical and structural triggers. This definition goes back to philosophical and semantic views of Frege
(1892) and Strawson (1952), later enriched with semantic-pragmatic analyses (cf. Karttunen 1973; Stalnaker 1973; Gazdar 1979). Though these characterizations of presupposition are still generally accepted (cf. Saeed 2009), many linguists have pointed to the
existence of counter-examples deeming the mainstream approach (necessarily) imperfect (cf. Huang this volume). In the present chapter, presupposition is treated as a phenomenon which does not rely for its existence on linguistic form alone, but on a complex
relation between the linguistic form, the speakers and the hearers (common) understanding of the world (cf. Grundy 1995: 78), as well as the speakers fulfillment of

Micropragmatics and macropragmatics

6.

7.

8.
9.

10.

11.
12.
13.

14.

15.

69

felicity conditions for performance of an act involving presupposition(s). This stance


integrates the standard view with several other research perspectives: the speech act approach to presupposition (cf. Grice 1981; Green 1989), the cognitive approach (cf. Carston 1996), the experiential approach (cf. Marmaridou 2000), or the common ground
approach (cf. Grundy 1995).
Examples of battles over the status of presupposition are Stalnaker (1973), Soames
(1989); while Horn (1996), Atlas (2004) and Huang (2007) offer well-balanced overviews.
In the present chapter the use of the term implicature is synonymous with conversational implicature. See Huang (this volume) for a comprehensive account of various
types of implicature.
We do not deal with conventional implicatures here (cf. e.g. Davis 1998; Bach 1999).
See reductionist attempts such as Horns (1984) Q-R model or Levinsons (2000) Q-I-M
principles; for supplements and general overviews see Levinson (2000), Horn (2007,
2009), Huang (2009a, b and this volume).
See Austin (1962/1975) and Searle (1975, 1979) for the foundations of the Speech Act
Theory. Tsohatzidis (1994), Thomas (1995), Marmaridou (2000), and Sadock (2004)
are excellent, objective overviews of both the classical ideas and later developments.
See also Collavin (this volume).
See Austin (1962/1975) and a review of Austins ideas in e.g. Alston (1994).
Cf. Austins (1962/1975) account of felicity conditions; see also Searle (1979).
Despite a temptation to the contrary, I save the term global for the discussion of macro
acts. I refer to this intention as larger not just to say that it is more indirect than the
intentions to greet, grasp students attention, or work out the number of absentees. My
point is that all of the latter can accumulate into a bigger intentional schema and that
there is no limit to the number of micro-intentional components making up such a
schema. Hence the use of the comparative larger, which allows for a downward
specification (i.e. the specification of what the larger intention is supposed to comprise), as well as an upward specification (i.e. the specification of what higher or highest level of intentionality it could contribute to). Interestingly, van Dijks (1980)
speech event approach to intentionality levels, as well as the activity type and genre
oriented approaches to intentions and context (cf. Levinson 1979; Bauman 1992; Luckmann 1995; Thibault 2003; discussed in section 4), are far from elucidating this bi-directional possibility, which is an important element of the micro-macro interface.
See van Dijk (1980) and Mey (1993). Levinson (1979) uses the term activity type. The
most frequent, however, is the term genre (cf. fn. 13). Recall (section 4) that [genre]
specifies [multiple] typical ways in which lexicogrammatical resources are deployed to
enact a communicative goal, so even such short one-utterance forms as (4) apply, since
the ways in which they could be supported by other forms (recall from the above the potentially infinite catalogue of expressions such as its getting late!) are controlled by
common, conventional, and thus generic constraints. Unlike speech events, which combine into macro speech acts (cf. van Dijk 1980), genres do not presuppose a methodologically superordinate concept. Thus, they constitute a more elegant framework in
terms of their downward specification, but fall short of describing the infiniteness of
consecutive layers in the whole intentionality structure.
See mainly van Dijks work, from classical contributions to text grammar (cf. van Dijk
1977, 1980), to recent research in critical discourse analysis (cf. e.g. van Dijk 2008).

70

Piotr Cap

See also his joint work with Walter Kintsch (e.g. van Dijk and Kintsch 1983), which in
my view represents the best step ever taken to integrate the concepts of discourse and
text in a pragmatic analysis.
16. Its protagonists include Emanuel Schegloff, Harvey Sacks, Gail Jefferson, Jonathan
Potter; cf. Ten Have (2007) for an overview.
17. The Whitehall Palace address, November 19, 2003. The analysis of (5) shows that macropragmatic studies often proceed in a top-down manner (cf. section 4); they draw on
social and cognitive context accessible to the analyst and employ the analysts expert
knowledge to an a priori formulation of the functional thesis. The source of the analysts expert knowledge is not only his/her cognitive experience, but also the data provided by disciplines outside the principal domain of linguistics. Thus, macropragmatics
fosters interdisciplinarity, a notable example of which is Critical Discourse Analysis.

References
Alston, William
1994
Illocutionary acts and linguistic meaning. In: Savas Tsohatzidis (ed.), Foundations of Speech Act Theory: Philosophical and Linguistic Perspectives,
2949. London: Routledge.
Atlas, Jay
2004
Presupposition. In: Laurence Horn and Gregory Ward (eds.), The Handbook of
Pragmatics, 2952. Oxford: Blackwell.
Attardo, Salvatore
1990
The violation of Grices maxims in jokes. In: Kira Hall, Michael Meacham and
Richard Shapiro (eds.), Proceedings of the 16th Annual Meeting of the Berkeley
Linguistics Society, 355362. Berkeley, CA: University of California Press.
Attardo, Salvatore
1993
Violation in conversational maxims and cooperation: The case of jokes. Journal of Pragmatics 19: 532558.
Austin, John
1962/1975 How to Do Things with Words. Oxford: Oxford University Press.
Bach, Kent
1999
The myth of conventional implicature. Linguistics and Philosophy 22:
327366.
Bateson, Gregory
1972
Steps of an Ecology of Mind. Toronto: Chandler Publishing Company.
Bauman, Richard
1992
Contextualization, tradition, and the dialogue of genres: Icelandic legends of
the kraftskald. In: Alessandro Duranti and Charles Goodwin (eds.), Rethinking
Context: Language as an Interactive Phenomenon, 125146. Cambridge:
Cambridge University Press.
Bazzanella, Carla
2002
The significance of context in interpretation: The we case. In: Bruce Edmonds
and Varol Akman (eds.), Context in Context. Special issue of Foundations of
Science 7, 239254.

Micropragmatics and macropragmatics

71

Beaver, David
2001
Presupposition and Assertion in Dynamic Sentences. Stanford: CSLI.
Bertolet, Rod
1994
Are there indirect speech acts? In: Savas Tsohatzidis (ed.), Foundations of
Speech Act Theory: Philosophical and Linguistic Perspectives, 335349. London: Routledge.
Bezuidenhout, Anne
1997
Pragmatics, semantic underdetermination and the referential/attributive distinction. Mind 106: 375409.
Bezuidenhout, Anne
2002
Truth-conditional pragmatics. Philosophical Perspectives 16: 105134.
Biber, Douglas
1988
Variation across Speech and Writing. Cambridge: Cambridge University
Press.
Blakemore, Diane
1992
Understanding Utterances: An Introduction to Pragmatics. Oxford: Blackwell.
Bublitz, Wolfram
2003
Emotive prosody: how attitudinal frames help construct context. In: Ewald
Mengel, Hans-Jrg Schmid and Michael Steppat (eds.), Anglistentag 2002
Bayreuth, Proceedings, 381391. Trier: Wissenschaftlicher Verlag.
Bublitz, Wolfram
2006
It utterly boggles the mind: knowledge, common ground and coherence. In:
Hanna Pishwa (ed.), Language and Memory: Aspects of Knowledge Representations, 359386. Berlin: de Gruyter.
Bhler, Karl
1934/1982 Sprachtheorie. Die Darstellungsfunktion der Sprache. Stuttgart/New York:
Fischer.
Cap, Piotr
2008
Towards the proximization model of the analysis of legitimization in political
discourse. Journal of Pragmatics 40: 1741.
Cap, Piotr
2010
Axiological aspects of proximization. Journal of Pragmatics 42: 392407.
Carston, Robyn
1996
Metalinguistic negation and echoic use. Journal of Pragmatics 25: 309330.
Carston, Robyn
2002
Thoughts and Utterances. The Pragmatics of Explicit Communication. Oxford: Blackwell.
Cole, Peter (ed.)
1978
Syntax and Semantics 9: Pragmatics. London: Academic Press.
Cummings, Louise
2005
Pragmatics: A Multidisciplinary Perspective. Edinburgh: Edinburgh University Press.
Davis, Wayne
1998
Implicature: Intention, Convention, and Principle in the Failure of Gricean
Theory. Cambridge: Cambridge University Press.

72

Piotr Cap

Duranti, Alessandro and Charles Goodwin (eds.)


1992
Rethinking Context. Language as an Interactive Phenomenon. Cambridge:
Cambridge University Press.
Elugardo, Reinaldo and Robert Stainton
2004
Shorthand, syntactic ellipsis, and the pragmatic determinants of what is said.
Mind and Language 19: 44271.
Fairclough, Norman
1995
Critical Discourse Analysis: The Critical Study of Language. London: Longman.
Fetzer, Anita
2002
Communicative intentions in context. In: Anita Fetzer and Christiane Meierkord (eds.), Rethinking Sequentiality: Linguistics Meets Conversational Interaction, 3769. Amsterdam: Benjamins.
Fetzer, Anita (ed.)
2007
Context and Appropriateness: Micro Meets Macro. Amsterdam: Benjamins.
Fillmore, Charles
1997
Lectures on Deixis. Stanford: CSLI.
Frege, Gottlob
1892
ber Sinn und Bedeutung. Zeitschrift fr Philosophie und Philosophische Kritik 100: 2550. Trans. as On sense and reference. In: Peter Geach and Max
Black (eds.), 1952 Translations from the Philosophical Writings of Gottlob
Frege. Oxford: Blackwell.
Gazdar, Gerald
1979
Pragmatics: Implicature, Presupposition and Logical Form. London: Academic Press.
Goffman, Erving
1986
Frame Analysis. Boston: Northeastern University Press.
Green, Georgia
1989
Pragmatics and Natural Language Understanding. Hillsdale, NJ: Lawrence
Erlbaum.
Grice, Paul
1975
Logic and conversation. In: Peter Cole and Jerry Morgan (eds.), Syntax and
Semantics 3: Speech Acts, 4158. New York: Academic Press.
Grice, Paul
1981
Presupposition and conversational implicature. In: Peter Cole (ed.), Radical
Pragmatics, 183198. New York: Academic Press.
Grice, Paul
1989
Studies in the Way of Words. Cambridge, MA: Harvard University Press.
Grundy, Peter
1995
Doing Pragmatics. London: Edward Arnold.
Gumperz, John
1992
Contextualization and understanding. In: Alessandro Duranti and Charles
Goodwin (eds.), Rethinking Context: Language as an Interactive Phenomenon, 229252. Cambridge: Cambridge University Press.
Gumperz, John
2003
Response essay. In: Susan Eerdmans, Carlo Prevignano and Paul Thibault
(eds.), Language and Interaction. Discussions with John J. Gumperz, 105
126. Amsterdam: Benjamins.

Micropragmatics and macropragmatics

73

Halliday, Michael and Ruqaiya Hasan


1976
Cohesion in English. London: Longman.
Halliday, Michael and Ruqaiya Hasan
2000
System and text: making links. Text 20: 201210.
Holdcroft, David
1994
Indirect speech acts and propositional content. In: Savas Tsohatzidis (ed.),
Foundations of Speech Act Theory: Philosophical and Linguistic Perspectives,
350364. London: Routledge.
Horn, Laurence
1984
Toward a new taxonomy for pragmatic inference: Q-based and R-based implicature. In: Deborah Schiffrin (ed.), Meaning, Form, and Use in Context: Linguistic Applications, 1142. Washington, DC: Georgetown University Press.
Horn, Laurence
1996
Presupposition and implicature. In: Shalom Lappin (ed.), The Handbook of
Contemporary Semantic Theory, 299320. Oxford: Blackwell.
Horn, Laurence and Gregory Ward (eds.)
2004
The Handbook of Pragmatics. Oxford: Blackwell.
Horn, Laurence
2007
Toward a Fregean pragmatics: Voraussetzung, Nebengedanke, Andeutung. In:
Istvan Kecskes and Laurence Horn (eds.), Explorations in Pragmatics: Linguistic, Cognitive and Intercultural Aspects, 3969. Berlin: Mouton de
Gruyter.
Horn, Laurence
2009
WJ-40: Implicature, truth, and meaning. International Review of Pragmatics
1: 334.
Huang, Yan
2007
Pragmatics. Oxford: Oxford University Press.
Huang, Yan
2009a
Neo-Gricean pragmatics and the lexicon. International Review of Pragmatics
1: 118153.
Huang, Yan
2009b
Neo-Gricean pragmatic theory of conversational implicature. In: Bernd Heine
and Heiko Narrog (eds.), The Oxford Handbook of Linguistic Analysis,
669696. Oxford: Oxford University Press.
Hymes, Dell
1974
Foundations in Sociolinguistics: An Ethnographic Approach. Philadelphia:
University of Philadelphia Press.
Karttunen, Lauri
1973
Presuppositions of compound sentences. Linguistic Inquiry 4: 169193.
Lakoff, George and Mark Johnson
1999
Philosophy in the Flesh: The Embodied Mind and its Challenge to Western
Thought. New York: Basic Books.
Langacker, Ronald
1987
Foundations of Cognitive Grammar. Vol. I: Theoretical Prerequisites. Stanford, CA: Stanford University Press.
Langacker, Ronald
2001
Discourse in Cognitive Grammar. Cognitive Linguistics 12: 143188.

74

Piotr Cap

Levinson, Stephen
1979
Activity types and language. Linguistics 17: 365399.
Levinson, Stephen
1983
Pragmatics. Cambridge: Cambridge University Press.
Levinson, Stephen
2000
Presumptive Meanings: The Theory of Generalized Conversational Implicature. Cambridge, MA: MIT Press.
Linell, Per
1998
Approaching Dialogue. Amsterdam: Benjamins.
Luckmann, Thomas
1995
Interaction planning and intersubjective adjustment of perspectives by communicative genres. In: Esther Goody (ed.), Social Intelligence and Interaction:
Expressions and Implications of the Social Bias in Human Intelligence, 175
188. Cambridge: Cambridge University Press.
Lyons, John
1995
Linguistic Semantics. Cambridge: Cambridge University Press.
Mann, William and Sandra Thompson
1983
Relational propositions in discourse. Technical Report ISI/RR-83115. University of Southern California: Information Sciences Institute.
Manning, Paul
2001
On social deixis. Anthropological Linguistics 43: 54100.
Marmaridou, Sophia
2000
Pragmatic Meaning and Cognition. Amsterdam: Benjamins.
Mey, Jacob L.
1993
Pragmatics. Oxford: Blackwell.
Nunberg, Geoffrey
1993
Indexicality and deixis. Linguistics and Philosophy 16: 144.
Recanati, Franois
1993
Direct Reference: From Language to Thought. Oxford: Blackwell.
Recanati, Franois
2001
What is said. Synthese 128: 7591.
Recanati, Franois
2004
Literal Meaning. Cambridge: Cambridge University Press.
Saeed, John
2009
Semantics. 3rd edition. Oxford: Blackwell.
Sadock, Jerrold
1978
On testing for conversational implicature. In: Peter Cole (ed.), Syntax and
Semantics 9: Pragmatics, 281298. London: Academic Press.
Sadock, Jerrold
2004
Speech acts. In: Laurence Horn and Gregory Ward (eds.), The Handbook of
Pragmatics, 5373. Oxford: Blackwell.
Saville-Troike, Muriel
1989
The Ethnography of Communication. Oxford: Blackwell.
Searle, John
1975
A taxonomy of speech acts. In: Keith Gunderson (ed.), Minnesota Studies in
the Philosophy of Science 9: 344369.

Micropragmatics and macropragmatics

75

Searle, John
1979
Expression and Meaning: Studies in the Theory of Speech Acts. Cambridge:
Cambridge University Press.
Soames, Scott
1989
Presupposition. In: Dov Gabbay and Franz Guenthner (eds.), Handbook of
Philosophical Logic, 553616. Dordrecht: Reidel.
Sperber, Dan and Deirdre Wilson
1986
Relevance: Communication and Cognition. Oxford: Blackwell.
Stalnaker, Robert
1973
Presuppositions. Journal of Philosophical Logic 2: 447457.
Ten Have, Paul
2007
Doing Conversation Analysis. A Practical Guide. Thousand Oaks: Sage.
Strawson, Peter
1952
Introduction to Logical Theory. London: Methuen.
Thibault, Paul
2003
Contextualization and social meaning-making practices. In: Susan Eerdmans,
Carlo Prevignano and Paul Thibault (eds.), Language and Interaction. Discussions with John J. Gumperz, 4162. Amsterdam: Benjamins.
Thomas, Jenny
1995
Meaning in Interaction: An Introduction to Pragmatics. London: Longman.
Tsohatzidis, Savas (ed.)
1994
Foundations of Speech Act Theory: Philosophical and Linguistic Perspectives.
London: Routledge.
Van Dijk, Teun
1977
Text and Context. London: Longman.
Van Dijk, Teun
1980
Macrostructures. An Interdisciplinary Study of Global Structures in Discourse,
Interaction, and Cognition. Hillsdale, NJ: Erlbaum.
Van Dijk, Teun
1995
Discourse, semantics and ideology. Discourse and Society 6: 243259.
Van Dijk, Teun
2002
Ideology: Political discourse and cognition. In: Paul Chilton and Christina
Schaeffner (eds.), Politics as Text and Talk, 3357. Amsterdam: Benjamins.
Van Dijk, Teun
2008
Discourse and Context. A Sociocognitive Approach. Cambridge: Cambridge
University Press.
Van Dijk, Teun and Walter Kintsch
1983
Strategies of Discourse Comprehension. New York: Academic Press,
Verschueren, Jef
1999
Understanding Pragmatics. London: Edward Arnold.
Wodak, Ruth and Paul Chilton (eds.)
2005
A New Agenda in (Critical) Discourse Analysis: Theory, Methodology and Interdisciplinarity. Amsterdam: Benjamins.

3.

Pragmalinguistics and sociopragmatics


Sophia Marmaridou

1.

Introduction

Within a broad definition of pragmatics as the scientific study of all aspects of linguistic behaviour, the distinction between pragmalinguistics and sociopragmatics
is intended to focus on two methodological approaches to pragmatic analysis.
Pragmalinguistics typically concerns the study of the particular resources that a
given language provides for conveying pragmatic meaning (illocutionary and interpersonal), whereas sociopragmatics relates pragmatic meaning to an assessment
of participants social distance, the language communitys social rules and appropriateness norms, discourse practices, and accepted behaviours. This methodological distinction, initially launched by Leech (1983) and Thomas (1983) and widely
adopted in subsequent work in pragmatics, is distinct from, though not totally unrelated to, earlier work in Marxist pragmatics, in which the term pragmalinguistics
essentially marks the pragmatic turn in linguistic analysis and is broadly associated, or even identified, with sociolinguistics, stylistics, or text linguistics (see
Mey 1979; Prucha 1983). Moreover, the term sociopragmatics has been used in
contrast to psychopragmatics (Dascal and Franozo 1989), also independently
from the previous distinction just mentioned.
In this paper I shall argue that, as defined above, the distinction between pragmalinguistics and sociopragmatics suggests a conceptual dichotomy in our understanding of language either as a system used for communication purposes, or as an
action domain reflecting, or co-constructing, social order in culturally sanctioned
ways. Regardless of whether this conceptual dichotomy is systematically reflected
in the methodological distinction it motivates, the latter has been particularly productive. In fact, it has both affected the internal organisation and theoretical refinement of other concepts, such as pragmatic failure and pragmatic transfer, and also
informed particular methodological approaches to second language teaching and
testing. Moreover, it has been of particular relevance to historical pragmatics and
historical corpus linguistics. The apparent usefulness of the distinction between
pragmalinguistics and sociopragmatics seems to be at odds with the fact that the
borderline between the two is often fuzzy, to the extent that they are often thought
of as the two end points of a continuum. Given the variety of studies straddling different points along this continuum, I propose to use the figure/ground schema afforded us by psychology to further argue that the two concepts under investigation
essentially reflect a methodological choice in current practices to focus on one area
of pragmatic research by conferring background status to another. The use of the

78

Sophia Marmaridou

figure/ground relationship in this context seeks to explain the bibliographically attested fluidity of the content of these concepts and to present them as distinct, but
interacting, instances of perspectivisation.
It has become obvious from the above that the discussion of the concepts
underlying the terms under investigation will adopt a historical approach to their
analysis. This is expected on the one hand to place pragmalinguistics and sociopragmatics in the wider framework of pragmatic study that was flourishing in the
two crucial decades of its theoretical formation, and, on the other, to show how
these concepts have gradually permeated more recent trends in pragmatic study.
In the next section reference will be made to the ways in which the terms pragmalinguistics and sociopragmatics have been used independently of each other and
prior to the methodological distinction they have been mainly associated with as
introduced by Leech (1983) and Thomas (1983). In particular, the term pragmalinguistics as used in Marxist linguistics in the late seventies and early eighties will
be shown to be broadly equivalent to pragmatic linguistics or societal pragmatics
(Mey 1993). In this framework, the term covers research which emphasises functional aspects of language use within various paradigms (Marxist, or otherwise).
Moreover, sociopragmatics, as a theory of the ways in which the non-linguistic environment affects language use, will be contrasted with psychopragmatics as a theory of how the linguistic environment affects thought (Dascal and Franozo 1989).
In the third section the distinction between pragmalinguistics and sociopragmatics
will be discussed from the perspective of Leech (1983) and Thomas (1983). More
specifically, the conceptual underpinnings of this distinction will be set off against
established research practices concerning issues of contrastive linguistics, pragmatic failure, pragmatic transfer, pragmatic development, and the teaching and
testing of pragmatics in second language learning settings. In the fourth section the
contribution of the pragmalinguistic and sociopragmatic perspective to historical
pragmatics and historical corpus linguistics will be examined with a view to highlighting more recent trends in pragmatic research. In the last section of this paper
an overall assessment of the concepts of pragmalinguistics and sociopragmatics
will be attempted and their distinction will be evaluated with respect to the fields of
study in which it has been adopted.

2.

The borders between pragmalinguistics and sociopragmatics

In order to appreciate the conceptual dichotomy encoded in the terms pragmalinguistics and sociopragmatics, one should be reminded of the two main trends that
have dominated the field of pragmatics ever since its inception as a distinct, established field of linguistic enquiry. In the Anglo-American tradition, pragmatics is
mainly concerned with speakers meaning as intentionally communicated to an addressee. As Jucker (2008: 894) observes, pragmatics in this tradition is a rescue

Pragmalinguistics and sociopragmatics

79

plan for syntax and semantics that deals with a systematic account of implicit
meanings and deals with intentional human action. Levinson (1983: 732), while
espousing this tradition, has identified, discussed, and amply demonstrated problems of defining pragmatics in this narrow sense, to conclude that the best way to
delimit this field of study is to actually observe what practitioners do, or provide a
list of the phenomena which a pragmatic theory must account for, namely deixis,
implicature, presupposition, speech acts, and aspects of discourse structure (Levinson 1983: 27). He also points out that the upper bound of pragmatics is provided
by the borders of semantics and the lower bound by sociolinguistics (and perhaps
psycholinguistics too), and admits that drawing a boundary between sociolinguistic and pragmatic phenomena is likely to be an exceedingly difficult enterprise
(Levinson 1983: 29).
Within the tradition of Levinsons topical definition of pragmatics, the term
pragmalinguistics has been used in the area of literary linguistics, or stylistics, to
refer to the study of Gricean maxims, speech acts and politeness phenomena in a
literary work. An example is Haverkates (1994) study of the dialogues of Don
Quixote de la Macha in Cervantess well known novel. Apparently, this pragmalinguistic frame of reference provides new insights into the interactional roles and
personality traits of Don Quixote and Sancho Panza, which traditional literary theory does not typically address.
The Anglo-American perspective on pragmatics has been criticised as leaving
out, or backgrounding, the larger social context of communication, which is of
central importance in the European tradition (Jucker 2008: 895). The latter derives more directly from Morriss (1938) definition of semiotics, distinguishing between syntactics as the formal relation of signs to one another, semantics as the relation of signs to the objects to which the signs are applicable, and pragmatics as
the relation of signs to interpreters. In Morriss account, the user/speaker is seen in
a wider perspective, since the human being, as an interpreter of signs, does not only
have a mental identity, but also a biological and social identity, which affects her
interpretation of signs (Marmaridou 2000: 18). In fact, Morris acknowledges the
dependence of semiosis on the human being as an entity that is simultaneously the
source and the subject of all psychological, biological and sociological phenomena
that have to do with the functioning of signs comprising language use:
It is a sufficiently accurate characterization of pragmatics to say that it deals with the
biotic aspects of semiosis, that is, with all the psychological, biological, and sociological phenomena which occur in the functioning of signs. (Morris 1938: 108)

Within the context of these two traditions in pragmatics, the terms pragmalinguistics and sociopragmatics have both been used to refer broadly to pragmatic phenomena that characterise the use of language for communication purposes. In what
follows, it will be shown that, while pragmalinguistics has been associated with a
societal and Marxist perspective on language use, sociopragmatics generally refers

80

Sophia Marmaridou

to the study of the external pragmatic factors affecting language use. In this sense,
sociopragmatics has also been juxtaposed to psychopragmatics, the latter focusing
on internal, cognitive operations such as reasoning and problem solving. It should
be pointed out that pragmalinguistics and sociopragmatics, as defined above, constitute two different and unrelated projects. Therefore they do not reflect the distinction and complementariness they have been associated with in the work of
other scholars, such as Leech (1983) and Thomas (1983) (see section 3). Rather,
the use of these two terms in the frameworks discussed below is indicative of the
various developments of pragmatics within the two decades of its initial expansion.
Focusing on sociological phenomena, Mey (1979) uses the term pragmalinguistics to redefine the object and goal of linguistic enquiry in a Marxist framework
as the study of language (the linguistic object) in a variety of pragmatic settings. In
his analysis he identifies pragmatics with the description of action in general and
pragmatic linguistics (or pragmalinguistics) with the study of linguistic action as
determined by the social conditions for speaking and understanding. As he claims,
from a pragmatic viewpoint, the conditions for speaking and understanding, for
production and consumption of language, cannot be divorced from the conditions of production and consumption in the society in general (Mey 1979: 11).
The totality of conditions that are active in the production and consumption of
texts, including co-text, constitute the context, which is dynamic, rather than static,
and creative, rather than passive.
Following Meys understanding of pragmalinguistics, Olsen (1979) raises the
issue of linguistic meaning as located in, and defined by, language use in its social
context. As he observes,
the meanings of words and utterances were not universals which merely were differently labelled in different languages. The meanings were dependent on, and part of, the
different social systems []. Utterances are produced and understood within the definition of the situation and it is the rules by which we define the situation that are shared
by the members of a social system (Olsen 1979: 247).1

That linguistic meaning is a function of language use and that it is being constructed all the time by the process of interaction are facts that, according to Olsen,
must be accounted for by pragmatic linguistics. It transpires that, in his analysis,
pragmalinguistics is to be broadly understood as the pragmatic turn, or a pragmatic
perspective on linguistics as informed by social theory.
As already mentioned, pragmalinguistics is also understood to generally cover
the study of functional aspects of language use as practised in Eastern Europe in
the late seventies. The relevant studies spread over the areas of functional stylistics, text linguistics, rhetorics, psycholinguistics, sociolinguistics, social communication theory, and semiotics (see Prucha 1983). Within such a broad framework, pragmalinguistics emerges as a theory of the pragmatic component of a
grammar and as a theory of verbal activity in social intercourse. These two per-

Pragmalinguistics and sociopragmatics

81

spectives are reflected in the analysis of the meaning of utterances as comprising


contextual and situational aspects respectively, as in Bang and Drs (1979) work.
Contextual meaning is determined by the lexicon and syntax, style and genre, and
textual implications, and it is to a high degree reproducible and general. Situational
meaning expresses an interpretation of the world, society and the human being, and
carries an evaluative orientation. To accept a situational meaning is to subscribe to
a certain set of values. In this understanding, contextual meaning belongs to the
pragmatic component of grammar, whereas situational meaning develops from
verbal activity in social intercourse. These two kinds of utterance meaning are said
to have a dialectal relation to one another and to define a continuum ranging from
the unique and unrepeatable (situational) to the general and reproducible (contextual) (Bang and Dr 1979: 46).
It transpires from the above that pragmalinguistics in the Eastern European and
Marxist traditions encompasses aspects of language use which range from grammatically and textually encoded elements to socially constructed discursive practices. The emphasis is typically placed on the latter, which figures prominently in
this tradition against the background of the grammatical and textual aspects of language. This theoretical turn is to be interpreted both as a reaction to explicitly formal accounts of language in the various structural approaches of the time (see, for
example, the work of Chomsky), and as an attempt to articulate an ideological
stance towards the epistemological foundations of linguistics. The realisation that
utterance meaning, as the domain proper of pragmalinguistic investigation, is a
function of linguistic means in interaction with, or in a dialectal relation to, the social conditions of language use creates a conceptual frame for what is later to be established as a distinction between pragmalinguistics and sociopragmatics, albeit
outside explicitly stated ideological agendas.
Apart from the societal perspective on language and the association of pragmatics with the social conditions of language use, the cognitive perspective on language contributes to its construal as an instrument of thought. The latter line of investigation has led to the use of the terms sociopragmatics and psychopragmatics
to carry the distinction between the use of language for communication purposes
and the use of language as an instrument of thought, respectively. More specifically, in their attempt to shed light on the pragmatic turn in psycholinguistics, Dascal and Franozo (1989) refer to pragmatics as the field of study concerned with
conversational maxims, politeness requirements, contextual circumstances of utterances, and other factors that impose constraints on the selection and understanding, in a given context, of a particular linguistic expression for a certain cognitive
content (Dascal and Franozo 1989: 5). In many cases, the pragmatic turn in psycholinguistics amounts to the study of the acquisition of pragmatics by young
children and, in particular, the study of how children acquire and perform speech
acts and how they establish interpersonal communicative relations (Bates 1976).
However, Dascal and Franozo (1989) observe that language plays an important

82

Sophia Marmaridou

role in our mental processes and higher level cognitive processes or states, such as
reasoning, problem solving, storing and retrieving information, believing, etc.
They further claim that the use of language for such purposes also belongs to pragmatics: we use language not only to communicate, but also to do our reasoning, to
solve problems, etc. They therefore use the term psychopragmatics to refer to the
study of such mental use of language and distinguish it from the study of its use for
communicative purposes, i.e. from sociopragmatics. By the former term they refer
to internal pragmatic factors that affect the performance of the cognitive operations
themselves which in some way involve language. By sociopragmatics they refer to
external pragmatic factors that concern the perception and the production of linguistic signs in a particular situation, such as indirectness in the performance of
speech acts. It follows that psychopragmatics is a theory of the way in which the
linguistic environment of thought influences thought, whereas sociopragmatics is a
theory of the ways in which the non-linguistic environment (which includes both
thought and situation) affects language interpretation and production (Dascal and
Franozo 1989: 13).
The proposed distinction between sociopragmatics and psychopragmatics
serves to compare and contrast language use in jokes and dreams, topics mainly addressed in the framework of Freuds writings (see Freud 1900; 1905). More specifically, Dascal (1985: 98) argues that jokes depend for their effectiveness on the
existence of the sociopragmatic phenomenon of indirectness; they are intended to
be understood as such by an audience and therefore perform a social function. As
opposed to jokes, dreams are mental objects, not meant to be communicated to an
audience, i.e. they have an asocial character. Therefore, they are not socially constrained, their only constraint being the existence of some kind of associative link,
perhaps arbitrary and fortuitous, that provides some path for the expression of
a problematic content. In this respect the psychopragmatic role of language
in dreams is [] similar to its use for another cognitive purpose namely, as a mnemonic device (Dascal 1985: 104). In this analysis, Dascals sociopragmatics
covers the broader area of general pragmatics in both the Anglo-American and the
European traditions with the aim of distinguishing it from psychopragmatics, the
latter illustrating a pragmatic turn in psycholinguistics.

3.

The pragmalinguistics/sociopragmatics distinction


in a cross-linguistic and cross-cultural perspective

The distinction between pragmalinguistics and sociopragmatics was essentially established within the framework of applied linguistics studies on which it was to
exert a strong influence. It was initially introduced in the work of Thomas (1981),
further explored in Thomas (1983), and explicitly spelt out as a methodological approach in pragmatics in Leech (1983). In the latters programme, General Prag-

Pragmalinguistics and sociopragmatics

83

matics is that aspect of human language which concerns the use of language in
communication. More specifically, the understanding of the nature of language
depends both on the study of grammar, as the abstract, formal system of language
(comprising phonology, syntax and semantics), and on the study of pragmatics, as
the interaction of the Cooperative and Politeness Principles partly characterizing
language use. Moreover, Leech (1983: 7) views semantics as a level of contact between grammar and pragmatics, so that linguistic meaning can be fruitfully studied
by a combination of semantic and pragmatic approaches, the former targeting the
sense of an utterance and the latter its force.
Within Leechs project, General Pragmatics is a fairly abstract study of the general conditions of the communicative use of language, excluding more specific
local conditions on language use: The latter may be said to belong to a less
abstract field of SOCIO-PRAGMATICS, for it is clear that the Cooperative Principle and the Politeness Principle operate variably in different cultures or language
communities, in different social situations, among different social classes, etc.
(Leech 1983: 10). Alongside sociopragmatics, Leech introduces PRAGMALINGUISTICS as the study of the more linguistic end of pragmatics where we consider the particular resources which a given language provides for conveying particular illocutions (Leech 1983: 11). It follows that sociopragmatic studies are
culture-specific, while pragmalinguistic studies are language-specific. Apparently,
Leechs observation that sociopragmatics is the sociological interface of pragmatics is implicitly an attempt to address an issue that Levinson (1983: 29) also
raises when he observes that the boundary between sociolinguistic and pragmatic
phenomena is hard to define.
Since Leech (1983) is concerned with establishing the Cooperative and Politeness Principles as the two cornerstones of General Pragmatics and with arguing
their universality, he does not pursue the pragmalinguistics/sociopragmatics distinction any further. Nevertheless, he views both principles as the natural constraints of universality. For example, while claiming the universality of the principles, he concedes that they do not apply in an identical manner in all societies:
Indeed, one of the main purposes of sociopragmatics [] is to find out how different societies operate maxims in different ways, for example by giving politeness a higher rating than cooperation in certain situations, or by giving precedence to one maxim of the
Politeness Principle rather than another (Leech 1983: 80).

Similarly, he observes that cultural stereotyping (e.g. the British are more tactful
that the Americans) is related to pragmalinguistic strategies such as strategies of
indirectness, and the norms observed in the performance of these strategies in different speech communities (Leech 1983: 231).
Leechs (1983) distinction between pragmalinguistics and sociopragmatics is
the natural outcome of his claim about the universality of the principles of pragmatics he proposes. By prioritizing conversational principles, he proposes a rhe-

84

Sophia Marmaridou

torical model of pragmatics referring to the study of the effective use of language
in communication and focusing on the goal-oriented speech situation, whereby the
speaker uses language in order to produce a particular effect in the mind of the
hearer. Leechs principles of general pragmatics are to be held constant and form
the basis, or the ground, against which all other language- and culture-specific
pragmatic variables can be studied as distinct figures. This position lends itself to
cross-linguistic and cross-cultural comparisons of communicative behaviour with
obvious applications for language teaching/learning and testing. Moreover, the distinction proposed within this framework seems compatible with the aims and
methods of the study of discourse and historical pragmatics in particular. In what
follows, these areas of investigation will be related to the conceptual and methodological distinction between pragmalinguistics and sociopragmatics as introduced in this section.

4.

Applications of the pragmalinguistics/sociopragmatics distinction

4.1.

Preliminary investigations in contrastive linguistics

The impact of the pragmalinguistics/sociopragmatics distinction has been particularly felt in the conceptual organisation of some theoretical constructs it has motivated, such as pragmatic failure, pragmatic transfer, and pragmatic development.
This development in the area of applied linguistics is to some extent foreshadowed
by Riley (1979) advocating a contrastive pragmalinguistics.
Taking the broad view of pragmalinguistics as pragmatic linguistics, Riley
(1979) defines its aim as the study of communicative acts. These may be realised
verbally, paralinguistically, or non-verbally, and rest on a theory of illocution; illocutionary force is extended in this system to all kinds of communicative acts. Focusing on linguistic acts, he notes that what is central to contrastive pragmalinguistics is language functions rather than linguistic structures, discourse, not
grammar, the communicative act in context, not the sentence in isolation (Riley
1979: 57). Drawing in this way a methodological line between form and function,
he highlights instances of same form and different function (e.g. the possibility of
using Youre not going out to express the illocutions of prohibiting, confirming,
threatening, expressing surprise, or stating), and instances whereby the same function is performed by different forms (e.g. the function of agreeing as expressed by
the forms Yes, of course, I agree, nodding, or repetition of the previous speakers utterance). In his model, form-function pairings constitute the ground against
which cross-linguistic contrasts are to be drawn.
In order to place this model in a wider framework encompassing all aspects of
pragmatic study, Riley (1979) further proposes cross-linguistic contrasts across
what he calls discourse structure, consisting of three sub-levels of structure:

Pragmalinguistics and sociopragmatics

85

Firstly, formal structure includes all aspects of message-bearing elements, e.g. verbal, paralinguistic, or non-verbal, in a particular situation. Secondly, illocutionary
structure consists of sequences of illocutionary acts, e.g. inviting accepting, confirming thanking, etc. Thirdly, interactive structure refers to linguistic organization in terms of interactional tactics such as turns (e.g. opening, reply, closing, address, exchange, transaction, etc.). Evidently, Rileys work is an attempt to include
in his model all topics of pragmatic interest and put them to the forefront of contrastive linguistics. However, his efforts to systematise the proposed model lead to
a schematic and idealised understanding of communicative acts. For example, the
sequences of illocutionary acts seem to refer to idealised prototypes. In reality, invitations are typically, but not always, followed by acceptances, or refusals, confirmations by thanks, etc.
This apparent theoretical weakness of Rileys model, though, is counterbalanced by its applicability in contrastive analysis and language teaching in particular. More specifically, the model allows for the identification and the teaching of interactional styles and illocutionary point in relation to sociolinguistic variables
such as roles, status, formality, and discursive control, as for example, in a typical
teacher student illocutionary sequence, or in a typical asking-for-street-instructions exchange between strangers. In this way, Rileys proposal addresses newly
created needs in the methodology of communicative language teaching research
that was being carried out at the time (see, e.g., Candlin, 1976; Brumfit and
Johnson, 1979; Widdowson, 1979). Perhaps the most important contribution of
Rileys work in introducing contrastive pragmalinguistics is that he creates an
awareness of the significance of pragmatic knowledge to language teaching and
emphasises the systematicity of such knowledge so that it can be actually taught.
His suggestions clearly encourage a methodological turn, to be further explored in
the context of the pragmalinguistics/sociopragmatics distinction introduced by
Thomas (1981; 1983) and Leech (1983).
4.2.

Pragmatic failure

Following Leechs (1983) division of linguistics into grammar and pragmatics and
adopting his goal-oriented understanding of language use (see above), Thomas
(1983) first construes linguistic competence as consisting of grammatical competence (abstract knowledge of phonology, syntax and semantics) and pragmatic
competence as the ability to use language effectively in order to achieve a specific
purpose and to understand language in context (Thomas 1983: 92). Within this
framework, pragmatic failure is defined as an inability to recognise the force of
the speakers utterance when the speaker intended that this particular hearer should
recognise it (Thomas 1983: 94). Notably, whereas grammatical error indicates the
speakers restricted linguistic proficiency, pragmatic failure reflects badly on her as
a person. It follows that, in the context of cross-cultural communication, pragmatic

86

Sophia Marmaridou

failure may lead to a misunderstanding of the foreign speakers intentions. Awareness-raising tasks aiming to develop foreign students metapragmatic ability to
consciously analyse language use are commendable in this case, but inadequate in
Thomass (1983: 99) view, unless informed by the distinction between pragmalinguistic and sociopragmatic failure.
The proposed categorisation of types of pragmatic failure, consistent with
Leechs (1983) dyad of language vs. culture specificity, is not intended to impose
rigid boundaries. Thomas (1983: 101) explicitly states that the two concepts form a
continuum from what is language-specific to what is culture-specific. Pragmalinguistic failure results from the speakers mapping of pragmatic force to an utterance
in a way that is systematically different from a native speakers. Sociopragmatic
failure results from the speakers miscalculation of the social conditions placed on
language in use, such as the size of the speakers imposition on the addressee in
performing a particular speech act, a cost/benefit scale of acts in the particular culture, the social distance between interlocutors, and relative, culture-specific rights
and obligations of interlocutors to each other in specific situations. As one moves
from the communicative intent of a speakers utterance to the interpersonal dynamics between speaker and addressee, one is also moving from the pragmalinguistic to the sociopragmatic end of the continuum.
In providing examples of pragmalinguistic failure, Thomas (1983: 101) concedes to the systematicity of formfunction pairings in a particular language and
culture in the sense that native speakers assign pragmatic force to certain utterances fairly predictably. Unlike Riley (1979), she seems to restrict the scope of this
systematicity to highly conventionalised utterances, such as the often-quoted can
you do X English utterance used to issue a request rather than ask for ones ability.
The fact that in other languages, such as French or Russian, the opposite is true
may lead the foreign language learner to an inappropriate transfer of a formfunction pair, and hence to pragmalinguistic failure. A further cause of pragmalinguistic failure may be the inappropriate transfer of speech act strategies from L1 to
L2, such as the use of direct vs. indirect forms of issuing requests politely. For
example, it is reported that polite, indirect requests for street instructions (e.g. excuse me, could you please tell me how to ) would be counterproductive in Russian, while they are pretty standard in English. Hence, the Russian learner of English using direct means to perform a request might be misunderstood as being
impolite (Thomas 1983: 102).
Politeness is indeed an area in which pragmalinguistic failure merges with sociopragmatic failure. Whereas the choice of polite forms and strategies seems to
belong to the pragmalinguistic end of the continuum, when and to whom to be polite concerns the sociopragmatics end. In Thomass view, the size of imposition of a
speech act, the awareness of taboos in a particular cultural context, and differing
assessments of relative power or social distance between interlocutors may lead to
sociopragmatic failure. One may add to these sources of sociopragmatic failure the

Pragmalinguistics and sociopragmatics

87

orientation of a culture to positive or negative politeness (see Sifianou 1992). However, it is not always clear whether an instance of pragmatic failure can be assigned
to the pragmalinguistic or to the sociopragmatic level. An example that Thomas provides has to do with the conception of what constitutes free goods in
the former Soviet Union as opposed to Britain. Asking for a cigarette in the Soviet
Union requires a minimal degree of politeness, as these items are free goods in
this cultural setting. A Russian asking for a cigarette in Britain and using a similar
strategy may have either encoded wrongly the correct amount of politeness she
intended, or misjudged the amount of imposition of her request. The former case
would represent an instance of pragmalinguistic, while the latter of sociopragmatic
failure. The effect of either of these two types of pragmatic failure may be the same
for the addressee and their causes difficult for the analyst to distinguish. However,
the distinction is essential in the context of language teaching, as it reflects two different types of decision-making. As Thomas observes, pragmalinguistic decisions
are language-specific and the teacher can correct them in a fairly straightforward
manner. Sociopragmatic ones are culture-specific and sensitive to the learners
own system of values; hence they should not be corrected, but only pointed out
and discussed (Thomas 1983: 109).
It is therefore clear that the discussion of pragmatic failure in the context of language teaching relies heavily on the distinction between pragmalinguistics and sociopragmatics. However, given that sociopragmatic decisions are social before
they are linguistic (Thomas 1983: 104), the importance of the distinction between
sociopragmatics and sociolinguistics becomes apparent. The identification of social roles and stereotypes, power relations between interlocutors relating to age,
social status, or dialect, socioculturally-based evaluations of situations, and linguistic value judgements have been typically considered social variables and have
traditionally concerned sociolinguistic research (see, e.g. Holmes 2008). How are
all these to be differently researched in sociopragmatics? Indeed, this question is
often raised, at least implicitly, in the relevant literature and is differently approached (see the discussion in Bou-Franch and Garcs-Conejos 2003; Roever
2006; Archer and Culpeper 2003, below). It transpires that the application of the
pragmalinguistics/sociopragmatics distinction to pragmatic failure on the one hand
enriches language teaching methodology, but on the other, raises the issue of the
theoretical distinction between sociopragmatics and sociolinguistics, to be addressed in the next section.
4.3.

Pragmatic transfer

The pragmalinguistic/sociopragmatics distinction is also reflected in the study of


pragmatic transfer within the wider context of second language acquisition research. The relevant literature strongly supports the position that non-native speakers understanding and production of linguistic action is influenced by their L1

88

Sophia Marmaridou

pragmatic knowledge (Kasper 1992; Kasper and Rose 2002). Prominent in this
kind of research is the concept of interlanguage as a stage on a continuum representing the second language learners development towards the L2 system. In Selinkers (1972) definition, interlanguage is a psycholinguistic phenomenon which
mainly refers to the learning process, rather than the outcome of this process. However, the dominant practice in interlanguage pragmatics concerns the outcome, i.e.
the collection and comparison of data from native speakers linguistic performance
in the target language, the learners use of her native language, and the learners interlanguage (i.e. her use of the target language at a particular stage of development).2 In this context, pragmatic transfer concerns the possible influences from
the learners L1 on her interlanguage performance.
Even though in practice pragmatic transfer has been typically associated with
transfer of speech act knowledge, indirectness, and politeness, it may also make
reference to sociolinguistic, or discourse parameters in the same instance of language use. For example, transfer of address terms and register concern social variation in language use and are therefore to be considered in the domain of sociolinguistics. Yet, their strategic exploitation in pursuing some illocutionary goal
places their investigation in the area of pragmatics proper. Moreover, some speech
acts are characterized by their position in discourse: greetings occur discourse-initially and refusals occur as seconds in a conversational exchange. Furthermore,
the speech act force of an utterance may become evident only after the utterance is
responded to by the interlocutor, which points to its negotiable character and conversational grounding.
As in the case of pragmatic failure, the pragmalinguistics/sociopragmatics distinction has enabled a more detailed discussion of pragmatic transfer and has added
clarity to this concept by maintaining the corresponding form/function dyad. In
Kaspers (1992) definition, pragmalinguistic transfer shall designate the process
whereby the illocutionary force or politeness value assigned to particular linguistic
material in L1 influences learners perception and production of form-function
mappings in L2 (Kasper 1992: 209). However, as shown in the discussion of pragmatic failure, politeness value does not only depend upon linguistic material, but
also upon social parameters of the speech event, such as the relative social distance
of interlocutors and the power relations between them. For example, L2 learners
may be familiar with address terms of varying degrees of deference or intimacy in
L2, but their use depends upon the learners assessment of social roles, settings,
etc., which may be based on the their own culture. Therefore, politeness is not only
a candidate for pragmalinguistic, but also for sociolinguistic transfer. As Kasper
(1992: 209) observes, the latter is operative when the social perceptions underlying language users interpretation and performance of linguistic action in L2 are
influenced by their assessment of subjectively equivalent L1 contexts. It follows
that the form-function pair does not guarantee appropriate constraints on the discussion of pragmalinguistic and sociopragmatic transfer. Whether politeness

Pragmalinguistics and sociopragmatics

89

transfer occurs at the pragmalinguistic or the sociopragmatic level is largely a


matter of perspectivisation. If construed as a transfer of (in)appropriate forms, the
latter are foregrounded against the background of the social conditions of politeness. If the social conditions of politeness are focused upon, the formal aspects
constitute the background to the analysis.
The issue of pragmalinguistic and sociopragmatic transfer is also associated
with the distinction between positive and negative transfer. Positive transfer of elements from L1 in L2 performance is methodologically difficult to identify, given
that acceptable L2 performance may have sources other than positive transfer of
L1 elements, for example the application of general pragmatic principles, generalisation of L2 specific knowledge, and teacher input, among others. As Kasper
(1992: 213) correctly points out performance data alone cannot tell us whether
[] learners did indeed consult their L1 knowledge or operated strictly on their interlanguage pragmatic competence. More specifically, positive transfer has been
associated with the possibility of pragmatic universals (as in Grice 1975; Leech
1983; Brown and Levinson 1987) that are assumed to function cross-culturally. Interlanguage research has shown that L2 learners have access to the same range of
strategies as native speakers in implementing linguistic actions, such as requests,
suggestions, invitations, refusals, complaints, compliments, etc. (Wolfson 1981;
Trosborg 1987; Beebe and Takahashi 1989; Blum-Kulka 1991; Olshtain and Weinback 1992). This has been taken as evidence of the existence of pragmalinguistic
universals. Similarly, L2 learners sensitivity to social factors such as interlocutors relative status, or degree of imposition, etc., possibly indicates the availability
of sociopragmatic universals. However, in spite of the fact that the evidence is
based on a variety of languages and cultures, the existence of pragmatic universals
at the pragmalinguistic or sociopragmatic level is by necessity an empirical matter,
especially since the universality of theoretical concepts in general pragmatics, such
as politeness, or the co-operative principle, have been seriously questioned (see,
e.g. Keenan 1976; Rosaldo 1982; Wierzbicka 1987; Eelen 2001). Negative pragmalinguistic transfer occurs when inappropriate forms or strategies are used in performing particular speech acts affecting the politeness value of the utterance, or
even its illocutionary force. For example, it has been reported that, in expressing
refusals, Japanese learners of English use strategies such as generalisations (I
never yield to temptation), statements of philosophy (to err is human), or suggestions for alternative action (why dont you ask someone else?) that are atypical of native English speakers (Beebe et al. 1990). At the sociopragmatic level, L2
learners perception of contextual factors may originate in their L1 and affect politeness style. It is reported that Japanese learners of English selected their refusal
strategies on the basis of whether the refusers status was higher or lower than the
interlocutors, whereas American native speakers choice was made on the basis of
status-equal or status-unequal speaker addressee relationships. In spite of the
great number of studies that address the issue of negative pragmatic transfer, the

90

Sophia Marmaridou

systematicity of the relevant experiments, and the careful assessment of their results, L2 learners inappropriate performance may also be a function of other factors, such as learning context, shifts in the available target variety of the L2, little
exposure to the target language and culture, varying L2 development stages, and
lack of contextualisation cues, among others. To the extent that the pragmalinguistics/sociopragmatics distinction shows the need for more precise experimental
techniques in the study of pragmatic transfer, it can contribute to the validation of
hypotheses and is therefore to be understood as a useful methodological tool in interlanguage pragmatics.
4.4.

Pragmatic development

A further contribution of the pragmalinguistics/sociopragmatics distinction to


pragmatic research concerns the issue of the development of grammar and pragmatics in interlanguage research. An important argument in interlanguage pragmatics evolves round the primacy of grammatical or pragmatic knowledge in
interlanguage development. While focus on the development of L2 learners
pragmatics as an autonomous component of communicative competence has had a
serious impact on second language acquisition research, its relation to other aspects of communicative ability has also attracted attention (see Bardovi-Harlig
1999 and references therein). In this context, the pragmalinguistics/sociopragmatics distinction serves to investigate what happens when existing grammatical
knowledge (such as the semantics of the progressive as a grammaticalised aspectual category) is not put to target-like pragmalinguistic use (e.g. to mitigate imposition), as in I was wondering if I could have a word (Kasper and Rose 2002: 180);
moreover, this distinction serves to investigate what happens when grammatical
knowledge leads to target-like pragmalinguistic use, but non-target-like sociopragmatic use. Robinson (1992) reports that, although a Japanese learner of English
had the appropriate knowledge to form and use the I would like to form to express preference, she nevertheless used I want to on the sociopragmatic assumption that the former form would be too polite when addressing her American
friend.
Several studies show that L2 learners may have knowledge of a particular
grammatical structure, which, however, they do not put to native-like pragmalinguistic use. For example, Bodman and Eisenstein (1988) report on L2 learners use
of sophisticated grammatical knowledge which they put to unconventional L2
pragmalinguistic use, as in expressions of thanking like May God grant you a long
life. It is hypothesised that such instances also testify to L1 pragmatic transfer
phenomena. Similarly, Takahashi (1996; 2001) has found that advanced Japanese
EFL students avoided bi-clausal I was wondering whether you could VP and favoured the mono-clausal Could you (please) VP? even though they knew the semantics of the bi-clausal structure. On this evidence Takahashi (2001: 173) con-

Pragmalinguistics and sociopragmatics

91

cludes that the Japanese EFL learners lack the L2 pragmalinguistic knowledge
that an English request can be mitigated to a greater extent by making it syntactically more complex. Sophisticated as the above studies may be, they cannot exclude the possibility that the particular adult L2 learners production is constrained
by sociopragmatic, rather than pragmalinguistic, considerations. For example, it
could be argued that Bodman and Eisensteins subjects may have chosen the particular non-native-like forms of thanking as a way of maintaining their cultural
identity. Besides, it is possible that the Japanese EFL students may not have assessed the particular request(s) as requiring a comparatively higher level of mitigation than others. It is therefore obvious that the lack of sharp boundaries between
pragmalinguistics and sociopragmatics is also testified in pragmatic development
studies.
Knowledge of a grammatical structure and its pragmalinguistic functions, accompanied by lack of familiarity with the sociopragmatic conditions of target-like
use, is also reported in the relevant literature. Bardovi-Harlig and Hartford (1991)
show that, in a student-advising session at an American University, a Japanese student wishing to reject the faculty advisors suggestion uses a series of questions as
a strategy of indirectly conveying negative intent. In the researchers analysis, the
student clearly had the grammatical knowledge to form questions and the pragmalinguistic knowledge to use questions in order to indirectly convey intent, but did
not have the sociopragmatic knowledge to assess when the particular strategy is
appropriate and effective. It is also reported that this question strategy was not observed in the rejections of American-English speaking students. The particular
data fall short of providing evidence for the appropriate use and effectiveness of
the question strategy in conveying intent by L2 learners or by native speakers of
the L2. For example, they do not show whether American-English speaking students employ the question strategy to indicate negative intent in other social settings, or to what extent this can happen. If it turned out that the particular strategy
to convey negative intent is not favoured by native speakers of the L2, then it is
possible that the Japanese student has erroneously associated a grammatical form
with a pragmatic strategy, which is evidence for insufficient development of her
pragmalinguistic rather than her sociopragmatic knowledge. Moreover, even if the
question strategy was associated with the same function by American speakers, it
is not clear whether the understanding of negative intent would result from the operation of sociopragmatic factors or from on-line inferencing heuristics, i.e. pragmalinguistic factors. As previously noted with respect to pragmatic failure and
pragmatic transfer, it is sometimes difficult to diagnose the source of a learners inappropriate use of a speech act strategy at any particular stage of L2 development.

92
4.5.

Sophia Marmaridou

Second language teaching and testing

The most widely stated aim of L2 teaching is to enable successful communication


in the foreign language. This involves not only the mastery of the phonology, syntax and semantics of the L2, but also its pragmatics. The question that is typically
addressed in this context is whether pragmatic knowledge can be taught in the
classroom, i.e. not necessarily by cultural immersion in the native environment,
and how it can be taught to learners of different ages, educational backgrounds, etc.
The applications of the pragmalinguistics/sociopragmatics distinction in second
language acquisition research and the consequent refinement of the concepts of
pragmatic failure and pragmatic transfer relate in an obvious way to second language teaching and testing. If this distinction is useful in, perhaps selectively, identifying causes of pragmatic failure and areas of pragmatic transfer, then L2 teaching and testing should be oriented towards the development of pragmalinguistic
and sociopragmatic appropriateness (Harlow 1990: 328).
Following Leech (1983), Kasper and Rose (2001) claim that communicative
action includes not only using speech acts (such as apologising, complaining,
complimenting, and requesting), but also engaging in different types of discourse
and participating in speech events of varying length and complexity. More importantly, speakers do not only want to perform speech acts in order to achieve
their goals, but also wish to attend to interpersonal relationships with the other
speech participants in the communicative event. Within this context, pragmalinguistics refers to the resources for conveying communicative acts and encoding interpersonal meanings. According to Kasper and Rose (2001: 2), these resources include pragmatic strategies such as directness and indirectness, speech act routines,
and the intensification or softening of communicative acts. The social perceptions
underlying participants interpretation and performance of communicative action
fall within the area of sociopragmatics and refer to speakers and hearers social
distance and social power, their respective rights and obligations, and the degree of
imposition involved in particular communicative acts.
It transpires from the above that in language teaching research, as in second
language acquisition research, pragmalinguistics and sociopragmatics focus on the
area of speech acts and, more specifically, on their linguistic realisation and their
strategic use in particular social circumstances. In this view, teaching L2 pragmatics primarily involves teaching the means and strategies of performing particular speech acts, i.e. the pragmalinguistics of L2 (see McNamara and Roever 2006),
and teaching politeness, i.e. teaching how to assess social distance between interlocutors, social power and degrees of imposition in the L2 sociocultural setting
(see Bou-Franch and Garcs-Conejos 2003). Clearly, to be pragmatically competent in L2, learners need knowledge both at the pragmalinguistic and the sociopragmatic levels. For example, learners need to know not only how to express a
particular speech act indirectly, but also the relation between indirectness and po-

Pragmalinguistics and sociopragmatics

93

liteness, as well as when and to whom to be polite, the degree of politeness that is
required in a particular context of interaction, etc.
Apart from the linguistic expressions themselves, strategies of communicative
actions may vary not only according to context within a particular language, but
also across languages and cultures. However, as Kasper and Rose (2001) observe,
adult learners get a considerable amount of L2 pragmatic knowledge for free, since
some pragmatic knowledge is universal, while other aspects of such knowledge
may be successfully transferred from the learners L1. At the pragmalinguistic
level, L2 teaching is facilitated by the cross-cultural availability of the major categories of speech acts (e.g. Searles (1979) directives, commissives, expressives,
etc.), and specific acts such as greetings, leave-takings, etc. Moreover, it has been
found that the major realisation strategies of some speech acts occur across several
ethnolinguistically distant speech communities (see e.g. House and Kasper 1987;
Faerch and Kasper 1989; Rose 2000, among others). It is shown that in early learning states, learners may not have acquired the linguistic means necessary to implement such strategies, but once linguistic obstacles are removed by teaching, they
will use the appropriate strategies without instruction. Similarly, corresponding
form-function pairings between L1 and L2 used in the same contexts facilitate
pragmalinguistic performance.
Positive transfer and awareness of universally existing social variables, such as
social distance, social power and degree of imposition, often affected by other social parameters such as sex, age, and familiarity between interlocutors (Harlow
1990), can also facilitate the learners task in acquiring sociopragmatic knowledge
in L2. However, pragmalinguistic and sociopragmatic conventions make reference
to the lexical and grammatical constructions of particular languages, while successfully performing communicative acts often depends on the degree to which
a strategy is conventionalised in a particular language and culture. These parameters, then, point to the necessity of classroom instruction in L2 pragmalinguistics and sociopragmatics. Especially concerning the latter, it has been argued
that the promotion of sociopragmatic awareness is also likely to improve production and comprehension of pragmalinguistic resources (Bou-Franch and GarcesConejos 2003).
The apparent usefulness of the pragmalinguistics/sociopragmatics distinction
in the study of L2 teaching is not proportionately reflected in L2 testing research.
As in the discussion of pragmatic failure above, it is often difficult in practice to determine whether a given error is due to pragmalinguistic or sociopragmatic deficits
(McNamara and Roever 2006: 55). For example, if politeness markers are missing
from an utterance, this could be either because the learner does not know these
markers, or because she is not aware that these markers should be used in the particular situation. The former indicates pragmalinguistic, whereas the latter sociopragmatic, inadequacy. Therefore, it is difficult to devise a test that would assess
pragmalinguistic knowledge to the exclusion of sociopragmatic knowledge, or the

94

Sophia Marmaridou

reverse. The problem seems more acute when testing pragmatic strategies, such as
indirectness. While the linguistic expressions used to encode indirectness belong
to the pragmalinguistic level, the social target to which they are addressed is a sociopragmatic matter. Moreover, sociopragmatic appropriateness seems to be less
dependent on linguistic proficiency than pragmalinguistic knowledge.3 L2 learners
with a high general L2 competence are likely to build their pragmalinguistic competence more easily, but high general L2 competence does not guarantee correct
judgements about sociopragmatic appropriateness.
The problem of testing pragmalinguistics and sociopragmatics independently
from each other also relates to the types of tests devised to test pragmatic knowledge in general, as there is a significant tension between the construction of authentic assessment tasks and practicality (McNamara and Roever 2006: 54). The
two most frequent types are discourse completion tests (DCTs) and role plays. In
attempting to test sociopragmatic knowledge, multiple choice DCTs were shown
not to be reliable, as it is extremely difficult to devise distractors (i.e. wrong
items) that are totally unacceptable by all members of the target language community without being wrong in a very obvious way (Brown 2001). Open DCTs can
only provide evidence of the L2 learners sociopragmatic ability concerning the
particular items in the test (e.g. apologies, requests, formulaic implicatures, etc.)
rather than her general sociopragmatic ability. Additionally, they do not provide
evidence of the learners ability to participate in conversations in real time, or to
take turns, especially given that speakers normally distribute strategies over various turns, monitoring their interlocutors reactions and adjusting their talk accordingly. Although role plays are more reliable in this respect, they are not preferred as
testing devices because they take substantial time to conduct, they are difficult to
monitor, and they require multiple rating. Besides, role play cannot establish context in the way that authentic communication does in the real world, nor does it represent the speakers or the addressees actual face needs in real terms. However, it
is reported that role plays can assess learners ability to make judgments about sociopragmatic appropriateness at discourse level (Cook 2001).
By contrast, it is possible to test L2 pragmalinguistic knowledge practically
and reliably, even though creating items for pragmalinguistic instruments is not
easy. Roever (2006), for example, constructed a test to assess interpretation of implicature, recognition of routine formulae, and knowledge of speech act strategies.
While these three types of knowledge differ, they are based on general pragmatic
and linguistic input. Interpretation of implicature was expectedly found to rely on
the learners proficiency and deduction abilities. The effect of L2 learners proficiency in their knowledge of routines was almost negligible, given that routines are
not creatively constructed, but rather learned as holistic items to be used in specific
situations. As opposed to routines, the production of speech acts and the pragmalinguistic strategies employed therein correlated with level of L2 proficiency, as
expected. Some of the limitations of assessing L2 learners pragmalinguistic

Pragmalinguistics and sociopragmatics

95

knowledge are similar to the ones of testing sociopragmatics. The written format of
the tests restricts the number of contextual cues that would be available in a real encounter. The lack of extended negotiation of speech acts in the context of a DCT is
another limiting factor. However, studies of pragmalinguistic assessment in L2 can
benefit both from the development of testing techniques in general and from
further cross-cultural study of pragmalinguistic parameters of communicative acts
in particular.
It becomes obvious from all the above that the pragmalinguistics/sociopragmatics distinction is better exploited in the area of L2 teaching methodology than
in studies of pragmatic failure, pragmatic transfer, pragmatic development and L2
testing. What the latter have in common is their diagnostic methodology in assessing L2 learners approximations to the target language. The sources of particular
approximations are not easy to detect. By contrast, a theoretical categorisation into
pragmalinguistic and sociopragmatic parameters of language use can fairly easily
be mapped onto teaching methods, materials and awareness raising tasks in which
L2 learners may be involved.

5.

Historical sociopragmatics4

The concepts of pragmalinguistics and sociopragmatics have also informed the


field of study generally known as Historical Pragmatics. Still a relatively young
field of linguistics, historical pragmatics takes a broad, sociologically based view
of patterns of intentional human interaction, as these are determined by the conditions of societies of earlier periods (see Jucker 2006 for a brief overview).5 One approach to historical pragmatics, pragmaphilology, describes the contextual aspects
of historical texts and is essentially synchronic. It refers to the identities of speakers and addressees, their social and personal relationship, the physical and social
setting of text production and reception and the goals of the text (Jacobs and Jucker
1995). Another approach, diachronic pragmatics, concerns form-to-function and
function-to-form mappings across different historical stages of the same language.
Apparently, form-to-function mappings correspond to a pragmalinguistic perspective on the analysis of phenomena such as deixis (Fries 1993), discourse markers
(Brinton 1990), or interjections (Taavitsainen 1995). Diachronic function-to-form
studies, adopting a sociopragmatic perspective, compare the realisation of speech
acts, politeness formulae, text types, or forms of dialogue, at different points in the
development of a language (Arnovick 1999; Jucker and Taavitsainen 2008; Fritz
1995; Biber and Finegan 1992). Since the mid-1990s the investigation of the relevant topics is based on available historical corpora (e.g. the Helsinki Corpus of
English Texts), which are either read and annotated in an old-fashioned and time
consuming way, or analysed by corpus-linguistic methods of data retrieval. Apart
from written genres, such as essays, fiction, personal letters, and manuals for good

96

Sophia Marmaridou

behaviour, speech-based genres have also been analysed, e.g. dialogues from plays
and fiction, court proceedings, political debates, public speeches, and sermons
(Biber and Finegan 1992; Jacobs and Jucker 1995).
Within this framework, historical sociopragmatics focuses on the interaction
between specific aspects of social context and particular historical language use
that leads to pragmatic meanings. More specifically, it concerns language use in its
situational, local context and the ways in which situational contexts generate
norms which interlocutors employ or exploit for pragmatic purposes. A synchronic
approach to historical sociopragmatics consists in showing how language use
shapes and is shaped by context at a particular historical period, while a diachronic
perspective involves the investigation of how shifts in language use affect shifts in
contexts, or how shifts in contexts shape language use over time. An important line
of investigation in historical sociopragmatics concerns the reconstruction of contexts on the basis of historical texts; the latter are viewed as carrying evidence of,
or projecting, their own contexts (Culpeper 2009: 182183).
As already mentioned, the local context, or the sociological context, of language use is of primary concern to sociopragmatics. For example, Nevalainen and
Raumolin-Brunberg (1995) have studied the sociopragmatics of terms of address
in Early English correspondence. Given that the way people address each other in
interaction depends on social variables such as age, status, dialect, etc., the question arises as to the relation between sociopragmatics and sociolinguistics. According to Nevalainen and Raumolin-Brunberg (2003), sociolinguistics comprises
three areas of research, namely, social dialectology, interactional sociolinguistics
and the sociology of language. Parallel to this distinction, Culpeper (2009: 180)
identifies three levels of what may constitute the context against which texts may
be understood: the most local, immediate text and co-text of interlocutors, the
medial level of social situation (including speech events, activity types, frames,
etc.), and the most general level, making reference to national and regional cultures, institutional cultures, etc. As the author notes (Culpeper 2009: 181)
sociopragmatics should primarily, though not exclusively, concern itself with the medial
context and the phenomena that constitute it. Social situations can provide a link between micro, more linguistically-oriented considerations (the typical focus of pragmalinguistics), and macro, more sociologically-oriented considerations (the typical focus
of a field such as Critical Discourse Analysis).6

According to this definition, sociolinguistics is concerned with mapping regular


patterns of usage in interaction, while sociopragmatics is concerned with the ways
in which these regular patterns are used and exploited in particular interactions.
Given the concern of historical sociopragmatics for interactional data and the
obvious non-availability of recorded forms of interaction of earlier periods, the
question arises as to the availability of historical corpora that lend themselves to
sociopragmatic research and the search tools required for this type of investigation.

Pragmalinguistics and sociopragmatics

97

Recent developments in historical corpus linguistics have allowed more pragmatic


research questions in a way that is amenable to corpus-linguistic methodologies
(see, e.g., Kohnen 2008; Jucker et al. 2008); but the need is still felt for the appropriate pragmatic tagging of relevant pragmatic units, such as those of interest to sociopragmatic research. Within this framework, Culpeper and Kyts compilation
of the Sociopragmatic Corpus, a subsection of the Corpus of English Dialogues:
15601760, has made sociopragmatic annotation possible. Facing the issue of the
sociopragmatic variability of context, Archer and Culpeper (2003) devise and implement a tagging system that allows the annotation of speech changes that potentially affect the social meaning of interaction in drama and trial proceedings.7 The
variables include relatively static, sociolinguistic values, such as status (nobility,
gentry, professional, ordinary commoners, lowest groups) and age (young, adult,
old), as well as dynamic values, such as activity role (witness, defendant, customer), kinship role (father, daughter, mother, son, father-in-law), social role (surgeon, baker, friend), and dramatic role (fool, villain, seducer, etc.). The significance of this type of work lies in capturing the utterance-by-utterance interaction
between speakers and their addressees in terms of sociopragmatic variables, which
enhance the dynamic aspects of the analysed texts.
Working with the same corpus, Archer and Culpeper (2009) develop the notion
of keyness as the identification of keywords, key parts-of-speech and key semantic
fields that are statistically characteristic of the speech of dyads in interaction, e.g.
master/mistress with servant and examiner with examinee in trials. Examples include the use of imperative verbs directed to servants by their masters in the partsof-speech category and the use of the semantic domain of documents and writing in
the speech of the same dyad. These social role dyads are a specific part of particular
social situations, the latter constituting the local context relative to all the social
roles and situations that make up the two genres under examination, i.e. drama and
trial proceedings. Therefore, keyness analysis is a tool for the identification of the
characteristic discourse norms of particular local contexts. While the authors use
the term sociophilology to refer to their approach, their concern for the particular
type of annotation and the analysis of the local conditions of language use indicates
the deployment of a sociopragmatic orientation to their work.
The concept of sociopragmatics has also informed further studies focusing on
the social conditions that affect the use of texts. Adopting the model of critical discourse analysis and frame analysis, Wood (2009) examines personal letters from
late 15th century English taken from the two volumes of the Paston Collection, and
Margaret Pastons letters in particular. Wood (2009: 188) addresses the issue of
authorship, which is important in historical sociopragmatics: since some of Margarets letters were penned by her own sons, might we not be justified in ascribing
the language of those letters to them, rather than representative of Margaret Paston
herself? Supporting the view that local contexts are culture specific, she shows
that local context does not only make reference to a particular interactional dyad,

98

Sophia Marmaridou

but also to a more extended speech community, to the discursive practices of that
community, the texts distribution and its consumption. Within this framework, she
shows that on the one hand these letters evidence the manipulation of power structures within a family and, on the other, they indicate Margaret Pastons control
over the form and content of the letters, thereby confirming authorship.
Viewed broadly, historical sociopragmatics includes the macro-level of social,
socio-cultural and sociological factors as well as the micro-level of personal, situational and stylistic factors. Nevala (2009) discusses the concept of person reference and social deixis in Late Modern English letters and journals. Her data consist
of the correspondence of some members of the Lunar Society of Birmingham and
the letters and journals of Agnes Porter, an eighteenth century governess. In particular, she concentrates on the interactional aspects of the referential term friend
as used by public figures and ordinary people, as well as on the way in which selfand addressee-oriented third person reference is used to convey contextual shifts in
interpersonal distance and authority. She demonstrates that the writers in the data
strategically use different terms in order to alternate between specific social positionings, but they also take into consideration the prevalent social and societal constraints.
Examining the letters subcorpus of the Network of Eighteenth-century English
Texts within a relevance theoretic perspective, Fitzmaurice (2009) focuses on the
sociopragmatic construction of implicature and inference in the illicit courtship
correspondence between Edward Wortley and Mary Pierrepont. She argues that
key historical and cultural reference points are necessary in understanding how
communicative practices are embedded in the local context. Such practices are
typically associated with certain activity types and social rituals including courting. Therefore, the understanding of implicated meanings in these letters makes
reference to the cotext created by the discourse of the letter, the situational context
constructed by the exchange of letters, and the broader historical context and the
social constraints in which the correspondence is embedded. Fitzmaurice (2009)
shows that it is at the discourse cotext that the participants in the interaction enact
their sociopragmatic roles and fight for dominance of their relationship.
It transpires from all the above that the concepts of pragmalinguistics and sociopragmatics have been particularly useful in historical pragmatics and enhance
the possibilities of historical corpus linguistics. Also notice that, within the framework of historical sociopragmatics, pragmalinguistic concerns are also accommodated, as, for example, the study of the forms used to perform a particular function
in Nevalas (2009) work. On this understanding, sociopragmatics is broadly construed as referring to both the micro and macro levels of pragmatic analysis. It is
possible that the success in the application of the two concepts under investigation
to historical pragmatics depends partly on the availability of appropriate corpora
and partly on appropriate retrieval methods that facilitate the analysis.

Pragmalinguistics and sociopragmatics

6.

99

Concluding remarks

As already shown, apart from Leechs (1983) and Thomass (1983) contrast and
distinction between pragmalinguistics and sociopragmatics, these terms have been
variously construed in the relevant literature and, used independently of each other,
they have emphasised the social conditions in which utterances are produced
and understood as verbal contributions in acts of communication. Regardless of
whether the pragmalinguistics/sociopragmatics distinction is theoretically justified, analytically possible, or ideologically desirable, it has motivated the corresponding methodological distinction in L2 acquisition research and L2 teaching
and testing in particular. The concepts of pragmatic failure, pragmatic transfer and
pragmatic development, relevant in these fields of research, explicitly make reference to this distinction. The research questions in the pragmatics of L2 teaching
and testing have to do with defining L2 learners verbal approximations to L2 as
detected in their performance and then attributing them to pragmalinguistic or sociopragmatic sources. The focus is on whether L2 learners utterances are appropriate communicative contributions in the social context in which they occur and
whether their inferencing is based on an understanding of social context. Appropriateness is an essentially evaluative term (see Fetzer 2004; 2007). In the framework of L2 acquisition research it refers to the learners speech actions, which are
constantly assessed as being in accordance with the linguistic norms and contextual constraints and requirements of particular communicative acts in the target language and culture. The appropriateness of L2 learners performance is not always
easy to evaluate on the basis of the pragmalinguistics/sociopragmatics distinction.
More specifically, the absence of sharp boundaries between these two concepts
weakens their methodological usefulness in diagnosing the source of pragmatic
failure, pragmatic transfer, or the stage of L2 learners pragmatic development, especially in the context of L2 testing, as already mentioned. It seems easier to maintain the pragmalinguistics/sociopragmatics distinction in the context of L2 teaching, in the sense that the methodology adopted, the materials used, and the tasks in
which learners are engaged can be designed so as to focus on this distinction and
raise the L2 learners awareness of the linguistic and social issues involved in their
use of L2.
In the area of historical pragmatics and historical corpus linguistics the situation is different. The texts examined belong either to the same period (as in synchronic studies) or to different periods (as in diachronic studies).8 Pragmalinguistic
diachronic studies proceed from form to function, whereas sociopragmatic diachronic studies follow the function-to-form process. Historical sociopragmatics
focuses on the interaction of context and language use either synchronically or
diachronically. In the context of historical pragmatics and historical corpus linguistics the distinction between pragmalinguistics and sociopragmatics can be
fruitfully maintained in spite of the lack of sharp boundaries between the two. His-

100

Sophia Marmaridou

torical texts are viewed as definitive and complete instantiations of a complex configuration of parameters, such as coparticipants and their social, interactional and
discursive roles, communicative action, genre, and cultural norms and strategies of
a speech community. Unlike historical pragmatics research, L2 acquisition research is based on texts as approximations of the target language, which makes it
extremely difficult to diagnose just which specific parameter is at issue in evaluating appropriateness.
On the basis of the above, it is fair to say that, even though the pragmalinguistics/sociopragmatics distinction was initially launched as a level of cross-linguistic application of the general pragmatic principles of cooperation and politeness (as in Leech 1983), and even though it was systematically taken up in the
context of L2 acquisition research, it seems to have been more relevant in the area
of historical pragmatics and historical corpus linguistics as a reliable methodological tool in the investigation of the pragmatics of historical texts. Needless to say,
further theoretical justification of the concepts of pragmalinguistics and sociopragmatics, and of the distinction between them, ultimately depends on future research
in pragmatics and related areas of linguistic enquiry.

Notes
1. It is interesting to note here that Olsens views remind one of Wittgensteins (1958) position that linguistic meaning resides in language use and Wierzbickas (1992) introduction
of a Natural Semantic Metalanguage to deal with the cultural relativity of word-meaning
pairings across languages. Evidently, the dependence of linguistic meaning on use in particular socio-cultural settings is a recurring theme in linguistic theory.
2. In fact, interlanguage, as an interim system of L2 learners, has some features of the L1
and some of the L2, but also features that are independent of their L1 and L2 (see Yule
2006: 167).
3. This is compatible with the view that sociopragmatic awareness is likely to improve pragmalinguistic performance.
4. See also Jucker A. H. and I. Taavitsainen (eds.) 2010. Historical Pragmatics, Vol. 8 of the
series Handbooks of Pragmatics (Berlin: Mouton de Gruyter).
5. Within the narrow, Anglo-American tradition to pragmatics, Traugott (2004: 538) defines
historical pragmatics as a usage-based, pragmatically motivated, approach to language
change.
6. Critical discourse analysis is fundamentally concerned with analysing opaque as well as
transparent structural relationships of dominance, discrimination, power and control as
manifested in language (Wodak 2001: 2).
7. These two text types offer interactive, face-to-face, speech-related data that only approximate authentic discourse. Drama consists of imaginary constructed dialogue and trial
proceedings constitute the record of a prior speech event.
8. Relevant to this point is the distinction between diachronic pragmatics and pragmaphilology (Jacobs and Jucker 1995) and diachronic sociopragmatics and sociophilology
(Archer and Culpeper 2009).

Pragmalinguistics and sociopragmatics

101

Acknowledgements
I would like to thank the three reviewers for their helpful suggestions and comments. I am also grateful to Jonathan Culpeper, Marcelo Dascal, Andreas Jucker,
and Carsten Roever for making their work available to me. This paper has benefited from very useful stylistic suggestions made by Eleni Antonopoulou, to whom
I am greatly indebted.

References
Archer, Dawn and Jonathan Culpeper
2003
Sociopragmatic annotation: New directions and possibilities in historical corpus linguistics. In: Andrew Wilson, Paul Rayson, and Tony McEnery (eds.),
Corpus Linguistics by the Lune: A Festschrift for Geoffrey Leech, 3758.
Frankfurt am Main: Lang.
Archer, Dawn and Jonathan Culpeper
2009
Identifying key sociophilological usage in plays and trial proceedings (1640
1760). An empirical approach via corpus annotation. Journal of Historical
Pragmatics 10/2: 286309.
Arnovick, Leslie K.
1999
Diachronic Pragmatics. Seven Case Studies in English Illocutionary Development. Amsterdam: Benjamins.
Bang, Jrgen Chr. and Jrgen Dr
1979
Language, theory, and conditions for production. In: Jacob L. Mey (ed.), Pragmalinguistics. Theory and Practice, 2150. The Hague: Mouton.
Bardovi-Harlig, Kathleen
1999
Exploring the interlanguage of interlanguage pragmatics: A research agenda
for acquisitional pragmatics. Language Learning 49: 677713.
Bardovi-Harlig, Kathleen and Beverly Hartford
1991
Saying no in English: Native and nonnative rejections. In: Lawrence F. Bouton and Yamura Kachru (eds.), Pragmatics and Language Learning, 4157.
Urbana-Champaign, IL: Division of English as an International Language,
University of Illinois.
Bates, Elizabeth
1976
Language and Context. The Acquisition of Pragmatics. New York: Academic
Press.
Beebe, Leslie M. and Tomoko Takahashi
1989
Sociolinguistic variation in face-threatening speech acts. In: Miriam Eisenstein (ed.), The Dynamic Interlanguage, 199218. New York: Plenum.
Bebee, Leslie M., Tomoko Takahashi, and Robin Uliss-Weltz
1990
Pragmatic transfer in ESL refusals. In: Robin C. Scarcella, Elaine Andersen,
and Stephen C. Krashen (eds.), Developing Communicative Competence in a
Second Language, 5573. New York: Newbury House.
Biber, Douglas and Edward Finegan
1992
The linguistic evolution of five written and speech-based English genres from
the 17th to the 20th centuries. In: Matti Rissanen, Ossi Ihalainen, Terttu Neval-

102

Sophia Marmaridou

ainen, and Irma Taavitsainen (eds.), History of Englishes. New Methods and
Interpretations in Historical Linguistics, 5573. Berlin: Mouton de Gruyter.
Blum-Kulka, Shoshana
1991
Interlanguage pragmatics: The case of requests. In: Robert Phillipson, Eric
Kellerman, Larry Selinker, Michael Sharwood-Smith, and Merrill Swain
(eds.), Foreign/Second Language Pedagogy Research, 255272. Clevedon:
Multilingual Matters.
Bodman, Jean and Miriam Eisenstein
1988
May God increase your bounty: The expression of gratitude in English by
native and non-native speakers. Cross Currents 15: 121.
Bou-Franch, Patricia and Pilar Garcs-Conejos
2003
Teaching linguistic politeness: A methodological proposal. International Review of Applied Linguistics 41/1: 122.
Brinton, Laurel
1990
The development of discourse markers in English. In: Jacek Fisiak (ed.), Historical Linguistics and Philology, 4571. Berlin: Mouton de Gruyter.
Brown, James Dean
2001
Six types of pragmatics tests in two different contexts. In: Kenneth Rose and
Gabriele Kasper (eds.), Pragmatics in Language Teaching, 301325. New
York: Cambridge University Press.
Brown, Penelope and Stephen C. Levinson
1987
Politeness: Some Universals in Language Usage. New York: Cambridge University Press.
Brumfit, Chris and Keith Johnson
1979
The Communicative Approach to Language Teaching. Oxford: Oxford University Press.
Candlin, Chris N.
1976
Communicative language teaching and the debt to pragmatics. In: Clea Rameh
(ed.), 27th Round Table Meeting, 237256. Washington, DC: Georgetown University Press.
Cook, Haruko Minegishi
2001
Why cant learners of Japanese as a foreign language distinguish polite from
impolite speech styles? In: Kenneth R. Rose and Gabriele Kasper (eds.), Pragmatics in Language Teaching, 80102. New York: Cambridge University
Press.
Culpeper, Jonathan
2009
Historical sociopragmatics: An introduction. Journal of Historical Pragmatics
10/2: 179186.
Dascal, Marcelo
1985
Language use in jokes and dreams: sociopragmatics vs. psychopragmatics.
Language and Communication 5/2: 95106.
Dascal Marcelo and Edson Franozo
1989
The pragmatic turn in psycholinguistics: Problems and perspectives. Theoretical Linguistics 15/1,2: 123.
Eelen, Gino
2001
A Critique of Politeness Theories. Manchester: St Jerome.

Pragmalinguistics and sociopragmatics

103

Faerch, Claus and Gabriele Kasper


1989
Internal and external modification in interlanguage request realization. In:
Shoshana Blum-Kulka, Juliane House, and Gabriele Kasper (eds.), Cross-cultural Pragmatics, 221247. Norwood, NJ: Ablex.
Fetzer, Anita
2004
Recontextualizing Context: Grammaticality Meets Appropriateness. Amsterdam: Benjamins.
Fetzer, Anita
2007
If I may say so: Indexing appropriateness in dialogue. In: Anita Fetzer (ed.),
Context and Appropriateness, 115145. Amsterdam: Benjamins.
Fitzmaurice, Susan M.
2009
The sociopragmatics of a lovers spat. The case of the eighteenth-century
courtship letters of Mary Pierrepont and Edward Wortley. Journal of Historical Pragmatics 10/2: 215237.
Freud, Sigmund
1900
The Interpretation of Dreams. London: Harvester Press, Hassocks.
Freud, Sigmund
1905
Jokes and their Relation to the Unconscious. London: The Hogarth Press.
Fries, Udo
1993
Towards a description of text deixis in Old English. In: Klaus R. Grinda and
Klaus-Dieter Wetzel (eds.), Anglo-Saxonica. Festschrift fr Hans Schabram
zum 65. Geburtstag, 527540. Mnchen: Fink.
Fritz, Gerd
1995
Topics in the history of dialogue forms. In: Andreas H. Jucker (ed.), Historical
Pragmatics. Pragmatic Developments in the History of English, 469498.
(Pragmatics and Beyond New Series 35). Amsterdam: Benjamins.
Grice, Paul
1975
Logic and conversation. In: Peter Cole and Jerry Morgan (eds.), Syntax and
Semantics, Volume 3: Speech Acts, 4158. New York: Academic Press.
Harlow, Linda L.
1990
Do they mean what they say? Sociopragmatic competence and second language learners. The Modern Language Journal 74: 328351.
Haverkate, Henk
1994
The dialogues of Don Quixote de la Mancha: A pragmalinguistic analysis
within the framework of Gricean maxims, speech act theory, and politeness
theory. Poetics 22: 219241.
Holmes, Janet
2008
An Introduction to Sociolinguistics. 3rd ed. London/New York: Longman.
House, Juliane and Gabriele Kasper
1987
Interlanguage pragmatics: Requesting in a foreign language. In: Wolfgang
Lrscher and Rainer Schulze (eds.), Perspectives on Language in Performance, 12501288. Tbingen: Narr.
Jacobs, Andreas and Andreas H. Jucker
1995
The historical perspective in pragmatics. In: Andreas H. Jucker (ed.), Historical Pragmatics. Pragmatic Developments in the History of English. (Pragmatics and Beyond New Series 35), 333. Amsterdam: Benjamins.

104

Sophia Marmaridou

Jucker, Andreas H.
2006
Historical pragmatics. In: Jan-Ola stman and Jef Verschueren (in collaboration with Eline Versluys) (eds.), Handbook of Pragmatics 2006, 114. Amsterdam: Benjamins.
Jucker, Andreas H.
2008
Historical pragmatics. Language and Linguistics Compass 2/5: 894906.
Jucker, Andreas H. and Irma Taavitsainen (eds.)
2008
Speech Acts in the History of English. Amsterdam: Benjamins.
Jucker, Andreas H. and Irma Taavitsainen (eds.)
2010
Historical Pragmatics. Handbooks of Pragmatics, Vol. 8. Berlin: De Gruyter.
Jucker, Andreas H., Gerold Schneider, Irma Taavitsainen, and Barb Breustedt
2008
Fishing for compliments: Precision and recall in corpus-linguistic compliment
research. In: Andreas H. Jucker and Irma Taavitsainen (eds.),
2008
Speech Acts in the History of English, 273274. Amsterdam: Benjamins.
Kasper, Gabriele
1992
Pragmatic transfer. Second Language Research 8/3: 203231.
Kasper, Gabriele and Kenneth R. Rose
2001
Pragmatics in Language Teaching. In: Kenneth R. Rose and Gabriele Kasper
(eds.), Pragmatics in Language Teaching, 110. New York: Cambridge University Press.
Kasper, Gabriele and Kenneth R. Rose
2002
Pragmatic Development in a Second Language. Oxford: Blackwell.
Keenan, Edward L.
1976
The universality of conversational postulates. Language in Society 5: 6780.
Kohnen, Thomas
2008
Tracing directives through text and time: Towards a methodology of a corpusbased diachronic speech-act analysis. In: Andreas H. Jucker and Irma Taavitsainen (eds.),
2008
Speech Acts in the History of English, 295310. Amsterdam: Benjamins.
Leech, Geoffrey N.
1983
Principles of Pragmatics. London/New York: Longman.
Levinson, Stephen C.
1983
Pragmatics. Cambridge: Cambridge University Press.
Marmaridou, Sophia S. A.
2000
Pragmatic Meaning and Cognition. Amsterdam: Benjamins.
McNamara, Tim and Carsten Roever
2006
The social dimension of proficiency: How testable is it? Language Learning
56/2: 4379.
Mey, Jacob L.
1979
Introduction. In: Jacob L. Mey (ed.), Pragmalinguistics. Theory and Practice,
1017. The Hague: Mouton.
Mey, Jacob L.
1993
Pragmatics. An Introduction. Oxford: Blackwell.
Morris, Charles W.
1938
Foundations of the theory of signs. In: Otto Neurath, Rudolf Carnap and
Charles W. Morris (eds.), International Encyclopaedia of Unified Science. Volume 1, 77138.

Pragmalinguistics and sociopragmatics

105

Nevala, Minna
2009
Altering distance and defining authority. Person reference in Late Modern
English. Journal of Historical Pragmatics 10/2: 238259.
Nevalainen, Terttu and Helena Raumolin-Brunberg
1995
Constraints on politeness: The pragmatics of address formulae in Early English correspondence. In: Andreas H. Jucker (ed.), Historical Pragmatics. Pragmatic Developments in the History of English, 5776. Amsterdam: Benjamins.
Nevalainen, Terttu and Helena Raumolin-Brunberg
2003
Historical Sociolinguistics. Harlow: Pearson.
Olsen, Svend Erik
1979
Psychopathology, interaction, and pragmatic linguistics. In: Jacob L. Mey
(ed.), Pragmalinguistics. Theory and Practice, 233253. The Hague: Mouton.
Olshtain, Elite and Liora Weinback
1992
Interlanguage features of the speech act of complaining. In: Gabriele Kasper
and Shoshana Blum-Kulka (eds.), Interlanguage Pragmatics, 195208. New
York: Oxford University Press.
Prucha, Jan
1983
Pragmalinguistics: East European Approaches. Amsterdam: Benjamins.
Riley, Philip
1979
Towards a contrastive pragmalinguistics. Papers and Studies in Contrastive
Linguistics 10: 5778.
Robinson, Mary Ann
1992
Introspective methodology in interlanguage pragmatics research. In: Gabriele
Kasper (ed.), Pragmatics of Japanese as native and target language Technical
report no. 3: 2782. Honolulu: University of HawaiI at Manoa, Second Language Teaching and Curriculum Centre.
Roever, Carsten
2006
Validation of a web-based test of ESL pragmalinguistics. Language Testing
23/2: 229256.
Rosaldo, Michelle Z.
1982
The things we do with words: Ilongot speech acts and speech act theory in philosophy. Language in Society 11: 203237.
Rose, Kenneth R.
2000
An exploratory cross-sectional study of interlanguage pragmatic development.
Studies in Second Language Acquisition 22: 2767.
Searle, John R.
1979
Expression and Meaning: Studies in the Theory of Speech Acts. Cambridge:
Cambridge University Press.
Selinker, Larry
1972
Interlanguage. International Review of Applied Linguistics 10: 209230.
Sifianou Maria
1992
Politeness Phenomena in England and Greece. A Cross-Cultural Perspective.
Oxford: Clarendon Press.
Taavitsainen, Irma
1995
Interjections in Early Modern English: From imitation of spoken to conventions of written language. In: Andreas H. Jucker (ed.), Historical Pragmatics.
Pragmatic Developments in the History of English, 439465. Amsterdam:
Benjamins.

106

Sophia Marmaridou

Takahashi, Satomi
1996
Pragmatic transferability. Studies in Second Language Acquisition 18: 189
223.
Tahahashi, Satomi
2001
The role of input enhancement in developing pragmatic competence. In: Kenneth R. Rose and Gabriele Kasper (eds.), Pragmatics in Language Teaching,
171199. New York: Cambridge University Press.
Thomas, Jenny
1981
Pragmatic failure. M.A. dissertation, Department of Language and Linguistics, University of Lancaster.
Thomas, Jenny
1983
Cross-cultural pragmatic failure. Applied Linguistics 4/2: 91112.
Traugott, Elizabeth Closs
2004
Historical Pragmentics. In: Lawrence R. Horn and Gregory Ward (eds.), Handbook of Pragmatics, 530561. Oxford: Blackwell.
Trosborg, Anita
1987
Apology strategies in natives/non-natives. Journal of Pragmatics 11: 147167.
Widdowson, Henry G.
1979
Explorations in Applied Linguistics. Oxford: Oxford University Press.
Wierzbicka, Anna
1987
English Speech Act Verbs: A Semantic Dictionary. Sydney/New York: Academic Press.
Wierzbicka, Anna
1992
Semantics, Culture, and Cognition. Universal Human Concepts in CultureSpecific Configurations. New York/Oxford: Oxford University Press.
Wittgenstein, Ludwig
1958
Philosophical Investigations. Oxford: Blackwell.
Wodak, Ruth
2001
What CDA is about. A summary of its history, important concepts and its
developments. In: Ruth Wodak and Michael Meyer (eds.), Methods of Critical
Discourse Analysis, 113. London: Sage Publications.
Wolfson, Nessa
1981
Compliments in cross-cultural perspective. TESOL Quarterly 15: 117124.
Wood, Johanna L.
2009
Structures and expectations. A systematic analysis of Margaret Pastons formulaic and expressive language. Journal of Historical Pragmatics 10/2: 187214.
Yule, George
2006
The Study of Language. 3rd ed. Cambridge: Cambridge University Press.

4.

Metapragmatics
Axel Hbler

Introduction
The term metapragmatics is slippery, many meanings have been ascribed to it,
too many perhaps. A restrictive or selective attitude, however, does not seem appropriate for a handbook article. On the contrary, the survey should accommodate
as many perspectives as are possible and justifiable. The problems mainly arise
from two interrelated facts. (1) The Greek loan prefix meta- has different meanings
(in between, after, later or beyond, among others) and thus lends itself to forming
all sorts of terms. (2) The formative meta- is popular. Wordspy, the internet window
administered by Random, lists 75 new meta-formations.1 Its attraction presumably derives from its particular air of something intellectual, detached and with an
impressively broad view.
To reduce the risk of a sell-out of this formative device, it may be prudent to
control its deployment and avoid it where it is not necessary. A case in point is the
use of the term metapragmatics, where the term pragmatics actually suffices. If we
define pragmatics as the discipline which has interpersonal communication as its
object, then it goes without saying that all abstractions and theoretical conclusions
which help explain communicative behavior form part of this discipline, in fact
they are defining features of the discipline. This holds true for all the answers given
to questions such as how to cooperate, how to be kind, polite, etc. and all the
conditions found to explain aspects of interpersonal communication, i.e., the conditions of thinkability (e.g., the constitutive rules of Searles speech acts), the
conditions of feasibility (e.g., the Gricean maxims) and the conditions of recognizability (e.g., the linguistic and paralinguistic expressive means) (cf. Caffi 2006:
84). To separate such theoretical stances or constructs from the discipline that developed them seems to make little sense. Communication can be act and object,
as Hagemann (1997: 32) subtly observes, pragmatics cannot. It is certainly true
that communication shows properties that can be labeled pragmatic, but such
labeling originates in an external view. This then is a reading of metapragmatics
that deserves mentioning,2 but will not be part of the focus of the article; the theoretical views and notions as such, however, will, of course, be used where necessary.
What remains to be unpacked are the following four readings of the term and
the four corresponding forms of linguistic practice: metapragmatics as the study of
explicit metacommunication (section 1); metapragmatics as the study of implicit
metacommunication (section 2); metapragmatics as the study of peoples abstract-

108

Axel Hbler

ing from interacting (section 3.1); and metapragmatics as metatheory of pragmatics (section 3.2).

1.

Metapragmatics as the (pragmatic) study of explicit


metacommunication

The object of a (meta)pragmatic analysis is metacommunication. In a first go, it


can be defined as communication about (selected aspects of) communication.
Usually, the communication talked about is neither communication as such (the
abstract and general possibilities and impossibilities of communication) nor some
definite previous or future communication, but the ongoing communication. Paul
(1999: 98) characterizes it immanently as practical speech reflection. Quite original is Antons view contrasting it with communication.
[W]e routinely and commonly attend not to speech but simply from it. That is, when we
listen with [sic!] one another in our everyday encounters, we routinely listen from our
speech to the thought so intended. (Anton 1998: 198199)

This behavior contrasts with what happens in particular circumstances, where


we explicitly attend to the speech itself. In Leders [1990] terminology, moments of
breakdown manifest speech dys-appearance.
Reconsider the case of reading a novel: I may be engrossed in the story, simply attending from the sentences to the story, but then, a misspelled word, a foreign word, or perhaps a too-recondite one appears. Now, I consciously and explicitly reflect to the word
[]. Speech, then, is routinely an absent body, an intentional arc which disappears for
the sake of the meaning so intended. (Anton 1998: 199)

What Antons circumscription brings out nicely is a kind of break separating communication from metacommunication, a sort of strained dialogical structure, which
is not attributable to a change in speakers (or voices), but to a change in the level of
communication. This is mainly brought about by a functional change from practical acting to reflecting.3
The change in level is often notable in that it involves a change from ordinary
language to metalanguage. Metalanguage is commonly understood as the language
referring to language; it represents what Verschueren (1998: 55) calls the object
notion of metalanguage.4 It is, however, not an autonomous language (not even
where linguists use a metalanguage). When members of a speech community engage in metacommunication, the metalanguage used does not consist of complicated terminology but of ordinary words, mainly nouns (such as chat or denial) and
verbs (such as promise or argue), whose feature that distinguishes them from other
words is their function to refer not to something in the world out there but to some
aspect of speech.5

Metapragmatics

1.1.

109

On full forms of metacommunication

Though metacommunication is always embedded in primary communication, its


extension can vary considerably. For the purpose of this article, it will be sufficient
to simply differentiate between full and reduced (or abbreviated) metacommunicative forms. A full form consists of at least one metacommunicative utterance;
anything below utterance level can no longer be considered metacommunication
proper, but is, of course, still metacommunicative.6 If primary communication is
the on-line event, then metacommunicative utterances (and more extended forms)
cause such an articulate deviation from it that they appear as off-line, as it were
(cf. Hbler and Bublitz 2007: 12). The boundary between primary communication
and metacommunication is usually not marked in strictly formal terms (except for
those cases in which a speaker refers to some previous utterance within the current
communicative situation by using some speech reporting frame (I said , you
said ); for references to utterances outside the current situation, cf. below). The
shift in level is identified, rather, in pragmatic terms.
The following questions are meant to delineate the main pragmatic parameters
that apply for analyzing metacommunication. They pertain to the issue/topic
chosen, to the intention pursued, and to some frame conditions of metacommunication (for a more detailed treatment and examples of full-length analyses, cf. Bublitz and Hbler (eds.) 2007).7
(1) What are the topics of metacommunication? Though always referring to
particular speech events, either single or serial, the topical scope can be generic as
well as specific. Metacommunicative topics of a more generic kind may, for instance, concern the problem of effability (I dont have words which could adequately express how I feel or Love is just a gross label), the mode chosen (You
dont have to touch me when you want something from me or Could you speak up a
bit?), or questions of participating in a conversation (Why do you always want to
dominate conversation? or You never take part in this kind of talk). They may be
comments that operate with general principles, norms and maxims; thus Grices
maxims find rich application (e.g., the maxim of quality in Are you telling the
truth?, the maxim of quantity in Just give us the gist of it, the maxim of relevance
in Come to the point, please! and the maxim of manner in Let me just finish this
story before we leave) as well as Leechs politeness principle (Stop being so patronizing!).
More specific metacommunicative topics may concern illocutionary functions
(Is that a complaint?), acts of reference (Who are you talking about?) or predication (Thats a very euphemistic formulation or What do you mean by insinuation?). Or they may concern matters of discourse organization. These include
performing or holding back certain speech acts (Give me your permission to do so,
please), announcing a speech act (And now I will disclose my next plans), or steering an ongoing discussion (Weve got three requests to speak; first Lila, then John,

110

Axel Hbler

then Karen). Comments relating to the text structure (such as But the real sensation
is coming now) may occur within longer turns.
(2) What for or why? These are the central questions for any pragmatic approach to communication. The purposes that metacommunicative utterances in
particular pursue are in part prefigured by their topic.8 Thus they concern the organization of discourse, including the textual structure of longer stretches of utterance (So far we have been describing a state of affairs; now we may turn to );
they help secure an adequate understanding, including attempts at rendering a
speech act function more precise (Is this a complaint or an accusation?) and other
kinds of monitoring. Their sense is to prevent or repair misunderstandings.
Another common purpose of metacommunicative comments is evaluative,
mainly critically, in nature, in as far as what is topicalized marks or implies a
negative deviance from what is to be expected (But I am repeating myself). Where
such comments are self-critical, they often represent a preventive measure anticipating criticism from others.
Besides general communicative norms and principles, which provide the standard for (metacommunicatively) evaluating a given utterance, other standards
seem to have been availed of recently as well. They are based on what has become
known under the label of critical language awareness (cf. Fairclough ed. 1992),
which Coupland and Jaworski (2004) characterize (rather critically in tone) thus:
The growing obsession with good communication (Cameron 2000) is filled by a communication industry happily producing prescriptive orders of metadiscourse which are
constitutive of new social orders. These meta-discourses are based on various regimes
of verbal hygiene (Cameron 1995) and codifications of communication skills. (Coupland and Jaworski 2004: 39)

The underlying intentions are certainly honorable: to contribute on the mental


level to the social goal of emancipation.
Farther reaching objectives, i.e., objectives that go beyond the associative link
with the communicative aspect topicalized, may consist in securing the participants receptivity or even in raising conflict, in a constructive or destructive key.
The last example (But I am repeating myself) and its variant (You are repeating
yourself) can well illustrate these purposes. And in a still wider perspective, such
metacommunicative comments may be used for defending or questioning communicative norms (the speaker subscribing to a culture-specific norm of brevity
and conciseness) or even creating/modifying personal identities as communicators
(profiling him- or herself as a critical, assertive person with a social status that entitles him or her to utter things of that sort).
(3) Who is the target? Though comments on utterances may appear completely
depersonalized, the speaker of the utterance commented on is nonetheless the ultimate target of such comments. In principle, any participant in a communication
can be the target; we thus can roughly differentiate between comments directed at

Metapragmatics

111

oneself and comments directed at some other speaker (cf. Hbler and Bublitz 2007:
1516).
In principle, metacommunicative actions are reciprocal and egalitarian,
wherever symmetrical relations between participants in an interaction are given
or should be given (cf. Habermas 1971). There are interactive frames, however,
which do not allow for symmetry; in the classroom or the courtroom, for example,
the rights for metacommunicative utterances are asymmetrically distributed. This
pertains, of course, primarily to the right of commenting on the contribution of the
other party. But even self-commenting may show restrictions, especially where the
possibility for a certain communicative move (such as changing the topic, or
launching into a longer discourse) is not readily available; one may have to choose
the form of asking for permission to add something not really to the point instead
of simply announcing that one is going to do that.
(4) What are the forms of realization? The examples illustrating the various
points have already provided a variety of examples of how a speakers metacommunicative intentions can be realized. Since any metacommunicative activity is
based on a reflexive and diagnostic analysis, a commenting utterance seems to be
the most direct equivalent, either a categorical (I am talking too much or You are
talking too much respectively) or a modalized one (Do I talk too much? Maybe, I
am talking too much or I think you are talking too much respectively). Another option is to operationalize the analysis in terms of a corresponding action to be taken
(I should not talk that much or Stop talking that much, Could you stop talking too
much? respectively). Where the target is not the speaker him/herself, such metacommunicative utterances, especially if they are critical, are likely to provoke
some reaction (Thats ok or Im always supposed to keep my mouth shut!), and this
could even lead to extended sequences.
As to the position of metacommunicative clauses, it is most common that they
follow the utterance that they refer to. Only where the speaker takes him/herself as
target may we find a reversed order, in which the utterance referred follows. This
restriction, of course, is not surprising; it is, after all, only in this circumstance that
the speaker of a metacommunicative utterance knows what it will be that s/he (cataphorically) refers to.
A systematic treatment of how the various aspects interact would certainly be
desirable, but is currently not feasible.
Certain meta-phenomena have so far not been taken into account for good
reasons: Though they consist of reflexive language (in the sense of Lucy 1993) and
though they contribute to the ongoing communication, they do not result in metacommunication. Their common denominator consists in their being reportive in
some way or another, i.e., they refer to instances of direct speech outside the current communicative situation and either reproduce them or sum them up.9
The most common forms are direct and reported speech, which are usually
framed (introduced or rounded off) by a phrase containing a verb of saying (He

112

Axel Hbler

said, or She told me that ), a performative verb (He promised to stop smoking), or one of a more interpretative character (He insisted that ). They lack, to
quote Meyer-Hermann (1978: 128), the defining feature that the object referred to
and predicated about is a communicative interaction (sequence) or part belonging
to the same interactional unit as the metacommunicative speech act. These metarepresentations, as Wilson (2000) calls them, refer to speech events outside a
given communicative situation and can therefore be considered extracommunicative (cf. Ungeheuer 1970). Due to their extracommunicative character, these
speech references get absorbed in the primary discourse either as instrument or object. Newspaper articles are a common source for illustration; examples of another
kind are literary analyses or philosophical language essays.
A phenomenon that is not extracommunicative and still represents hardly an instance of metacommunication is the so-called echo-utterance, where the speaker
echoes the preceding speaker by repeating some linguistic material yet giving a
specific turn to it; cf. Sperber and Wilson (1986), Wilson (2000); Graf (2007).
Echo-questions (Youve got frequent-flyer status with which company?) try to reelicit a certain part of the preceding utterance that the addressee either did not
understand or feels provoked by. Echo-statements such as in the following example
He: Its a lovely day for a picnic
[They go for a picnic and it rains]
She: (sarcastically): Its a lovely day for a picnic, indeed.
(Sperber and Wilson 1986: 239)

usually just convey attitudes10 towards the propositional state of affairs quoted/
echoed. In neither case are we confronted with instances of metacommunication;
the echo-utterances are information-oriented (in that they elicit information or
evaluate information) and do not concern some formal aspect of communication.
Where they cause a break, as in the information-seeking reading of the echo-question above, they do not transpose the ongoing communication onto the meta-level.
Another quotative variant, which is not extracommunicative either but functions within an ongoing communicative situation, is at least a strong candidate for
metapragmatics. The reference is to parody, which is defined as imitating the characteristic style of a speaker (or an author or a work of his/hers) with the intention to
ridicule or achieve a comic effect. We usually associate a literary genre with it, but
parody also occurs in ordinary life and everyday conversation. The illustration that
follows is taken from a piece of literature, but the parodist here is not the author
himself but one personage in his (realistic) drama.
Nick:
George:
Nick:
George:
Nick:
George

I try not to
Get involved. Um? Isnt that right?
Yes thats right.
Id imagine not.
I find it embarrassing.
(sarcastic): Oh, you do, hunh?

Metapragmatics
Nick:
George

113

Yes. Really. Quite.


(mimicking him): Yes. Really. Quite. (Then aloud, but to himself:) Its disgusting!
(Edward Albee, Whos Afraid of Virginia Woolf? New York: Pocket
Books, 1966: 90)

The parodistic mimicking performs the metacommunicative act inexplicitly, but


could be made explicit as is illustrated in the given example by the aside with
which Georges turn ends; both modes convey his disgust at the way in which Nick
expresses himself. As is also shown by the example, parody at least in everyday
conversation where it is used only occasionally shows something similar to, but
weaker than, the break which marks the boundary between communication and
metacommunication: the ongoing dialogic flow comes to a halt. With Bally (1965),
we could characterize this inexplicit type of metacommunication as mode vcu,
contrasting with the mode pur of explicit metacommunication (cf. below).
Stylization, as treated by Coupland (2004), and some forms of irony provide
other instances of such inexplicit metacommunication. They represent doublevoiced utterances in the sense of Bakhtin (e.g., 1981) in that they are moving on
two different levels (the level of primary communication and the level of metacommunication, which consists in exaggerated imitation) and yet are united in one
utterance.
1.2.

On abbreviated forms of metacommunication

The subtitle already links this part directly to section 1.1 where full forms of metacommunication were tackled. The same (meta-)pragmatic aspects are applicable
here as well. They will, consequently, be expounded by the same guiding questions; merely the order will be slightly altered, starting this time with the most distinctive feature, i.e., the formal make-up.
(1) What are the forms of realization? The means allowing for abbreviated
metacommunication are quite easy to delineate. They mainly consist in adverbials
that typically operate on the whole proposition of the given utterance. The most
concise forms are disjuncts (such as frankly); slightly more voluminous are prepositional phrases/clauses (such as in short), infinitive clauses (such as to tell you in
a few words) and participial clauses (such as bluntly speaking). They usually can
be expanded to a fully fledged metacommunicative utterance (Ill be very frank
with you; I will sum it up in a few words; I dont want to fuss around but say bluntly
what I have in mind). Being briefer than their elaborated counterparts, they are
better integrated into the dialogical flow; the communicative break they cause is
more subtle, and at best syntactically marked off from the main clause providing
the propositional content of what is primarily communicated. While the full forms
of metacommunication were characterized as off-line phenomena (seen from the
perspective of ongoing primary communication), the abbreviated variants keep the

114

Axel Hbler

communication on-line, in a manner of speaking. It is therefore no wonder that


they are unlikely to open up metacommunicative (side-)sequences.
(2) What are the topics that the adverbials cover? Quirk et al.s (1985) classifications of adverbs/adverbial provide a comprehensive overview. Of prime relevance are style disjuncts and with some restrictions conjuncts.
According to Quirk et al., style disjuncts convey the speakers comment on
the style and form of what he is saying, defining in some way under what conditions he is speaking as the authority for the utterance (Quirk et al. 1985: 615).
The metalinguistic character can (often) be brought forth explicitly by transforming the disjunct into a clause in which the adverbial (turning into a process adjunct)
is linked to a verb of speaking with I as subject. (e.g., Frankly, I am tired I tell
you frankly, I am tired).11
In line with Quirk et al. (1985: 615 ff.), we can subclassify style disjuncts. One
group of adverbs that is of interest focuses on the manner in which an utterance is
made; with their help, a speaker can either refer to the style of his/her utterance
(e.g., bluntly, briefly, simply) or include some modal overtones (e.g., truthfully).
Recall that longer variants are also available, such as prepositional phrases, or
(in-)finitive and participial clauses (cf. above). The other group of interest consists
of adverbs (and adverbial expressions) which characterize the make-up of a formulation, whether it has to be taken, say, figuratively or literally. Again, some expressions may be tinged with modal overtones. Thus, an utterance like Hawkins
was not, strictly speaking, a traitor may address not only the issue of whether the
word traitor is a suitable term for Hawkinss behavior but also the issue of whether
or not Hawkins is a traitor. In addition, there are some adverbs of degree (belonging to the category of subjuncts) that can serve the same metalinguistic purpose:
compromisers (such as kind of, sort of) and approximators (such as almost, virtually) provide indications as to the reliability and adequacy of the lexical expression chosen. In uttering, for example, He is kind of a traitor or He virtually
stole the money the speaker may want to let the listener know that the verdictive
terms traitor and steal respectively are not to be taken literally, but come close.
These adverbials can, in general, be linked up again with some conversational
principle, norm or maxim. A speaker using truthfully in his/her utterance echoes
Grices maxim of quality, using briefly his maxim of quantity, while with figuratively he alludes to his maxim of relevance (by flouting it), and with frankly to his
maxim of manner.12 Adverbs like sort of pay tribute to Leechs politeness principle
in as far as they may render the utterance more palpable.
The second set of adverbials that in part can be interpreted in a metacommunicative key are conjuncts/conjunctions. They are text-structural markers that
specify how an utterance fits into the surrounding discourse in which what is to
follow is systematically connected to what has gone before (Halliday and Hasan
1976: 227). Not all of the 7 conjunctive roles (and sub-roles) between which Quirk
et al. (1985: 634 ff.) distinguish (i.e., listing, summative, appositional, resultive, in-

Metapragmatics

115

ferential, contrastive, transitional) are metacommunicative. The ones that do not


qualify for inclusion are those that establish the co-textual link purely in terms of
content, not of form. The ones whose use allow for a pragmatic interpretation as
being metacommunicative are adverbs of enumerative listing (such as first, second, third; to begin with, next, to conclude), summative expressions (such as in
conclusion, to summarize), contrastive adverbs with a reformulatory (such as
more accurately, alias, in other words) or replacive meaning (as in alternatively,
rather; better, worse) and finally, adverbials, such as incidentally or by the way,
which mark discoursal transition.13
The preceding classification of metacommunicative adverbials shows more
openly perhaps than the treatment of full metacommunicative utterances does a
clear-cut topical dividing line between personal and textual thematic foci. With respect to narratives, McNeill (1992: 185 ff.) differentiates in my opinion quite felicitously between two levels outside the narrative proper: the meta-level, which
includes the language clues characterizing and structuring the narrative as text, and
the para-level which relates to the narrator listener axis, comprising elements of
an interpersonal nature (e.g., providing orientation to the listener, comments and
attitudes of the narrator etc.) Applied to communication in general, we could distinguish between a meta-communicative level that would pertain to the structural
aspects of dialogues and turns and the para-communicative level that would account for those non-primary forms of communication which are person-centered
and focus on the interactional aspects involving the speaker and/or hearer.14 More
recent treatments draw similar distinctions, though less categorically. del (2006)
starts out from the notion of metadiscourse embracing various configurations that
enfocus either the text, the writer (or speaker), and the reader (or hearer) and distinguishes in a second step between metatext, which relates to the text or code
proper, and writer-reader interaction. Also Hyland (2005) uses metadiscourse
as cover-term; he defines it as basically interpersonal and then draws a functional
distinction between means that help to guide the reader (or listener) through the
text (interactive function) and those that involve the reader (or listener) in the text
(interactional category). Some such terminological differentiation could help to
disentangle the complexity of the concept at issue.
(3) The question as to who is the target (and who has got the right to freely select the target) is easy to answer for the adverbials with a metacommunicative
function. Where a speaker uses style disjuncts, s/he has necessarily him/herself as
target in that his/her metacommunicative comment refers pro- or retrospectively to
his or her own primary conversational contribution. Whereas the speaker as metacommunicative target is oftentimes quite transparent with such style disjuncts (cf.
frankly speaking), s/he seems to withdraw from being target where conjuncts that
are reflexive of discourse structure come into play. The sociological question concerning the right to select the target, in this case, boils down to the question of
whether any speakers in any communicative situation may take the opportunity to

116

Axel Hbler

comment on him- or herself as communicator in such brief (adverbial) terms. This


is such an elementary (and minimalistic) question that it can either be answered in
strictly categorical moral terms along the line of Habermas or in empirical terms on
the basis of rich data sets.
(4) The final question to face concerns, again, the raison detre, the why or
what for? The purposes that a pragmatic view can ascribe to these reduced forms of
metacommunication cannot be expected to be essentially different from those
identified for the extended forms with the speaker as target. But they can be expected to be somewhat reduced, according to the iconic principle of a proportionality between quantity of form and quantity of meaning. This seems, indeed, to be the
case. In one way or another, they help the speaker optimize the ongoing primarylevel conversation. The style disjuncts serve him/her as (prophylactic) lubricant to
avoid clashes with interactants (and subsequent repairs). The textual conjuncts facilitate the reception of the primary message emitted.

2.

Metapragmatics as the study of implicit metacommunication

Auer (1986: 22) has hailed it as a new research paradigm and Verschueren (1998:
60) holds that it is the central area, the proper domain of metapragmatic studies:
peoples metapragmatic communicative behavior based on their metapragmatic
awareness of situation and context as co-determinants of communication. No specific explicit repertoire is available to meet this meta-function. The approach conceives metalanguage as a dimension of language use and does not relate it to the
object notion of metalanguage (Verschueren 1998: 55); cf. section 1. What we look
for and find in utterances, instead, are indicators that testify to a metapragmatic
awareness of speakers, a kind of self-monitoring activity which constantly calibrates (to use a Silversteinian term) the communicative intention with situational
and contextual conditions. The significance is highlighted by Silverstein, to whom
Verschueren repeatedly refers in his argumentation.
Without a metapragmatic function simultaneously in play with whatever pragmatic
function(s) there may be in a discursive interaction, there is no possibility of interactional coherence, since there is no framework of structure here, interactional text
structure in which indexical origins or centerings are relatable one to another as aggregated contributions to some segmentable, accomplishable event(s). (Silverstein
1993: 3637)

The corresponding type of reflexive language are indexical forms, which change
their value depending on the actual event of speaking (Lucy 1993: 10). The specific segment of reflexive language consists of shifters, contextualization cues
(prosodic patterns, code switching, etc.) or implicit voices (cf. Verschueren 1998:
61).15

Metapragmatics

117

Communicatively speaking, there is, consequently, nothing that could be paraphrased in explicit metacommunicative terms (cf. 1.2). And likewise, no neat
break or boundary between a primary and a secondary level of communication can
be found, and no turn to off-line communication. But at least one metacommunicative feature is preserved: a reader/listener will be lead to attend to the words instead of attending from them, to recall Antons (1998) subtle play with prepositions
(quoted in section 1).
The attending to words, more specifically to indexical words and their metapragmatic dimension, shall be further elaborated and illustrated in two points. The
first point concerns the so-called shifters themselves, e.g., personal pronouns, deictic expressions, or tenses whose meanings shift due to their dependency on the
individual single text and situation.16 The pronoun you, for example, acquires its
specific referential meaning according to the communicative situation. This situation-dependent meaning assignment then is the result of the metapragmatic reasoning, on the speakers as well as the receivers side. More interesting and complex are such metapragmatical assignments in the case of some text-cohesive
devices, i.e., reference, substitution, or ellipsis. Halliday and Hasan (1976) provide
characterizations in which the metapragmatics behind their use and understanding
is (unvoluntarily) well highlighted. All three are forms of presupposition, devices
for identifying something by referring it to something that is already there known
to, or at least recoverable by, the hearer (Halliday and Hasan 1976: 144).17 With
respect to reference, we read that [b]oth exophoric and endophoric reference embody an instruction to retrieve from elsewhere the information necessary for interpreting the passage in question (Halliday and Hasan 1976: 33). More concretely,
a reference item signals supply the appropriate instantial meaning, the referent in this
instance, which is already available (or shortly to become available); and one source of
its availability is the preceding (or following) text. (Halliday and Hasan 1976: 227)

In substitution and ellipsis (as a special case of substitution), the presupposition is


at the level of words and structures. When a substitute is used, it signals that the
actual item required, the particular word or group or clause, is recoverable from the
environment (Halliday and Hasan 1976: 145).
The second point pertains to contextualization cues, whose metapragmatic dimension is as Gumperz (1982: 131) states rarely consciously noted. He gives
the following definition:
[A] contextualization cue is any feature of linguistic form that contributes to the signaling of contextual presuppositions. Such cues may have a number of such linguistic realizations depending on the historically given linguistic repertoire of the participants. The
code, dialect and style switching processes, some [] prosodic phenomena [] as well
as choice among lexical and syntactic options, formulaic expressions, conversational
[] strategies can all have similar contextualizing functions (Gumperz 1982: 131).

118

Axel Hbler

Among his examples, we find one where a graduate student is supposed to interview a black housewife in a low income, inner city neighborhood. After an appointment has been arranged over the phone by someone from the university, the
student goes to the black home and is met by the husband, who opens the door,
smiles and says to him So yre gonna check out ma ol lady, hah? What he utters is
a formulaic phrase identifiable through co-occurrent selections of phonological,
prosodic, morphological and lexical features (Gumperz 1982: 133); it is marked
for ethnicity and signals both expectations about what is to be accomplished and
about the form that replies must take (Gumperz 1982: 134). That such expectations exist comes best to the fore where they are not met by the interactant and
subsequently cause irritation, break-down or the like. In the example, the student
does not meet the husbands expectations by (inappropriately) replying Ah, no,
I only came to get some information. They called from the office. Thereupon the
husband dropped his smile and disappeared without a word while calling his
wife.
Sometimes, the metapragmatic reasoning behind an utterance may not be interactionally oriented, but may rather serve the maintenance and construction of identity (cf. Tajfel 1982). Identity is not only to be understood as a fixed set of properties or operations residing in the individuals cognitive make-up, where language is
regarded as a largely docile medium through which dimensions of a persons
identity [] may receive occasional expression (Wooffitt and Clark 1998: 107); it
is also (and above all) an accomplishment with respect to interactional and inferential concerns generated by the trajectory of verbal exchanges [], [whose] ascription is inextricably tied to the details of talk-in-interaction (Wooffitt and Clark
1998: 107). Take the utterance Hang on, love. The term of address used could be
indicative of an informal (cross-gender) relationship between speaker and addressee. In the actual instance, however, the addressee is not the speakers wife, but
a (female) mediator the couple has turned to in their divorce case. The speaker here
constitutes his relationship with [the addressee] in a way which is at odds with the
mediator-disputant relationship they have thus far collaboratively sustained
(Greatbatch and Dingwall 1998: 128). With this utterance, the speaker is questioning the addressees professional identity.
What the metapragmatic devices under consideration all have in common and
whereby they differ from the reflexive types taken into consideration in the preceding two sections is the mode of reflection. Bally (1965) distinguishes between the
mode pur and the mode vcu; it is the latter one which we have been confronted
with in the present section, whereas we were concerned with the former one previously. Ballys distinction is worth quoting in greater detail.
Si le langage nest pas une cration logique, cest que la vie dont il est lespression na
que faire des ides pures. Si lon me dit que la vie est courte, cet axiome ne mintresse
pas en lui-meme, tant que je ne le sens pas, tant quil nest pas vcu [] (Bally 1965:
15)

Metapragmatics

119

If language is not a logical creation, it is due to the fact that life, whose expression it
represents, does not simply produce pure thoughts: If I am told that life is short, I am not
interested in this idea unless I feel it, unless I experience it.

The following illustration fits well in our context, since the Italian pur(e) is a discourse marker.
Le jugement intellectuel La terre tourne se change en jugement de valeur dans la bouche
de Galile seciant devant ses juges: E pur si muove! Ce nest plus une vrit scientifique, cest laffirmation dune valeur attache cette vrit [] (Bally 1965: 15)
The logical judgement The globe revolves becomes an evaluative judgement in the
mouth of Galileo screaming into the face of his judges E pur si muove! Then it is no longer
a simple scientific truth, rather it is the affirmation of a value attached to this truth.

Uncovering the metapragmatic dimension of language and language use in indexical forms is the objective of the corresponding field of scientific activity, which, in
turn, derives its name from the very objective. Strictly speaking, metapragmatics is
in this sense then the study of the metapragmatic dimension of language and
speech rather than the study of metacommunication. If we were to continue in the
attempt at terminologically distinguishing the various conceptualizations of metapragmatics, the term syncommunicative could well serve this purpose; the prefix
syn- (of Greek origin like the prefixes meta- and para- in the complementary expressions) conveys quite adequately the fact that this reflexive level is not isolatable but inextricably linked to the primary level of communication.18

3.

Beyond interacting

Leaving ongoing conversations and taking a more generic view instead opens up
perspectives to be dealt with now. Observing communication in its general design
and different patterns and describing and systematizing the observations is the perspective that constitutes a first-level step away from conversation; theorizing about
the conditions, presuppositions and implications of such endeavors constitutes a
second-level step. While the second level is reserved for the professional linguist
(cf. 3.2), the first level is open to both the professional and the ordinary conversationalists (cf. 3.1).
3.1.

Metapragmatics as the study of peoples abstracting from interacting

Through their participation in all sorts of conversations, members of a speech community gather in the course of time knowledge about conversation in general, their
genres, patterns, styles, norms, etc. Since this knowledge relates to the pragmatic
dimension of communication, it may be specified as metapragmatic knowledge.
Some similar metapragmatic knowledge may be obtained by the scientist who

120

Axel Hbler

studies the conversations of this speech community as an outsider yet in the role of
a participating observer. In other words, it is hardly possible to clearly distinguish
between the scientific metapragmatic knowledge and the metapragmatic knowledge of the users of a language. The assumption of a continuum between both
stances, as made by Caffi (2006: 87), might therefore seem quite appropriate;19 the
difference consists merely in a smaller or greater distance from the object, viz.: the
use of the language and a lower or higher degree of explicitness.
Folk theories, however, call for the mediating linguist, the linguist, that is, who
elicits folk knowledge or elaborates on their shorthand versions. Gumperz (1982)
identifies as mediating linguists in particular those working in the fields of discourse analysis or ethnography of communication. Folk-linguistics is a separate
discipline (cf. Niedzielski and Preston 2000), but has roots in both. In any case the
task is to come to grips with the varying ways in which linguistic behavior is conceptualized by those engaged in it (Verschueren 1998: 6061). Of the folk-linguistic topics, speakers awareness of and knowledge about pragmatic issues is, of
course, particularly relevant for the current context. Two areas and modes (cf. Preston 1996) shall be considered in greater detail.
(1) A major source of information about peoples metapragmatic knowledge
could be the lexical repertoire they have at their disposal for referring to aspects of
speech; it is not an autonomous language, but a subset of the ordinary language.
Speech act verbs are a favorite research area (where folk taxonomies merge with
the empirical-conceptual approach of speech act theories and the ethnography of
speaking, cf. Caffi 2006: 85). But other lexical expressions relating to other selected aspects of communication could also be common ground where folk- and
empirical-conceptual linguistics meet. Lexical expressions, in general, can be interpreted as a means that a speech community uses for coding cognitive and cultural models, which consist in more or less coherent sets of concepts for structuring
experience (cf. also Gee 1999: 40 ff.). Experience also includes acting by communicating, and this, in turn, calls for adequate conceptualizations of a metapragmatic kind and their storage in words.
One of the most comprehensive collections of English speech act verbs is the
dictionary compiled by Ballmer and Brennenstuhl (1981).20 Its comprehensiveness
derives from two decisions. (a) They apply a much wider definition of speech act
verb by including all those verbs which designate (aspects of) speech activities
(Ballmer and Brennenstuhl 1981: 3), no matter whether they can be used performatively or not; thus lie and persuade, for example, are speech act verbs just as
well as admit or promise. (b) They include literal as well as metaphorical expressions, as, for example, contradict and reject. Of particular interest is the authors ordering of the gathered material into a huge word field and its many subfields,
because these can be understood as identifying the metapragmatic aspects of communication that the English speaking community takes recourse to. The word
fields in their entirety unfold a global metapragmatic conceptualization whose

Metapragmatics

121

main constituents delineate models entering into an integral system of speech activities. Each model has as its linguistic correlate a set of verbs, which again have
substructures. The enaction model, for example, consists of verbs, of which wish,
aim at, or anticipate express VOLITION , alert s. o., indicate, or refer to PUTTING INTO
FOCUS , while expressions of offering, asking, commissioning, ordering, warning
etc. are instances of ENACTION , and verbs like accept, refuse, or obey instances of
REACTIONS ON ENACTIONS . At the most refined level, we find sets of verbs that show
a high degree of synonymy; for the enactment category SURPRISING , for instance,
the following verbs are listed: amaze, astonish, astound, bewilder, disconcert, flabbergast, shock, stagger, startle, strike, surprise. Taken together they delineate, if
we will, how the English language community conceptualizes acts of surprising
somebody by words; seen in a differentiating way, each single item contributes a
special nuance to the overall-concept, in comparison to the others.
Along these lines, Goossens (1987) has made an analysis of some such nearsynonymous verbs. His treatment of the speech act verbs say, tell, talk and speak
can well illustrate how much an in-depth analysis is able to reveal about peoples
metapragmatic models, though it explicitly tries just to answer a typically Fillmorean question, namely, how do linguistic action verbs frame the scene of linguistic (inter)action? (Goossens 1987: 95). Thus, native English speakers assign
different roles to receivers, encoding in tell and say the receptor who plays a more
passive part, and in talk the interactor who plays a more active, participative part;
in the case of speak, the English speaker accounts for the two roles by using distinct prepositions, with where the receptor acts as interactor, to where as receptor
(cf. Goossens 1987: 103). The four verbs reveal, furthermore, that native English
speakers differentiate messages according to the degree to which they have been
condensed. This is reflected in the facts that, for example, say freely combines
with direct enunciations, while tell is considerably restricted in taking them
(Goossens 1987: 103); the differentiation is in principle even reflected where the
distributional pattern is reversed, in other words where more extreme condensations go with tell, but not with say (e.g., tell the truth vs. *say the truth).
Lexical expressions that belong to certain metaphorical concepts offer particularly rich information about peoples cognitive metapragmatic models.21 Metaphors have in general attracted much attention by linguists because they seem to
offer insights into how the human mind operates. As Lakoff and Johnson state in
their pivotal study of 1980: [M]etaphor is pervasive in everyday life, not just in
language but in thought and action. Our ordinary conceptual system, in terms of
which we both think and act, is fundamentally metaphorical in nature (Lakoff and
Johnson 1980: 3); and metaphors that conceptualize communication in one respect
or another have been paid attention to right from the very beginning.
For an illustration, we may turn to two of the most widely known examples,
i.e., the so-called conduit metaphor, identified by Reddy (1979) and differentiated
later by Vanparys (1995) and Semino (2006), and the argument is war concept

122

Axel Hbler

proposed by Lakoff and Johnson (1980) and modified by Grady (1998) and Semino
(2006). The results of these investigations, derived from a close analysis of sets of
metaphorical expressions in general use and subsequently supplemented by corpus
data, can be understood as reconstructions of native English speakers conceptualizations of certain aspects of how they communicate and what their form of communication is like. These conceptualizations are metapragmatic in nature.
As far as communication between communicators in general is concerned, the
dominant meta-pragmatic model people have developed shows three main constituents: the speaker/sender puts an idea-object into a word-container and sends it
(through a conduit) to the receiver, who extracts from the word-container the ideaobject.
Discussion/argumentation represents a specific form of communication/discourse. The corresponding metaphors widely in circulation reveal that English
speakers share a (metapragmatic) model of argumentative discourse that centers
around physical conflict; debatants have two positions, attacking the others
opinion, while defending ones own; we have phases of maneuvering and retreat;
the outcome amounts to truce or victory/surrender.
(2) Another important source for gaining insights into peoples understanding
of communication would consist in information that researchers are able to elicit
directly from members of the speech community; this is a method which is often
applied by discourse analysts, where the aim is to produce ideotypical descriptions [of concrete discourses] that can be dissected into significant components
(Gumperz 1982: 157). But the results are on the whole quite unsatisfying; ordinary
conversationalists are obviously poor descriptors (Gumperz 1982: 157),22 and
this should hold true even more, were they confronted with a more abstract and
complicated task. Alternatively, an experimental way of eliciting metapragmatic
knowledge may be feasible (similar to the method that sociologists, e.g., Goffman,
occasionally deploy, for discovering certain norms which people subscribe to in
relations in public, i.e., by confronting them with transgressions), but has not, to
my knowledge, been pursued. The use of questionnaires seems to be the relatively
more reliable method.
The study by Simon-Vandenbergen (1995) shall be used to further illustrate the
point at stake, even if it is not exactly made for such a purpose. Its aim was to obtain assessments of linguistic behavior from English native speakers. The method
consisted in having natives evaluate selected aspects and forms of communicative
behavior through evaluating corresponding (metaphorical) expressions, which, as
we saw, embody parts and aspects of folk models of behavior. The questionnaire
applied included expressions covering social functions of talk (e.g., conversation,
chat, or prattle), turn-taking (e.g., run on, cut someone short, step in), topic management (e.g., bring up, move unto, meander), and manner of speaking (e.g., rattle,
babble, or air, thunder, or tell someone flat out, wrap up ones meaning). The questions/tasks that were to stimulate statements about peoples metapragmatic views

Metapragmatics

123

had three formats: 1. (social) assessments in terms of positive negative neutral,


2. (aesthetic) assessments in terms of light heavy neither, and 3. the same as
format 1 but supplemented by (a) a question asking why the subject rated the expression the way s/he did and (b) the request to give an example sentence.
While certainly all tasks can be interpreted as throwing light on peoples understanding of communication, it is the supplementary task (a) which should elicit the
most; unfortunately, Simon-Vandenbergen does not provide any cue as to what and
how subjects met this task. Nevertheless, some of the results she obtained allow
conclusions. Speech tempo is a good case in point. This is the picture emerging for
native English conversationalists (cf. Simon-Vandenbergen 1995: 108): Speed is
an important criterion; high and low speeds are often negatively evaluated as too
fast and too slowly respectively, in which case they tend to be talked about in
metaphorical terms.
One could, of course, object and claim that the negative evaluations elicited by
the questionnaire were just triggered by the metaphoricity of the expressions themselves. Any such objection, however, would not be justified. As Simon-Vandenbergen already seems to imply, it is the perceived marked value of the communicative phenomenon in relation to norms and standards in a given community that
triggers metaphorical expressions and assessments, and not the metaphorical expression that triggers a corresponding evaluation. Besides, in the case of speed,
there is even some independent support from social-psychological research for the
claim that the negative evaluations relate, indeed, to speed and are not simply triggered by the metaphoricity of the lexical expression. Smith et al. (1975), for
example, had subjects listen to recordings of speakers performing one and the same
text with different speeds; in reality, it was just one speaker whose natural speech
tempo had been technically manipulated. The subjects had to judge the speakers
as to benevolence, a social criterion which aligns well with Simon-Vandenbergens positive-negative scale. The authors found that [t]he benevolence/rate
plot reveals an inverted U-relationship, with the mean for normal voices [] being
rated more benevolent (Smith et al. 1975: 150), while high- and low-speed
speakers scored significantly low in benevolence.
The insights obtained by metapragmatic research outlined so far can be enriched (and maybe even relativized) by widening the scope so as to also include
possibilities of variation, always applying the same analytical parameters.23 Of the
three main types of variation, i.e., social, developmental, and historical, the social
variant has not yet received due attention. Illustrations will therefore be restricted
to the other two, turning to the lexicological source first.
As Geeraerts and Grondelaers (1995) point out,
if cognitive models are also cultural models, they are also cultural institutions, and as
such, they carry their history along with them: their institutional nature implies their historical continuity. It is only by investigating their historical origins and their gradual
transformation that their contemporary form can be properly understood. (1995: 177)

124

Axel Hbler

In line with this stance, Arnovick (1999) has examined changes in the conceptualization of promising:
The frequency of promises rendered defective at inception or follow-through has created a practical (if not moral) dilemma, a problem with promises. People do not always
trust the promises they hear. As a result, the need to convince the hearer is often assumed by the speaker. To perform a promise that satisfies the hearer, the speaker often
has to emphasize his or her commitment. (Arnovick 1999: 60)

She notes that even a performative use of promise, which is already a very strong
form of promising (cf. Traugott 1997), is nowadays often felt to be insufficient and
to require some further declarations, such as those observed with George Bush:
Read my lips, I guarantee it, Believe you me. A simple I will nowadays works
only in a promissory situation and an extra-linguistic institution for its force (Arnovick 1999: 58), in contrast to OE times when both the use of sculan (expressing a
subjects obligation) and wilan (expressing a subjects intention) were binding on
the speaker. Nowadays, promises usually undergo relativizations and rationalizations: promises merely express honorable intentions; changing circumstances
can render vows unrealistic; insincerity is politically motivated, deception is committed for the sake of the task pursued. Such views are further warranted by references to independent proposals (by Muck 1989 and Rawls 1955) for constructing
theories of promising that incorporate a limit on speaker responsibility (Arnovick 1999: 60).
The historical and culture-bound relativity of metaphorical concepts on the
pragmatics of language can well be illustrated by two investigations of mine. In
Hbler (1998), I showed that the predominant spatial metaphorization relating to
the notion of express, which represents the first constituent within the conduit
paradigm resulted from a change that had taken place around the turn from the 16th
to the 17th centuries. The argument centered around the change in the use of prepositions going with the notion of express; it was a change from the instrumental
with to the spatial in (words). The former conceptualization aligns well with
and foreshadows the tools-paradigm proposed by Reddy (1979) as a possible alternative to the predominance of the conduit metaphor. And in Hbler (ms), I examined data from the 16th and 17th centuries which show that argument was conceived
during that period in less fierce and more playful terms, more specifically in terms
of a tennis match; Gradys (1998) findings can be taken as a diluted late version of
this early concept.
A discoursal source is tapped on by Gotti (2006). He shows for the 16th/17th
century that neologizing (in contrast to the practice nowadays) was still a matter to
talk and write about; it was the time when a scientific language developed in England, by taking over elements from the common code. The following passage can
be read as an instance of the public discourse about it, particularly about using new
words of foreign origin in English.

Metapragmatics

125

But now as touching myne entent in writing this treatise in the English. Though this
cause might seme sufficient to satisfy many men that I am an englysh man, and therefore
may more easely and plainly write in my natyue tonge, rather then in any other: yet vnto
them that know the hardness of the mater, this answer shuld seme vnlykely: considering
that it is more harder to translate into such a tonge, wherein the arte hath not ben written
before, then to write in those tongues that are accustomed, and (as I might say) acquainted with the termes of the science (Recorde 1547, quoted in Gotti 2006: 218)

Recorde was a physician and mathematician. His topic in this passage (probably
from the preface) is metalinguistic and concerns the (un-)availability of English
words for certain concepts. But the discourse itself could be a case for metapragmatics (in the sense discussed in section 1); the passage is metacommunicative in
that the writer raises the problem of effability in English and thus justifies his use
of un-English words in the treatise to follow. In the current context, however, the
quotation can be seen as documenting metapragmatic knowledge about the limits
and chances of informing an uninformed readership about specific and new states
of affairs. Representing the metapragmatic knowledge of a member of an educated
16th-century English speech community, the document could also give grounds for
comparing it with (in-)effability treatments by speakers from different social-cultural backgrounds; but I do not know of any such investigation.
The developmental type of variation in metapragmatic knowledge is, in
contrast, well investigated. For the purpose of illustration, the article by Bernicot
and Laval (1996) has been chosen, because it focuses, again, on promises. Their
subjects are children of three age groups, around 3, around 6 and around 10 years
of age. The childrens metapragmatic knowledge is defined as their capacity to talk
and think about acts of keeping and breaking promises and tested by examining
their verbal comments stimulated by a series of cartoon stories; each story consists
of a promise (by a child or a parent) and the subsequent fulfilment or non-fulfilment; the effect of sadness (as sign of dissatisfaction) and happiness (as sign of satisfaction) respectively, which the (non-)fulfilment has on the person to whom the
promise was made, has then to be determined by the subjects themselves and subsequently justified. The authors summarize their results thus:
Metapragmatic knowledge was found to evolve with age. At the ages of 3 and 6,
childrens metapragmatic knowledge mainly concerns the execution of the action. At
age 6, the listeners desires start being added in cases where [this] preparatory condition
is not satisfied. At age 10, explanations pertaining to action accomplishment completely
disappear, and explanations about the speakers intentions alone or about both the
speakers intentions and the listeners desires appear. (Bernicot and Laval 1996: 120)

The metapragmatic knowledge of adults would at least also comprise the sincerity
condition of promises, i.e., that the speaker intends to accomplish the future action
(cf. Gibbs and Delaney 1987).

126
3.2.

Axel Hbler

Metapragmatics as metatheory of pragmatics

So far, we have dealt with cases where speakers provide the data/material, which
linguists then process, refine (as was the case with metacommunication) or elaborate on (as in the case of speakers theoremes). We now leave the sphere where linguists occupy themselves with other peoples products and turn to their own activity, subjecting it to reflection.24 That portion of such reflections which becomes
manifest constitutes a professional theoretical dialogue (mainly in the written
medium) about pragmatics, with the aim of defining the field, supporting, supplementing, or justifying certain theoretical stances or challenging them. Such theoretical discussions show, in principle, two orientations. They either operate introspectively or involve external considerations, and sometimes these perspectives
merge. Since pragmatics as such is already the most complex language-related discipline, it is easy to understand that the meta-discourse about it is immense and
open-ended; problems abound. The few topics that will subsequently be addressed
can, therefore, do nothing else but give some indications.
The problems already start on the most elementary of levels, including even the
question what pragmatics is all about. What, in other words, is the object of pragmatics? Is it human communication or just the verbal part of it? The answer may
depend on the answer given to an even more elementary question: How to define
the scope of activity for a linguist working in the field of pragmatics? Is it the discipline that defines the scope or is it the object as such? In that particular case,
pragmatics originated in philosophy but was subsequently fully absorbed by linguistics; should it therefore remain a linguistic discipline and restrict its activities
accordingly? According to what? Was it not one of the major figures of linguistics,
de Saussure, who embedded linguistics in semiotics, which consequently would
license any occupation with the non-verbal as well? If, on the other hand, it is the
object which defines for scholars of pragmatics the scope of their activities, then
the non-verbal with its phonetic and kinesic domains would be included anyway
because natural communication, even any single utterance, can only be fully
understood as a trimodal phenomenon. It is not even justifiable to claim that the
verbal part plays the main role because intonation and/or gestures oftentimes modify in an important way what is verbally said; they sometimes even override the
verbally mediated meaning. Even if one were to subscribe to the view that, ideally
speaking, the nonverbal modes should be attended to as well, it could still be argued that such a holistic approach is simply not practicable; several disciplines
should therefore participate and join in a common endeavor. But how much compartmentalizing does the subject matter tolerate? Interdisciplinary research may
be a good answer, but transdisciplinarity may still be unavoidable up to a certain
degree.
Linguistics and communication research did not fall out of the sky; its predecessor was rhetoric, the art of persuasion. Traces of this tradition of thinking can

Metapragmatics

127

still be found in Searle (1969) in that he keeps the link alive that exists between a
speech act and its effect, the perlocution. Though the ensuing research paradigm
widened the scope so as to treat the speech act as part of speech act sequences, it
may cause wonder why perlocution never played a role. Why has research in indirect communication strategies been restricted to what the speaker plots with the
addressee underdetermined; why has research not systematically investigated the
(perlocutionary) effects that certain strategies are likely to cause in receivers? Because social psychology is doing this job? At some US universities, departments of
Communication (not tantamount to mass communication) have been established
whose activities are at the cross-roads of various lines of research, including linguistics, rhetoric and persuasion, and social psychology. Is that a more adequate
discipline for coming to terms with interpersonal communication? Or what do they
lack that the (European) language-biased approach has to offer?
Questions of involving other disciplines do not only represent boundary issues (Caffi); they may also be raised and pursued with pleasure when they help
dignify central views or important findings of ones own discipline. The dignifying effect results from their proving that the views or results are not isolated, but
align well with what is considered to be more fundamental in the sense that it provides the foundations of ones own discipline, its overarching rationale. A good
case in point is the ever growing theoretical horizon associated with pragmatics.
The cognitive turn in the humanities necessitated a cognitive foundation also for
pragmatics. Since the cognitive sciences, however, have (neuro-)biological foundations, to what degree are they to be paid tribute? Is it enough to allude every now
and then to such deep connections? Is it acceptable to do so only when it seems opportune, i.e., when it serves the argument at issue, and neglect them otherwise?
McNeills treatment of (everyday conversational) narratives provides an interesting example. A central part of his production theory, which (to varying degrees)
takes all three modes of realization into consideration (i.e., language, intonation,
and gestures), operates with a model of self-organization that obviously has its
roots in the theory of autopoietic (or self-referential, self-organizing) systems,
which goes back to Maturana (1970a, 1970b). McNeill, however, does not discuss
(metapragmatically!) these roots. Was it because the theory is difficult for students
of interpersonal communication to digest?25 Where are the limits, and how far do
we have to go, in how much detail?
Many of the questions raised (as well as others that could be raised) still await
answers. Whatever the answers are or will be, the fundamental nature of these issues (and even the questions themselves) are rarely free of any ideological bias,
whether one wants to admit it or not. A sensibility for the ideological dimension of
research in pragmatics cannot be taken for granted; it seems, in fact, that it still
needs to be called in. [A] constant monitoring of linguistic rhetoric in view of the
ideological underpinnings of theories and analyses is not a luxury but a prerequisite for the advancement of linguistic pragmatics (Verschueren 1998: 67). How

128

Axel Hbler

subtle the dangers of ideological pitfalls can sometimes be is well highlighted by


Caffi.
[O]ne has to avoid the risk of new forms of hypostasis [], brought about by certain,
apparently harmless, metalinguistic uses. For instance, the use of a definite noun phrase
can be idealizing: the risk of an idealization exists whenever one refers to the speaker,
who could end up by being an hyperuranian subject, or a lunar effet-sujet. (Caffi 2006:
83)

But more tangible issues have, of course, to be subjected to such self-examination


as well. What is the researchers stance to the speech community that s/he wants to
investigate about, even towards the individual informant? Does the stance have
any political implications? And on a methodological level, are the questions asked
(orally or by means of a questionnaire) politically correct?
A subspecies of this kind of metatheoretical problem consists in the question of
whether the methods of tackling pragmatically relevant data is neutral enough to
cope with different yet comparable sets of data. The problem of culturally biased
approaches is addressed here as well. Among the English speech act verbs, for
example, we find none that would have lexicalized any specification as to time and
location (as in Tenejapa Tzeltal, where concepts like talk that has taken place just
a short time ago or talk occurring in a grassy area are lexicalized), but we do find
English speech act verbs of directing, such as appoint, convene or ordain, which
include specifications as to social setting and institution (cf. Verschueren 1987:
131133). A simple speech act classification would not do because it would neglect the specificities. In the face of such shortcomings, Verschueren proposes to
avoid attempts at classifying linguistic action verbs (and the corresponding acts)
and [] concentrate instead on the careful contrastive analysis of the semantic dimensions involved in the lexicalization of linguistic action (Verschueren 1987:
126). The descriptive framework he offers is supposed to allow for accommodating
(by way of comparing) intra-linguistic, cross-linguistic and cross-cultural differences.
A (partial) translation of Verschuerens program into action is Meys analysis
of a (by now familiar) speech act, promise.
[T]he context in which a promise is made is of the utmost importance for its status as a
promise and for its binding effects. Take the case of a young person promising his or her
parents not to smoke before the age of eighteen. In this case, the societal conditions surrounding the execution of such a promise can be exceedingly difficult (peer-group
pressure, work conditions, etc.). (Mey 1993: 170)

But it serves the author to underline a metapragmatic argument: the view that a
philosophical speech act theory la Austin (21990) and Searle (1969) may not suffice for linguistic purposes.
Both Austin and Searle operate on the one sentence one case principle. That is to say,
in order to illustrate their theory, they use examples that are characteristic of what they

Metapragmatics

129

see as the case under discussion, such as, e.g., a particular kind of speech act. I will, in
the following, show the shortcomings of the case approach (Mey 1993: 170).

The example also demonstrates that metapragmatic discourse does not always
have to be the raison detre for an entire (theoretical) text. It often has, rather, a
more limited purpose, i.e., to profile ones own theoretical assumptions against the
backdrop of what is widely held to be the case. This pattern is pervasive and can be
found in almost all scholarly work in whatever discipline; but it is of course only in
pragmatics that such metatheoretical discourse is metapragmatic. The illustration
to follow is taken from a prominent source.
A more radical difference between Grices approach and relevance theory [as proposed
by Sperber and Wilson, A.H.] is this. Grices principle and maxims are norms which
communicators and audience must know in order to communicate adequately. Communicators generally keep to the norms, but may also violate them to achieve particular
effects; and the audience uses its knowledge of the norms in interpreting communicative
behaviour. The principle of relevance, by contrast, is a generalisation about ostensiveinferential communication. Communicators need no more know the principle of relevance to communicate than they need to know the principles of genetics to reproduce
[] (Sperber and Wilson 1986: 162).

The quotation is an excerpt from a short metapragmatic passage of one and a half
pages, where Sperber and Wilson confront their concept of relevance with that of
Grice (e.g. 1975). The metapragmatic discourse in this case is even sparked off by
a metalinguistic problem, i.e., the fact that both parties use one and the same term,
but associate it with different concepts. The passage, however, is not metacommunicative because concepts, methods and the like are the issue, not the communicative form or aspects of it. This is a feature that generally applies to any form in
which pragmatics is theoretically reflected. It is a type of reflection that is at best
accidentally language-reflexive. This sets it apart from all the other phenomena
that have been taken account of in this article and aligns it with the genre metatheory. Metapragmatics may be, in the end, an all too generous term.

Notes
1. On July 31, 2009.
2. This is basically what Caffi (2006) takes account of as metapragmatics 2.
3. Applying the typology of metamessages proposed by Jakobson (1971 [1957] 130133),
we are mainly concerned with type (M/M), i.e., messages about messages. Certain attention will also be given to type (C/M), that is to those metamessages in which so-called
shifters (as the code involved) incorporate a message about a message. Messages about
code (M/C) as well as the meta-message type (C/C), where proper names are the protagonists, fall outside the scope of metacommunication proper. Cf. also Lucy (1993: 16).
4. In section 2, a different concept will be introduced, which conceptualizes metalanguage
as a dimension.

130

Axel Hbler

5. Metalanguage in the sense outlined, however, is part of but not equivalent to reflexive
language as described by Lucy (1993) just as speech reflexive activity/ reflexive language use is not tantamount to metacommunication.
6. Such reduced forms will be tackled separately below.
7. Quite extensive studies along these lines have been conducted earlier by Meyer-Hermann (1980 [unpublished]). Lumping together topic and function, he identified 72
metacommunicative speech act types (on the basis of French and Portuguese material).
The 11 most frequently used types are: requesting the performance of speech acts, acknowledging the performance of speech acts, announcing the performance of speech
acts, describing communicative interaction, guiding discussions, criticizing communicative interaction, restating a communicative function more precisely, eliciting receptive readiness, securing comprehension, evaluating communicative interaction, and anticipating sanctions prophylactically (cf. Welte 1990: 178180). They all have been
included in one way or another in the present outline.
8. Meyer-Hermann (1980), in fact, tackles functions only, the topics being implied.
9. Quoting, in fact, is the phenomenon that Jakobsons meta-message type (M/M) was obviously meant to mainly accommodate.
10. As Wilson (2000: 148) points out, [the] attitudes conveyed by echoic utterances can be
very rich and varied: the speaker may indicate that she agrees or disagrees with the original, is puzzled, angry, amused, intrigued, skeptical etc., or any combination of these.
11. Often the disjunct may have a paraphrase with a verb of saying taking the form of a
finite clause (If I may say so without offence, your writing is immature).
12. Hagemann (1997) conducted a detailed investigation on German diction-characterizing adverbs in Gricean terms.
13. I am not sure about the status of adverbs such as well which can either function as discourse initiators (cf. Quirk et al. 1995) or continuation markers (cf. Halliday and Hasan
1976).
14. The topical distinction as such has been highlighted, under different labels, by various
authors; cf. Bamford and Biondi (2005: XVII), Hagemann (1997: 2728), Geissner
(1981: 206). Watzlawick et al. (1968) and Bateson (1972) focus exclusively on the interpersonal dimension and reserve for this aspect the term metacommunication.
15. In Jakobsons typology (cf. footnote 3), they would represent metamessages of the type
(C/M).
16. Jakobson (1971: 131), who adopted the concept from Jespersen (1922: 123), discusses
shifters at some length.
17. Halliday and Hasans use of the term presupposition differs, of course, from its usual
semantic-pragmatic reading; they use it in the wider sense that something is to be supplied, or understood (Halliday and Hasan 1976: 144).
18. Cf. also analogous term formations such as synsemantic.
19. But caution is advised. Speakers awareness [] does not have to match the linguists
metapragmatic descriptions. Naively confusing these two may contribute to the furthering of a specific kind of folk metapragmatics inherent in western linguistics itself. Assuming that such nave confusion can be avoided, at least some of my further comments
will be based on results obtained in this line of research (Verschueren (1998: 6061).
20. In certain respects complementary, but of a narrower scope, is Schmid (2000) with his
investigation of shell nouns, a subset of which overlaps with notions taken account of
by Ballmer and Brennenstuhl (1981).

Metapragmatics

131

21. For obvious reasons, I simply take both kinds for granted, though it is certainly a matter
worth discussing; cf. Kvecses (2005: 193 ff.) and his discussion of the question
whether cultural models can exist without prior metaphorical understanding, as claimed
by Quinn (1991).
22. When participants report on actual verbal encounters, they tend to do so by mentioning
some item of content, or by referring to what people were getting at (Gumperz 1982:
157).
23. For questioning the parameters, cf. section 3.
24. This overlaps with what Caffi characterizes (under metapragmatics 1) as concern for
the epistemological implications of what researchers assume pragmatics is about (Caffi
2006: 83).
25. The reader may test it, in the original or in a version such as Benseler et al. (eds.) 1980,
which has communication scientists as its target audience.

References
del, Annelie
2006
Metadiscourse in L1 and L2 English. Amsterdam: Benjamins.
Anton, Corey
1998
About talk: The category of talk-reflexive words. Semiotica 121/3,4:
193212.
Arnovick, Leslie K.
1999
Diachronic pragmatics. Seven case studies in English illocutionary development. Amsterdam: Benjamins.
Auer, Peter
1986
Kontextualisierung. Studium Linguistik 19: 2247.
Austin, John L.
1990
How to do things with words. Edited by James O. Urmson and Marina Sbis.
2nd. ed. Oxford: Oxford University Press.
Bakhtin, Michail M.
1981
The dialogic imagination. Edited by Michael Holquist translated by Emerson,
Caryl and Michael Holquist. Austin, TX: University of Texas Press.
Ballmer, Thomas and Waltraud Brennenstuhl
1981
Speech act classification. A study in the lexical analysis of English speech activity verbs. Berlin: Springer.
Bally, Charles
1913 [1965] Le langage et la vie. Genve: Ed. Atar, [reprinted Genve: Librairie Droz].
Bamford, Julia and Marina Biondi
2005
Introduction. In: Julia Bamford and Marina Biondi, Dialogue within discourse
communities. Metadiscursive perspectives on academic genres. Tbingen:
Niemeyer. IXXIV.
Bateson, Gregory
1972 [1955] Steps to an ecology of mind. New York: Balantine.
Benseler, Frank, Peter M. Hejl and Wolfram K. Kck (eds.)
1980
Autopoiesis, communication and society. The theory of autopoietic system in
the social sciences. Frankfurt, New York: Campus.

132

Axel Hbler

Bernicot, Josie and Virginie Laval


1996
Promises in French children: Comprehension and metapragmatic knowledge.
Journal of Pragmatics 24: 101122.
Bublitz, Wolfram and Axel Hbler (eds.)
2007
Metapragmatics in use. Amsterdam: Benjamins.
Caffi, Claudia
2006
Metapragmatics. Encyclopedia of Language and Linguistics. Amsterdam:
Elsevier. 8288.
Cameron, Deborah
1995
Verbal hygiene. London: Routledge.
Cameron, Deborah
2000
Good to talk? Living and working in a communication culture. London: Sage.
Coupland, Nikolas
2004
Stylised deception. In: Adam Jaworski, Nikolas Coupland and Dariusz Galasinski (eds.), Metalanguage. Social and ideological perspectives, 249274.
Berlin: Mouton de Gruyter.
Coupland, Nikolas and Adam Jaworski
2004
Sociolinguistic perspectives on metalanguage: Reflexivity, evaluation and
ideology. In: Adam Jaworski, Nikolas Coupland and Dariusz Galasinski (eds.),
Metalanguage. Social and ideological perspectives, 1551. Berlin: Mouton de
Gruyter.
Fairclough, Norman (ed.)
1992
Critical language awareness. London: Longman.
Gee, James P.
1999
Discourse analysis. London, New York: Routledge.
Geeraerts, Dirk and Stefan Grondelaers
1995
Looking back at anger: Cultural traditions and metaphorical patterns. In: John
R. Taylor and Robert E. MacLaury (eds.), Language and the cognitive construal of the world, 153179. Berlin: Mouton de Gruyter.
Geissner, Hellmut
1981
Sprechwissenschaft. Theorie der mndlichen Kommunikation. Knigstein im
Taunus: Scriptor.
Gibbs, Raymond W. and Suzanne M. Delaney
1987
Pragmatic factors in making and understanding promises. Discourse Processes
10: 107126.
Goffman, Erving
1972
Relations in public: Microstudies of the public order. New York: Harper and
Row.
Goossens, L.
1987
Dealing with linguistic action verbs in depth. In: Jef Verschueren (ed.), Linguistic action: Some empirical-conceptual studies. Norwood, NJ: Ablex.
95106.
Gotti, Maurizio
2006
Metalinguistic considerations in 17th-century specialized discourse. In: Anna
Duszak and Urszula Okulska (eds.), Bridges and barriers in metalinguistic
discourse, 217231. Frankfurt: Peter Lang.

Metapragmatics

133

Grady, Joseph
1998
The Conduit metaphor revisited: A reassessment of metaphors for communication. In: Jean-Pierre Koenig (ed.), Discourse and cognition: Bridging the
gap. Stanford, CA: CSLI Publications. 205218.
Greatbatch, David and Robert Dingwall
1998
Talk and identity in divorce mediation. In: Charles Antaki and Sue Widdicombe (eds.), Identities in talk, 121132. London: Sage.
Graf, Claudia
2007
Echo und Zitat. In: Elke Brendel, Jrg Maibauer and Markus Steinbach (eds.),
Zitat und Bedeutung. Linguistische Berichte, Sonderheft 15: 89109.
Grice, H. Paul
1975
Logic and conversation. In: Peter Cole and Jerry L. Morgan (eds.), Syntax and
semantics. Vol. 3: Speech acts. New York: Academic Press.
Gumperz, John J.
1982
Discourse strategies. Cambridge: Cambridge University Press.
Habermas, Jrgen
1971
Vorbereitende Bemerkungen zu einer Theorie der kommunikativen Kompetenz. In: Jrgen Habermas and Niklas Luhmann, Theorie der Gesellschaft oder
Sozialtechnologie, 101141. Frankfurt: Suhrkamp.
Hagemann, Jrg
1997
Reflexiver Sprachgebrauch: Diktumscharakterisierung aus Gricescher Sicht.
Opladen: Westdeutscher Verlag.
Halliday, Michael A.K. and Ruquaia Hasan
1976
Cohesion in English. London: Longman.
Hbler, Axel
1998
The expressivity of grammar. Grammatical devices expressing emotion across
time. Berlin: Mouton de Gruyter.
Hbler, Axel
o.J.
The conceptualization of argument in terms other than war. A socio-cultural
case study. (Ms)
Hbler, Axel and Wolfram Bublitz
2007
Introducing metapragmatics in use. In: Wolfram Bublitz and Axel Hbler
(eds.), Metapragmatics in use, 126. Amsterdam: Benjamins.
Hyland, Ken
2005
Metadiscourse. Exploring interaction in writing. London, New York: Continuum.
Jakobson, Roman
1971 [1957] Shifters, verbal categories, and the Russian verb. In: Selected writings II:
Word and language, 130147. The Hague: Mouton.
Jespersen, Otto
1922
Language. London: Allen and Unwin.
Kvesces, Zoltan
2005
Metaphor in culture. Universality and variation. Cambridge: Cambridge University Press.
Lakoff, George and Mark Johnson
1980
Metaphors we live by. Chicago: The University of Chicago Press.

134

Axel Hbler

Leder, Drew
1990
The absent body. Chicago: The University of Chicago Press.
Lucy, John A.
1993
Reflexive language and the human disciplines. In: John A. Lucy (ed.), Reflexive language. Reported speech and metapragmatics. Cambridge: Cambridge
University Press. 932.
Maturana, Humberto R.
1970a
Biology of cognition. Report 9. Biological Computer Laboratory, Department
of Electrical Engineering. University of Illinois, Ill.
Maturana, Humberto R.
1970b
Neurophysiology of cognition. In: Paul L. Garvin (ed.), Cognition: A multiple
view, 323. New York: Spartan.
McNeill, David
1992
Hand and mind. What gestures reveal about thought. Chicago, London: The
University of Chicago Press.
Mey, Jacob L.
1993
Pragmatics. An introduction. Oxford, Cambridge: Blackwell.
Meyer-Hermann, Reinhard
1978
Aspekte der Analyse metakommunikativer Interaktionen. In: Reinhard MeyerHermann (ed.), Sprechen Handeln Interaktion, 103142. Tbingen: Niemeyer.
Meyer-Hermann, Reinhard
1980
Studien zur Funktion von Metakommunikation am Beispiel portugiesischer
und franzsischer Alltagssprache. Habilitationsschrift Universitt Bielefeld.
(Ms.)
Muck, Bruce
1989
Promising rules and intentions. PhD Diss. University of Wisconsin.
Niedzielski, Nancy A. and Dennis R. Preston
2000
Folk linguistics. Berlin: Mouton de Gruyter.
Paul, Ingwer
1999
Praktische Sprachreflexion. Tbingen: Niemeyer.
Preston, Dennis R.
1996
Whaddayaknow? The modes of folk linguistic awareness. Language Awareness 5: 4077.
Quinn, Naomi
1991
The cultural basis of metaphor. In: James Fernandez (ed.), Beyond metaphor:
The theory of tropes in anthropology. Stanford, CA: Stanford University Press.
5693.
Quirk, Randoph, Sidney Greenbaum, Geoffrey Leech and Jan Svartvik
1985
A comprehensive grammar of the English language. London, New York: Longman.
Rawls, John
1955
Two concepts of rules. Philosophical Review 64/1: 332.
Reddy, Michael
1979
The conduit metaphor: a case of frame conflict in our language about language.
In: Andrew Ortony (ed.), Metaphor and thought, 164201. Cambridge: Cambridge University Press.

Metapragmatics

135

Schmid, Hans-Jrg
2000
English abstract nouns as conceptual shells. Berlin: de Gruyter.
Searle, John R.
1969
Speech acts. Cambridge: Cambridge University Press.
Semino, Elena
2006
A corpus-based study of metaphors for speech activity in British English. In:
Anatol Stefanowitsch and Stefan T. Gries (eds.), Corpus-based approaches to
metaphor and metonymy, 3662. Berlin: Mouton de Gruyter.
Silverstein, Michael
1993
Metapragmatic discourse and metapragmatic function. In: John A. Lucy (ed.),
Reflexive language, 3358. Cambridge: Cambridge University Press.
Simon-Vandenbergen, Anne-Marie
1995
Assessing linguistic behaviour. A study of value judgements. In: Louis
Goossens, Paul Pauwels, Brygida Rudzka-Ostyn, Anne-Marie Simon-Vandenbergen and Johan Vanparys (eds.), By word of mouth, 71124. Amsterdam:
Benjamins.
Smith, Bruce L., Bruce L. Brown, William J. Strong and Alvin C. Rencher
1975
Effects of speech rate on personality perception. Language and Speech 18:
145152.
Sperber, Dan and Deirdre Wilson
1986
Relevance. Communication and cognition. Oxford: Blackwell.
Tajfel, Henri
1982
Social identity and intergroup relations. Cambridge: Cambridge University
Press.
Traugott, Elizabeth
1997
Subjectivication and the development of epistemic meaning: The case of
promise and threaten. In: Toril Swan and Olaf J. Westvik (eds.), Modality in
Germanic languages: Historical and comparative perspectives, 185210. Berlin: de Gruyter.
Ungeheuer, Gerold
1970
Kommunikative und extrakommunikative Betrachtungsweisen in der Phonetik. In: Proceedings of the Sixth International Congress of Phonetic Sciences,
7386. (Prag 1967). Prague.
Vanparys, Johan
1995
A survey of metalinguistic metaphors. In: Louis Goossens, Paul Pauwels,
Brygida Rudzka-Ostyn, Anne-Marie Simon-Vandenbergen and Johan Vanparys (eds.), By word of mouth, 134. Amsterdam: Benjamins.
Verschueren, Jef
1987
Metapragmatics and universals of linguistic action. In: Jef Verschueren (ed.),
Linguistic action. Some empirical-conceptual studie,125147. Norwood, NJ:
Ablex.
Verschueren, Jef
1998
Notes on the role of metapragmatic awareness in language use. In: Adam
Jaworski, Nikolas Coupland and Dariusz Galasinski (eds.), Metalanguage.
Social and ideological perspectives, 5373. Berlin: Mouton de Gruyter.
Watzlawick, Paul, Janet H. Beavin and Don D. Jackson
1968
Pragmatics of human communication: A study of interactional patterns, pathologies, and paradoxes. London: Faber and Faber.

136

Axel Hbler

Welte, Werner
1990
Alltagssprachliche Metakommunikation im Englischen und Deutschen. Frankfurt: Lang.
Wilson, Deirdre
2000
Metarepresentation in linguistic communication. In: UCLWPL 11: 127161.
Wooffitt, Robin and Clark Colin
1998
Mobilizing discourse and social identities in knowledge talk. In: Charles
Antaki and Sue Widdicombe (eds.), Identities in talk, 107120. London: Sage.

Part II
Theoretical foundations

5.

The rise of pragmatics: a historiographic


overview
Wataru Koyama

1.

Introduction

This article investigates the theoretical, philosophical, social institutional, and interpersonal genealogies of several key concepts of contemporary linguistic pragmatics, such as presupposition, implicature, speech act, and performativity, as well
as perhaps more fundamental ones, viz., context, indexicality, function, referential
and non-referential practice, metapragmatics, and reflexivity. More specifically,
after briefly dealing with the genealogy of the concepts context and indexicality in Section 2, the article discusses the genealogies of presupposition and
effect (presupposing and creative functions) in Section 3, function (multifunctionality) in Sections 4, 5, and 6, and performativity, metapragmatics, and
reflexivity in Section 7.
In so doing, this article tries to articulate a historiographic overview of how linguistic pragmatics was established and developed before it became an institutionalized international discipline. It specifically focuses on two critical developments since the late nineteenth century that led to the rise of pragmatics in the
1970s: i.e., the linguistic turn, to be followed by the pragmatic turn. The
former, which took place from the 1880s to the 1930s, gave rise to structural-functionalist linguistics. This historical transformation was closely linked to the neoKantian movement, marking the empirical turn of Kantian philosophy, and the
phenomenological movement, focusing on the synchronic here and now of communicative events (see Sections 5 and 6). On the other hand, the linguistic turn also
led to the emergence of analytic philosophy, with its focus on the propositional,
logical-semantic aspects of language. Structural linguistics, interacting with analytic philosophy, took the formalist turn in the 1940s, turning away from context
and (communicative) function and moving to concentrate on the (morpho)phonological, (morpho)syntactic, and semantic aspects of language. Gradually, however,
especially in the 1950s and 60s, there was increasing awareness of the importance
of context, often discussed under the rubrics of indexicality and pragmatics, even
in logico-analytic philosophy (see Section 2). This led to the pragmatic turn and the
institutional establishment of linguistic pragmatics in the 70s, linguistics and philosophy moving from the analytically-oriented study of de-contextualizable regularities of language toward the empirical study of the contextualized use of language. Thus, since the 1970s, the theoretical notions articulated or suggested by
the older traditions of structural-functional linguistics, such as context, indexical-

140

Wataru Koyama

ity, function, and metalanguage, have been re-discovered and further explored. In
addition, the pragmatic turn has allowed the convergence of pragmatics with the
empirical social sciences, which are traditionally concerned with actions and interpretations in context. As is described in Section 3, this convergence characterizes
the current state of linguistic pragmatics, where the earlier emphasis on regularities
in language use is increasingly supplemented with the focus on the pragmatically
achieved (trans)formation of group identities and power-relations of language
users in specific sociocultural contexts.
In such a way, this article tries to present a historiographic account showing
that the true home of linguistic pragmatics lies in the Continental tradition (cf.
Huang 2007: 45), which tries to provide a contextual, pragmatic view of language
and communication in our life-worlds, not in the Anglo-American tradition,
which sees pragmatics as a narrowly delimited component supplementary to the
ahistorical, universalizing studies of syntax and semantics, abstractable from referential regularities. As this article will demonstrate, pragmatics came out of the
Continental tradition in the late nineteenth century and, after the formalist lapse at
the mid-twentieth century, the discipline has been moving back from the AngloAmerican to the Continental tradition, further exploring referential and non-referential practices, i.e., what we do, with or without words, in our life-worlds (see
Section 5 for details).

2.

Genealogy of the concepts context and indexicality

Let us begin our historiographic reconstruction with one of the most basic notions
of pragmatics, namely, context, which is closely related to the phenomenological
notion of life-world and which includes not only the tokens of linguistic types
but also communicative-event participants and their pragmatic acts (cf. Mey
[1993] 2001: 206235). It would not be too far-fetched to say that pragmatic
means contextual(ized), as the discipline itself is differentiated by its focus on
context, from structural linguistics and semantic logic, which are mainly concerned with de-contextualized regularities abstractable from referential practices.
Clearly, the historiography of pragmatics must be written around this theme: how
context has been discovered as a theoretical topic, and theoretically dealt with (cf.
Nerlich and Clarke 1996).
Of course, all empirical studies of language or communication have been
at least implicitly concerned with context, so that we might argue that modern
theorization on context(uality) started with the emergence of modern empiricism,
as suggested by John Lockes (16321704) treatise on human understanding
and communication (cf. Locke [1689] 1975), which anticipated Thomas Reids
(171096) pragmatic commonsense philosophy and thus paved the way for the
development of modern semiotics, pragmati(ci)sm, and pragmatics.1

The rise of pragmatics: a historiographic overview

141

Yet, from the linguistic (structural) point of view, which focuses on de-contextualizable linguistic categories, it is, of course, personal and demonstrative pronouns, moods, modalities, tenses, and other deictic categories (deixis) or indexical
types that most clearly show contextuality, often called context-dependency (cf.
Fillmore 1975; Levinson 1983: 5496); hence, the specifically linguistic origins
of todays linguistic pragmatics might be found in empirical investigations and
theorizations of the deictic categories. Thus, the roots of linguistic pragmatics
may be sought, as Nerlich and Clarke (1996) have done, in the structural-functionalist traditions: i.e., the (Austro-)German tradition, which led to Karl Bhlers
(18791963) theory of indexical expressions (Zeigwrter), the deictic field (Zeigfeld; vs. Symbolfeld), and Organon model, developed in 193334; the French
one, culminating in Benvenistes (190276) theory of pronouns, subjectivity, and
smantique, or what would become known as nunciation (Benveniste, 1966); and
the British tradition, which eventually led to Lyonss (1977) semantics (in the
broad sense, including linguistic pragmatics) and Hallidays text linguistics, the
latter synthesizing the Firthian theory (London School) and the Praguean theory
of referential (informational) communication and multi-functionalism (see Section 6).2
We shall follow these functionalist movements in Europe, anticipating the
worldwide rise of pragmatics in the latter half of the twentieth century, later in this
article (see Sections 46). Here, we should take a brief look at the development on
the other side of the Atlantic. That is, in the United States, the Praguean functionalist theory became combined with the Peircean theory of semiotic pragmati(ci)sm
by Jakobson (18961982), who articulated a contextually-anchored, that is, speech
event-centered account of grammatical categories, including what Bally called
actualisateurs or Jespersen (18601943) named shifters (Jespersen 1922),
which Jakobson (1957) characterized as denotational-indexicals. As is well
known, the notion of indexicality came from Peirce (see note 1), who defined it as
the kind of relationship between signs (e.g., tokens of linguistic types) and their
objects (e.g., contextual referents) that is pragmatically motivated by the empirical
principle of contiguity in the context of signification (communication). Note that
indexicality is a general semiotic notion, characterizing not only deixis, but
also pointing gestures, gazes, spatial arrangements (e.g., proximity), and any other
modes of communicative practices based on the principle of contiguity in the contexts of communicative events (cf. Peirce [18931910] 1955). Clearly, indexicality
is a key to the proper theorization of context and contextualized communication, as
duly recognized by the micro-sociologist (ethnomethodologist) Garfinkel (see
Section 3) and, later, the social pragmatician Mey ([1993] 2001: 5260).
The significance of indexicality for pragmatics had been also acknowledged, in
a different disciplinary context, in Bar-Hillel (1954), which would lead to Richard
Montagues (193071) attempt at a formal pragmatics. As this indicates, Bar-Hillel
was an analytic philosopher generally working in the logical syntax framework

142

Wataru Koyama

of Rudolf Carnap (18911970), closely associated with the behavioristic pragmatist Charles Morris (190179), i.e., two figures involved in the logical-empiricist
project of Unified Science (Einheitswissenschaft), or the Unity of Science Movement, led by Otto Neurath (18821945; a member of the Vienna Circle), in the
early-to-mid-twentieth century (cf. Carnap [1934] 1937; Morris 1938). In particular, Morris, partially following Peirces semiotic pragmati(ci)sm (see immediately
above), had articulated, in 1937, the famous semiotic triad of syntax, semantics,
and pragmatics,3 to be adopted by Carnap. Coming out of such a tradition, Bar-Hillels work on indexicality, pragmatics (as wastebasket of syntax and semantics),
and implying (what Grice would later call implicature) in the 1940s and 1950s
(cf. Bar-Hillel 1946; Chapman 2005: 92) anticipated, in retrospect, not only Montague and Grice, but also the pragmatic turn in analytic philosophy in the 1960s
and 70s, which saw Kripkes (1972) theory of rigid designation, causal theory
of reference on proper names (cf. Putnam 1975), speakers reference (true reference; Donnellan 1966, 1978; Kripke 1979), and Kaplans (1978) theory of
Dthat on demonstratives. With these developments, by which analytic philosophy became increasingly focused on indexicality and pragmatics, the discipline
turned post-analytic and came to merge with neo-pragmatism in the 1970s. Also,
whereas the Carnapian logical syntax tradition was associated with neo-Bloomfieldian structural linguistics and its derivatives like Chomskyan generativism, the
newer, post-analytic developments became associated with Generative Semantics
and the emerging discipline of linguistic pragmatics in the 1970s (cf. Donnellan
1978; Davis 1991).
Thus, indexicality, along with its associated concepts, has been a key element
for the constitution of pragmatics and the proper theorization of context. Further,
the Peircean notion of indexicality and Bhlers ideas of deictic field (Zeigfeld) and deictic center (origo) became theoretically combined in the linguisticanthropological tradition, derived from not only Boas (18581942) and Sapir
(18841939), but also Peirce, Jakobson, Benveniste, and Kuryowicz (1895
1978), and led by a disciple of Jakobsons: namely, Silverstein (b. 1945). As documented by Pressmans (1994) contribution to the historiography of pragmatics
(also, Caton 1987), Silverstein (1976a) demonstrated, in his famous Jakobsonian
work on case-marking and the Noun Phrase (NP) hierarchy, that grammatical categories are universally, systematically anchored onto the deictic center, located at
the center of the context of communicative interaction (i.e., Jakobsons (1957)
speech event). Further, Silverstein (1976b) made clear that this pragmatic, indexicality-based theory of linguistic structure was part of the general linguistic-anthropological theory of sociocultural communication, covering not only linguistic
structure, but also referential and non-referential communicative practices, the
latter pertaining to social indexicality, or the indexing of the group identities and
power-relations of communicative-event participants. Thus, indexicality has been
shown to be a key concept in the pragmatic study of contextuality, across the struc-

The rise of pragmatics: a historiographic overview

143

tural, referential, and non-referential domains of language and communication (see


Section 6, below, for more on indexicality; cf. Hanks 1990; Duranti and Goodwin
1992; Agha 2007).

3.

Genealogy of the concepts (contextual) presupposition and effect

Having briefly seen how the notions of indexicality and deictic center have
been made use of to account for the communicative event-centered constitution of
context and linguistic structure, we now move to explore historical developments
in the theoretical understanding about pragmatic processes that take place in, and
(re-)create, context. Here, it appears that the intuitive notions of situational (contextual) appropriateness and effectiveness (or efficacy) of actions have led to,
and still lie behind, more elaborate theorizations of this theme, in terms of, for
example, presupposition and perlocution(ary effect), respectively.
Let us first deal with theoretical notions related to contextual appropriateness.
As noted immediately above, the concept of presupposition, be it semantic or
pragmatic (cf. Karttunen 1974), is a theoretical elaboration of the intuitive notion
appropriateness. The concept was developed by Frege (18481925; see Section 5) and Russell (18721970; a leading figure in the logicist tradition of British
analytic philosophy), and was later picked up by Peter Strawson (19192006; a
major figure in the Oxford Ordinary Language School). Subsequently, the
schools theoretical framework and concepts were discovered and integrated into
linguistics in the late 1960s and 70s by (post-Chomskyan) Generative Semanticists: viz., George Lakoff, McCawley, Robin Lakoff, Ross, Sadock, et al. (cf. Lakoff 1989; Harris 1993; Koyama 2000). As these were immediate predecessors of
linguistic pragmatics (which became established in the 1970s; cf. Mey [1993]
2001: 1927),4 presupposition became a main theme in the new discipline, especially in the 1970s and early 1980s (cf. Gazdar 1979: 89128; Levinson 1983:
167225). Generally speaking, since the 1970s, the studies of presupposition have
been characterized by the increasing focus on pragmatic, rather than semantic, presupposition and the increasing awareness of the centrality of pragmatic context in
matters of presupposition in general.
Similarly, the notion of felicity condition, apparently concerning contextual
appropriateness and derived from the works of Austin (191160) and Searle (b.
1932), became a prominent theme in the new discipline. (Austin was the leading
figure of the Oxford philosophy of language, and Searle was an American-born
student of Austin and Strawson at Oxford in the 1950s). As Austin must have been
at least dimly aware, the theme of felicity condition, which concerns the contextual conditions that must obtain in order for a (speech) act to occur validly, was
actually an old theme in law (see note 7), and, we should also observe, in other disciplines concerned with (judicial or other) rituals, such as sociology and anthropol-

144

Wataru Koyama

ogy, which had the long tradition of investigating the social or cultural (contextual)
appropriateness of rituals and other actions (see Section 7). Thus, in the 1970s and
early 80s, felicity condition, an old theme recently re-thematized by Austins
disciple Searle, became a topic not only in the new discipline of pragmatics, but
also in the social sciences (see immediately below), thereby facilitating the interaction and partial merger between pragmatics, micro-sociology, and linguistic
anthropology. In other words, it appears that this theme may have helped the new
discipline to move in the direction of sociocultural: that is, toward the microsociology of conversation analysis (cf. Levinson 1983: 284370) and, beyond
that, linguistic anthropology, social semiotics, social pragmatics, critical pragmatics, and critical discourse analysis, which have been gaining prominence in
pragmatics since the late 1970s and especially 80s and 90s (cf. Mey [1993]
2001). Note that, in the 1970s and early 80s, when felicity condition was hotly
discussed, the micro-sociologist Goffman (192282) wrote an important article on
it (Goffman 1983), and a variety of concepts coming out of his uvre, such as
face (cf. Brown and Levinson [1978] 1987), frame (cf. Tannen 1986), and
footing (cf. Irvine 1996), began to exercise great influence among pragmaticians, discourse analysts, and linguistic anthropologists (cf. Goffman 1967, 1974,
1981, 1983). On the other hand, in the same period, the more individual-psychological aspects of felicity condition, such as the speakers intention and sincerity,
articulated by Searle, became critically examined from cross-cultural perspectives
by linguistic anthropologists (cf. Rosaldo 1982; Du Bois 1987; Duranti 1993),
working in the tradition of ethnography of speaking/communication (see Section 6, below, for details; cf. Gumperz and Hymes [1972] 1986; Bauman and
Sherzer [1974] 1989). This move anticipated the increasing appreciation of crosscultural differences in pragmatics especially since the 1980s (cf. Mey [1993] 2001:
262288).
Indeed, the social anthropological concern with situational (contextual) appropriateness of actions goes back to a founder of British social anthropology,
Malinowski (18841942), who advanced the notions of context of situation
and context of culture,5 adopted in the 1930s by his close associate, J. R. Firth
(18901960), i.e., the leading figure of the London school of functional linguistics.
The modern anthropological idea, advanced by Malinowski, Radcliff-Brown
(18811955), and Boas (18581942) in the early twentieth century, that any action,
be it linguistic or non-linguistic, should be understood in relation to the situational
or cultural context in which the action takes place, was intertwined with the anthropological doctrine of cultural relativism, which urged that actions be seen and
understood from natives point of view, from within, or emically in the terminology of Pike (1967: 3772). Beyond the disciplinary matrix of twentieth-century social/cultural anthropology, the idea of the cultural relativity of actions, especially linguistic actions, was also shared by Wittgenstein (18891951) and Peter
Winch (192697), the former sometimes considered a precursor of the Ordinary

The rise of pragmatics: a historiographic overview

145

Language School, including Austin, Grice, and Searle, and subsequently, linguistic
pragmatics (cf. Levinson 1983: 227).
Importantly, the (situational or cultural) contextualist doctrine, advanced by
modern anthropology and later adopted by much of linguistic pragmatics, not only
emphasized the role of context in determining the meanings of actions, but also
tended to take context to be an independent macro-social variable determining the
meanings of micro-social actions, the latter considered as dependent variables. In
other words, the doctrine failed to pay sufficient attention to how actions create
and change the contexts (situations and cultures) in which they take place. This
static view of culture was perhaps most pronounced in Radcliff-Browns Durkheimian functionalism, where the meanings of actions were explained by the
(macro-)social functions they served. In this view, which came close to macrosocial determinism, the meanings of micro-social actions (what we do, with or
without words) were equated with the functions they served in the macro-social
order (contextual reality) presupposed to exist independently of actions.
To be sure, in Malinowskis functionalism, which was less Durkheimian and
more Freudian than Radliff-Browns (cf. Kuklick 1991), it was not the macro-social order, but the psychological needs of socially contextualized individuals that
were taken as presupposed variables, teleologically explaining the meanings
(functions) of their actions; all the same, the meanings of actions were functionally explained by the conformity (fit, or appropriate correspondence) of the actions with their presupposed ends, be they psychological or (macro)sociological.
The functionalist doctrine was espoused by much of linguistic functionalism
and the nascent discipline of linguistic pragmatics in the 1970s. For instance, in
these disciplines, actions (utterances, or speech acts) were often explained or theorized by appealing to the notions such as (social or psychosocial) maxims and
conditions, which were, as in Malinowskian functionalism, taken to be independent variables explaining the meanings (implicatures) or efficacy of utterances and
other practices. This functionalist way of seeing speech and other social actions
had been dominant in the post-World War Two behavioral sciences, which became
challenged in the social sciences in the 1960s and 70s, when the idea that society
and culture are contingently (re-)created in and by daily actions and, thus, can be
changed by actions, gained momentum, especially in the anti-establishment,
youth, peace, and (second-wave) feminist movements. In the social sciences, even
before the 1960s, the ethnomethodologist Harold Garfinkel (b. 1917) had begun
frontally to attack the functionalist doctrine, attributed to Durkheim (18581917),
and explore how micro-social, contextualized actions create, and indeed lie at the
foundation of, social organization (cf. Heritage 1984). Interacting with the microsociological studies carried out by Goffman (mentioned above, in this Section),
ethnomethodology subsequently saw a new development called conversation
analysis, explored by Sacks, Jefferson, Schegloff, and their followers, and influencing pragmatics (cf. Levinson 1983: 284370). Meanwhile, since the 1970s,

146

Wataru Koyama

second-wave feminism, critical of the functionalist view of society, has given rise
to womens studies, gender studies, and later, queer studies, a series of developments in which feminist linguistics/pragmatics have been involved (cf. Lakoff
1973; Cameron and Kulick 2003), and in which the term performativity, taken
from Austin, has played a crucial role (cf. Butler 1990). As this indicates, the
speech act theories of Oxonian philosophy and early pragmatics were not exclusively concerned with the question of contextual appropriateness. Rather, they investigated conditions, maxims, and other presupposed contextual variables, at
least partially so as to explain the efficacy, or the creative (performative) function,
of actions, the intuitive appreciation of which was, indeed, the point of departure
for them. And, it was this insight of performativity that attracted theoretical interests of many literary critics, historians, social scientists, and feminists, to pragmatics (see Section 7 for more on performativity). Thus, interacting with them,
pragmaticians moved to construct pragmatic theories focusing on the creative,
rather than presupposing, function of language use, especially as related to powerrelations and group identities of language users. As a result, not only feminist pragmatics, but also critical linguistics/pragmatics, associated with the Frankfurt
School, the New Left, and neo-Marxism, came into being in the late 1970s (cf. Mey
1979, 1985, [1993] 2001), in critical opposition to the functionalist view of society.
Eventually, the 1980s and 90s saw the rise of critical discourse analysis in Western
and Central Europe. Even sociolinguistics, in the wake of criticisms against the
functionalist doctrine (cf. Williams 1992), has begun to move in the direction of
social constructivism, investigating how social realities and systems are dynamically created by discursive interactions (cf. Coupland, Sarangi, and Candlin 2001).

4.

Genealogy of the idea (communicative) functions


(multi-functionality)

Thus, appropriateness and effectiveness, that is, the presupposing and creative
functions of interactional practices, constitute one dimension of pragmatic processes in context, around which the history of pragmatics can be described. Crosscutting this is another dimension of pragmatic processes in context: viz., the dimension of what one might call the functional (indexical) orientations of
communication, such as the referential (descriptive), expressive (emotional), and
appeal functions (cf. Bhler [1934] 1990). Jakobsons (1960) six-functions
model, one of the most solid models of the multi-functionality of communication,
identified the emotive, conative, referential, metalingual (metalinguistic), poetic,
and phatic functions, respectively characterized by the six factors constituting
the event of communicative practice, i.e., addresser, addressee, referent, code,
message, and channel (contact; see Section 6 for more details). Indeed, the term
function has been usually used in this sense in linguistic functionalism and prag-

The rise of pragmatics: a historiographic overview

147

matics, and the history of pragmatics has been often written by tracing the history
of studies dealing with these functions in linguistics, philosophy, psychology, law,
and the social sciences (cf. Nerlich and Clarke 1996). Although Austins theory of
speech act seems to have arisen more or less independently of these functionalist
studies, which had blossomed from the 1880s to the 1930s (see Section 5), it is
clear that they anticipated the basic insight of Wittgenstein and the Ordinary Language School (including Austin, Warnock, Urmson, Grice, Strawson), to be absorbed by Generative Semantics and linguistic pragmatics from the late 1960s onward, that language includes many functions other than the referential function or
propositional meaning, which appears to have been the exclusive concern of logic,
structural linguistics, and generative syntax.
That is, as Nerlich and Clarke (1996) have shown, the social, psychological,
linguistic, and philosophical studies of the non-referential functions (multi-functionality) of language and communication, were vigorously pursued especially
from the 1880s to the 1930s. This was the period of the linguistic turn (cf. Rorty
1967; Apel 1994), before the rise of formalist structural linguistics, which concentrated on the formal and semantic, but not pragmatic or indexical aspects of language, in the 1940s. To be sure, the functionalist traditions were kept alive in the
works of Jakobson, Benveniste, Kuryowicz, Pragueans, and (neo-)Firthians, to
name but a few important figures and schools, but the dominance of formalists
(such as neo-Bloomfieldians and Chomskyans) in structural and generative linguistics, associated with the philosophical project of logical syntax, in the United
States and beyond, continued until the 1970s (see Section 2). As a result, the earlier
functionalist theories and their achievements became largely forgotten, so that,
when Generative Semanticists re-discovered the importance of context (pragmatics) and started to rebel against the orthodox Chomskyan doctrine of autonomous syntax in the late 1960s, they made appeal to the philosophical theories of
Austin, Grice, Strawson, and Searle, but not to the older functionalist traditions (cf.
Harris 1993). Linguistic pragmatics, which arose in the 1970s in the wake of Generative Semantics, was no less strongly influenced by the philosophical theories,
but it also actively interacted with the offspring of the older functionalist traditions,
i.e., a variety of functionalist approaches to grammar, text, and discourse, which
saw a renaissance in the 1970s (cf. Bolinger 1968; Comrie [1981] 1989; Halliday
and Hasan 1976; Leech 1983: 4678). Thus, the importance of the notions advanced by the older traditions, especially deictic field (Zeigfeld), indexicality,
and metapragmatics (the latter two derived in part from Jakobsons (1957, 1960)
code-and-message model and metalingual function, respectively; cf. Silverstein 1976b), has become more fully appreciated, and a number of attempts have
been made to explore and elaborate on the implications of these important notions,
especially in the recent literature of pragmatics (see Sections 2 and 7; cf. Mey
[1993] 2001; Verschueren 1999; Bublitz and Hbler 2007).

148
5.

Wataru Koyama

The linguistic turn and the philosophy of language: functionalism,


phenomenology, and neo-Kantianism

The older traditions of functionalism, as noted above, blossomed from the 1880s to
the 1930s, the period when philosophy took the linguistic turn. Here, focusing on
the disciplinary development of philosophy, which was closely linked to the rise of
semantics and pragmatics, we note that the late nineteenth century saw the returnto-Kant movement, as can be seen in the Kant-inspired pragmatic philosophy of
Peirce in the United States and, especially, the epistemology of neo-Kantians in
Germany, whose opponents included the logico-mathematician Frege (cf. Coffa
1991). This suggests that logico-analytic philosophy, other kinds of the philosophy
of language, neo-Kantianism, and phenomenology were, directly or indirectly, all
involved in the linguistic turn, and interacted with each other, as well as with the
older traditions of functionalism (cf. Richardson 1998; Friedman 2000). Let us
briefly explore historical connections among them, taking Frege as our starting
point.
Needless to say, Frege (18481925) was one of the founders of modern predicate logic and semantics, and his distinction between sense (Sinn) and reference
(Bedeutung) anticipated the later division of semantics and pragmatics, articulated
by Morris and Bar-Hillel (see Section 2, above). No less important, Freges differentiation between proposition and force (Kraft), anticipated by Condillac among
others,6 was a precursor of the distinctions between constatives and performatives,
and between locutionary, illocutionary, and perlocutionary forces, advanced by
Austin, who was a translator of Freges Die Grundlagen der Arithmetik (The
Foundations of Arithmetic). As is well known, these distinctions underlay the formula F(p) (where p and F stand for proposition and illocutionary force, respectively), articulated by Austins student, Searle (cf. Nerlich and Clarke 1996:
22, 76, 202204). Interestingly, the later Searle (1983) advanced a phenomenologically-oriented theory of speech, as suggested by the title Intentionality (a notion
derived from Husserl (18591938), Brentano (18381917), and, ultimately, the
Scholastic concept of intentio); and Searles master, Austin, intended his philosophy of language to be a kind of linguistic phenomenology.7
Thus, there appears a link between Freges logic(ism) and Husserlian phenomenology, although Husserl, who started his career as a philosopher of mathematics
and logic, moved away from logic towards transcendental phenomenology and,
eventually, the phenomenology of the life-world (Lebenswelt). Subsequently, phenomenology inspired various kinds of the philosophy of language: e.g., not only
Austins and Searles, but also the French dialectician Merleau-Pontys (1960),
who exchanged his views with Oxonians, viz., Ryle, Austin, and Urmson, at a colloquium in 1958. Further, phenomenology also influenced the critical social theory
of Habermas (b. 1929; cf. Mey [1993] 2001: 315), who has adopted the notion of
Lebenswelt, and, earlier, the sociology of Alfred Schutz (18991959), who carried

The rise of pragmatics: a historiographic overview

149

it from Vienna to New York. In North America, phenomenological sociology was


then carried out by Berger and Luckmann (cf. Mey [1993] 2001: 115) and by ethnomethodologists, including Garfinkel (cf. Giddens [1976] 1993; Bourdieu 1980).
As linguistic pragmatics has been variously influenced by ethnomethodology, conversation analysis, and Critical Theory, it can be said that the phenomenological
orientation has entered linguistic pragmatics not only from the Ordinary Language
School (see immediately above), but also from phenomenological sociology.
Hence, the phenomenological strand is fairly strong in linguistic pragmatics. After
all, the discipline was established in reaction against formalist structuralism in linguistics, which was, unlike the phenomenologically-oriented functional theories of
Bhler, Jakobson, and the Prague School (cf. Holenstein 1977), non- or even antiphenomenological.
Thus, although analytic logic and phenomenology may appear polar opposites,
there is a linkage between them, as indicated by the careers of Husserl and Austin,
as well as by the history of linguistic pragmatics, which has been, since the late
1970s, moving from de-contextualized logic and maxims towards contextualized
events, i.e., social interactions in historical and cultural life-worlds.
Further, this strongly suggests a certain affinity between linguistic pragmatics
and the philosophical movement of neo-Kantianism. Here, observe that, just like
pragmatics, neo-Kantianism seems located between logic and phenomenology, as
witnessed by the fact that Peirce, who belonged to the Return-to-Kant Movement,
invented a semiotically-based predicate logic, distinct from Freges arithmetic
one, and called his pragmati(ci)sm phenomenology and phaneroscopy.8 Also,
one may note that a neo-Kantian philosopher par excellence, namely Cassirer
(18741945), who belonged to the Marburg School of neo-Kantianism and emigrated to New York, contributed an important paper (Cassirer 1945) on the status
of structural linguistics to the New York-based journal Word, run by Martinet
(190899), Jakobson, and other structural-functionalists, who were opposed to
native-born neo-Bloomfieldian formalists in the United States. This points to the
linkage between neo-Kantianism and functionalist linguistics, and the latter was,
as we have seen, intertwined with phenomenology, especially in the works of Jakobson and Bhler.9
As a matter of fact, Cassirers career, characterized by the application of
(neo)Kantian philosophy, in particular epistemology, onto empirical investigations
into myths, languages, and the history of ideas, was part of the historical transformation, or empirical turn, of Kantian philosophy, which moved from pure and
practical (praktisch in Kants terminology) reason, i.e., critical-philosophical,
transcendental principles of epistemology and praxiology, towards more empirical,
historical, anthropological, pragmatic (pragmatisch) studies of knowledge and actions, at the turn of the twentieth century. Crucially, we should see that, in general,
when knowledge and actions are studied empirically, it becomes clear that they are
mediated by language and communication. Here, recall from Section 2, that Locke

150

Wataru Koyama

(16321704), one of the founders of modern empiricism, felt compelled to write a


treatise on human understanding, in particular, language, communication, and semiotics; and Thomas Reid (171096), deeply influenced by Locke (as well as by
Hume, a Scotsman like Reid), was a precursor of modern pragmatics and, perhaps,
phenomenology (cf. Mulligan 1987: 3334). Subsequently, Peirce (18391914),
influenced by Locke, Reid, and Kant (17241804), developed a semiotic philosophy of communication, which he called pragmati(ci)sm, based on Kants term
pragmatisch (see immediately above). The term was adopted by the behavioristic
pragmatist Morris, who coined the term pragmatics in his famous trichotomy of
communication (see Section 2). We still operate with Morriss terminology, i.e.,
syntax, semantics, and pragmatics; and this fact suggests the intermingling of not
only the terminological but also the conceptual genealogies of todays linguistic
pragmatics and the neo-Kantian movement in philosophy.
Thus, the empirical turn of (neo)Kantianism seems to lie at the roots of linguistic pragmatics. Here, notice that the heyday of neo-Kantianism in philosophy
coincided with the linguistic turn, i.e., the period in which the older traditions of
functionalism blossomed, from the 1880s to the 1930s (cf. Nerlich and Clarke
1996). This cannot be by chance. By the linguistic turn, judgment, a concept integral to Kantian philosophy, became replaced with proposition in post-Fregean
logic, and, more generally, with language in twentieth-century philosophy, including logico-analytic philosophy and the Ordinary Language School;10 at the
same time, (neo)Kantianism turned empirical, giving rise to pragmati(ci)sm and
the empirical science of language, i.e., linguistic functionalism, studying the synchronic, living state of language in action, context, and communication. This
movement, replacing the nineteenth-century historical comparative linguistics
(over)systematized by Neo-Grammarians, was what was called (functionalist)
structuralism (cf. Jakobson 1973), which would become overshadowed by formalist structuralism, the latter turning away from context and pragmatic functions to
formal structure and semantics in the 1940s and 50s.
We shall soon come back to the synchronic turn in the studies of language,
corresponding to the linguistic turn in philosophy: i.e., the rise of structuralism as
part of linguistic functionalism, between the 1880s and the 1930s. But, first, we
must observe that the empirical turn of (neo)Kantian philosophy meant the emergence of the empirical sciences of not just language, but also the mind, society, and
culture: i.e., modern twentieth-century psychology, sociology, and anthropology.
Note that William James (18421910), Wundt (18321920), Max Weber (1864
1920), Simmel (18581918), Durkheim, Mauss, Boas, Malinowski, and RadcliffBrown, who are usually counted as founders of these disciplines, worked in this
period, from the 1880s to the 1930s, and some of them were explicitly associated
with pragmatism and philosophical neo-Kantianism, while others directly or indirectly communicated with them. Further, whether or not they were pragmatists
or neo-Kantians in the narrow sense, their empirical sciences and theories gen-

The rise of pragmatics: a historiographic overview

151

erally shared the epistemological framework of neo-Kantianism, such as the methodologically-based distinction between nomothetic sciences, investigating laws
and other regularities, and idiographic (or historical) sciences, exploring unique
and singular, sociohistorically contextualized phenomena (cf. Stocking 1996). Together with the related opposition between collectives and individuals, this distinction underlay, among others, the Saussurean notions of langue and parole, the
proto-Weberian Methodenstreit concerning the status of economics as a socio-historical (idiographic) or ahistorical (nomothetic, classical Newtonian-Laplacean)
science, and the Durkheimian understanding of sociology in opposition to individualist psychology.
Thus, (neo-)Kantianism was the philosophical foundation for the rise of the
human sciences (cf. Foucault 1966), including linguistics and, later, pragmatics.
The linguistic turn in philosophy and the (functionalist) structuralist turn in linguistics were part of the (broadly conceived) neo-Kantian movement, which also
included the rise of phenomenology. That is, in the studies of language, the linguistic turn at the beginning of the twentieth century consisted of the move towards
the synchronic, living state of language in action (i.e., pragmatics); and the rise
of structuralism should be understood as part of the functionalist, phenomenological movement, oriented towards actions and other happenings taking place hic et
nunc, or the dectic center (origo) of such pragmatic occurrences (see Section 2). As
Jakobson was never tired of pointing out (cf. Jakobson 1973; Holenstein 1977),
structuralism, i.e., functionalist (vs. formalist) structuralism, was not opposed to
phenomenology; they constituted a unified whole (holistic Gestalt), as it were. The
philosophy underlying structuralism was the neo-Kantian one, represented by the
idea of the primacy of (human) (ap)perception, advanced against nineteenthcentury positivism and mechanical empiricism, which had culminated in the NeoGrammarians comparative (historical) linguistics in the late nineteenth century,
and which had earlier served to suppress the proto-linguistic turn, achieved by
Bernhardi and Humboldt in Germany, Condillac and the Idologues in France, and
Reid and the followers of his commonsense philosophy such as Dugald Stewart
(17531828) at the turn of the eighteenth to the nineteenth century (see notes 1 and
2, above; cf. Nerlich and Clarke 1996). The primacy of human perception, or subjective apperception, clearly underlies the structuralist notions of phoneme and
sound pattern, advanced in contradistinction to mechanical physical sounds,
studied by the empiricist, positivist science of nineteenth-century phonetics.
Further, phonemes were considered part of not just subjective (individual), but
also intersubjective, sociocultural reality (see Section 6). The insiders of a community sharing (by degrees) a linguistic structure, i.e., members of a linguistic
community, (ap)perceive the (more or less) same phonemes, but not phones;
they have phonemes as part of their psychological reality (cf. Sapir 1933), that
is, intersubjective phenomenological reality, a pattern of (ap)perception not shared
by the outsiders of the (Durkheimian) socio-cultural, linguistic-structural commu-

152

Wataru Koyama

nity. Thus understood, it becomes less mysterious why the phenomenologist Merleau-Ponty (1945, 1960) appreciated both Gestalt psychology and Saussures linguistics, celebrating the latters discovery of synchrony and parole, and why
Saussure was closely associated with Bral (18321915), a founder (precursor) of
modern twentieth-century semantics and pragmatics, in Paris (cf. Aarsleff 1982;
Nerlich 1992). For all of them, including even Saussure, language was, au fond, an
intersubjective, phenomenological reality, pragmatically unfolding in front of our
eyes, that is, emerging at the synchronic hic et nunc of communication. Thus, the
historical and theoretical roots of modern structural linguistics and linguistic pragmatics converge in the rise of structuralism, i.e., functional structuralism, in the
period from the 1880s to the 1930s, that is, the days of the older traditions of linguistic functionalism, to be eclipsed when structural linguistics became dominated
by neo-Bloomfieldian, Chomskyan, and other kinds of dogmatic formalism after
the achronic (vs. synchronic) turn in the 1940s.

6.

Intersubjectivity and interaction: functionalism, Ordinary Language


philosophy, and the social turn in pragmatics

Intersubjectivity involves not only (1) the speaker (addresser) and (2) the hearer
(addressee), but also (3) the channel (contact) between them, (4) the sign (message) that is sent from the former to the latter, (5) the (intersubjectively) shared
code(s) for interpretations of the sign, and (6) the (intersubjectively) understood
referent(s) of the sign, located in the micro- or macro-context of the interaction between the speech- or communicative-event participants, such as the speaker and
addressee. This is, of course, the contextual, pragmatic matrix of communication,
including language, which was articulated in the Praguean functionalist theory of
Jakobson (1960), mentioned in Section 4. As part of the code, linguistic structure is
embedded in this functional matrix of communication. In other words, structure
emerges through communicative practices; there is no structure (langue) without
pragmatics (parole). According to this Jakobsonian theory, as noted earlier, there
are six basic functions of communication, corresponding to the six aspects of the
contextual matrix: viz., (1) emotive, (2) conative, (3) phatic, (4) poetic, (5) metalingual (metalinguistic), and (6) referential functions.11 Although linguistic structure, as part of the code, is primarily used to interpret the sign (token) to identify
(i.e., intersubjectively construct) its referent(s) and thus primarily involved in the
metalinguistic, poetic, and referential functions (respectively corresponding to the
code, sign, and referents), it inevitably interacts with the other functions as well,
because it exists only as part of the code pragmatically used for intersubjective interpretations of the sign indexically exchanged in the context of communication.
Thus, language, including linguistic structure, i.e., (morpho)phonology, (morpho)syntax, and semantics, should be understood, theorized, and analyzed by tak-

The rise of pragmatics: a historiographic overview

153

ing into proper consideration the six functions, as well as the six aspects, of communicative practices (cf. Silverstein 1976a, 1976b, 1993).
But, of course, the contextual, functional matrix of communication, articulated
in the six-functions model, is not limited to, or centered around, linguistic structure.
For example, the code includes not just linguistic structure, but also maxims, interpretive rules of thumb, passing theories, metapragmatic frames, and models
of interpretation vis--vis (not necessarily linguistic) signs, including gestures, facial expressions, and other paralinguistic phenomena (cf. Bolinger 1968), all of
which occur in specific, sociohistorical contexts of communicative practices (see
Section 7; cf. Mey [1993] 2001: 206235). And such contextualized practices
index12 not just referents (in the referential function) and the semantic and pragmatic codes being used (in the metalinguistic function). They also index the social identities of, and power-relations among, communicative-event participants
(e.g., in the emotive and conative functions). That is, contextualized practices
index not just referents, but also social identities and power-relations, which can be
read in the emergent texts of sociocultural practices of communication. Such
texts are not to be understood as static objects, but as the dynamically emergent,
socioculturally interpretable organizations of referential and non-referential (interactional) practices (mainly corresponding to the poetic, phatic, and metapragmatic functions), that is, organizations studied in conversation analysis (cf.
adjacency pairs; Levinson 1983: 284370), discourse analysis (cf. frames of interaction; Goffman 1974), linguistic anthropology (cf. Silverstein and Urban 1996;
Blommaert 2005), and, of course, pragmatics (cf. Mey [1993] 2001).
Thus, in a sense, the Praguean/Jakobsonian functional theory provided a general framework in which todays empirical approaches to language, listed immediately above, operate (also see Section 2). As this indicates, the rise of structuralism
in the late nineteenth and early twentieth centuries served to prepare the proper theorization of the functional, contextual matrix of language, which would emerge in
the 1970s after the formalist lapse from the 1940s to 60s. This lapse, however,
made obscure the pragmatic, functionalist foundation of structural linguistics. Indeed, by the late 1960s, when Generative Semanticists, who were immediate predecessors of linguistic pragmaticians, started to rebel against formalist structuralism (Chomskyan generativism) and delve into the realm of pragmatics, North
American linguistics had become more closely linked to the philosophy of language, especially logico-analytic philosophy, than to the early twentieth-century
functionalist traditions of structuralism, so that, in trying to find theoretical tools to
deal with pragmatic phenomena, Generative Semanticists discovered and adopted
the theories of speech act, implicature, and presupposition, articulated by Ordinary
Language philosophers like Austin, Grice, and Strawson (see Section 3, above; cf.
Levinson 1983). The publication of Austins (1962; second edition, 1975) William
James Lectures, originally delivered in 1955 at Harvard, which is located in the
same city as MIT (Cambridge, Mass.), i.e., the home of generativism, made the

154

Wataru Koyama

theories of the Ordinary Language philosophers more visible to generativists interested in pragmatic matters (also cf. Grice 1967; Chapman 2005: 100). And Searle
(1969) provided a desideratum, i.e., the desired systematization of speech act theory, making it amenable to the linguists formal, syntactic treatment of pragmatic
phenomena.13 As Generative Semantics yielded to linguistic pragmatics in the
1970s, the early years of the new discipline were marked by the preoccupation with
the issues articulated by the Ordinary Language philosophers, and many of the culture-specific assumptions of their philosophy, i.e., post-Lockean modern British
philosophy, concerning human communication, seem to have been directly borrowed into linguistic pragmatics.
Here, recall that Locke, a founder of modern British empiricism, discussed,
like Saussure, the arbitrary nature of human language (communication), but, unlike the latter, Locke meant by the term the volitional, intentional, willful character of the individuals use of language, whereas Saussure meant the conventional
(non-volitional), extensionally unmotivated (intensional) yet socially existing (i.e.,
Durkheimian) association of the signifiant and the signifi. As this indicates, British and Anglo-American philosophy, especially the empiricist tradition thereof
(see Section 2), has tended to theorize communication (more generally, human
practice) in terms of volitional, intentional actions or, in sociological terms of Max
Weber, the purposive function (cf. Silverstein 1976b; Apel 1994: 63). Ordinary
Language philosophy, an empiricist philosophy of language and communication
concerned with what we do with words and what is meant, is no exception. It
sees language and communication as purposive activities, as explicitly articulated
by G. E. M. Anscombe, G. H. von Wright, and Grice, who tried to explain the social, interactional fact of communication in terms of what we intend to do or what
we intend to mean, that is, individuals conscious intention (cf. Chapman 2005).
Notwithstanding, as Weber and other social scientists have shown, actions have
consequences (perlocutionary effects) which are unintended by agents (subjects),
and which may be more important, in terms of historical consequences (i.e., pragmatic effects), than actors mere intentions (cf. Silverstein 1976b). Further, any
practice at least partially consists of unintentional, non-conscious elements; obviously, we are not conscious of everything we do. For example, we, as ordinary
language users, are usually more aware of surface segments (continuous, segmentable units), such as phonemes, words, and expressions, than of allophones and
discontinuous syntactic units (cf. Lucy 1992); yet, the latter two are often part of
dialectal/sociolectal varieties and thus index their users social identities, pertaining to class, gender, status, etc., and power-relations among them, as can be readily
seen in the sociolinguistic literature, Labovian or otherwise. Thus, there are nonpurposive pragmatic functions in discourse, which are too important to be left out
in any theory of pragmatics.
In addition, cross-cultural comparisons suggest that many societies place much
less value on speakers intentions in the evaluation of (sociocultural) meanings

The rise of pragmatics: a historiographic overview

155

of their linguistic practices than the culture of Ordinary Language philosophers


does, a point poignantly raised and substantiated by linguistic anthropologists,
who have been critical of the intentionalist and individualist bias of the modern
Standard Average European speech act theories ever since the 1970s and early
80s.14 Subsequently, partly as a result of this cross-cultural (anthropological)
critique, linguistic pragmatics has come to appreciate and focus on cross-cultural
differences in speech acts and, more generally, linguistic practices and their
metapragmatic understandings, the latter now discussed under the rubric of language ideology.15
No less important, the individualist assumption of the speech act theories was
undermined by empirical studies of discourse, which showed that the meaning(s)
of an action (speech act) cannot be determined out of context, i.e., without considering a series of interactions which contains the action (cf. Hancher 1979; Levinson 1983). This means that, to the extent that the meaning(s) of an action is
determinable, it is determined by the contextualized interaction of speech-event
participants, not solely by the individual speakers intention. Contributions of conversation analysis (cf. Levinson 1983: 284370) and discourse analysis (cf. Mey
[1993] 2001: 191), focusing their attention on interaction rather than on solitary individual actors minds, have been important in this interactional turn (social
turn) of pragmatics, which has characterized the discipline since the 1980s or so
(cf. Thomas 1996). Pragmatics today is much more interaction- or discourse-centric than in the 1970s, when the new discipline was more intention- and individualcentric under the influence of Ordinary Language philosophy and modern British
empiricism.

7.

Genealogy of performative utterances and other multi-functional


co-incident happenings in the speech event

Let us finally move to the question of performativity, another central theme in linguistic pragmatics. Clearly, this is a complex notion, only a part of which is accounted for by the creative function (efficacy), discussed in Section 3. That is, in
addition to the creative function, performativity involves self-referentiality (cf.
Recanati 1979),16 which is itself a composite notion, made up of reflexitivity and
(transparent) referentiality. The latter means the (relatively) transparent relationship between the referential function (what is said) and the non-referential, interactional function (what is done), as observed when we use a referentially explicit
metapragmatic utterance, such as When I said that, I meant it to be a compliment
(cf. Caffi 1984; Mey [1993] 2001: 173205). On the other hand, reflexivity as
such is just the mutual indexing of some semiotic elements, linguistic or non-linguistic, co-occurring in the event of communication. For example, a pointing gesture and a token of the deictic expression this book, occurring together, mutually

156

Wataru Koyama

index each other and function reflexively, as the gesture and the linguistic utterance
coincide in the same context (deictic, indexical field; see Section 2). Mutual, reflexive indexing also obtains when a defiant smile, or a serious demeanor, occurs
with an utterance Im sorry, or when a judge, properly dressed or completely undressed, utters I declare you guilty. Clearly, these are examples of what Gregory
Bateson ([1955] 1972) called metacommunication (cf. Ruesch and Bateson
1951; Caffi 1984) and what Gumperz (1982) called contextualization cue (cf.
Auer and Luzio 1992; Duranti and Goodwin 1992), notions coming out of the anthropological, psychiatric, and ethological traditions, independently of Austins
discussion of performatives.
Austins (1962) notion of performativity, following the discussions by the phenomenologist Adolf Reinach (see note 7) and others in the study of law, did not
clearly separate these two dimensions of performativity or explore each dimension in depth. Originally, linguistic pragmatics was under the influence of Austins
way of formulating the problem, but it has gradually moved to see that the problem
of performativity should be dealt with by separating out different principles underlying the problem, such as the principle of metapragmatics, which is becoming a
major topic in our discipline (cf. Ducrot 1984; Verschueren 1999; Mey [1993]
2001). Like the cognate notion metacommunication, the concept of metapragmatics came out of the anthropological literature, in particular Silverstein
(1976b), who coined the term to designate the pragmatic (vs. semantic) aspect of
the metalingual function,17 articulated by his teacher, Jakobson (1957, 1960),
who, in turn, derived it from the discussion of metalanguage (vs. object language) by the mathematician David Hilbert (18621943), Kurt Gdel (190678),
Tarski and other logicians, e.g., Russell, the early Wittgenstein, and Carnap (cf.
Apel 1994). The metapragmatic function obtains when a sign, be it an utterance, a
gesture, or whatever, indexes another sign (speech, non-linguistic action, etc.) in
context. When the two signs co-occur (coincide) in the same context, they are said
to show reflexively calibrated metapragmatics (cf. Lucy 1993; Mey [1993]
2001). According to Silverstein (1993), the reflexive metapragmatic function
serves to create texts, be they referential or non-referential (see Section 6 on
text). For instance, the reflexive co-indexing of referential signs, such as (thematic) anaphor and its (rhematic) antecedent, in the same unit of discourse creates
referential textuality, as studied in Praguean functional structuralism, text linguistics, and discourse analysis (cf. Halliday and Hasan 1976). Similarly, the reflexive co-indexing of non-referential signs, such as the two units of an adjacency
pair (e.g., request-compliance or -rejection; cf. Levinson 1983: 284370), in the
same phase of discourse creates interactional, non-referential textuality.
Clearly, the idea of textualization came from, among other places, Bhlers
notion of Sprachwerk language work and, in particular, Jakobsons poetic function (see Section 6), i.e., the communicative function that is focused on messages
(i.e., signs) occurring in the context of interaction and thus gives textual organiz-

The rise of pragmatics: a historiographic overview

157

ation to them, as can be most clearly seen in parallelisms and repetitions abundantly observed in poetry and other rhetorically elaborate verbal arts (e.g., folklore
and myths; cf. Bauman 2004). Importantly, basic units of poetic text are often nonreferential segments, such as meters and rhymes, in contradistinction to referential
units such as rhemes, themes, and syntactic segments like phrases, clauses, sentences. Thus, poetic texts are, at least partially, non-referential texts, i.e., texts
of non-referential interaction. And hence, poetic texts tend to become endowed
with non-referential meanings, such as emotional (emotive) and persuasive
(conative) forces (functions). The same characteristics are found in rituals,
which also typically show the abundance of parallelisms and repetitions in the linguistic and non-linguistic, interactional modes (cf. Caton 1990). To be sure, in ordinary conversations too, we also find such interactional poetic parallelisms and
repetitions especially when they contain what Goffman (1967) called interaction
rituals (cf. greetings and other adjacency pairs, mentioned above). Yet, the characteristic is most salient in rituals, in which, we should recall, Austin (1962) discovered (explicit) performatives: e.g., christening rituals, legal rituals, etc. Performative utterances are the part of rituals characterized by the use of little ritual
one-liners, i.e., performative sentences, which are, thus, socially recognized formulaic types, recursively co-occurring with the non-linguistic elements of ritual
settings (cf. Volosinov [1929] 1973; Silverstein 1993).
Thus, performatives show reflexivity, i.e., the reflexive metapragmatic
function, which is observed more broadly in rituals, poetry, metacommunication,
contextualization cues, etc. On the other hand, performatives involve another
principle, that is, referentiality, or relatively pronounced degrees of transparency between the referential function (what is said) and the non-referential, interactional function (what is done), as noted above. Clearly, it is this principle that
has been the principal focus of pragmatic discussions on performativity and
speech act, most clearly observable in Searles theory of indirect speech act and
Grices theory of implicature. These two notions were formulated to account
for non-explicit or implicit relationships between the referential and non-referential functions of utterances, and they have been main themes of linguistic pragmatics since the 1970s. Much of linguistic pragmatics still operates with these
concepts. For example, Brown and Levinsons ([1978] 1987) politeness theory
and, inter alia, Sperber and Wilsons Relevance Theory seem squarely located in
this tradition.
Clearly, the theme of referential explicitness vs. implicitness goes back to the
old tradition of rhetoric, where literal expressions and tropes were understood in
such terms (cf. Ricur [1975] 1977). When Austin discovered non-referential
functions in explicit performatives, he did not depart from this age-old tradition, as
he focused on non-referential functions only insofar as they appear together with
the referential function, as in performatives, where the two kinds of function
happen to coincide. Just as in ancient rhetoric, Austin and his followers moved

158

Wataru Koyama

from the referential function, as studied in logic and formalist structuralism, to


non-referential functions. That is, Austins was a reference-centric approach to
pragmatic functions, the severe limit of which is indicated by the comparison with
multi-functionalist approaches to pragmatics found in, e.g., Jakobsons theory (see
Section 6), its developments in linguistic anthropology (see Section 2 and immediately above), and the theory of pragmatic act, advanced in social pragmatics (cf.
Mey [1993] 2001: 206235; also see Section 3, above). The rise of social pragmatics and linguistic anthropology in linguistic pragmatics since the 1990s seems
to suggest that the discipline is not only recovering but also further elaborating the
multi-functionalist, social interactionist insights of the older traditions of linguistic
functionalism, which blossomed from the 1880s to the 1930s (see Section 2).

Notes
1. More specifically, Reids philosophy influenced Benjamin Smart (17861872), Adolphe
Garnier (180164), and Henry Sidgwick (18381900), and developed into Charles
Peirces (18391914) semiotic pragmati(ci)sm, Charles Morriss (190179) behaviorist
pragmatism, Sidgwicks student George Stouts (18601944) empiricist and behavioristic philosophy, the latters student G. E. Moores (18731958) commonsense empiricism, and Grices (191388) theory of non-natural meaning and implicature in the
nineteenth and twentieth centuries (cf. Nerlich and Clarke 1996: 2021, 105106; Chapman 2005: 22, 25, 7172).
2. The German tradition ran from Bernhardi (17691820) and Wilhelm von Humboldt
(17671835) to Philipp Wegener (18481916), the later Karl Brugmann (18491919),
Anton Marty (18471914), and Bhler, whereas the French one ran from the Idologues
to Bral, (18321915), Meillet (18661936), Mauss (18721950), Bally (18651947),
Guillaume (18831960), and Benveniste, to mention only a few of the major figures.
3. This was preceded by the medieval trivium (artes sermocinales), i.e., grammar, logic,
and rhetoric. Note that Morriss predecessor Peirce was thoroughly familiar with Scholastic philosophy.
4. The Journal of Pragmatics was launched in 1977 (cf. Haberland and Mey 1977). Subsequently, major textbooks of pragmatics became published in the early 1980s (cf. Leech
1983; Levinson 1983), the International Pragmatics Association (IPrA) was founded in
1986, and Pragmatics was started in 1991, while the second wave of major, and more
Continental-European (vs. Anglo-American), textbooks appeared in the 1990s (cf. Mey
[1993] 2001; Verschueren 1999).
5. These ideas had been suggested and, in part, advanced by Philipp Wegener (18481916).
6. For instance, before Condillac (171480), Grammaire gnrale et raisonne had made
the distinction between dictum and modus, roughly corresponding to proposition and
force, respectively.
7. Also, note that Adolf Reinach (18831917), whose theory of social acts was one of the
precursors of Austins theory of speech acts and performativity, was a phenomenologist. Reinach was a student of Husserls and, with Johannes Daubert (18771947), belonged to the Munich School of phenomenology. The Scottish commonsense philos-

The rise of pragmatics: a historiographic overview

8.

9.

10.

11.
12.
13.
14.
15.
16.

17.

159

ophy of Thomas Reid (see note 1), which underlay Ordinary Language philosophy and
much of twentieth-century British empiricism, also exercised influence over the early
phenomenologists, such as Brentano (18381917), one of the predecessors of Reinach.
(Reinach was under the influence of the (individualistic) functionalist Anton Marty
(18471914), a student of Brentanos.) Further, it should be underlined that Reid, Reinach, and Austin all used the philosophy of law as a common source of reference for their
philosophies of language (see Section 7, below; cf. Mulligan 1987: 3334; Nerlich and
Clarke 1996: 12, 104, 211215).
Peirce and Husserl were both influenced by Bolzano (17811848), who had been, in
turn, influenced by Johann Lambert (172877) and Locke; the latter two also directly influenced Peirce (cf. Nerlich and Clarke 1996: 21).
Bhler was also closely associated with the Vienna Circle, including Carnap. As this
suggests, and as Friedman (2000) has shown, logical positivism (of the Vienna Circle)
and phenomenology shared common themes (e.g., intersubjectivity, construction of the
world, etc.) and common roots (e.g., neo-Kantianism). For instance, the early Carnap
made appeal to Gestalt psychology and epistemological holism, which were intimately
associated with structuralism, phenomenology, and especially, the neo-Kantian notion
of constitution.
Generally speaking, the linguistic turn in philosophy was made up of (1) the Russellian critique of ordinary language by means of logico-mathematical analysis, (2) the
Carnapian critique of metaphysical dogmas (later, epistemology), such as realism and
phenomenalist positivism, by means of logic, and (3) the Austinian critique of philosophy, including logic, by means of the commonsense analysis of ordinary language.
Austins philosophy of language was anticipated by Scottish commonsense philosophy,
especially Reid, who influenced Peirce, Moore, and Grice, among others (see note 1,
above).
See Section 7, below, for the metalinguistic (more specifically, metapragmatic) and
poetic functions.
See Section 2 for indexicality.
Recall Rosss and other Generative Semanticists performative hypothesis (cf. Sadock 1974; Levinson 1983: 247263, 271273).
See Section 3, above; cf. Silverstein 1976b; Rosaldo 1982; Du Bois 1987; Duranti 1993.
Cf. Schieffelin, Woolard, and Kroskrity [1992] 1998; Verschueren 1999; Mey [1993]
2001: 173205, 262288; Jaworski, Coupland, and Dariusz 2004.
As Nerlich and Clarke (1996) have pointed out, the property of self-reference (soirfrence) had been discovered, in 1929, by Erwin Koschmieder (18951977), who
started to discuss Koinzidenzfall the case of coincidence in the 1940s.
See Section 6, above.

References
Aarsleff, Hans
1982
From Locke to Saussure: Essays on the Study of Language and Intellectual
History. Minneapolis, MN: University of Minnesota Press.
Agha, Asif
2007
Language and Social Relations. Cambridge: Cambridge University Press.

160

Wataru Koyama

Apel, Karl-Otto
1994
Karl-Otto Apel: Selected Essays, Volume 1: Towards a Transcendental Semiotics. Edited by Eduardo Mendieta. Atlantic Heights, NJ: Humanities Press.
Auer, Peter and Aldo di Luzio (eds.)
1992
The Contextualization of Language. Amsterdam: Benjamins.
Austin, John L.
1962
How to Do Things with Words: The William James Lectures Delivered at Harvard University in 1955. Edited by James O. Urmson. Oxford: Clarendon
Press.
Bar-Hillel, Yeoshua
1946
Analysis of correct language. Mind 55: 328340.
Bar-Hillel, Yeoshua
1954
Indexical expressions. Mind 63: 359376.
Bateson, Gregory
[1955] 1972 A theory of play and fantasy. In: Gregory Bateson, Steps to an Ecology of
Mind, 177193. New York: Ballantine Books.
Bauman, Richard
2004
A World of Others Words: Cross-Cultural Perspectives on Intertextuality. Oxford/Malden, MA: Blackwell.
Bauman, Richard and Joel Sherzer (eds.)
[1974] 1989 Explorations in the Ethnography of Speaking, 2nd edition. Cambridge:
Cambridge University Press.
Benveniste, Emile
1966
Problmes de linguistique gnrale, I. Paris: Gallimard.
Blommaert, Jan
2005
Discourse: A Critical Introduction. Cambridge: Cambridge University Press.
Bolinger, Dwight
1968
Aspects of Language. New York: Harcourt Brace Jovanovich.
Bourdieu, Pierre
1980
Le sens pratique. Paris: Minuit.
Brown, Penelope and Stephen C. Levinson
[1978] 1987 Politeness: Some Universals in Language Usage. Cambridge: Cambridge
University Press.
Bublitz, Wolfram and Axel Hbler (eds.)
2007
Metapragmatics in Use. (Pragmatics and Beyond, n. s. 165.) Amsterdam: Benjamins.
Bhler, Karl
[1934] 1990 Theory of Language: The Representational Function of Language. Amsterdam: Benjamins. [Sprachtheorie: Die Darstellungsfunktion der Sprache.
Jena: Fischer.]
Butler, Judith
1990
Gender Trouble: Feminism and the Subversion of Identity. London: Routledge.
Caffi, Claudia (ed.)
1984
Special Issue on Metapragmatics. Journal of Pragmatics 8(4).
Cameron, Deborah and Don Kulick
2003
Language and Sexuality. Cambridge: Cambridge University Press.

The rise of pragmatics: a historiographic overview

161

Carnap, Rudolf
[1934] 1937 The Logical Syntax of Language. London: Kegan Paul. [Logische Syntax
der Sprache. Vienna: Springer.]
Cassirer, Ernst A.
1945
Structuralism in modern linguistics. Word 1: 99120.
Caton, Steven C.
1987
Contributions of Roman Jakobson. Annual Review of Anthropology 16: 223
260.
Caton, Steven C.
1990
Peaks of Yemen I Summon: Poetry as Cultural Practice in a North Yemeni
Tribe. Berkeley/Los Angeles/London: University of California Press.
Chapman, Siobhan
2005
Paul Grice, Philosopher and Linguist. Basingstoke/New York: Palgrave Macmillan.
Coffa, J. Alberto
1991
The Semantic Tradition from Kant to Carnap: To the Vienna Station. Cambridge: Cambridge University Press.
Comrie, Bernard
[1981] 1989 Language Universals and Linguistic Typology: Syntax and Morphology,
2nd edition. Chicago: University of Chicago Press.
Coupland, Nikolas, Srikant Sarangi and Christopher N. Candlin (eds.)
2001
Sociolinguistics and Social Theory. Harlow: Pearson Education.
Davis, Steven (ed.)
1991
Pragmatics: A Reader. Oxford/New York: Oxford University Press.
Donnellan, Keith S.
1966
Reference and definite descriptions. Philosophical Review 75: 281304.
Donnellan, Keith S.
1978
Speaker reference, descriptions and anaphora. In: Peter Cole (ed.), Syntax and
Semantics, Volume 9: Pragmatics, 4768. New York: Academic Press.
Du Bois, John W.
1987
Meaning without intention: Lessons from divination. IPrA Papers in Pragmatics 1: 80122.
Ducrot, Oswald
1984
Langage, mtalangage et performatifs. In: Le dire et le dit, 117148. Paris:
Minuit.
Duranti, Alessandro
1993
Intentions, self, and responsibility. In: Jane H. Hill and Judith T. Irvine (eds.),
Responsibility and Evidence in Oral Discourse, 2447. Cambridge: Cambridge University Press.
Duranti, Alessandro and Charles Goodwin (eds.)
1992
Rethinking Context: Language as an Interactive Phenomenon. Cambridge:
Cambridge University Press.
Fillmore, Charles J.
1975
Santa Cruz Lectures on Deixis, 1971. Bloomington, IN: Indiana University
Linguistics Club.
Foucault, Michel
1966
Les mots et les choses: une archologie des sciences humaines. Paris: Gaillmard.

162

Wataru Koyama

Friedman, Michael
2000
A Parting of the Ways: Carnap, Cassirer, and Heidegger. Chicago/La Salle, IL:
Open Court.
Gazdar, Gerald
1979
Pragmatics: Implicature, Presupposition, and Logical Form. New York: Academic Press.
Giddens, Anthony
[1976] 1993 New Rules of Sociological Method: A Positive Critique of Interpretative
Sociologies, 2nd edition. Stanford: Stanford University Press.
Goffman, Erving
1967
Interaction Ritual: Essays on Face-to-Face Behavior. New York: Anchor
Books.
Goffman, Erving
1974
Frame Analysis: An Essay on the Organization of Experience. New York:
Harper and Row.
Goffman, Erving
1981
Forms of Talk. Philadelphia: University of Pennsylvania Press.
Goffman, Erving
1983
Felicitys condition. American Journal of Sociology 1: 153.
Grice, H. Paul
1967
Logic and conversation. Unpublished manuscript of the William James Lectures, Harvard University. Published in Studied in the Way of Words. Cambridge, MA: Harvard University Press [1989].
Gumperz, John J.
1982
Discourse Strategies. Cambridge: Cambridge University Press.
Gumperz, John J. and Dell Hymes (eds.)
[1972] 1986 Directions in Sociolinguistics: The Ethnography of Communication. Oxford: Basil Blackwell.
Haberland, Hartmut and Jacob L. Mey
1977
Editorial: Pragmatics and linguistics. Journal of Pragmatics 1(1): 116.
Halliday, Michael A. K. and Ruqaiya Hasan
1976
Cohesion in English. London: Longman.
Hancher, Michael
1979
The classification of cooperative illocutionary acts. Language in Society 8:
114.
Hanks, William F.
1990
Referential Practice: Language and Lived Space among the Maya. Chicago:
University of Chicago Press.
Harris, Randy Allen
1993
The Linguistic Wars. Oxford: Oxford University Press.
Heritage, John
1984
Garfinkel and Ethnomethodology. Cambridge: Polity Press.
Holenstein, Elmar
1977
Jakobsons contribution to phenomenology. In: Daniel Armstrong and Cornelis H. van Schooneveld (eds.), Roman Jakobson: Echoes of his Scholarship,
145162. Lisse: The Peter de Ridder Press.

The rise of pragmatics: a historiographic overview

163

Huang, Yan
2007
Pragmatics. Oxford/New York: Oxford University Press.
Irvine, Judith T.
1996
Shadow conversations: The indeterminacy of participant roles. In: Michael
Silverstein and Greg Urban (eds.), Natural Histories of Discourse, 131159.
Chicago: University of Chicago Press.
Jakobson, Roman
1957
Shifters, verbal categories, and the Russian verb. Cambridge, MA: Harvard
University Russian Language Project.
Jakobson, Roman
1960
Closing statement: Linguistics and poetics. In: Thomas A. Sebeok (ed.), Style
in Language, 350377. Cambridge, MA: MIT Press.
Jakobson, Roman
1973
Main Trends in the Science of Language. London: George Allen and Unwin.
Jaworski, Adam, Nikolas Coupland and Galasinski Dariusz (eds.)
2004
Metalanguage: Social and Ideological Perspectives. Berlin: Mouton de
Gruyter.
Jespersen, Otto
1922
Language. London: Allen and Unwin.
Kaplan, David
1978
Dthat. In: Peter Cole (ed.), Syntax and Semantics, Volume 9: Pragmatics,
221243. New York: Academic Press.
Karttunen, Lauri
1974
Presupposition and linguistic context. Theoretical Linguistics 1: 181194.
Koyama, Wataru
2000
How to be a singular scientist of words, and worlds, and other (possibly) wonderful things. Journal of Pragmatics 32: 651686.
Kripke, Saul
1972
Naming and Necessity. Cambridge, MA: Harvard University Press.
Kripke, Saul
1979
Speakers reference and semantic reference. In: Peter A. French, Theodore E.
Uehling Jr. and Howard K. Wettstein (eds.), Contemporary Perspectives in the
Philosophy of Language, 627. Minneapolis, MN: University of Minnesota
Press.
Kuklick, Henrika
1991
The Savage Within: The Social History of British Anthropology, 18851945.
Cambridge: Cambridge University Press.
Lakoff, Robin T.
1973
Language and womans place. Language in Society 2(1): 4580.
Lakoff, Robin T.
1989
The way we were. Journal of Pragmatics 13: 939988.
Leech, Geoffrey N.
1983
Principles of Pragmatics. London: Longman.
Levinson, Stephen C.
1983
Pragmatics. Cambridge: Cambridge University Press.
Locke, John
[1689] 1975 Essay on Human Understanding. Edited by Peter H. Nidditch. Oxford:
Oxford University Press.

164

Wataru Koyama

Lucy, John A.
1992
Language Diversity and Thought: A Reformulation of the Linguistic Relativity
Hypothesis. Cambridge: Cambridge University Press.
Lucy, John A. (ed.)
1993
Reflexive Language: Reported Speech and Metapragmatics. Cambridge University Press: Cambridge.
Lyons, John
1977
Semantics (2 volumes). Cambridge: Cambridge University Press.
Merleau-Ponty, Maurice
1945
Phnomnologie de la perception. Paris: Gallimard.
Merleau-Ponty, Maurice
1960
Signes. Paris: Gallimard.
Mey, Jacob L.
1979
Zur kritischen Sprachtheorie. In: Jacob L. Mey (ed.), Pragmalinguistics:
Theory and Practice, 411434. The Hague: Mouton.
Mey, Jacob L.
1985
Whose Language?: A Study in Linguistic Pragmatics. Amsterdam: Benjamins.
Mey, Jacob L.
[1993] 2001 Pragmatics: An Introduction, 2nd edition. Oxford/Malden, MA: Blackwell.
Morris, Charles W.
1938
Foundations of the theory of signs. In: Otto Neurath, Rudolf Carnap, and
Charles W. Morris (eds.), International Encyclopedia of Unified Science,
Volume 1, No. 2, 77137. Chicago: University of Chicago Press.
Mulligan, Kevin (ed.)
1987
Speech Act and Sachverhalt: Reinach and the Foundations of Realist Phenomenology. Dordrecht: Martinus Nijhoff.
Nerlich, Brigitte
1992
Semantic Theories in Europe: From Etymology to Contextuality. Amsterdam:
Benjamins.
Nerlich, Brigitte and David D. Clarke
1996
Language, Action, and Context: The Early History of Pragmatics in Europe
and America, 17801930. Amsterdam: Benjamins.
Peirce, Charles S.
[18931910] 1955 Logic as semiotic: The theory of signs. In: Justus Buchler (ed.),
Philosophical Writings of Peirce, 98119. New York: Dover.
Pike, Kenneth Lee
1967
Language in Relation to a Unified Theory of the Structure of Human Behavior,
2nd, rev. edition. The Hague: Mouton.
Pressman, Jon F.
1994
Pragmatics in the late twentieth century: Countering recent historiographic
neglect. Pragmatics 4(4): 461489.
Putnam, Hilary
1975
Philosophical Papers, Volume 2: Mind, Language, and Reality. Cambridge:
Cambridge University Press.
Recanati, Franois
1979
La transparence et lnonciation: pour introduire la pragmatique. Paris:
Seuil.

The rise of pragmatics: a historiographic overview

165

Richardson, Alan
1998
Carnaps Construction of the World: The Aufbau and the Emergence of Logical Empiricism. Cambridge: Cambridge University Press.
Ricur, Paul
[1975] 1977 The Rule of Metaphor: Multi-Disciplinary Studies of the Creation of
Meaning in Language. Toronto: University of Toronto Press.
Rorty, Richard M. (ed.)
1967
The Linguistic Turn: Recent Essays in Philosophical Method. Chicago: University of Chicago Press.
Rosaldo, Michelle Z.
1982
The things we do with words: Ilongot speech acts and speech act theory in philosophy. Language in Society 11: 203237.
Ruesch, Jrgen and Gregory Bateson
1951
Communication: The Social Matrix of Psychiatry. New York: Norton.
Sadock, Jerrold M.
1974
Toward a Linguistic Theory of Speech Acts. New York: Academic Press.
Sapir, Edward
1933
La ralit psychologique des phonmes. Journal de Psychologie Normale et
Pathologique 30: 247265.
Schieffelin, Bambi B., Kathryn Woolard and Paul V. Kroskrity (eds.)
1998
Language Ideologies: Practice and Theory. New York/Oxford: Oxford University Press. [Originally, 1992, Special Issue on Language Ideologies,
Pragmatics 2(3).]
Searle, John R.
1969
Speech Acts: An Essay in the Philosophy of Language. Cambridge: Cambridge
University Press.
Searle, John R.
1983
Intentionality: An Essay in the Philosophy of Mind. Cambridge: Cambridge
University Press.
Silverstein, Michael
1976a
Hierarchy of features and ergativity. In: Robert M. W. Dixon (ed.), Grammatical Categories in Australian Languages, 112171. Canberra: Australian Institute of Aboriginal Studies.
Silverstein, Michael
1976b
Shifters, linguistic categories, and cultural description. In: Keith H. Basso and
Henry A. Selby (eds.), Meaning in Anthropology, 1155. Albuquerque, NM:
University of New Mexico Press.
Silverstein, Michael
1993
Metapragmatic discourse and metapragmatic function. In: John A. Lucy (ed.),
Reflexive Language: Reported Speech and Metapragmatics, 3358. Cambridge: Cambridge University Press.
Silverstein, Michael and Greg Urban (eds.)
1996
Natural Histories of Discourse. Chicago: University of Chicago Press.
Stocking, Jr., George W. (ed.)
1996
Volksgeist as Method and Ethic: Essays on Boasian Ethnography and the
German Anthropological Tradition. (History of Anthropology, Volume 8.)
Madison, WI: University of Wisconsin Press.

166

Wataru Koyama

Tannen, Deborah (ed.)


1986
Framing in Discourse. Oxford/New York: Oxford University Press.
Thomas, Jenny
1996
Meaning in Interaction: An Introduction to Pragmatics. London: Longman.
Verschueren, Jef
1999
Understanding Pragmatics. London: Arnold.
Volosinov, Valentin N.
[1929] 1973 Marxism and the Philosophy of Language. Cambridge, MA: Harvard University Press.
Williams, Glyn
1992
Sociolinguistics: A Sociological Critique. London: Routledge.

6.

Semiotic foundations of pragmatics


Winfried Nth

1.

Semiotics as a general framework of pragmatics

Levinson (1983: 15) gives the standard account of how semiotics, the general
study of signs, is considered to be fundamental to linguistic pragmatics. The author of the Cambridge Textbook on Pragmatics begins with Charles W. Morris
(19011979), the founder of a general semiotics first outlined in 1938. Morriss
science of signs (1938: 12) is both broader and narrower than the scope of modern linguistic pragmatics.1 It is broader insofar as it is not restricted to the study of
verbal communication but deals with signs in all their forms and manifestations,
whether in animals or men, whether normal or pathological, whether linguistic
or nonlinguistic, whether personal or social (1964: 1), acoustic, visual, olfactory,
gustatory, or tactile. It is also broader since pragmatics, as conceived by Morris,
comprises branches of language studies which today fall into the domain of other
disciplines of linguistics, such as psycholinguistics, sociolinguistics, neurolinguistics, and much besides (Levinson 1983: 2). It is narrower insofar as Morris
aims at a science of signs on a biological basis and specifically within the framework of the science of behavior (1946: 80) indebted to the positivist paradigm of
psychological and social behaviorism. As Levinson (1983: 24) points out, the
scope of pragmatics became still narrower in the frameworks of logical positivism
and analytical philosophy of ordinary language: Carnap (1942) restricts pragmatics to the logical study of language use, for Bar-Hillel (1954) pragmatics is the
study of language use involving indexical words, and Montague (1970: 68) conceives pragmatics as a contextual theory of truth conditions founded on intensional
logic.
Sign behavior, according to Morris, involves three main factors: that which
acts as a sign [the sign vehicle], that which the sign refers to [the designatum], and
that effect on some interpreter in virtue of which the thing in question is a sign to
that interpreter [the interpretant] (1938: 3). Based on this triad, Morris (1938:
67) defines semiotics as a field of study of the following three domains corresponding to three well-known branches of modern linguistics: syntax (or syntactics), the study of the relation between sign vehicles, semantics, the study of the relations between sign vehicles and their designata, and pragmatics, the study of the
relation between sign vehicles and their interpreters (cf. Posner 1985).
20th century semiotics has developed in many directions, and some of the
schools of semiotic research have largely ignored the pragmatic dimension of sign
processes. This is especially true of structuralist semiotics in the tradition of Fer-

168

Winfried Nth

dinand de Saussure (18571913) and Louis Hjelmslev (18991965), whose focus


is entirely on language as a system and not on language use. It is no wonder, therefore, that Jacob L. Mey, once Hjelmslevs PhD student in Copenhagen, in his Introduction to Pragmatics, disregards semiotics entirely as a framework of linguistic pragmatics and that Dan Sperber and Deirdre Wilson, while reducing
semiotics to a theory of codes, dismiss its relevance to pragmatics altogether
(Sperber and Wilson 1986: 7, 2).
The structuralist and behaviorist heritage of semiotics has been an important
obstacle to a more fruitful interdisciplinary dialogue between general semiotics
and linguistic pragmatics. For a long time, it has impeded the spread of the seminal
ideas which the semiotic writings of Charles S. Peirce (18391914) offer to linguistics in general (Nth 2002a) and pragmatics in particular. Only relatively recently has the relevance of these ideas to pragmatics become acknowledged,2 but
so far mostly only by philosophers of language and hardly by linguists (e. g., Levinson 1983: 1, fn. 1).
When Morris conceived the triadic subdivision of semiotics into syntactics, semantics, and pragmatics, he was clearly influenced by Peirce. Peirce had also proposed a triadic framework for the study of signs, but the three branches of semiotics which Peirce set up describe a much wider field of study. Peirce was
inspired by the medieval canon of the three liberal arts, the trivium of grammar,
logic, and rhetoric.3 In contrast to the Saussurean tradition, in which linguistics is a
branch of semiology (cf. Nth 2000: 7273), Peirce considers linguistics and semiotics as two different branches of the system of sciences. Whereas linguistics, in
Peirces detailed classification of the sciences of 1902 (CP4 1.203283), is a
branch of the empirical or special sciences, semiotic(s), the quasi-necessary, or
formal, doctrine of signs, belongs to the normative and formal sciences (CP 2.227,
c. 1897). In contrast to the empirical sciences, which study what is in the actual
world, semiotics, as a normative science, aims at inquiring into what must be the
character of all signs and what would be true of signs in all cases (CP 2.227).
The normative approach is particularly evident in logic since to decide whether a
statement or conclusion is true or false is a normative decision. This is why, to
Peirce, semiotics was only another name for logic (CP 2.227, c. 1897).
The three branches of semiotics, in Peirces redefinition of the trivium of the
liberal arts, are the following: (1.) speculative (or formal) grammar, (2.) logic
(proper), and (3.) pure, formal or speculative rhetoric.5 Speculative grammar, the
precursor of Morriss syntactics, is the study of signs as such; it inquires into the
general conditions of signs being signs (CP 1.444, c. 1896), and it is a general
theory of the nature and meanings of signs, whether they be icons, indices, or symbols (CP 1.191, 1903). Of the branches of modern linguistics, not only syntax, but
also phonology and morphology belong to this branch of semiotics. Logic proper,
the second branch of semiotics, which Peirce also calls critic, is the precursor of
Morriss semantics and certainly also comprises linguistic semantics. According to

Semiotic foundations of pragmatics

169

Peirce, it deals with the relations of signs to the objects that they represent, being
the science of what is quasi-necessarily true of signs [] in order that they []
may be true; [ it] is the formal science of the conditions of the truth of representations (CP 2.229, c. 1897).
Speculative rhetoric, the precursor of Morriss pragmatics and with it of linguistic pragmatics, deals with the relations of signs to their producers and interpreters; it studies the effects created by signs in the participants of the process of
semiosis. Speculative rhetoric is the study of the necessary conditions of the
transmission of meaning by signs from mind to mind, and from one state of mind to
another (CP 1.444, c. 1896); it deals with the formal conditions of the force of
symbols, or their power of appealing to a mind (CP 8.342, 1908). Speculative
rhetoric is broader than Morriss pragmatics in two respects. First, to study how
signs are transmitted from one state of mind to another means that this branch of
semiotics does not only study signs in communication but also in thought not communicated to any other interpreter. Thinking has pragmatic dimensions since it always proceeds in the form of a dialogue a dialogue between different phases of
the ego (CP 4.6, 1906). Second, speculative rhetoric is also conceived as a rhetoric of fine art, a rhetoric of practical persuasion, and a rhetoric of science (EP
26: 329, 1904). Peirce had the vision that the rhetoric of science, which he also
called methodeutics (Santaella 1999), was destined to grow into a colossal doctrine which may be expected to lead to most important philosophical conclusions
(CP 3.454, 1896).

2.

Semiotic foundations of communication

Linguistic pragmatics studies how language is used in communication, especially,


how verbal messages are exchanged between speakers and hearers in dialogue.
Merely incidentally transmitted utterances, monologues, and other forms of language use without an addressee are of no concern to the field of study (cf. Levinson
1983: 16).
Communication is a central issue of semiotics; its study is also called semiotics
of communication (Santaella and Nth 2004). The general study of signs and sign
processes may thus provide a theoretical framework for linguistic pragmatics.
Among its topics of relevance to linguistic pragmatics are: (1) models and functions of communication, (2) the instrumental view of language in communication,
(3) the problem of intentionality, (4) the nature of cooperation in conversation, and
(5) the dialogical nature of the sign and its relevance to the study of communication.

170
2.1.

Winfried Nth

Models and functions of communication

Most theories of pragmatics take a very rudimentary model of communication as


the point of departure of their analyses of speech acts. With reference to Grice
(1957), Levinson considers the triad S-U-R, sender (speaker, writer), utterance,
and receiver (hearer, reader), to be fundamental: Communication consists of the
sender intending to cause the receiver to think or do something, just by getting
the receiver to recognize that the sender is trying to cause that thought or action (Levinson 1983: 16). The utterance, in this triad, is a linguistic token; U has
meaning, which becomes mutual knowledge to S and R in the process of communication, but initially, U may have two kinds of meaning, possibly in conflict, a
speaker meaning and a sentence meaning (Levinson 1983: 16).
Leech, in his Principles of Pragmatics, proposes a flow diagram of communication according to which an S uttering U to H is engaged in three kinds of semiotic
transaction: (1) an interpersonal transaction, or discourse, (2) an ideational transaction or message transmission, and (3) a textual transaction or text (Leech 1983:
59). As to its three constituents, S-U-H, and its notion of message transmission, this
model is certainly inspired by the technical model of the flow of information from
S to H by Shannon and Weaver (1949). This model is inadequate for the study of
human communication, since it is linear, unidirectional, and assigns a merely passive role to the recipient of the message, making the sender the only active agent in
communication. It disregards Hs autonomy in the process of interpreting Ss message as well as the important elements of circularity by which S and H are connected
in communication through feedback loops by which H contributes to the production
of Ss message from the moment of its articulation on (cf. Nth 2000: 24446).
A much cited communication model in linguistic pragmatics is the semiotic
model of the six factors and functions of communication which Roman Jakobson
proposed in extension of Bhlers triadic organon model of language (Bhler
1933). Jakobson defines the factors involved in a process of communication as follows: An addresser sends a message to an addressee. This message refers to a context (the referent of the verbal message) seizable by the addressee. Addresser
and addressee have a common code, such as the vocabulary and grammar of the
language in which they communicate, and finally, there is a contact which serves as
a physical channel and psychological connection between the addresser and the
addressee, enabling both of them to enter and stay in communication (Jakobson
1960: 353).
Each of these six factors is the determinant of a specific communicative function; one and the same message may have several functions, but one is usually predominant. Messages which focus on the referent (context), such as informative, descriptive, or narrative texts, have a predominantly referential function. The
emotive or expressive function is predominant when the focus is on the speakers
attitude in relation to the meaning of the message. The use of interjections, a tremb-

Semiotic foundations of pragmatics

171

ling voice, or the frequent use of the pronouns I or me are examples of signs of an
utterance in which the emotive function of language is predominant. The conative
function is oriented toward the addressee; its most typical grammatical expressions
are the vocative and the imperative. The phatic function predominates in messages
which, without necessarily conveying any particular meaning at all, serve to establish, to prolong, or to discontinue communication, to check whether the channel
works (Hello, do you hear me?), to attract the attention of the interlocutor or to
confirm his continued attention (Are you listening?) (1960: 355). The metalinguistic function predominates when language is used to refer to language and verbal communication; metalanguage is language about language, e.g., as in grammar
books or in the discussion of questions of terminology. Finally, the poetic function,
according to Jakobson, predominates when the focus of verbal communication is
on the qualities of language irrespective of its message, e.g., on the sound pattern,
rhythm, or meter of a text.
Although Peirce considers the model of the sign to be more fundamental than
the model of communication, it is not true that the founder of modern semiotics had
little to say about the topic of communication, as Habermas (1995) believes. Elements of a Peircean theory of communication have meanwhile been reconstructed.7 In his manuscript on Pragmatism of 1907, Peirce defines communication
of a speaker with a hearer as follows: Signs mostly function between two minds,
or theatres of consciousness, of which one is the agent that utters the sign (whether
acoustically, optically, or otherwise) while the other is the patient mind that interprets the sign (EP 2: 403). In this description, the characterization of communication as a theater, a staged scenario of acting agents, is more than a mere metaphor. Actors on stage do not act autonomously, but they represent actions and act
on behalf of other agents. But who should be the agents in communication if not S
and H? This question is the topic of sections 2.2 and 2.3.
2.2.

Are verbal signs instruments of those who communicate?

Most approaches to linguistic pragmatics are based on the assumption that verbal
messages are instruments by which senders seek to influence receivers. Language
use is regarded as a means to some end. Searle (1969: 48), e.g., writes: The sentence uttered by a speaker who knows its meaning then provides a conventional
means of achieving the intention to produce certain illocutionary effects in the
hearer. Leech (1983: 36), too, postulates that speaking is a means to an end and
hence an instrument for a purpose: spoken or written messages are transferred or
transmitted from a speaker to a hearer, and the speakers purpose in this transaction is to convey a particular illocutionary force to the hearer (Leech 1983:
5860).
The view that verbal signs are used as instruments has been defended since antiquity. It is central to one of the earliest treatises on the philosophy of language,

172

Winfried Nth

Platos Cratylus. In this dialogue on the relation between words and the objects for
which they stand, Socrates compares the practical utility of the craftsmans tools
with the utility which words have for a speaker. A tool, says Socrates, is naturally
adequate and appropriate to the task for which it serves. To the weaver, the right
and natural (391E) instrument is the shuttle, to the blacksmith, the anvil and the
hammer serve best to shape the iron, and to the shoemaker, the awl is the natural
tool for piercing leather. As any practical action requires its proper tool, so does the
act of speaking. The speakers appropriate instruments for communicating their
ideas are their words (388a-389c). Practical instruments have a practical utility,
whereas words serve a semiotic utility. As the weavers shuttle is useful for separating the warp from the woof, so is the word useful to the speaker who wants to
communicate or to organize his or her ideas.
In the history of semiotics, the instrumental theory of the sign has lived on in
many variants (cf. Nth 2009). In the context of pragmatics, Wittgensteins (1953)
revival of the topic is of special interest. Instrument, or tool, are appropriate
characterizations of verbal signs in Wittgensteins philosophy of language since
these terms illustrate that the proper study of verbal signs is the study of their use.
What language and instruments have in common is that they are both used and
have functions: Think of the tools in a box: there is the hammer, pliers, a saw,
screwdriver, a glue-pot, nails and screws. The functions of words are as diverse
as the functions of these objects ( 11). Whereas Socrates focuses on the utility of
verbal signs in communication, Wittgensteins focus is on their meaning and the
way meaning is revealed in the use of signs. Not only is language in general an instrument, but also its concepts are instruments ( 569). Instruments as well as
language have their sense in their employment: Look at a sentence as an instrument and its sense as its employment ( 421).
Although the view that language serves as an instrument in communication is a
proverbial commonplace, Peirce contests it. A tool is only a means, i.e., an efficient
cause, but not an end, i.e., not a final cause, in the craftsmans process of producing
an effect. Words used in communication, by contrast, are more than mere tools
since they have purposes of their own, which is not their speakers purpose. What a
word means is not determined by its user. The truth conditions of a proposition are
not determined by the utterer of the sentence which it contains either; they are determined by the reality to which it refers and by the consequences the utterance
has. Whether an argument is valid or not is determined by the logic of language and
not by the person who argues, and whether a speech act is felicitous or not depends
on whether it fulfills certain felicity conditions that cannot be ignored by the utterer. Hence, to use words means to be constrained by restrictions determined by
language and the consequences of its use.
Instead of saying that it is the speaker who produces the utterance in accordance with the rules of syntax, semantics, and pragmatics, we can also say that these
rules determine the speaker to speak the way he or she speaks, but the pragmatist

Semiotic foundations of pragmatics

173

account of language and meaning goes a step further. According to Peirces pragmatic maxim (see 3.1), it is not the rules of language and logic that determine our
utterances, but the anticipated feedback which the success or failure of our utterances gives us whenever we speak. The anticipation of the effects of our words
makes us utter them in this or that way. Nor can it be argued that the utterers
thoughts in which his or her intention to speak is conceived use the ensuing utterance as instruments to put the speakers intentions into effect. Utterances which express thoughts are interpretants of these thoughts, which make the interpretants the
final causes of the thoughts. Since the utterance is itself a sign with the purpose of
creating further interpretants in the hearers mind it cannot be merely instrumentally (or efficiently) caused by the thoughts in which they were conceived.
Nothing could be less in accordance with Peirces semiotic premises than a
dualist assumption of internal thoughts becoming merely externalized as spoken or
written words in communication. Sperber and Wilson (1986: 1), for example, make
this assumption when they reflect on the relationship between their own thoughts
and their expression in writing as follows: what we put down on paper are little
dark marks []. As for our thoughts, they remain where they always were, inside
our brains. Peirce, by contrast, argues that signs exert their semiotic agency inside
as well as outside of human brains, and perhaps even more so outside than inside.
He underlines the latter view by saying that it is even much more true that the
thoughts of a living writer are in any printed copy of his book than that they are in
his brain (CP 7.364, 1902). Hence, thought-signs do not remain contained in the
container of a human brain just as written words do not remain fixed in their meaning once they are written down. Words are symbols which live and grow outside of
human brains in the semiotic effects which they have, and this growth takes place
both in thought and in communication. What they have in common is their dialogical nature:
Before the sign was uttered, it already was virtually present to the consciousness of the
utterer, in the form of thought. But [] a thought is itself a sign, and should itself have
an utterer (namely, the ego of a previous moment) to whose consciousness it must have
been already virtually present, and so back. Likewise, after a sign has been interpreted,
it will virtually remain in the consciousness of its interpreter, where it will be a sign []
and, as a sign should, in its turn, have an interpreter, and so on forward. (EP 2: 403,
1907; see below, 2.4 for the argument of the infinite regress.)

In sum, if signs have purposes, or final causes, and exert a semiotic agency of their
own, they cannot be the mere instruments, that is, instrumental (or efficient)
causes, of other signs; and since both thoughts and utterances are signs, utterances
cannot be the instruments of the thoughts in which they were first conceived.
Although the sign is not an instrument, instruments are necessary for the dissemination of signs. We need a pen and paper to write down words, and our voices
are necessary instruments by which our words reach their hearers, but these words
themselves are not our instruments. Not words are our semiotic instruments, but

174

Winfried Nth

our voices, tongues and lips, pens, paper, typewriters, telephones, or computers
are. The instruments by which thought-signs exert their semiotic agency are the
neurons, the brain lobes, and the perceptual organs, eyes and ears, by means
of which thoughts are processed. This is probably Peirces most radical thesis:
our brain is not the only cause of our utterances; it is merely the efficient cause
of the thoughts and of their expression. The ideas on which this approach to semiosis is based have parallels in current quasi-biological positions of the philosophy
of mind known as memetics and teleosemantics; the biolinguistic assumption is
here that words in culture spread like genes spread in biological species (cf. Nth
2009).
In order to deconstruct the autonomy of the speaking subject, Peirce, for the
sake of the argument, may exaggerate here and there when he speaks of the purpose of the sign which is not the purpose of the user. Elsewhere, he acknowledges
two purposes, the purpose of the sign and the purpose of the sign user, in a mutual
feedback control circuit. In an evolutionary perspective, he describes the loop in
which sign users using signs and signs using sign users interact in terms of a process of co-evolution:
Man makes the word, and the word means nothing which the man has not made it mean,
and that only to some man. But since man can think only by means of words or other external symbols, these might turn round and say: You mean nothing which we have not
taught you, and then only so far as you address some word as the interpretant of your
thought. In fact, therefore, men and words reciprocally educate each other. (CP 5.313,
1868)

There is a certain affinity between this Peircean account of the relationship between the signs and their users on the one hand and Saussures theories of the immutability of the verbal sign and the determination of each signifier by the system
of which it is part on the other. As Stetter (1989: 159) has pointed out in his Saussurean interpretation of the above quote: we are not the masters of our signs who
determine the interpretation of terms but it is rather determined by presuppositions
inherent in the system of language, but Peirces object of the sign is of course
more than the system of which the sign is a part.
The co-evolution of signs and their users can only take place if purposes are ascribed to both. But this does not mean that the sign user is the autonomous agent
external to the signs as which the intentionalists describe him (see 2.3). Peirce
solves the apparent paradox of the conflict between the purposes of the sign user
and the one of the sign with the argument that sign users are themselves objects of
signs (see 2.5) so that their intentions are, in turn, purposes of signs. By no means
do sign users pursue nonsemiotic purposes, for they themselves are and act as signs
in the flow of thought-signs and public signs which constitute them as signs. This is
the line of argument of the famous passage in which Peirce defines the human
being as a sign:

Semiotic foundations of pragmatics

175

There is no element whatever of mans consciousness which has not something corresponding to it in the word; and the reason is obvious. It is that the word or sign which
man uses is the man himself. For, as the fact that every thought is a sign, taken in conjunction with the fact that life is a train of thought, proves that man is a sign; so, that
every thought is an external sign, proves that man is an external sign. That is to say, the
man and the external sign are identical, in the same sense in which the words homo and
man are identical. Thus my language is the sum total of myself; for the man is the
thought. (CP 2.314, 1868)

If the sign user is a sign, his or her intentions are signs, too, and in the speakers utterances the underlying intentions have to be interpreted as objects of signs. As the
object determines the sign so does the sign determine its interpretants. Words only
stand for the objects they do, and signify the qualities they do, because they will
determine, in the mind of the auditor, corresponding signs (CP 2.92, 1902). As
any object of a sign determines the sign which it represents so do the speakers intentions, since they are objects of signs. This is how the speakers intention participates in the creation of an interpretant.
2.3.

Intentionality of communication and the purpose of verbal signs

The classics of linguistic pragmatics agree in considering intentionality as a prerequisite of verbal communication. Levinson (1983: 16) summarizes as follows:
Communication consists of the sender intending to cause the receiver to think
or do something, just by getting the receiver to recognize that the sender is trying to cause that thought or action. So communication is a complex kind of intention that is achieved or satisfied just by being recognized.
Searle (1983: 16466) postulates two intentions. First, there must be an intention to represent on the speakers side, but since it is possible to represent without
the purpose to address or to influence anybody, a second intention is required: the
intention to communicate. Searle (1983: 166) explains that it is impossible to inform a hearer that it is raining without intending that this message represents
truly or falsely the weather condition expressed. Not only the speakers intention
but also the hearers awareness of this intention is a prerequisite of communication. Furthermore, the belief produced in the hearers mind as the result of the
message must have its cause in the speakers intention (cf. Sperber and Wilson
1986: 23).
In structural semiotics, intentionality is the criterion by which communication
is distinguished from signification. The distinction has its roots in the writings of
Buyssens (1943).8 All signs which humans can interpret, including the signs of
physical nature, have signification, but not all of them are intended; only those
signs constitute acts of communication which have an addresser whose intention it
is to convey a message to an addressee. In the tradition of analytical language philosophy, Grice (1957) interprets the difference in question in terms of the dichot-

176

Winfried Nth

omy of natural (i.e., unintentional) and nonnatural (i.e., intended) meanings.


The latter term is unfortunate, though. To call the signs of natural language nonnatural, as Grice does, fails to do justice to the continuity of semiosis in the evolution of life from nature to culture (cf. Nth 2008).
Prieto (1966: 3338) anticipates Searles theory of the two communicative intentions when he distinguishes between two indications which signs have in communication, a notificative and a significative indication. Significative indication is
the one by which a sign indicates whatever it means (i.e., its signification). Notificative indication, by contrast, indicates that the sign is being communicated with
the intention of a sender. The content of a notificative indication is so to speak: Attention! This sign is intended to convey a message to you! (cf. Hervey 1982: 71).
If signs can only be intended or not, intentionality cannot be a matter of degree.
However, intentions are volitions, and wishes and desires may vary in strength.
Addressers may be divided as to their intentions, they may have doubts and hesitate whether they should communicate or not, or they may merely pretend to communicate and act as if they communicated (cf. Parret 1994). Such factors of intentionality are rarely taken into consideration in linguistic pragmatics.
Psychoanalysis, by contrast, has shown how verbal and nonverbal messages
may also communicate hidden intentions which the utterer may want to conceal
but fails to do, e.g., when he or she, impatient with the hearer, loses control over his
or her voice, sounding angry despite the intention to appear calm. In such a scenario, the speakers loss of control over his or her voice is the sign of a conflict between two intentions, one determined by the unconscious mind, the other by the
conscious self. The conflict of intentions is thus the symptom of a divided self in
which a conscious mind aims at concealing what the unconscious mind intends to
express.
Linguistic pragmatics also disregards what the communication theorists
Watzlawick, Beavin, and Jackson (1967) say about the ubiquity of communication
in everyday life. In their study Pragmatics of Human Communication, the authors
postulate the metacommunicative axiom that it is impossible not to communicate,
be it verbally or nonverbally: Behavior has no opposite. There is no such thing as
nonbehavior []. One cannot not behave, and any form of behavior in social interaction is of semiotic relevance since one cannot not communicate (Watzlawick et al. 1967: 4849). Even silence and doing nothing are messages.
Peirces semiotic theory of intentionality (cf. Kappner 2004) and his theory of
the purpose of signs in sign processes are rather incompatible with the views of intentionality held in mainstream linguistic pragmatics.9 Peirce defines the sign as a
medium (MS10 339: 283r, 1906), but not as one which mediates between an addresser and an addressee; instead, the sign logically mediates between its object,
which it represents, and its interpretant, which is the interpretative effect it creates.
The object of the sign is the verbal or nonverbal knowledge, feeling, behavior, or
habit which the sign user associates with the sign. The knowledge of its object is a

Semiotic foundations of pragmatics

177

prerequisite for the interpretation of the sign. While the object of the sign includes
aspects of what linguists define as meaning (see 3.3), the interpretant includes the
illocutionary force and the perlocutionary effect of a speech act, as defined by Austin (1962). In Peirces definition, the interpretant is the cognitive, emotional, or behavioral effect which results from the interpretation of the sign.
According to Peirce, the sign is a semiotic agent with its own intention (cf. 2.2).
Signs continue to exert their agency in the absence of those who first produced
them. This topic is also discussed by Derrida (1988: 8), who uses the metaphor of
the machine for the agency which the written sign exerts irrespective of its producer: To write is to produce a mark that will constitute a sort of machine which is
productive in turn, and which my future disappearance will not, in principle, hinder
in its functioning, offering things and itself to be read and to be rewritten. Peirces
view of the agency of the sign goes beyond the mere assumption of its interpretability in the absence of its writer. Signs have a life of their own, which can be
studied in processes of language change and in the way texts are interpreted in new
ways as time passes and signs change. In 1904, Peirce writes about the purpose of
the symbol: The symbol, by the very definition of it, has an interpretant in view.
Its very meaning is intended. Indeed, a purpose is precisely the interpretant of a
symbol (EP 2: 308).
With purpose, Peirce thus does not mean the sign users intention, but the
signs intention to represent its object and to create and interpretant, i.e., to be interpreted in another sign (MS 1476, 1904). Purpose is thus a semiotic teleology
inherent in the sign. Not only uttered or written signs have purposes, but also
thought-signs. Their purpose is to act in a mental dialogue in which one thoughtsign is translated or interpreted in a subsequent one (CP 5.284, 1868).
On the one hand, the sign represents its object to the mind of an interpreter, on
the other, the object represented by the sign is in a sense the cause, or determinant,
of the sign even if the sign represents its object falsely (CP 6.347, 1909; cf. Parmentier 1985). To justify the autonomy of the sign by its being determined sounds
contradictory, but it describes the process of an autopoietic self-determination of
the sign. Since the object of the sign is not external, but internal to the sign, to say
that the sign is determined by its object means that it is determined by a dynamics
inherent in itself and not by an agency outside of the sign (cf. Colapietro 1995).
The way in which a sign is determined by its object depends on its modes of relation to its object and its interpretants. Rhematic signs, such as words in isolation,
are determined to represent their objects as such, without affirming, questioning, or
negating their existence. Dicentic signs, such as affirmations, are determined by
the truth of what they represent, and arguments are determined by the laws of logic
and reasoning in general. Symbols are determined by the habits of their users, by
which they have become symbolic signs. Indices are determined by their objects
because they are related to them by causal, temporal, or spatial relations, whereas
icons represent their objects by means of qualities of their own. All these factors

178

Winfried Nth

determine the sign independently of their use. For example, a rhematic symbolic
legisign such as the word apple, uttered by a speaker, is determined by the habits of
the speakers of English to represent the fruit of the apple tree by means of this
word; the word represents its object as a general kind without affirming the existence of any particular fruit. The sign user may try to escape from his or her determination by the object of this sign and use this word to refer to a pear, but in the
long run, words cannot be used in this way because the speech acts will fail. A sign
user may also lie or argue against the laws of logic, but in the long run, to ignore the
truth and the laws of reasoning will not pay. Furthermore, verbal signs are determined by biolinguistic and evolutionary constraints, which have restricted the
speakers possibilities of articulation and perception (cf. Deacon, 1997: 116).
2.4.

Conversational cooperation, commonage, and the common ground

The model of communication underlying the classical theories of linguistic pragmatics is one of rational discourse.11 The aim of dialogic communication is guided
by the ideal of mutual consensus; the speakers purpose must be achieved. Communication is modeled as being cooperative; its goal is not only comprehension but
also the full acceptance of the intended message on the part of the hearer; speech
acts can only succeed or fail, and successful speech acts fulfill felicity conditions;
the speakers intentions are described as being sincere, and the messages are always relevant. Such accounts of verbal exchange between speakers and hearers
model communication according to the ideals of a desirable symmetry between the
agents interacting in dialogue. The underlying assumption is that a full congruence
between the speakers intended meaning and the hearers understanding of it is not
only possible but also desirable.
To reach consensus is indeed an ideal of dialogic interaction in rational discourse. The very concept of communication implies, in its root, the notion of making something common to those who communicate. Morris (1946: 195) expresses
this normative goal of communicative interaction in a definition according to
which communication is the use of signs to establish a commonage of signification. The assumption that commonage is an essential characteristic of communication is not undisputed. In everyday language, e.g., it is contradicted by the
military metaphors by which speakers represent their arguments as positions
which need be defended (Lakoff and Johnson 1980: 7). In the sociological theory
of communication, Luhmann (1984) starts from the basic assumption that the differences between the utterers and the interpreters frames of mind instead of their
common ground, is the source of all dialogic interaction, whereas Habermas
(1981) founds his Theory of Communicative Action on the assumption that consensus is constitutive of rational communicative interaction.
As a prerequisite for, and means of, reaching the normative ideal of dialogic
commonage, linguistic pragmatics recognizes such principles as Grices maxims

Semiotic foundations of pragmatics

179

of conversational cooperation and its derivative, the principle of relevance advocated by Sperber and Wilson. A precursor in the study of these normative principles is Peirce (Pietarinen 2004, Short 2007: 21314). Although by no means restricted to the study of rational discourse (Oehler 1995: 26768), Peirces
speculative rhetoric describes conversational cooperation as follows:
Honest people, when not joking, intend to make the meaning of their words determinate,
so that there shall be no latitude of interpretation at all. That is to say, the character of
their meaning consists in the implications and non-implications of their words; and they
intend to fix what is implied and what is not implied. They believe that they succeed in
doing so, and if their chat is about the theory of numbers, perhaps they may. (CP 5.447,
1905)

However, immediately after this description of rational conversational cooperation, Peirce also points out that commonage can only serve as a regulative idea, an
ideal which can rarely be reached in ordinary conversation. The sentence after the
above passage goes on to say: But the further their topics are from such presciss12,
or abstract, subjects, the less possibility is there of such precision of speech. Insofar as the implication is not determinate, it is usually left vague (CP 5.447,
1905). Furthermore, Peirce also argues that, despite all efforts of cooperation, dialogues also evince a fundamental divergence of interest between the participants.
From this perspective, Peirce calls the participants of dialogue opponents (MS
515: 25, s. d.; cf. Hilpinen 1995: 293) and describes dialogic interaction in terms of
a metaphorical scenario of war: The utterer is essentially a defender of his own
proposition and wishes to interpret it so that it will be defensible. The interpreter,
not being so interested, and being unable to interpret it fully without considering to
what extreme it may reach, is relatively in a hostile attitude, and looks for the interpretation least defensible (MS 9: 34, c.1903).
Relevance, so much discussed in linguistic pragmatics since Grice, is another
principle about which Peirce has written (cf. Pietarinen 2004). Peirce discusses it
as follows: If the utterer says Fine day! he does not dream of any possibility of
the interpreters thinking of any mere desire for a fine day that a Finn at the North
Cape might have entertained on April 19, 1776. He means, of course, to refer to the
actual weather, then and there, where he and the interpreter have it near the surface
of their common consciousness (MS 318: 3233, 1907). Whereas Grice would
describe the elliptic utterance Fine day as one which requires filling the gap of
what remains unsaid by means of nonconventional conversational implicatures,
Peirce would consider most conversational implicatures unnecessary since communication, studied from the semiotic perspective, is not only verbal communication. What linguistic pragmatics describes as gaps between what is said and what
is left out may be a gap in the verbal message, but interpreters of a dialogue do not
only interpret words; they also interpret nonverbal signs and the situational circumstances of the utterance. What appears a gap in the verbal message may not be
a gap in the multi-channel message of the dialogic exchange. Furthermore, not

180

Winfried Nth

everything that is left open and undetermined in conversation needs to be filled by


implicatures either. It is wrong to assume that the speakers message is always
clearly determined and that the interpreters only task is to fill the gaps in order to
restore the intentions which seem unclear. Vagueness and the impossibility of resolving it may be the speakers purpose and the interpreters source of creative
thinking (see 3.3).
In Peirces speculative rhetoric, commonage is not only a normative goal; it is
also its prerequisite of the dialogue. As the prerequisite of communication, it belongs to the domain of the object of the sign; as its goal, it pertains to the interpretant created by the sign. The object of the sign can only be known if both the utterer
and the interpreter have collateral experience of the Real or Dynamical Object of a
sign. In order to know what a sign represents, we need to have previous acquaintance with what the sign denotes (CP 8.179, s. d.). The sign itself cannot express its
dynamical object, it can only indicate and leave the interpreter to find out by collateral experience what it represents (EP 2: 198, 1909). There is nothing nonconventional in this process of interpretation, as Grice assumes with his implicatures. Interpretation, according to Peirce is always a natural process of abductive,
inductive, and deductive inferencing.
An aspect of the object of the sign, i.e., that which must be presupposed for
communication to be successful, is the common ground of experience and knowledge which the utterer and the interpreter share. Both Grice and Searle have described it as a prerequisite of conversation. Pietarinen (2004: 302305) shows in
which respect Grices and Searles theories of the preconditions of successful communication were anticipated by Peirce. In 1908 Peirce formulates it as follows:
No man can communicate the smallest item of information to his brother-man unless they have [] common familiar knowledge [] such that each knower knows
that every other familiarly knows it, and familiarly knows that every other one of
the knowers has a familiar knowledge of all this (MS 614: 12, in Pietarinen
2006: 438). To make clear that no circulus vitiosus is involved in this endless reciprocity, Peirce underlined that this infinite regress is a logical and not a psychological one: Of course, two endless series of knowings are involved; but knowing
is not an action but a habit, which may remain passive for an indefinite time (MS
614: 12, in Pietarinen 2006: 438).
Insofar as commonage is the goal of communication, it pertains to the interpretant of the sign. Peirce calls the meaning effects common to the speaker and the
hearer the cominterpretant of the sign. At the same time, he recognizes that there
may be effects of meaning which are specific to the utterer and the interpreter, the
former being the intentional, the latter the effectual interpretant:

Semiotic foundations of pragmatics

181

There is the Intentional Interpretant, which is a determination of the mind of the utterer;
the Effectual Interpretant, which is a determination of the mind of the interpreter; and
the Communicational Interpretant, or say the Cominterpretant, which is a determination
of that mind into which the minds of utterer and interpreter have to be fused in order that
any communication should take place. This mind may be called commens. It consists of
all that is, and must be, well understood between utterer and interpreter at the outset, in
order that the sign in question should fulfil its function. (EP 2: 478, 1906)

The commind, as Peirce calls the fusion of the two minds in one and the same
commens elsewhere, is a normative ideal. In practice, the ideal of a commind can
never be fully reached since mutual understanding is only possible in a fragmentary way. In 1907 (MS 318), Peirce points out that no interpreter can be said to have
access to the utterers mind. Knowledge about what a speaker means can only be
obtained in fragments, and such fragments are mere copies of a scrap torn out of
anothers life. By such scraps we can only supplement the ideas of [our own]
life (MS 318: 194, 1907; cf. Johansen 1993: 198202). As interpreters we must
match those fragments found in the signs of the utterance with our own discourse
universe and find out where they can be inserted or recopied in our own panorama of universal life (MS 318: 194, 1907). Interpretation is thus a semiotic
patchwork put together by abductive, inductive and deductive reasoning.
2.5.

Dialogue as the prototype of verbal communication, and the dialogical


nature of the sign

Peirce not only considered the dialogue as a wonderfully perfect kind of signfunctioning (EP 2: 391, c. 1906), he also had derived his first model of the sign together with the first version of his three categories, the I, the IT, and the THOU
from the three elementary constituents of a dialogue (W 113: xxvii, 174, 1865).14
Fisch explains how the model of the sign is derived from the model of the dialogue:
Peirce began where most of us begin, with a model, which, taken by itself, would
suggest too narrow a definition; namely the model of conversation between two
competent speakers of the same natural language say, English. [] It goes without saying that words are signs; and it goes almost without saying that phrases,
clauses, sentences, speeches, and extended conversations are signs (Fisch 1986:
357).
In Peirces fully developed semiotics, a sign, or representamen, is something
which stands to somebody for something in some respect or capacity. It addresses
somebody, that is, creates in the mind of that person an equivalent sign, or perhaps
a more developed sign. That sign which it creates I call the interpretant of the first
sign. The sign stands for something, its object (CP 2.228, c.1897). Sign, object,
and interpretant must be understood as following each other in time. The object
comes first insofar as it precedes the sign; it is that which must be known before the
sign can be understood. The sign comes next since it represents the object to an in-

182

Winfried Nth

terpretant, and the interpretant follows last since it is created by the sign. This temporal sequence makes clear how the three correlates of the sign correspond to the
three correlates of a dialogue, addresser (or sender), message, addressee (or receiver): the sign (or Representamen) is the message; it addresses the interpretant as
its receiver, and the sender of this message is its object, which determines the sign
(cf. Bergman 2003: 11, Pietarinen 2006: 426).
Despite the close association of the interpreter and the interpretant in the process of interpretation, the interpretant must not be confused with the mind of the
addressee of a message. In its full definition, the interpretant is any significant effect of the sign (CP 2.303, 1902), be it a feeling, action, or thought, and since significant effects of signs are not restricted to individuals and to the present moment
of sign interpretation, the interpretant can also be a memory, a habit, a fashion, an
innovation, a revolution, a war, or a peace treaty. In all cases, the interpretant is
itself a sign.
The object of the sign corresponds to its sender in an equally impersonal understanding (cf. 2.2 on the utterers intention as an object of the sign). Ransdell (1977)
explains why Peirce equates the two instances of conversation which linguists are
used to distinguishing as the referent of the sign on the one hand and the sender of a
message on the other: to the degree that the reality, the phenomena which we perceive, experience, or think about, are interpretable phenomena, they are objects of
signs. Reality is a sign producer since it is the cause of the signs which we have of
it; after all, we interpret reality by means of its signs, which in turn create interpretants. Both the sender of a message who, although not a creation of this message,
is nevertheless the source of its distribution, and the signs object precede the
moment and are the causes of the production of the sign.
Not only the sender but also the sign itself has an agency in the process of semiosis, as shown above (2.23). Peirce calls it the action of the sign (e.g.,
CP 5.47273, 1905), but again, this agency, which is due to the purpose of the sign,
is neither restricted to the purpose of a human being nor must it be the effect of a
verbal sign. There are even signs which have no sender at all. These are evidently
the natural signs (cf. Nth 2008). Since there can be no isolated sign, Peirce attributes a Quasi-utterer to (natural) signs without a real utterer (CP 4.551, 1906).
There are also signs without an interpreter but a Quasi-interpreter (CP 4.551,
1906), for example, if nobody takes the trouble to study the record (EP 2: 404,
1907). According to Peirces dialogic semiotics, it is essential that the utterer or
quasi-utterer as well as the interpreter or quasi-interpreter are not instances exterior to, but can be found within the sign, in which they must nevertheless be distinct although they are, so to say, welded in the sign (CP 4.551, 1906). A summary of the resulting simultaneity of the presence and absence of the utterer in the
dialogical sign is the following:

Semiotic foundations of pragmatics

183

The action of a sign generally takes place between two parties, the utterer and the interpreter. They need not be persons; for a chameleon and many kinds of insects and even
plants make their livings by uttering signs, and lying signs at that. Who is the utterer of
signs of weather, which are not remarkably veracious, always? However, every sign certainly conveys something of the general nature of thought, if not from a mind, yet from
some repository of ideas, or significant forms, and if not to a person, yet to something
capable of some how catching on. (MS 318, S. 17, 1907)

In the dialogic process of conversation, the continuous alternation between the role
of the utterer and the one of the hearer involves a permanent transformation of
signs into interpretants and interpretants into the objects of new signs creating new
interpretants. This continuous process of alternations and transformations is a
source of creativity in semiosis. Since not only the utterers but also the objects of
the signs participate in the dialogue in the sense that they determine its interpretants, not only the voices of two participants but a plurality of voices is audible.
Dialogue is a Bakhtinian polyphonic chorus with a plurality of participants speaking the voices of many discourses (cf. Nth 2009).

3.

Pragmatism and Peirces semiotics of meaning and reference

Peirces semiotics is also relevant to issues of linguistic pragmatics such as the relationship between and the respective scopes of semantics and pragmatics, the difference between speaker, hearer, and utterance meaning, the nature of reference
and indexicality in language, and the theory of speech acts. Even the concept of
pragmatics has its origin in Peirces writings.
3.1.

Pragmatics, pragmatism, and pragmaticism

The term pragmatics, as introduced by Morris (see 1.), was clearly inspired by the
philosophy of pragmatism, whose main representatives in the first decades of the
20th century were William James and John Dewey, and whose founder was Peirce
(cf. Murphy 1990). In a letter of 1912, Peirce explained why and how he coined the
term in the 1870s (Houser 1998: xxi-xxii): When I gave the doctrine of pragmatism the name it bears [] I derived [ it] from , behavior, in order
that it should be understood that the doctrine is that the only real significance of a
general term lies in the general behavior which it implies (in: Eisele 1987: 95).
Morriss change of the concept of pragmatism to pragmatics made the term more
compatible with the names of the neighboring fields of pragmatics in linguistics
syntactics and semantics as well as with the term linguistics itself.
One of the main aims of pragmatism is to overcome the dualism between
thought and action (Colapietro 1992: 43031). Peirce formulated the essence of
pragmatism, as he conceived it, in several versions of his pragmatic maxim for-

184

Winfried Nth

mulated in several versions since 1878. In 1902, the maxim stated the following:
Consider what effects, that might conceivably have practical bearings, we conceive the object of our conception to have. Then, our conception of these effects is
the whole of our conception of the object (CP 5.9, c. 1905). Peirce gives the following account of how this maxim should be applied: Pragmatism is the principle
that every theoretical judgment expressible in a sentence in the indicative mood is
a confused form of thought whose only meaning, if it has any, lies in its tendency to
enforce a corresponding practical maxim expressible as a conditional sentence
having its apodosis in the imperative mood (CP 5.18, 1903). Colapietro (1992:
433) gives an example of how this principle can be followed in practice: To learn
most clearly the meaning of an assertion e.g., of Sugar is soluble requires us
to translate the assertion into a maxim such as If you place this substance into
water or similar liquids it will dissolve.
Disappointed with how his word coinage became abused in the merciless way
that words have to expect when they fall into literary clutches (CP 5.414, 1905) in
a trend that culminated in the deterioration of pragmatism into behaviorism
(Nadin 1993: 223), Peirce, in 1905, began to replace the term pragmatism with the
neologism pragmaticism, a word ugly enough to be safe from kidnappers, as its
author remarked (CP 5.414, 1905).
Parret (1983: 3) believes that there are three main lessons which the philosophy
of pragmatism has taught to linguist pragmatics: Meaning is context bound,
rationality is discourse bound, and signifying should be seen as understanding.
What linguistic pragmatics has learned from the tradition of pragmatism is certainly the importance of the study of language with respect to its consequences for
human action and behavior.
A common ground of Peirces semiotics and the classics of linguistic pragmatics is the normative approach to communication. Among the other insights
which Peirce has contributed to linguistic pragmatics is the pragmatic principle of
expressibility often attributed to Searle. Searles (1969: 220) definition is: For
any meaning X and any speaker S whenever S means (intends to convey, wishes to
communicate in an utterance, etc.) X then it is possible that there is some expression E such that E is an exact expression of or formulation of X. Peirce reduces it to the form: thought and expression are really one (CP 1.349, 1903).
3.2.

Reference and indexicality

Reference and deixis are key issues of linguistic pragmatics; in Peirces semiotics,
in which the term reference is hardly used (cf. Nth 2006), these topics are studied
as indexical signs. Searle (1969: 2629, 7296) studies reference as a speech act.
Expressions may be used to refer or they may not be used to refer. A man came is
used to refer to a man who came; John is a man is not used referentially, but
predicatively (Searle 1969: 27). Necessary conditions of a referring speech act are:

Semiotic foundations of pragmatics

185

whatever is referred to must exist (1969: 77); There must exist one and only one
object to which the speakers utterance of the expression applies, and: The hearer
must be given sufficient means to identify the object from the speakers utterance
of the expression (Searle 1969: 82).
Existence, singularity, and referential selectivity are criteria of the genuine
index in Peirces semiotics. In fact, all of Searles conditions of reference are conditions of the genuinely indexical sign: A genuine Index and its Object must be
existent individuals (whether things or facts) (CP 2.283, c. 1902), and indexical
signs are directions of what to do to find the object (CP 2.289, c. 1893). In
contrast to Searle, Peirce also studies to what degree nonverbal signs can be genuine indices and to what degree language may be insufficient for the purpose of referring to an individual object. Furthermore, he also studies another kind of index,
which he calls degenerate index. Such indices do not refer directly to individual
existing object; instead, they refer to symbols or mental objects (cf. Nth 2000:
187). Their prototype is the relative pronoun, which is a sign referring to another
sign.
Since not all expressions are used to refer, according to Searle, not all of them
have a referent. In Peirces semiotics, too, only genuine indices have a referent in
the sense of referring indexically to an individual existent object, but although not
all signs have a referent, all signs have an object.15 Peirces semiotics is utterly incompatible with the dualistic theories of referential semantics which split verbal
reference into two mutually independent domains, one being external to the human
mind, the other being mental and hence internal to it. According to Peirce, not reference and sense are the correlates of the sign, but the object and the interpretant,
and all signs have an object as well as an interpretant. All words are signs which
represent an object and determine an interpretant, even the conjunction and. Savan
(1994: 189) describes its object as our experience of how of two or more entities
are combined to form a group, a set, or otherwise a whole, but since the object of
the sign need not be a material thing, we should add that the idea of a symmetrical
conjunction is among the objects of this sign.
A sign without an object would be a contradiction in terms in Peirces semiotics. In contrast to the referent of logical semantics, which is defined as an individual or class of existents, Peirces object of the sign may also be a feeling, an
experience, a cognition, a thought, an imagination, or even a fictional event. Unlike
the sense or meaning of a word, the interpretant is not necessarily a mental concept, a thought, or an idea corresponding to the sign; it can again be a feeling, an
action, or even be a material thing, provided it is a result of a sign. For example, the
cake baked according to a cooking recipe is among the interpretants of the recipe
practically interpreted by the cook. The object of the sign may also be something
believed formerly to have existed or expected to exist or something of a
general nature (CP 2.232, 1910). Only rarely is it a thing. It is rather that information, knowledge, or experience which an observer of the sign must have in

186

Winfried Nth

order to be able to interpret the sign. Peirce had adopted the term object from 13th
century scholastic terminology where objectum meant a creation of the mind in its
reaction with a more or less real something [] upon which cognition is directed
(MS 693A: 33; Pape 1996b: 115).
The index is defined in opposition to the symbol and to the icon as a class of
signs comprising natural but also conventional signs. An index is really affected
(CP 2.248, c. 1903) by, or physically connected with, its object; the sign and its
object make an organic pair, but the interpreting mind has nothing to do with
this connection except remarking it, after it is established (CP 2.299, c. 1895).
Among the features of Peirces index are that it directs the interpreters attention
towards the object and that it asserts nothing, but shows the object (cf. Goudge
1965: 5354). Examples of verbal and nonverbal indexical signs are: a weathercock, a yardstick, a photograph, a rap on the door, a pointing finger, any other gesture that indicates a present emotion, a cry for help, deictic words, proper names,
possessive, relative, personal, and selective pronouns.
All verbal utterances contain explicit or implicit indices referring to, or identifying, the situational circumstances, the time, place, or persons in whose context
they were uttered. The adequate interpretation of these indices is necessary to distinguish successfully between the relevant and irrelevant interpretation of a sign
(cf. 2.4):
If, for example, a man remarks, Why, it is raining! it is only by some such circumstances as that he is now standing here looking out at a window as he speaks, which
would serve as an Index (not, however, as a Symbol) that he is speaking of this place at
this time, whereby we can be assured that he cannot be speaking of the weather on the
satellite of Procyon, fifty centuries ago. (CP 4.544, 1906)

Morris did not adopt the term index, but his category of identifiors corresponds to
Peirces index (1946: 154, 362). But in contrast to Peirce, Morris restricted the
class of identifiors to spatio-temporal deixis. Identifiors signify locations in space
and time (locata) and direct behavior toward a certain region of the environment.
The identifior has a genuine, though minimal, sign status; it is a preparatorystimulus influencing the orientation of behavior with respect to the location of
something other than itself. Morris distinguished three kinds of identifiors: indicators, which are non-language signals, descriptors, which describe a spatial or
temporal location, and namors, which are language symbols, and hence, substitute signs synonymous with other identifiers.
3.3.

Meaning in dialogue and its vagueness

Several meanings are involved in the utterance of a speaker addressed to a hearer.


Searle (1969: 4548) distinguishes between intended and conventional meanings,
Leech (1983: 6) between pragmatic and semantic meanings, and Levinson (1983:

Semiotic foundations of pragmatics

187

1617) between speaker meanings and utterance or sentence meanings. The various kinds of meaning distinguished by these authors partly overlap, but some of the
criteria according to which these distinctions are made differ significantly.
Leechs semantic meaning roughly corresponds to Levinsons sentence
meaning and Searles conventional meaning. All three kinds of meaning are opposed to what is studied in linguistic pragmatics, i.e., how meaning is intended by a
speaker and understood by a hearer. Leech defines semantic meaning as a meaning defined purely as a property of expressions in a given language, in abstraction
from particular situations, speakers, or hearers (1983: 6). Can such a meaning
abstracted from the circumstances of its use be called conventional, as Searle
suggests when he opposes conventional to intended meaning? The idea that
intended meaning is opposed to conventional meaning is rejected by Strawson
(1971), who points out that conventional meaning is not only semantic meaning
but pertains to pragmatic meaning, too, since speech acts, like words and sentences, are also determined by conventions.
The study of pragmatic meaning, according to Leech (1983: 14), begins with
the study of utterances: In fact we can correctly describe pragmatics as dealing
with utterance meaning and semantics as dealing with sentence meaning. With
this definition, Leechs dichotomy of semantic vs. pragmatic meaning is no longer
compatible with Levinsons dichotomy of utterance meaning and speaker meaning, which suggests that the study of speaker meaning is the domain of pragmatics
and not the study of utterance meaning.
Furthermore, Levinsons distinction between speaker meaning and utterance
meaning raises the question of how speaker meaning can differ from utterance
meaning at all. Is not the utterance meaning the meaning conveyed by the utterer,
i.e., speaker? Can an utterer, by means of an utterance, convey any other meaning
than his own or her own meaning?
The distinctions drawn by the three authors thus do not only overlap, they also
contradict each other. If speaker meaning is the utterers and thus also the intended
meaning, one can no longer oppose it to utterance meaning (as Levinson does).
Speaker meaning can only be intended meaning since whatever a speaker utters
must be intended, but if utterance meaning is speaker meaning and speaker meaning is intended meaning, the congruence of theses meanings makes Leechs distinction between semantic and pragmatic meanings invalid. On the other hand, if
speaker and utterance meanings are intended meanings, and if utterances are determined by conventions, too, Searles distinction between intended and conventional
meanings is invalid.
If there is speaker meaning there should also be hearer meaning. Grice (1989)
takes it into consideration but the dichotomies set up by Leech, Searle and Levinson neglect it. Sbis and Fabbri (1980: 314) criticize this neglect for reducing the
interpretive work of the addressee [] to the mere recognition of a set of coherent
intentions on the part of the agent/speaker. The neglect of the category of a hearer

188

Winfried Nth

meaning in linguistic pragmatics is due to its normative approach to the study of


conversation. Like Chomsky, who pursues a normative approach to syntax when
he postulates the idealized competent speaker as one who commits no mistakes in
uttering sentences, Searle (1969: 20) explicitly adopts a normative approach to the
description of the speakers meanings in conversation when he declares that cases
where the speaker does not say exactly what he means the principal cases of
which are nonliteralness, vagueness, ambiguity, and incompleteness are not theoretically essential to linguistic communication.
Speakers can mean more than they say (Leech 1983: 9). According to Searle
(1969: 45) the difference between what a speaker means and what he or she says is
a difference between the meanings intended by the speaker and those conventionally inherent in his or her utterances. Meaning is more than a matter of intention; it
is also at least sometimes a matter of convention, says Searle (1969). The theory
of implicatures is based on the assumption of such differences between meaning
and saying.
Peirces theory of meaning is pragmatic throughout (cf. Hilpinen 1995: 297).
Meaning is the idea which the sign attaches to its object (CP 5.6, 1905), but
meanings are not a priori inherent in signs; they can only reveal themselves in processes of semiosis (Pape 1996a: 308). Words only have meaning in so far as we
are able to make use of it in communicating our knowledge to others and in getting
at the knowledge that these others seek to communicate to us (CP 8.176, 1903).
Thus, no semantic or conventional meaning can be found in Peirces assumptions. In accordance with the pragmatic maxim, the meaning of an utterance reveals itself in its interpretations as well as in the actions and habits which are its
consequences. In one of his later definitions, Peirce concludes that the meaning of
any [assertion] is the meaning of the composite of all the propositions which that
[assertion] would under all circumstances empower the interpreter to scribe (MS
280, ca. 1905; cf. Pietarinen 2005: 1769). Since utterances have no inherent meaning independent of their use in speech acts, Peirce does not distinguish between utterance and speaker meaning. Furthermore, he distinguishes between many more
kinds of meaning. Some aspects of meaning are associated with the object of the
sign insofar as the sign presupposes collateral knowledge of its object and a common ground of knowledge shared by the speaker and the hearer. Other meanings
pertain to the interpretant, the effect of meaning created by the sign (cf. Bergman
2000: 234 for both). In the theory of the interpretant discussed above, we find the
intentional interpretant as a counterpart to speaker meaning and the effectual interpretant as the counterpart to Grices hearer meaning, but Peirce distinguishes
other kinds of interpretants (cf. Johansen 2002: 48). His immediate interpretant,
the meaning which the sign indicates before it is interpreted, comes close to
Searles conventional meaning, but the interpretant is not always the meaning of a
sign in the linguistic sense. For example Peirce distinguishes between an emotional, and energetic, and a logical interpretant. The first is an emotional, the sec-

Semiotic foundations of pragmatics

189

ond a bodily reaction to a sign. Only the third comes close to the kind of meaning
which linguists usually describe since logical meaning is the meaning of a concept
but since not all words are concepts, words also have meanings which are not logical interpretants.
Peirce also acknowledges the possibility of differences between saying and
meaning so much discussed in linguistic pragmatics, but his account of how meaning is conveyed in the dialogue of an utterer and an interpreter is based on other
premises. Meaning is not determined by rules, conventions, and speakers intentions, as Searle (1969: 4345) sees it, but by habits, common grounds, and the collateral experience of the object of the sign. Meaning is not only vague by exception, but it is essentially vague (Nth and Santaella 2009). If meaning is fuzzy in
language because of the generality and vagueness of words, it is even vaguer if the
feelings, desires, and inner conflicts are considered which speakers and hearers
have when they communicate. Emotions, e.g., are never certain and clear; they are
necessarily vague, merely expressible and interpretable in the form of allusions,
conjectures, surmises, guesses, intuitions, and communicative negotiations. Therefore, no communication of one person to another can be entirely definite, i.e., nonvague. [] Wherever degree or any other possibility of continuous variation subsists, absolute precision is impossible. Much else must be vague, because no mans
interpretation of words is based on exactly the same experience as any other mans.
Even in our most intellectual conceptions, the more we strive to be precise, the
more unattainable precision seems (CP 5.506, c. 1905).
Vagueness and generality are the two major causes of indeterminacy in conversation, but this indeterminacy leads not only to misunderstandings or even failures
of communication. Instead, indeterminate signs are also a source of creativity to
both utterers and interpreters. The interpreter can and must exert his or her own
imagination in determining what the vague or general sign means. He or she can
specify what is left undetermined by associating it with meanings from his or her
own textual universe. The utterer of a vague sign has the advantage that he or she
can suggest meanings without being held responsible for them. Hence vagueness is
not only a risk that endangers communication; it also offers a useful potential of
creativity to those who communicate. Peirce develops these ideas of the advantages and risks of taking and giving freedom of semantic choice with an example of
a generality and one of referential vagueness:
A sign [] is objectively general in so far as it extends to the interpreter the privilege of
carrying its determination further. Example: Man is mortal. To the question, What
man? the reply is that the proposition explicitly leaves it to you to apply its assertion to
what man or men you will. A sign that is objectively indeterminate in any respect is objectively vague in so far as it reserves further determination to be made in some other
conceivable sign, or at least does not appoint the interpreter as its deputy in this office.
Example: A man whom I could mention seems to be a little conceited. The suggestion
here is that the man in view is the person addressed; but the utterer does not authorize

190

Winfried Nth

such an interpretation or any other application of what she says. She can still say, if she
likes, that she does not mean the person addressed. Every utterance naturally leaves the
right of further exposition in the utterer; and therefore, in so far as a sign is indeterminate, it is vague, unless it is expressly or by a well-understood convention rendered general. (CP 5.447, 1905)

Elsewhere, Peirce gives indefinite pronouns as the example of the risks and
chances which generality and vagueness bring to the semantic choices made by the
utterer and the interpreter: Some means that the speaker is to select an instance,
while Every or Any means that a second person is to perform the selection. Of
course, it is easier to satisfy the conditions of a statement if one can select ones
own examples, except for this, that he who undertakes to find an example guarantees that there is one, while if he leaves the selection to another, and there is none,
his statement is not broken down (CP 2.523, fn 1, 1893).

4.

Semiotics of speech acts: Peircean and Greimasian contributions

The theory of speech acts is the branch of linguistic pragmatics to which semiotics
has both contributed as a precursor and to which it has offered alternative approaches. The semiotic precursor of speech act theory is again Peirce; the alternative approach to be presented in the following is an approach based on the semiotics of A. J. Greimas. Only passing reference can be made to the semiotic
elements of a pragmatic theory of speech acts which may be found in the typology
of discourse of Charles Morris (1946: 203205; cf. Nth 2000: 94) and to the rather
different elements of a semiotics of some speech acts developed in the framework
of M. A. K. Hallidays systemic semiotics by Ventola (1987).
4.1.

Peirces contributions to speech act theory

Peirces contributions to speech act theory have been the focus of interest of several studies.16 In wider contexts they have also been discussed by Hilpinen (1995,
1998) and Pietarinen (2004, 2006). Actually, Peirce has even anticipated key terms
and issues of speech act theory, among them Austins thesis that by speaking we
do things and that this doing is an act. In a fragment of 1908, Peirce describes
the speech act of taking an oath in these terms: Taking an oath is not mainly an
event of the nature of a [] Vorstellung, or representing. It is not mere saying, but
is doing. The law, I believe, calls it an act (CP 5.546).
In anticipation of the distinction between an illocutionary act and its propositional content of an utterance later drawn by Searle (1969: 2329), Peirce defines a
proposition as the logical content of a sentence, irrespective of the particular
mood in which it is uttered (cf. Brock 1981a: 281). Sentences differ from utterances as types differ from tokens:

Semiotic foundations of pragmatics

191

A sentence, in the sense here used, is a single object. Every time it is copied or pronounced, a new sentence is made. But a proposition is not a single thing and cannot
properly be said to have any existence. Its mode of being consists in its possibility. A
proposition which might be expressed has all the being that belongs to propositions although nobody ever expresses it or thinks it. It is the same proposition every time it is
thought, spoken, or written, whether in English, German, Spanish, Taglog, or how. A
proposition consists in a meaning, whether adopted or not, and however expressed. (MS
599: 56, ca. 1902)

Among the speech acts which Peirce describes as moods of propositions are assertion, judgment, question, denial, doubt, command, teaching, taking an oath, laying
a wager, lying, poetic language, and forms of politeness.17 Expressing a proposition in one of these moods involves an act, an exertion of energy, and is liable to
real consequences, or effects (CP 5.547, ca. 1908). The exertion of energy associated with these moods is evidently equivalent to Searles illocutionary force of
speech acts. Examples and illustrations of the consequences which such speech
acts have for their utterers and interpreter are: An assertion is an act by which a
person makes himself responsible for the truth of a proposition. Nobody ever asserted that the moon is made of green cheese; yet this is a familiar proposition
(MS 599: 5, ca. 1902). The person who took an oath has only made an assertion if
the denotation of the proposition is a fact (CP 4.500, ca. 1903):
[An oath] would be followed by very real effects, in case the substance of what is asserted should be proved untrue. This ingredient, the assuming of responsibility, which is
so prominent in solemn assertion, must be present in every genuine assertion. For
clearly, every assertion involves an effort to make the intended interpreter believe what
is asserted, to which end a reason for believing it must be furnished. But if a lie would
not endanger the esteem in which the utterer was held, nor otherwise be apt to entail
such real effects as he would avoid, the interpreter would have no reason to believe the
assertion. (CP 5.546, 1908)

By contrast, a conversation solely determined by the conventions of politeness has


no consequences as to its truth: Nobody takes any positive stock in those conventional utterances, such as I am perfectly delighted to see you, upon whose falsehood no punishment at all is visited (CP 5.546, 1908). Among the speech acts
whose consequences depend on the truth of the proposition expressed are not only
assertions and oaths, but also laying a wager. Peirce describes the consequences of
these speech acts as follows:
What is the difference between making an assertion and laying a wager? Both are acts
whereby the agent deliberately subjects himself to evil consequences if a certain proposition is not true. Only when he offers to bet he hopes the other man will make himself
responsible in the same way for the truth of the contrary proposition; while when he
makes an assertion he always (or almost always) wishes the man to whom he makes it to
be led to do what he does. Accordingly in our vernacular I will bet so and so, is the
phrase expressive of a private opinion which one does not expect others to share, while
You bet is a form of assertion intended to cause another to follow suit. (CP 5.31, 1902)

192

Winfried Nth

Peirce considers not only the conditions of publicly expressed speech acts but also
such mental acts, as a judgment or a doubt. A judgment is a mental act by which
one makes a resolution to adhere to a proposition as true, with all its logical consequences (CP 5.31, 1902). Among the mental acts expressing propositions which
are not asserted is the doubt: I may state it to myself and worry as to whether I
shall embrace it or reject it, being dissatisfied with the idea of doing either
(CP 5.31, 1902). Peirce also specifies several of those conditions of speech acts
which Searle later called preparatory conditions (cf. Brock 1981a: 282). For
example, Peirces formulation of what Searle calls the preparatory condition of a
speech act is: The only way to get a sufficient understanding of a proposition is to
imagine one person communicates in some language to a second. The utterer does
not, properly speaking, communicate it to the interpreter unless the latter is at the
time ignorant (or oblivious) of its truth (MS 284: 41, 1905; Brock 1981a).
The distinction between the illocutionary and the perlocutionary effects of a
speech act has almost no counterpart in Peirces theory of speech acts. A consequence of the pragmatic maxim is that speech acts do not only sometimes have perlocutionary effects; they have always such effects. The study of these effects is the
study of the interpretant, which is a necessary constituent of all signs (see 2.4).
4.2.

Speech acts as narrative scenarios: The Greimasian approach

Elements of the structural semiotics of A. J. Greimas18 are the foundation of a theory of speech acts developed by Sbis and Cooren.19 Sbis (1983: 100101) criticizes Searles speech act theory for being based on a causal theory of action (cf.
Davidson 1980) which results in a too strictly speaker centered approach to the
study of dialogues. According to Sbis, the causal approach to speech acts means a
regrettable return of the subject, the triumph of the empirical speaker and the resuscitation of the use of introspection as a criterion of sense (1983: 100101). To
liberate pragmatics from its mentalist bias, she proposes a semiotic conception of
acts and actions. In accordance with the definition proposed by Greimas and
Courts (1979: 5), Sbis defines an act as a fair-tre, a make-be, i.e., the producing of a state (or of a change of state). She argues that the agent of an act thus
defined is not the cause of the speech act but simply its presupposition. Instead of
the study of speech acts, Sbis proposes the study of interactions between speakers
and hearers. The focus should be on the hearer as an active participant of the dialogue; it should be taken into consideration that the hearer has the choice (i) to select an acceptable interpretation of the speech act, and (ii) to either accept the
speech act, under such an interpretation, as a successful act, or to completely or
partly reject it as more or less inappropriate and unhappy (Sbis and Fabbri
1980: 305).
Instead of investigating the utterers agency in dialogic interaction, Sbis
(1983: 102, 1987: 259) deals with action as an ascribed concept: the interpreter

Semiotic foundations of pragmatics

193

ascribes a certain change-in-the-world to the responsibility of the [utterer as the]


agent. In this perspective, agency shifts from the utterer to the interpreter, whose
decision it is to ascribe intentions and meanings to an utterer or not. With this shift
of perspective, the analysis can take into consideration ascriptions of agency to
quasi-utterers in Peirces sense (see 2.5). According to Sbis (1987: 259), the utterer of a sign, whose agency is now merely an agency ascribed by the hearer, can
thus be a human individual, an infra-individual instance (e.g., the Unconscious), a
super-individual construction (class, party, society, church, and the like), and even
a personified natural agent. Hence, in a semiotic speech act theory adopting these
premises, not only speakers are agents, but even beings other than humans can be
said to do things with words (Cooren 2006: 8). Two common examples of assertive and commissive speech acts in which agency is ascribed to nonhuman agents
are: This decision contests his result and This signature commits you to payment (Cooren 2006: 17).
Based on her assumption that acts result in transformations of states, Sbis
(1985: 529) arrives at a redefinition of basic concepts of speech act theory: (1) A
speech act is the production of an utterance insofar as it brings about a change for
which the utterer is taken to be responsible []. (2) An illocutionary act is the
production of a change of context at the conventional level of the addressers and
the addressees modal competences. (3) A perlocutionary act is the production of
a change of context at the material (psychological or praxeological) level of the addressees reactions. An example of her Greimasian approach to the illocutionary
force of speech acts is Sbiss (1985: 530) study of the discursive figure of manipulation, which she defines as the agency of a manipulating addresser who puts the
manipulated addressee into a position of not being able not to do. The application
of Greimass theory of modalities leads Sbis (1984: 100) to a typology of modal
changes of the speakers and hearers states in speech acts, e.g.: in exercitives (such
as granting or permitting), the speaker addresses the addressee in the deontic mood
of being allowed (can), and the effect in the addressee is one of the modality of
obligation (ought); in commissives, these two modal ascriptions are inverse.
Taking up Sbiss and Greimass elements of a semiotic pragmatics, Cooren
(2000: 81169) proposes a model based on the premise that speech acts involve a
narrative scenario typically involving three actants: the speaker as the agent, the
hearer as the recipient, and the utterance as the object of transfer (Cooren 2000:
86). Uttering is seen as an act of production in which the object produced by the
agent, i.e., the utterance, is irremediably severed or dissociated from its origin
so that the producer no longer has control over the object produced and acquires
an agency of its own (Cooren 2000: 8283). A speech act is always embedded in a
plurality of verbal and nonverbal texts; it does not only refer to, it also produces a
situation. Continuing to act in absence of its utterer, the speech act becomes the
product of multiple texts that speak on behalf of their producer, creating one or
several different situations (Cooren 2000: 87).

194

Winfried Nth

Following Greimass discourse grammar, which describes narrative roles by


means of the actantial model of verb valency of Tesnire (1959), Cooren develops
a typology of illocutionary acts based on verb valencies. According to this model,
assertives are bivalent and performatives are trivalent speech acts (2000: 89). The
latter involve an agent, an object, and a recipient. The tetravalent actantial model
of the factive verbs, which involve four actants, then allows opening up the traditional speaker/hearer schema by acknowledging action from a distance, which
could be called tele-action (Cooren 2006: 2). The utterance John asked Mary to
send his report to Bill is an example of a tetravalent teleactive speech act. To telecommunicate, thus means to delegate a speech act to another agent who communicates it on behalf of the delegating agent: one discursive agent mobilizes other
agents that serve as representative or delegates of the delegating agent (Cooren
2000: 6).
In view of the fundamental differences between Peirces pragmaticist semiotics
and Greimass structural discourse semiotics, it is remarkable that both schools of
semiotics, for very different reasons, agree in their criticism of the speaker-centered approaches to speech acts. It is equally remarkable that, quite independently
of each other, both schools of semiotics postulate models of semiotic agency in
verbal interaction radically distinct from those proposed by the classics of linguistic pragmatics. Nevertheless, the fundamental difference between the two semiotic approaches in this latter respect must not be ignored. Whereas Greimass
model attributes agency to actants in discourse, Peirce attributes agency and autonomy to the sign itself.

Notes
1. Cf. Posner (1991) and Nth (2000: 433438).
2. See Parret (1983), Posner (1997), Nadin (1987, 1993), Deledalle, ed. (1989), Driel, ed.
(1991), Hilpinen (1995, 1998), Pankow (1995), Pietarinen (2004, 2005, 2006), Rellstab
(2008), and Andersen (2009).
3. Cf. Liszka (1996: 315), Pietarinen (2004), and Houser (2009: 91).
4. CP refers to Peirce (193158).
5. Cf. Fisch (1986: 339), Liszka (1996: 109), and Bergman (2007).
6. EP 2 refers to Peirce (1998).
7. Oehler (1990, 1995), Colapietro (1995), Bergman (2000, 2004), Santaella and Nth
(2004), and Pietarinen (2006).
8. See also Prieto (1966, 1975; cf. Nth 2000: 228).
9. See 2.2 and Burkhardt (1990).
10. MS refers to Peirce (1979).
11. Cf. Allwood (1978: 4) and Levinson (1983: 103).
12. For Peirces distinction between presciss (from prescind) and precise see Spinks (1991:
1923) and CP 5.449, 1905. Roughly, prescission is giving attention to one element
and neglect of the other (CP 1.549, 1867).

Semiotic foundations of pragmatics


13.
14.
15.
16.
17.
18.
19.

195

W 1 refers to Peirce (1981).


Cf. Oehler (1995: 268) and Johansen (1993).
Cf. Santaella (1988, 1990, 1994), Joswick (1996), and Pape (1996b).
Brock (1975, 1981a, b), Martens (1981), and Thibaud (1997).
Cf. Brock (1981b) and Hilpinen (1998).
For Greimas, see Nth (2000: 11219).
Sbis and Fabbri (1980), Sbis (1983, 1984, 1985, 1987, 1994, 2002), and Cooren
(2000, 2006).

References
Allwood, Jens
1978
On the Analysis of Communicative Action (Gothenburg Papers in Theoretical
Linguistics 38). Gothenburg: University, Department of Linguistics.
Andersen, Christiane
2009
Semiotics. In: Jef Verschueren and Jan-Ola stman (eds.), Key Notions for
Pragmatics, 217228. Amsterdam: Benjamins.
Austin, John
1962
How to Do Things with Words. London: Oxford University Press.
Bar-Hillel, Yehoshua
1954
Indexical expressions. Mind 63: 35974.
Bergman, Mats
2000
Reflections on the role of the communicative sign in semeiotic. Transactions of
the Charles S. Peirce Society 36.2: 225254.
Bergman, Mats
2003
Peirces derivations of the interpretant. Semiotica 144: 117.
Bergman, Mats
2004
Fields of Signification. Explorations in Charles S. Peirces Theory of Signs.
(Philosophical Studies from the University of Helsinki 6), Helsinki, PhD dissertation.
Bergman, Mats
2007
The secret of rendering signs effective: the import of C. S. Peirces semiotic
rhetoric. The Public Journal of Semiotics 1.2: 211.
Brock, Jarrett
1975
Peirces conception of semiotics. Semiotica 14: 124141.
Brock, Jarrett
1979
Principal themes in Peirces logic of vagueness. Peirce Studies (Lubbock,
Texas) 1: 4150.
Brock, Jarrett
1981a
An introduction to Peirces theory of speech acts. Transactions of the Charles
S. Peirce Society 17: 319326.
Brock, Jarrett
1981b
Peirce and Searle on assertion. In: Kenneth Ketner, Joseph M. Ransdell, Carolyn Eisele, Max H. Fisch and Charles S. Hardwick (eds.), Proceedings of the
C.S. Peirce Bicentennial International Congress. 281287. Lubbock, TX:
Texas Tech Press (Graduate Studies Texas Tech University 23).

196

Winfried Nth

Bhler, Karl
1933
Die Axiomatik der Sprachwissenschaft. Frankfurt am Main: Klostermann.
Burkhardt, Armin, (ed.)
1990
Speech Acts, Meaning and Intentions. Berlin: Walter de Gruyter.
Buyssens, Eric
1943
La communication et larticulation linguistique. Bruxelles / Paris: Presses
Universitaires (revised edition 1964).
Carnap, Rudolf
1975
Introduction to Semantics and Formalization of Logic, 2 vols. in 1. Cambridge:
Harvard University Press [1942/43].
Colapietro, Vincent
1992
Purpose, power, and agency. The Monist 75: 423444.
Colapietro, Vincent
1995
Immediacy, opposition, and mediation: Peirce on irreducible aspects of
the communicative process. In: Leonore Langsdorf and Andrew R. Smith
(eds.), Recovering Pragmatisms Voice: The Classical Tradition, Rorty, and
the Philosophy of Communication, 2348. Albany, NY: State University
Press.
Cooren, Franois
2000
The Organizing Property of Communication. Amsterdam: Benjamins.
Cooren, Franois
2006
Between semiotics and pragmatics: Opening language studies to textual
agency. Paper presented at the annual meeting of the International Communication Association, Dresden. <http://www.allacademic.com/meta/p87081_
index.html, access 05-25-2009>.
Davidson, Donald
1980
Essays on Actions and Events. Oxford: Oxford University Press.
Deacon, Terrence
1997
The Symbolic Species. New York, NY: Norton.
Deledalle, Grard (ed.)
1989
Semiotics and Pragmatics. Amsterdam: Benjamins.
Derrida, Jacques
1988
Limited Inc. Evanston, IL: Northwestern University Press.
Driel, Herbert van (ed.)
1991
Het semiotisch pragmatisme van Charles S. Peirce. Amsterdam: Benjamins.
Eisele, Carolyn
1987
Peirces pragmatics. In: Herbert Stachowiak (ed.), Der Aufstieg des pragmatischen Denkens im 19. und 20. Jahrhundert (Pragmatik: Handbuch des pragmatischen Denkens, vol. 2.), 8397 Hamburg: Meiner.
Fisch, Max H.
1986
Peirce, Semeiotic, and Pragmatism. Ed. Kenneth L. Ketner and Christian
J. W. Kloesel. Bloomington: Indiana University Press.
Gorl, Dinda
1993
Evolving through time: Peirces pragmatic maxims. Semiosis 71/72: 313.
Goudge, Thomas A.
1965
Peirces index. Transactions of the Charles S. Peirce Society 1: 5270.

Semiotic foundations of pragmatics

197

Greimas, Algirdas J. and Joseph Courts


1979
Smiotique: Dictionnaire raisonn de la thorie du langage. Paris: Hachette.
Trad. 1982. Semiotics and Language: An Analytical Dictionary. Bloomington,
IN: Indiana University Press.
Grice, Paul
1957
Meaning. Philosophical Review 64: 377388.
Grice, Paul
1989
Studies in the Way of Words. Cambridge, MA: Harvard University Press.
Habermas, Jrgen
1981
Theorie des kommunikativen Handelns, 2 vols. Frankfurt am Main: Suhrkamp.
Habermas, Jrgen
1995
Peirce and communication. In: Kenneth Laine Ketner (ed.), Peirce and Contemporary Thought. Philosophical Inquiries, 243266. New York: Fordham
University Press.
Hervey, Sndor G. J.
1982
Semiotic Perspectives. London: Allen and Unwin.
Hilpinen, Risto
1995
Peirce on language and reference. In: Kenneth Laine Ketner (ed.), Peirce and
Contemporary Thought. Philosophical Inquiries, 272303. New York: Fordham University Press.
Hilpinen, Risto
1998
Pragmatics and pragmatism: On C.S. Peirces pragmatic theory of meaning. In:
Paul Weingartner, Gerhard Schurz, and Georg Dorn (eds.), Die Rolle der Pragmatik in der Gegenwartsphilosophie The Role of Pragmatism in Philosophy,
151164 (=Akten des 20. Internationalen Wittgenstein-Symposiums. Schriftenreihe der Wittgenstein-Gesellschaft, vol. 26) Wien: Hlder-Pichler-Tempsky.
Houser, Nathan
1998
Introduction. In: Nathan Houser and Christian J. W. Kloesel (eds.), The Essential Peirce. Vol. 1, xvii-xxxviii. Bloomington: Indiana University Press.
Houser, Nathan
2009
Peirce, phenomenology and semiotics. In: Paul Cobley (ed.), The Routledge
Companion to Semiotics, 89100. London: Routledge.
Jakobson, Roman
1960
Linguistics and poetics. In: Thomas A. Sebeok (ed.), Style in Language,
350377. Cambridge, MA: MIT Press.
Johansen, Jrgen Dines
1993
Dialogic Semiosis. Bloomington: Indiana University Press.
Johansen, Jrgen Dines
2002
Literary Discourse: A Semiotic-Pragmatic Approach to Literature. Toronto:
University of Toronto Press.
Joswick, Hugh
1996
The object of semeiotic. In: Vincent M. Colapietro and Thomas M. Olshewsky
(eds.), Peirces Doctrine of Signs, 93102. Berlin: de Gruyter.
Kappner, Stefan
2004
Intentionalitt aus semiotischer Sicht: Peirceanische Perspektiven. Berlin: de
Gruyter.

198

Winfried Nth

Lakoff, George and Mark Johnson


1980
Metaphors We Live By. Chicago: University Press.
Leech, Geoffrey N.
1983
Principles of Pragmatics. London: Longman.
Levinson, Stephen C.
1983
Pragmatics. Cambridge: University Press.
Liszka, J. Jakb
1996
A General Introduction to the Semeiotic of Charles S. Peirce. Bloomington: Indiana University Press.
Liszka, J. Jakb
2009
Rethinking the pragmatic theory of meaning. Cognitio (So Paulo) 10.1:
6180.
Luhmann, Niklas
1984
Soziale Systeme. Frankfurt am Main: Suhrkamp.
Martens, Ekkehard
1981
C. S. Peirce on speech acts. In: Kenneth Ketner, Joseph M. Ransdell, Carolyn
Eisele, Max H. Fisch and Charles S. Hardwick (eds.), Proceedings of the
C.S. Peirce Bicentennial International Congress, 289292. Lubbock, TX:
Texas Tech Press (=Graduate Studies Texas Tech University 23).
Montague, Richard
1970
Pragmatics and intensional logic. Synthese 22: 6894.
Morris, Charles W.
1938
Foundations of the Theory of Signs (Foundations of the Unity of Science:
Towards an International Encyclopedia of Unified Science, vol. 1.2.) Chicago,
IL: University Press [1970].
Morris, Charles W.
1946
Signs, Language, and Behavior. New York: Braziller. (Quoted from the reprint
in C. W. Morris. 1971. 73398.)
Morris, Charles W.
1964
Signification and Significance: A Study of the Relations of Signs and Values.
Cambridge, MA: MIT Press.
Morris, Charles W.
1971
Writings on the General Theory of Signs. The Hague: Mouton.
Murphy, John P.
1990
Pragmatism from Peirce to Davidson. Boulder, CO: Westview Press.
Nadin, Mihai
1987
Pragmatics in the semiotic frame. In: Herbert Stachowiak (ed.), Der Aufstieg
des pragmatischen Denkens im 19. und 20. Jahrhundert, 148170. (Pragmatik: Handbuch des pragmatischen Denkens, vol. 2.) Hamburg: Meiner.
Nadin, Mihai
1993
Semiotics in action: The pragmatic level. In: Herbert Stachowiak (ed.), Spachphilosophie, Sprachpragmatik und formative Pragmatik, 219250. (Pragmatik: Handbuch des pragmatischen Denkens, vol. 4.) Hamburg: Meiner.
Nth, Winfried
2000
Handbuch der Semiotik, 2nd rev. ed. Stuttgart: Metzler.
Nth, Winfried
2002a
Charles Sanders Peirce, pathfinder in linguistics. Interdisciplinary Journal for
Germanic Linguistics and Semiotic Analysis 7.1: 114.

Semiotic foundations of pragmatics

199

Nth, Winfried
2002b
Wrter als Zeichen: Einige semiotische Aspekte der Sprache. In: Jrgen
Dittmann and Claudia Schmidt (eds.), ber Wrter, 932. Freiburg: Rombach.
Nth, Winfried
2006
Representation and reference according to Peirce. Caderno. 9a Jornada do
Centro de Estudos Peirceanos / IIIrd Advanced Seminar on Peirces Philosophy
and Semiotics, 7183. (So Paulo, PUC-SP), So Paulo.
Nth, Winfried
2008
Natural signs and the natural ingredients of cultural signs. In: Martin Dring,
Hermine Penz and Wilhelm Trampe (eds.), Language, Signs and Nature: Ecolinguistic Dimensions of Environmental Discourse, 2137. Tbingen: Stauffenburg.
Nth, Winfried
2009
On the instrumentality and semiotic agency of signs, tools, and intelligent machines. Cybernetics and Human Knowing 16.12: 1136.
Nth, Winfried and Lucia Santaella
2009
Die Relevanz der Peirceschen Semiotik des Vagen fr die Theorie der Kommunikation. In: Siegfried J. Schmidt (ed.), Die (Un)mglichkeit der Kommunikation. Frankfurt am Main: Peter Lang (Special Issue SPIEL: Siegener Periodicum zur Internationalen Empirischen Literaturwissenschaft 26.1 [2007],
7396).
Oehler, Klaus
1990
Peirce oder Habermas? In: Elisabeth Walther and Udo Bayer (eds.), zeichen
von zeichen fr zeichen: festschrift fr max bense, 4247. Baden-Baden: Agis.
Oehler, Klaus
1995
A reponse to Habermas. In: Kenneth Laine Ketner (ed.), Peirce and Contemporary Thought, 267271. New York: Fordham University Press.
Pankow, Christiane
1995
Semiotics. In: Jef Verschueren, Jan-Ola stman and Jan Blommaert (eds.),
Handbook of Pragmatics Manual, 469476. Amsterdam: Benjamins.
Pape, Helmut
1996a
Charles Sanders Peirce. In: Tilman Borsche (ed.), Klassiker der Sprachphilosophie, 307324. Mnchen: Beck.
Pape, Helmut
1996b
Object and final cause in Peirces semeiotic. In: Vincent M. Colapietro and
Thomas M. Olshewsky (eds.), Peirces Doctrine of Signs, 103117. Berlin: de
Gruyter.
Parmentier, Richard J.
1985
Signs place in medias res: Peirces concept of semiotic mediation. In: Elizabeth Mertz and Richard J. Parmentier (eds.), Semiotic Mediation, 2348. Orlando, FL: Academic Press.
Parret, Herman
1983
Semiotics and Pragmatics. Amsterdam: Benjamins.
Parret, Herman (ed.),
1994
Pretending to Communicate. Berlin: de Gruyter.

200

Winfried Nth

Peirce, Charles Sanders


193158 Collected Papers, Vols. 16. ed. Charles Hartshorne and Paul Weiss, Vols. 78,
ed. Arthur W. Burks. Cambridge, MA: Harvard University Press (quoted as
CP).
Peirce, Charles Sanders
1979
The Charles S. Peirce Papers, 30 reels, 3rd microfilm edition. Cambridge, MA:
The Houghton Library, Harvard University, Microreproduction Service
(quoted as MS).
Peirce, Charles Sanders
1981
Writings of Charles S. Peirce. Vol. 1: 18571866. Bloomington, IN: Indiana
Univ. Press (quoted as W 1).
Peirce, Charles Sanders
1998
The Essential Peirce. Selected Philosophical Writings. Vol. 2, 18931913).
Ed. by the Peirce Edition Project. Bloomington, IN: Indiana University Press
(quoted as EP 2).
Pietarinen, Ahti-Veikko
2004
Grice in the wake of Peirce. Pragmatics and Cognition 12.2: 295315.
Pietarinen, Ahti-Veikko
2005
Relevance theory through pragmatic theories of meaning. In: Bruno G. Bara,
Lawrence Barsalou and Monica Bucciarelli (eds.), Proceedings of the XXVII
Annual Conference of the Cognitive Science Society, 17671772. Mahwah, NJ:
Lawrence Erlbaum,
Pietarinen, Ahti-Veikko
2006
Signs of Logic. Peircean Themes on the Philosophy of Language, Games, and
Communication. Dordrecht: Springer.
Plato
1953
The Dialogues of Plato, transl. Benjamin Jowett, 4th ed. (1st ed. 1871). Oxford:
Clarendon. Also: http://www.gutenberg.org/dirs/1/6/1/1616/1616.txt.
Posner, Roland
1985
La syntactique, sa relation la smantique et la pragmatique, la morphologie
et la syntaxe, et la syntagmatique et la paradigmatique. In: Herman Parret and
Hans-Georg Ruprecht (eds.), Exigences et perspectives de la smiotique. Recueil dhommages pour A.J. Greimas, vol. 1, 7396. Amsterdam: Benjamins.
Posner, Roland
1991
Research in pragmatics after Morris. Dedalus (Lisbon) 1: 115156.
Prieto, Luis J.
1966
Messages et signaux. Paris: Presses Universitaires.
Prieto, Luis J.
1975
Etudes de linguistique et de smiologie gnrales. Genve: Droz.
Ransdell, Joseph
1977
Some leading ideas of Peirces semiotic. Semiotica 19: 157178.
Rellstab, Daniel
2008
Peirce for linguistic pragmatics. Transactions of the Charles Sanders Peirce
Society 44.2: 312346.
Santaella, Lucia
1988
Charles Sanders Peirces object (of the sign). Versus 49: 5358.

Semiotic foundations of pragmatics

201

Santaella, Lucia
1990
An introduction to Peirces concept of object. Cruzeiro semitico 13: 7582.
Santaella, Lucia
1994
Why things are objects of the sign. Cruzeiro semitico 18: 3746.
Santaella, Lucia
1999
Methodeutics, the liveliest branch of semiotics. Semiotica 124: 377395.
Santaella, Lucia and Winfried Nth
2004
Comunicao e semitica. So Paulo: Hacker.
Savan, David
1994
C. S. Peirce and American semiotics. In: Michael Shapiro and Michael
C. Haley (eds.), The Peirce Seminar Papers 2, 179208. Providence, RI: Berghahn.
Sbis, Marina
1983
Actes de langage et (acte d)nonciation. Langages 18.70: 99106.
Sbis, Marina
1984
On illocutionary types. Journal of Pragmatics 8: 93112.
Sbis, Marina
1985
Manipulation et sanction dans la dynamique des actes de langage. In: Herman
Parret and Hans-Georg Ruprecht (eds.), Exigences et perspectives de la smiotique. Recueil dhommages pour A.J. Greimas, Vol. 1, 529538. Amsterdam:
Benjamins.
Sbis, Marina
1987
Speech acts and context change. In: Thomas T. Ballmer and Wolfgang Wildgen
(eds.), Process Linguistics, 254280. Tbingen: Niemeyer.
Sbis, Marina
1994
Language and dialogue in the framework of the analytical philosophy of ordinary language. In: Edda Weigand (ed.), Concepts of Dialogue, 159170. Tbingen: Niemeyer.
Sbis, Marina
2002
Cognition and narrativity in speech act sequences. In: Anita Fetzer and Christiane Meierkord (eds.), Rethinking Sequentiality, 7197. Amsterdam: Benjamins.
Sbis, Marina and Paolo Fabbri
1980
Models ( ? ) for a pragmatic analysis. Journal of Pragmatics 4: 301319.
Searle, John R.
1969
Speech Acts: An Essay in the Philosophy of Language. Cambridge: University
Press.
Searle, John R.
1983
Intentionality. Cambridge: University Press.
Shannon, Claude E. and Warren Weaver
1949
The Mathematical Theory of Communication. Urbana: University of Illinois
Press.
Short, Thomas L.
2007
Peirces Theory of Signs. Cambridge: University Press.
Sperber, Dan and Deirdre Wilson
1986
Relevance: Communication and Cognition. Oxford: Blackwell.

202

Winfried Nth

Spinks, Cary W.
1991
Peirce and Triadomania. Berlin: de Gruyter.
Stetter, Christian
1989
Some remarks concerning the theory of speech acts and the idea of pragmatics. In: Gerard Deledalle (ed.), Semiotics and Pragmatics, 155165.
Amsterdam: Benjamins.
Strawson, Peter
1971
Intention and convention in speech acts. In: John R. Searle (ed.), The Philosophy of Language, 2338. Oxford: University Press.
Tesnire, Lucien
1959
Elments de syntaxe structurale. Paris: Klincksieck.
Thibaud, Pierre
1997
Between saying and doing: Peirces propositional space. Transactions of the
Charles Sanders Peirce Society 33: 270327.
Ventola, Eija
1987
The Structure of Social Interaction: A Systemic Approach to the Semiotics of
Service Encounters. London: Frances Pinter.
Watzlawick, Paul, Janet H. Beavin, and Don D. Jackson
1967
Pragmatics of Human Communication. New York: Norton.
Wittgenstein, Ludwig
1953
Philosophische Untersuchungen Philosophical Investigations. Oxford:
Blackwell.

7.

Pragmatics in modern philosophy of language


Nikola Kompa and Georg Meggle

Introduction
According to what might be called the traditional paradigm in the philosophy of
language, linguistic interpretation is a linear and serial process. In particular, pragmatic interpretation (such as the working out of Gricean implicatures) is claimed
to commence only after the semantic interpretation of a given sentence has been
worked out. This traditional paradigm has come under attack from various directions. Some philosophers contend, for instance, that there is no such thing as a
purely semantic content or a semantically expressed proposition (or truth-condition) but that pragmatic inferences are necessary at every level of linguistic interpretation: this claim will be discussed in the first part (chapter 1) of this article. One
may even go one step further and maintain that semantic notions need to be spelled
out pragmatically in the first place, that linguistic interpretation requires comprehension of a particular type of action and that semantics is thus simply a part of
pragmatics: this will be explored in the second part (chapter 2) of our article.

1.

Pragmatic inferences and semantic rules

1.1.

Two approaches to language and meaning

In 1938, Charles Morris distinguished three different dimensions of semiosis in


his theory of signs:1
One may study the relations of signs to the objects to which the signs are applicable.
This relation will be called the semantical dimension of semiosis, symbolized by the
sign DSEM; the study of this dimension will be called semantics. Or the subject of study
may be the relation of signs to interpreters. This relation will be called the pragmatical
dimension of semiosis, symbolized by the sign DP; the study of this dimension will be
named pragmatics. One important relation of signs has not yet been introduced: the formal relations of signs to one another. [] This third dimension will be called the syntactical dimension of semiosis, symbolized by the sign DSYN, and the study of this dimension will be named syntactics. (Morris 1938: 2122)

Syntax studies the relations of signs to other signs, semantics is concerned with the
relation between signs and objects, and pragmatics takes the relation between signs
and their interpreters as its subject matter. To partition the domain this way has
been common ever since. Morriss characterization, however, leaves the three
areas severely underdetermined. For example, pragmatics is commonly supposed

204

Nikola Kompa and Georg Meggle

to cover such diverse phenomena as speech acts, deixis, presuppositions or implicatures, their unifying feature being only that they have something to do with the
relation between signs and interpreters (and that they cannot easily be accommodated within a semantic theory). Also, the characterization may be not as clear-cut
as Morris and others hoped it to be. In particular (and as a consequence), where
exactly to draw the line between semantics and pragmatics is a hotly contested
issue in current philosophical and linguistic debate.2 Moreover, the difficulties one
encounters when trying to sharpen the distinction between semantics and pragmatics point to a deeper problem about how to theoretically approach language in
general. It is common in the philosophy of language to contrast two different approaches to questions concerning language, meaning and interpretation. Robert
Brandom, for example, puts the difference thus: some philosophers view language
as a kind of practice or activity, others prefer a semantic Tarskian approach to language (cf. Brandom 2002: 4041). The former focus on how language is used; they
study language as a kind of practice, as something speakers do in order to achieve
certain aims. The latter, on the other hand, strive to develop a theory of content that
is closely tied to Alfred Tarskis truth-theoretical considerations. Underlying the
two different approaches to language are different views of what a language is,
which David Lewis aptly characterizes thus:
What is a language? Something which assigns meanings to certain strings of types of
sounds or of marks. It could therefore be a function, a set of ordered pairs of strings and
meanings. [] (Lewis 1975/1983: 163)

Those who favor this view of what a language is think of language as an abstract
system that systematically pairs certain strings of sounds or marks with certain
other entities, called meanings. The mapping can be done, it seems, without taking
language users into account. In the contrasting view, on the other hand, the language user takes center stage:
What is a language? A social phenomenon which is part of the natural history of human
beings; a sphere of human action, wherein people utter strings of vocal sounds, or inscribe strings of marks, and wherein people respond by thought or action to the sounds
or marks which they observe to have been so produced. (Lewis 1975/1983: 164)

Which of these two views of language will prove theoretically fruitful is the point
at issue between semantically and pragmatically oriented accounts of language and
linguistic interpretation. Very roughly, the pragmaticist will advocate a usagebased approach to language. S/he will, accordingly, think of language as a human
activity, a social phenomenon that helps us interact and communicate with our fellow human beings. His/her opponent, the semanticist, takes a language to be a
formal system that can be studied fairly independently of how human beings actually use language. His/her aim is to come up with a formal semantic theory that
allows us to assign every sentence of language L its meaning in L. More specifically, s/he will try to pair syntactically analyzed sentences (of a language) with

Pragmatics in modern philosophy of language

205

their truth-conditions, as s/he typically takes the meaning of a sentence to be its


truth-condition (cf., e.g., Portner and Partee 2002: 3).
Semantic interpretation was commonly taken to be independent of pragmatics.
Some even claimed that semantics comes first (cf. Manor 2001: 67f). Pragmatic interpretation (such as the derivation of Gricean implicatures) comes in only after the
semantic interpretation of the expressions in question has been worked out. But
new work on the context sensitivity of natural languages questions not only the semantics first-view but the idea of semantic independence as such. More specifically, in recent years the project of a formal semantic theory has come under attack
from what came to be called contextualist approaches to language and meaning.
Very roughly, the contextualist can be characterized as someone who takes context sensitivity to be a ubiquitous feature of natural languages, a feature much more
pervasive than deixis or anaphora and much less well-behaved than implicatures,
for example. S/he favors a pragmatically oriented, usage-based approach to meaning, arguing that linguistic (even truth-conditional) interpretation involves pragmatic inferences all the way down. Her/his opponent, on the other hand, advocates
a formal semantic approach to meaning according to which truth-conditional interpretation is a purely syntactic-semantic endeavor. S/he would, presumably, agree
with the following characterization:
[] in general, formal theories can be characterized as fundamentally syntax-driven
theories, which claim that it is possible to deliver an account of the propositional or
truth-conditional content of a sentence in natural language simply via formal operations
over the syntactic features of that sentence, that is, over the lexical items it contains and
their mode of composition. (Borg 2004b: 3)

The pragmatically oriented contextualist and his/her semantically oriented opponent negotiate what was traditionally taken to be key semantic notions: the notion
of what is said in the utterance of a sentence, the notion of the proposition expressed by the utterance, the notion of a sentences truth-conditions, and at the
most basic level the notion of meaning itself. Semanticists try to save one or another of these notions from pragmatic intrusion.
1.2.

Formal semantics

Let us first try to get a better grasp of what form a formal semantic theory might
take before we consider, in a second step, the objections the contextualist raises
against this kind of approach. Very roughly, formal semantic theories try to develop, for as large a fragment of natural language as possible, a formal representation that contains all syntactically and semantically relevant information. In a
first step, the logical form of each sentence in the relevant fragment of the language
in question has to be worked out. In a second step, the elements of logical form
have to be interpreted, i.e., each element has to be assigned a semantic value in ac-

206

Nikola Kompa and Georg Meggle

cordance with the respective interpretation rules. More specifically, the idea is that
there are interpretation rules that assign semantic values to the basic expressions in
the language. Composition rules then tell us which value to assign to complex expressions, the semantic value of a complex expression being a function of the semantic values of its parts. In this manner we work our way up until, after finitely
many steps, we are able to assign semantic values (i.e., truth-conditions) to whole
sentences. Compositionality obviously plays an important role here, for if language were not compositional, no recursive procedure for assigning semantic
values to expressions of arbitrary complexity would be available. Yet such a recursive procedure is exactly what the semanticist is looking for: a procedure that
yields for every well-formed sentence (of finite length) its meaning, i.e., the proposition expressed or its truth-conditions, after finitely many steps. And to interpret
an expression is to follow such a recursive procedure. Here is a somewhat lengthy
quote from Jason Stanley, a (moderate) proponent of formal semantics, to illustrate
the point:
Successful interpretation involves assigning denotations to the constituents of the logical form, and combining them in accord with composition rules that do not vary with
extra-linguistic context. The denotations that successful interpreters will assign to constituents of a logical form will be constrained by the linguistic conventions governing
those elements. [] What results from a successful application of this first stage of
interpretation is a unique proposition, a fully truth-evaluable entity. (Stanley 2002:
149150)

One of the attractive features of a formal semantic approach is that it seems to provide a good explanation of the productivity of language. Obviously, competent
speakers are able to produce novel sentences and to understand sentences they
have not encountered before. This fact stands in need of explanation. The formal
semanticist seems able to provide an explanation by pointing out the recursive
character of linguistic interpretation.
Yet what semantic value do interpretation rules assign to the basic expressions
of a language? What is the semantic value, i.e. the meaning, of a linguistic expression? It is the answer to this question on which philosophical theories of meaning differ. According to the reference theory of meaning it is the expressions reference, often called its extension: what the expression is used to talk about or refer
to. Unfortunately, the reference theory of meaning faces a couple of serious and
well-known objections, some of which already attracted Gottlob Freges attention
and brought him to distinguish between an expressions reference (what he called
Bedeutung) and its sense (Sinn, cf. Frege 1891, 1892a and 1892b). Freges insights, in turn, loomed large in the development of so called intensional semantics.
Rudolf Carnap, who attended some of Freges lectures in Jena, strived to find a
suitable method for the semantic analysis of meaning (Carnap 21956: 2) and proposed the concept of extensions and intentions of an expression as a suitable tool.
This distinction between an expressions extension and intension soon caught on.

Pragmatics in modern philosophy of language

207

Very roughly, the idea is this: It may not be adequate to equate an expressions
meaning with its actual extension because the latter may depend on the way the
world is, which may be subject to change. Yet isnt the task of an expressions
meaning to pursue this kind of dependence? In other words, isnt it the case that
meaning determines the expressions extension in any possible world? Such functions from possible worlds to extensions are called intensions. The intension of an
expression is supposed to determine its extension (i.e. its truth-value, in the case of
whole sentences) in any possible world. In other words, intensions take possible
worlds as arguments and provide extensions as truth-values. Hence, intensions
seem to be better candidates than extensions for what we intuitively think of as an
expressions meaning.
One of the most influential later developments in intensional semantics is Montague Grammar. It was originally developed by Robert Montague who elaborated
on ideas introduced by Rudolf Carnap, Alfred Tarski and Alonzo Curch and subsequently modified and expanded by other logicians, philosophers and linguists
(cf. Montague 1970a, 1970b, 1973, and also Partee 1973, 1975; Porter and Partee
2002).
In the course of these considerations it became more and more obvious that the
extension of an expression not only depends on the way the world is but also, in
some cases at least, on quite specific features of the context in which the expression
is used. The most prominent examples are indexical expressions such as I,
here or now. What these expressions refer to depends on the context of utterance; similarly for demonstrative pronouns such as this or that. Insight into
this double-dependence of extension on context and world respectively has led to
the development of what came to be called two-dimensional semantics. According
to a renowned view held by David Kaplan (1989), for example, the overall import
of a linguistic expression breaks down into two components: its character and its
content. Both can be construed as functions, the character of an expression being a
function from the context of utterance into the expressions content, the content
being a function from possible circumstances of evaluation (very roughly, something like a possible world or an index, terminology differs here) into the extension
of the expression in the respective circumstances.3 This opens up a new way of accommodating context sensitive expressions within a formal semantic theory: just
associate with each context sensitive expression a function or rule (a character)
that determines, for any possible context of its use, the expressions content (its extension, in the case of indexical or demonstrative expressions) in that context.
Thought of in this way, linguistic (truth-conditional) interpretation is (context-sensitivity notwithstanding) still a purely syntactic-semantic enterprise. No pragmatic
processes seem to be required; a result that nicely fits in with the formal semantic
project as sketched above.

208
1.3.

Nikola Kompa and Georg Meggle

Context-sensitivity and pragmatic inferences

Still, as mentioned above, the formal semantic project came under attack when it
became obvious that there is much more context-sensitivity in natural language
than the formal semanticist is willing to acknowledge. Moreover, it is far from
clear that she could accommodate all the context-sensitivity within her formal
semantic approach even if she wanted to. The main problem is that pragmatic inferences seem to be required after all in order to truth-conditionally interpret utterances containing context-sensitive expressions and to assign these expressions a
semantic value in context. Yet the nature of pragmatic inferences does not square
well with the allegedly deductive, algorithmic nature of semantic interpretation
(cf. Borg 2004a: 218219).
1.3.1.

The non-monotonic nature of pragmatic inferences

The need for pragmatic inferences in linguistic interpretation has long been acknowledged in philosophical and linguistic theorizing. Ever since Paul Grice developed his theory of conventional and conversational implicatures the need for
pragmatic inferences had become obvious. According to Grice (1989), what a
speaker means by an utterance divides up into what she says and what she implicates. What she says is (subject to certain qualifications) closely tied to the conventional meaning of the words used. The implicatures (at least the conversational ones), on the other hand, have to be pragmatically worked out. Usually, in
generating an implicature, a speaker exploits certain maxims her hearers expect
her to comply with, maxims to the effect that she should try to be relevant, succinct, truthful and informative. By violating or flouting one of these maxims, a
speaker can communicate something other or even more than what she explicitly
said.
To use one of Grices own examples, if you ask me whether our common friend
Smith has a girlfriend at the moment, and I reply by saying that he has been paying
a lot of visits to Zurich lately, you might reason as follows. What she said was not
directly relevant for my question. But then, I have no reasons to assume that she is
not cooperative. So she must have meant to communicate something other than
what she said. What could that be? Maybe she meant to suggest that Smith has a
girlfriend in Zurich. The conclusion is defeasible, though. It might need revision
in the light of new information. And this line of reasoning is an example of a pragmatic inference. Since Grice, it has been common to think of pragmatic inferences
as being restricted to cases such as these. More specifically, pragmatic inferences
are commonly taken to come into play in linguistic interpretation only after the
semantic, truth-conditional content of the sentences has been worked out. Yet that
is exactly what the contextualist denies. She claims that pragmatic inferences are
required all the way down. And the problem is that the defeasible nature of prag-

Pragmatics in modern philosophy of language

209

matic inferences clashes with the deductive nature of semantic interpretation. This
is problematic for various reasons:
Firstly, because in these cases it is not enough to just follow any allegedly semantic rule or algorithm, because there is no semantic rule or algorithm to follow.
(Though that is not to say that there might not be any syntactic rule, i.e. a rule of
grammar).
Secondly, a different conception of context is called for than the one invoked in
syntactic-semantic interpretation if our aim is to accommodate all different forms
of context-sensitivity. According to a very common conception of context (e.g.
Lewis 1980; Kaplan 1989), an utterance context is taken to comprise a neat list of
contextual features such as a world, a speaker, a time and a place of utterance, and
thus to determine at least the values of indexical expressions.4 Yet such a conception of context will not do for present purposes. For one thing, because the idea of
context determining anything stands in dire need of explanation. More importantly,
what is needed is a conception of context according to which a context is seen as
something that provides information: context has to provide all the information
that speaker and interpreters have to draw on in order for communication and interpretation to be successful. That seems to be the idea behind Kent Bachs notion
of wide context:
Wide context does not literally determine anything. It is the body of mutually evident information that the speaker exploits to make his communicative intention evident and
that his audience relies upon, taking him to intend them to do so, to identify that intention. (Bach and Bezuidenhout 2002: 285)

Bach emphasizes that wide context only enables the hearer to figure out the semantic value of a given expression but does not determine it. Content is ascertained
on the basis of information provided by wide context. This epistemic notion of
context as providing speaker and hearer with information about how to interpret a
given utterance seems to be better suited for modelling all different forms of context-sensitivity.5
Thirdly, and most importantly, pragmatic inferences often take the form of an
inference to the best explanation or rather to the best interpretation.6 As a result,
they are liable to be defeated by new information. And there is no limit to the
amount of information that may turn out to be relevant (cf. Borg 2007: 356 and
Borg 2004b: 79; Recanati 2004: 54). More specifically, pragmatic inferences are
defeasible, non-monotonic, in the following sense: It may be the case that although
conclusion C follows (not in the deductive sense, of course) from a set P of premises, it no longer follows from a set M, such that M P, i.e. such that P is a proper
subset of M. In other words, adding new premises might turn a good inference
into a bad one. (Defeasible reasoning is, of course, not confined to cases of linguistic interpretation. There is, by now, a huge literature on this mode of reasoning;
see for an overview Koons 2009). But if that is so, then there may not be an algo-

210

Nikola Kompa and Georg Meggle

rithmic procedure for pragmatic interpretation as there is no guarantee that the task
can be completed after finitely many steps. Inferences concerning speaker intentions are examples of pragmatic inferences. Here is an example of Borgs to illustrate the defeasible nature of intentional interpretation:
So, imagine that you see Sally filling a glass of water from the tap. Then you might reason as follows: Sally is getting a glass of water form the tap. The best explanation for
this action is that Sally is thirsty and wants a drink; therefore Sally is thirsty and wants a
drink. Clearly this is a non-demonstrative piece of reasoning and it is susceptible to the
influence of an open-ended range of contextual factors. For instance, say that you know
that Sally has just come in with Sourav and that Sourav is wearing running gear and
looks out of breath, then the best explanation for Sallys action might be that Sourav is
thirsty and Sally is getting a drink for him. Or imagine that Sally has just glanced at her
potted plant, then the best explanation might be that she wants to water her plant. (Borg
2004b: 7879)

Interpreting a persons actions in the way described clearly requires pragmatic inferences. And the interpretation of linguistic utterances is similar in that respect to
the interpretation of actions. For many linguistic expressions are such that in order
to figure out what has been said, speaker intentions at least to the extent that
speakers manage to make them manifest need to be taken into account. Actually,
even indexical and demonstrative expressions seem to require pragmatic inferences for their interpretation; what place or time exactly a speaker refers to with an
indexical expression seems to depend, partially at least, on which region in space
or what stretch of time he intended to refer to. (But let us put these problems aside
for the time being.) Yet intentional interpretation is not the only case where pragmatic inferences come into play. In interpreting utterances, hearers not only draw
on contextually provided information about (manifest) speaker intentions; other
contextual information has to be taken into account, too.
1.3.2.

Context-sensitivity some examples

For the purpose of illustration, let us consider a few more examples of context-sensitivity in natural language. Suppose a speaker says
The suitcase is too heavy.
at one time while trying to put a suitcase into the trunk of a car, at another time
while packing his hand luggage for a flight to Paris (for a similar example, cf. Bach
1994: 268). Or suppose a speaker says:
Everyone came to the party. (Stanley 2005: 223224)
at one time while discussing whether all the important people in his department
came to the party, at another time when commenting on how crowded the party
was.
As opposed to sentences such as triangles have three sides or bachelors are
unmarried, our example sentences are context-sensitive in that they can be used to

Pragmatics in modern philosophy of language

211

say different things relative to different contexts of utterance; and their truth-conditions differ accordingly. Here is another example.
Jill cant continue. (Borg 2004b: 228)
This sentence can be used to say that Jill cannot continue school, or that she cannot
continue dance classes, or university education, and so on. What it is used to say on
a particular occasion of utterance usually depends on context and can, accordingly,
be figured out by taking context into account. Moreover, whether what is thereby
said is true or false will depend on context, too. Similarly if a speaker says
Tom is tall.
The speaker might have used the sentence to say that Tom is tall for a fifth grader or
that he is tall for an NBA player, etc. Moreover, his utterance may well be true in
the first case but false in the second.7
Accordingly, the contextualist will claim that in order to figure out what has
been said (not what has been implicated in the Gricean sense!) on a particular occasion of their use, most natural language sentences require pragmatic inferences
on the part of the interpreter. Moreover, the contextual clues the interpreter has to
draw on include not only information about the speakers intentions but also information about the purpose or point of the conversation, the participants shared
background assumptions and interests, etc.
Even more interesting from a contextualist point of view are still subtler forms
of context sensitivity. Many natural-language expressions can be used to depict
slightly different actions or entities in different contexts of their use (owing to the
fact that language is used for classification, and classification itself is context-sensitive). Accordingly, they can be interpreted differently, again depending on the
purpose or point of the conversation, the participants shared background assumptions, etc. That is just to say that there is ample polysemy in natural language,
where [p]olysemy is understood here in a broad sense as variation in the construal
of a word on different occasions of use (Croft and Cruse 2004: 109). Compare the
following utterances:
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)

Jane opened the window


Bill opened his mouth.
Sally opened her book to page 56.
Mike opened his briefcase.
Pat opened the curtains.
The child opened the package.
The carpenter opened the wall.
The surgeon opened the wound. (Carston 2002: 325)

To open a wound is not the same thing as to open a book. And a wall is usually
opened not in the way one opens ones mouth. The word open can be used to depict a variety of different actions; and an interpreter has to contextually work out,
i.e., pragmatically infer the appropriate understanding. There may be default inter-

212

Nikola Kompa and Georg Meggle

pretations, though, depending on the object of the verb. But these preferred interpretations are not inevitable. Also, one might think that in all these cases there is
something like a semantic core meaning. In the above example, manipulate such
that the inside/interior/thing underneath can be seen might be a promising candidate. But then the question arises of whether open is supposed to be synonymous
with manipulate such that the inside/interior/thing underneath can be seen. If it is
not supposed to be synonymous, the meaning of open differs from its core meaning, a puzzling result. Yet if it is supposed to be synonymous then one might
wonder what the point of exact synonyms could possibly be in natural language?
(And let us put aside the problem that giving a synonym is not the same as giving
the meaning of an expression, for one might know that a and b are synonymous, without knowing the meaning of either a or b.) Moreover, new work on
language acquisition seems to show that children, when learning a language, operate on the assumption that every difference in form (of linguistic expressions)
marks a difference in meaning (cf., e.g., Clark 1993: 64; Tomasello 2003: 7275ff;
Bloom 2000: 65). Otherwise language acquisition would be a mystery. So the idea
of exact synonyms and, consequently, of core meanings in the case of polysemy is
rather problematic.
Whether a sentence containing a polysemic expression is correctly applicable
to a particular situation depends on the participants interests and purposes, their
background assumptions and interests, etc. Given such and such purposes, interests, etc, it may be correctly applied, given different purposes, interests, etc., it may
be not correctly applied. Yet couldnt we specify further and try to make some of
the purposes, interests or common assumptions explicit? The problem is, as Anne
Bezuidenhout has pointed out, that
[t]here is no sentence that we can produce that can settle all questions about how some
original sentence is to be understood, since language doesnt function that way. It is not
self-interpreting. (Bezuidenhout 2002: 113)

In performing pragmatic inferences, one may have to draw on contextual information concerning the circumstances of utterance, speaker intentions (to the extent that they are manifest), the participants interests and background assumptions, the purpose or point of the conversation, and so on. More generally, we
interpret someones utterance against the background of a shared system of knowledge and a common view of how nature works and how our culture works
(Searle 1980: 226227).
This idea has been elaborated, among others, by Charles Travis (e.g., 1985,
1996, 1997), Julius Moravcsik (1998), and Anne Bezuidenhout (e.g, 2002), to
name just a few. Moravcsik, for example, takes polysemy to affect almost all natural language expressions (cf. Moravcsik 1998: 33).8 Here is an example of his to illustrate the point:

Pragmatics in modern philosophy of language

213

The verb walk admits of polysemy. On the one hand, there is a common meaning core,
locomotion with legs in appropriate position. But one has to add: appropriate distance,
covered in an appropriate time. For what counts as a walk for a toddlers first attempts
does not count as a walk for a normal, healthy adult, and the walk of a recovering patient
in a hospital is still a different matter. What counts as a walk depends on different senses
of the word. (Moravcsik 1998: 3536)

How the verb walk is to be understood, i.e., what counts as a walk, depends on
context, more specifically on the various ways we can and do interact with the
world, on our purposes, intentions, etc.: Our understanding of words is influenced
by the variety of ways in which we relate to the world. (Moravcsik 1998: 91) And
there is no limit to the different applications a given word may enjoy; there is no
way of telling in advance what the different senses of a word are that we need it to
have in order to be able to talk about everything we want to talk about (cf. Moravcsik 1998: 37). There are as many different uses to which we may put the expressions of our language as there are different purposes to which we may interact
with the world. The meanings of our words provide rough and ready guidelines for
their own future applications, in the light of new interests, purposes, assumptions
etc. Yet, as Moravcsik emphasizes, we cannot predict or exactly specify all past
and future human interactions with nature that have or will affect our uses of the
words of our language.
To echo John Austin, the question is: Why do we call different things by the
same name? (Austin 31979a: 69) A sensible answer seems to be: It allows us to use
a finite vocabulary to talk about or describe an (in principle) infinite array of situations. Jonathan Cohen puts the point thus:
Rather, it is an enormous convenience that the same word can often be uttered in one or
other of several different though related senses. Instead of having to learn a very much
larger number of words, each with fixed and context-independent meanings, we can
learn a relatively small number of words with variable meanings and then exploit their
verbal or situational contexts of utterance in order to disambiguate their actual occurrences. (Cohen 1985: 132)

Our expressions are flexible enough to fit all the various situations we encounter almost perfectly. Jon Barwise and John Perry speak of the efficiency of language
(Barwise and Perry 1983: 5; cf. also Goodman 21976: 80). The expressions of our
language answer to our human needs, they serve our practical purposes. They can be
contextually adjusted, if need be. And need there is as we never encounter exactly
the same situation twice over. And when we encounter a situation we have not encountered before, our expressions allow us to conceptually structure it by using familiar vocabulary. Sometimes, though, we have to unduly stretch the meaning of a
given expression so that it will become applicable to the case at hand. That yields
polysemy or metaphor, depending on the extent to which the new situation differs
from old ones. Yet metaphor, just as polysemy, seems to defy formal treatment. Both
phenomena seem to require pragmatic inferences for their interpretation.

214

Nikola Kompa and Georg Meggle

Metaphor in particular is commonly used to conceptualize abstract and mental


phenomena, often with the help of vocabulary familiar from the realm of sense experience. Metaphor is, as Nelson Goodman nicely put it, a matter of teaching an
old word new tricks. (Goodman 21976: 69) That is why we are feeling blue, or
complain about someone being cold, and so on. We also use metaphors in the explanation of human behavior. We say that he broke down under pressure thereby
exploiting the metaphor the mind is a brittle object (cf. Lakoff and Johnson 1980:
28). We say that she is feeling up, that her spirits rose, or that she sank into a coma.
With the help of these orientational metaphors (as Lakoff and Johnson 1980: ch.
4 call them), we conceptualize mental phenomena by giving them a spatial orientation. But abstract phenomena too are commonly conceptualized with the help of
metaphorical expressions. We construe theories and arguments as buildings: The
theory needs more support, it lacks a foundation; the argument collapsed, it was not
solid enough (cf. Lakoff and Johnson 1980: ch. 10). Or we think of love as if it
were a journey: look how far we have come; yet now we are at a crossroad; from
here on we have to go separate ways as this isnt going anywhere; we are in a deadend street. We commonly use expressions from the concrete realm of sense experience to conceptualize abstract and mental phenomena (cf. also Keller and Kirschbaum 2003: 36 and 99). So if one were to say, as formal semanticists with their
fondness for a clear-cut distinction between literal and figurative meaning would
be inclined to do, that metaphorical utterances are always strictly speaking false,
that would render most of our talk about abstract or mental phenomena false. Polysemy and metaphor are central features of natural languages without which we
would not be able to structure the multifariousness of our experience and to conceptually copy with mental and abstract phenomena.
Some philosophers take these and similar considerations to show that word
meanings (or concepts) are not ready-made entities but have to be constructed in
context. According to Robyn Carston, for instance, often in linguistic interpretation an ad hoc concept is constructed and functions as a constituent of what is
explicitly communicated. (Carston 2002: 357) And, presumably, pragmatic inferences are required in the construction of these ad hoc concepts. But one might even
go one step further and try to do without something like linguistic meaning (as traditionally understood) altogether (cf., e.g., Recanati 2004: 146ff; Croft and Cruse
2004: 97ff). The idea would be that although words encode semantic information,
the information itself does not amount to something like a full-fledged meaning.
Rather, it provides the starting point for pragmatic processes, which result in a contextually specified interpretation of the expressions in question. Linguistic meaning would provide information that constrains interpretation, while pragmatic
processes are required to yield the contextually relevant contribution the expression makes on a particular occasion of its use. According to this conception of
linguistic interpretation, the notion of linguistic meaning itself needs to be pragmatically spelled out.

Pragmatics in modern philosophy of language

1.4.

215

Conclusion

The preceding considerations cast doubt on the idea of a functional, systematic pairing of sentences with meanings (truth-conditions) as propagated by formal semantic
approaches to language. This way of approaching language seems to ignore the
highly context-sensitive nature of linguistic interpretation. Many natural language
expressions require for their proper interpretation that context be taken into account.
This kind of context-sensitivity rests on a conception of context according to which
context provides information about the participants shared interests, intentions, purposes, background assumptions and so on. The information is extracted by pragmatically inferring the contextually appropriate interpretation of a given context-sensitive utterance. Pragmatic inferences are indispensable in guring out what a given
expression contributes to the truth-conditions of a sentence containing it (i.e., to the
proposition expressed). Accordingly, the value assigned to the expression hardly deserves to be called a (purely) semantic value. And the process of interpretation performed hardly deserves to be called (purely) semantic interpretation. Rather, pragmatic processes seem to intervene at (almost) all stages of linguistic interpretation.
And maybe even the notion of linguistic meaning itself is a pragmatic notion.

Pragmatic semantics

2.1.

The center of pragmatic semantics

Just cancel the second word (maybe) of the last sentence in the preceding paragraph, and you have already reached the very center of Pragmatic Semantics (PS).
Whereas traditional semantics (including formal truth-conditional semantics as its
most developed form) starts with linguistic expressions and then asks how to define the various kinds of their respective linguistic meanings, PS starts just the
other way round, namely with the understanding (i.e., the knowing of the meanings) of actions. From the point of view of PS, semantics is a special part of pragmatics (Action Theory), in the ideal case therefore a part of Action Logic.
Throughout, we will have to make use of the following two-fold distinction
Type
Token

Action
(3)
(4)

Action-Product
(1)
(2)

or of its already language-related specification, respectively:


Utterance
Type
Token

Utterance-Product
Expression (form)
(Concrete) Expression

216

Nikola Kompa and Georg Meggle

Whereas traditional semantics starts from the right side, (1) above, PS starts from
and is primarily focussed on the left one.
Correspondingly, pragmatics is covering quite differing programs. Whereas,
as exemplified in chapter 1 above, according to traditional semantics it may mean
anything which goes beyond field (1), the program of Pragmatic Semantics as
sketched here is to define in the end even (1)-meanings themselves in terms of action-theoretical concepts.
In PS actions (and their meanings) come first. For this semantic paradigm,
there even need not be any expressions to start from. Actions, think of gestures, for
example, do not need to be product-producing actions (utterances of expressions)
in order to have meanings. Thus, in this approach expressions themselves may
be recognized or introduced as what they are: constructions on the basis of (some
continuity in) the meanings of the corresponding actions. Thus, PS is primarily a
Semantics of Actions.
2.2.

Rule-based versus intention-based pragmatic semantics

In Pragmatic Semantics we also find the division between nominalists and Platonists, depending on whether type-meanings (meanings of action-types, field (3) in
the schema above) can themselves be defined in terms of token-meanings ((4) in
our schema) or not. Since type-meanings (of the PS-relevant sort) are social facts
(i.e., state of affairs which, relative to population P, are facts iff it is common believe in P that they are facts), whereas the respective meanings of action-tokens are
just meanings for some particular subject, this (Platonist versus nominalist) division is a special case of the more general division within Social Philosophy between collectivist versus individualist approaches, according to whether or not
meanings as social facts are taken to be conceptually inexplicable with reference to
individualistic terms.
This distinction is very often identified with a distinction between the two possible readings of Wittgensteins basic equation of meaning = use, 9 which one may
refer to as the rule-based reading versus the intention-based reading. Whereas according to the rule-based reading we have, in order to understand (i.e. to know the
meaning of) a sign A, just to know the rules we have to follow in order to use A
correctly, the intention-based reading is referring to what in a particular situation A
is being used for. Which alternative will be the most fruitful one? This has been the
most debated question in the whole field of pragmatic semantics up to now (see
2.52.7 below).
2.3.

Intention-based semantics

This overview will start from Intention-Based Semantics, which is definitely following a strictly nominalistic / individualistic strategy by embedding PS into gen-

Pragmatics in modern philosophy of language

217

eral individualistic Action Theory including Rational Decision Theory as its most
important part (see Jeffrey 1983). The positions of the main collectivist alternatives can then easily be located by referring to their denials of and their alternatives to the respective aspects of Intention-Based Semantics.
For Intention-Based Semantics, as sketched by Herbert P. Grice (1957) and
further developed mainly by Stephen Schiffer (1972), Jonathan Bennett (1976),
David Lewis (1969), Franz von Kutschera (1975) and Georg Meggle (2010),
human languages are primarily used for interpersonal communication. Accordingly, linguistic meanings as social facts are primarily defined in terms of the communicative regularities or conventions holding in the respective social groups
(populations). But as the meanings of communicative action tokens may and often
do diverge from the meanings of the exemplified action type, communicative actions have first to be defined more generally, i.e. on the token level. Thus, the Intention-Based Semantics-program (even when restricted to the action side, i.e., not
yet including some expressions as their action-products) has to involve two steps:
(i) explication of general concepts of communicative acts (as tokens); and (ii) explication of (meanings of) communicative action types by means of (i). Whereas
step (ii) is at the very center of Intention-Based Semantics, (i) is its necessary preliminary stage defining communicative acts in terms of their respective communicative intentions.10
2.3.1.

General communication theory

Communicative acts are instrumental acts, with the most basic types of communicative acts being either directives or informatives, depending on whether Ss (the
speakers) primary communicative aim is to get H (the hearer) either to do or to believe something. But notice that this way of conveniently speaking of speaker
and hearer does not imply that the respective communicative acts are restricted
to acoustic utterances. The differentia specifica of communicative acts versus instrumental acts in general is that communicative acts are acts which, at least from
the point of view of S, will be successful if and only if they will be understood by
their addressee H. This entails the so called reflexivity-condition of communication: For an act to be a communicative one, its actor S has to intend it to be
understood by H (as being a communicative act). A General Communication Theory has to explicate these postulates in a non-circular way.
For this explication, one best starts from the more general communicative
acts: Instrumental acts are acts done for some end (or, to put it differently, done
with specific intentions). More precisely (but already strongly adapted to the
special cases of decisions under certainty, i.e., cases where the deciding person
expects the results of her actions with a maximal degree of probability), person X
is doing f with the intention of bringing it about that the state of affairs A will be
the case: (1) X does f, (2) X wants that A, and (3) X believes (expects) that A

218

Nikola Kompa and Georg Meggle

will be brought about as and only as a consequence of her doing f. Instrumental


acts will thus be successful iff the belief (3), let us call it the expectation of success, turns out to be true, i.e., if A will in fact come about in the way expected
by X, i.e., as and only as a consequence of Xs doing f. And you understand this
act iff you know its ends, i.e., iff you know the intentions with which f is or was
being done by X.11
As there are no restrictions as to what ends one may try to achieve by means
of communication, we have to distinguish between the intentions with which S is
performing her communicative act (i.e. her communication related intentions) in
general versus her acts communicative intentions in particular, this latter term
referring only to those intentions that are necessary for an act to be a (respective)
communicative act.
Unlike instrumental acts, communicative acts do not only involve the expectation of success, but also the expectation of understanding (i.e., the expectation
that H will understand Ss communicative act), both expectations connected in the
differentia specifica way already mentioned above: that S expects that success will
be brought about as and only as a consequence of understanding. Thus, for the
communicative act to be successful this expectation must turn out to be true, i.e.
communicative success must in fact come about in the way expected by S, i.e., as
and only as a consequence of Hs understanding.
The most crucial and contested question to be settled is: What do we have to
know in order to understand Ss communicative act? To answer this question, one
has to notice that the Reflexivity-Condition (RC) above already implies what we
might call the absolute openness of communication. Whatever condition has to be
satisfied in order for an act to be a communicative one (lets say condition ), it follows from RC that this very condition  itself has to be intended to be recognized
by H as being satisfied as well, which is itself a necessary condition for the act to be
a communicative one, etc. Logically, this openness implied by RC will have no
upper end. For such an unrestricted (and in this sense absolute) openness lets
write openness* for short. From RC we get straight to O*:
O*

Communicative intentions must be open* ones.

On the grounds of this condition, Grices first proposal that from Ss point of view
only Hs recognition of Ss primary communicative intention is necessary, can immediately be seen to be inadequate. Grices communications are not open enough.
But Grices Basic Model (BM) can easily be improved:12
BM

Ss doing is an attempt at communication, addressed to H, and of the content that H is to do r, iff (i) S intends, by doing the act, to make H do r; and
(ii) S believes (expects) that she will achieve this aim iff she is letting H
openly* know that S has this intention (i)

Pragmatics in modern philosophy of language

219

Communication is thus regarded as proposed by Grice plus the factor of openness*. This kind of openness*, however, relates only to communicative intentions,
i.e., to those intentions which are necessary conditions for an act to be a communicative act. Sincerity is not a necessary condition for an act to be a communicative
one; communicative openness (being a case of openness*) is therefore absolutely
compatible with the insincerity of the communicative act in question (including
downright lies).
This concept of communicative acts (as defined in BM) is a very broad one.
Though it involves the expectations of understanding and of success, it implies nothing about the reasons these expectations may or will have to be based on in order
to be (rationally to be expected by S to be) satisfied. This is fine as far as communicative acts (and their subjective meanings) in general are concerned. But things
change when we come to the special cases of communication by means of (types
of) actions with some intersubjective communicative meaning.
2.3.2.

Meanings of action types

Meaning and understanding (i.e. knowledge and recognition of meaning) are two
sides of one and the same coin. This holds both for subjective and intersubjective
meanings. (Inter-) subjective meanings of (the type f of) an action are relative to
the (class of) persons who understand the actions (of type f) as having these (inter-)
subjective meanings. In addition, meanings of both kinds are relative to (types of)
situations.
The fact, that some gesture (of type) f, lets say thumbs up, as used in the
population P in situations of type , means p (e.g.: you are fine) may be
equated roughly with the following regular meaning condition of f in P relative to
: Whenever (or when in sufficiently many cases) in a -situation a S (as a member
of P) addresses H (as another member of P), making a gesture (of type) f, it holds
that by doing f, S is trying to communicate to H that p (You are fine). Since
(type-) meanings (as social facts) have to be known to be (intersubjective) meanings, the above regular-meaning-definiens will have to be postulated as being common knowledge in P (regular meaning in a narrow sense of f in P relative to ).13
Regular meanings of f (in P for ) are, in -situations and for all members of P, the
best possible reasons for the respective Ss expectation of understanding.
Conventional meanings result from the likewise rationally founded expectation
of success: H will believe what S is trying to make her believe by this very action f
(namely that p) iff H will understand Ss communicative action f. Normally this
will hold only if (S expects that) H believes that, as far as p is concerned, neither S
will be deceptive to her (H) nor that S himself will be at fault.
And these expectations will be stable (within P and relative to -situations) in
the long run only if they will always (or in most or at least in sufficiently many
cases) turn out to be satisfied. And this again will be true only if both S and H will

220

Nikola Kompa and Georg Meggle

be following a joint strategy, which in turn will be stable only if this strategy is
based on joint interests, as for example, that H is to believe p iff it is in fact the case
that p. Iff, in P and relative to -situations, the communicative expectations of success are based in this way, then the regular meanings of the respective communicative actions will also be conventional ones. This explication is roughly the same
as David Lewiss explication of signalling-conventions (in Lewis 1969).
Now, and this is the decisive step from actions to expressions as their products,
the meanings of (complete or unstructured) signs or expressions (you may think of
the gesture thumbs up again) can be equated with the conventional meanings of
the communicative acts of producing these expressions.
2.3.3.

Linguistic meaning

Unfortunately, when turning to languages and linguistic meanings of structured


signs, Lewis opted out of the whole Intention-Based Semantics-program by putting his proposals on the shoulders of traditional realistic semantics. How this predicament may be avoided in relation to a simple language (of first order predicate
logic) is shown in Kutschera (1975), who provides us with an illustration of the
schema according to which a grammar can be constructed. For a discussion of
such a pragmatic reconstruction see Bennett (1976) and Meggle (2010), the former
trying to couch it in more behavioristic terms, the latter in more formal terms. Following this approach, the meanings of so called realistic semantics (even in its
most developed form of intensional semantics) can be shown to be successfully explicable in terms of Intention-Based Semantics. But, as in many other places, explicability does not imply reduction.
2.4.

Classical speech act theory

Though rightly looked at as being a paradigm case of pragmatics, (classical)


speech act theory was definitely not (intended to be) a special case of PS. This was
clear from the start because John L. Austin (1962), in his basic distinction between
locutionary, illocutionary and perlocutionary (aspects of) acts, was referring to the
use of a sentence in its literal meaning in the sense of Frege. From Austins point
of view, speech act theory should only be interested in the illocutionary functions
of speech acts. This view is shared by his most influential interpreter, John R.
Searle. In this sense, Searles slogan that a theory of language is part of a theory of
action (Searle 1969: 17) can only be misleading.
As to the class of illocutionary acts, Austin did not give us an explicit definition, only an open list of examples. But there are two elements in his characterizations of illocutions which obviously contradict the explication of communicative
acts (in the wide sense of directives and informatives) as sketched in 2.3.1 above.
According to Austin, illocutionary acts are (i) essentially conventional acts; and, as

Pragmatics in modern philosophy of language

221

such, (ii) they have to exclude any perlocutionarity.14 Whereas (i) may be
weakened (cf. Bach and Harnish 1979) by distinguishing between essentially conventional illocutionary acts (including Austins favorites of marrying or baptizing
someone) on the one side, and mere communicative actions on the other side, the
non-perlocutionarity of illocutions remained the central dogma of classical
speech-act-theory (even with regard to communicative acts).
There is some truth in this dogma, but this truth can be easily reduced to the distinction between communication related intentions in general (i.e. perlocutionary
intentions in a broad sense) versus communicative intentions in particular (i.e. perlocutionary intentions in the narrow sense) (cf. 2.3.1 above). With regard to the
broad sense of perlocutionary, the non-perlocutionary dogma is (by definition)
true; but in this sense it is (as implied in 2.3.1) no objection to the presented definition of (the communicative variant of) illocution at all.
2.5.

Speech act based PS

It was William P. Alston (1964) who first tried to build a PS on speech act theory.
His basic idea was that two (unstructured) expressions  and can be said to have
the same meaning iff they have the same illocutionary potential, i.e., could be used
to perform the same illocutionary acts, where the same illocutionary acts was
meant to refer not only to acts of the same illocutionary types (assertions, directives, questions etc.) but also to these acts as having also the same contents (assertions that p, directives to bring about q, questions as to whether it is the case etc.).
In other words: Illocutionary potential in Alstons sense is already a content-determined illocutionary act potential (CDIAP). And though his basic idea was presented by him only in the form of an identity criterion for the meanings of  and , it
was clear that Alston wanted to identify and thus also define the meanings of expressions by means of their CDIAPs. That was the essential difference: Whereas in
Austins theory meanings of expressions are presupposed as being given, Alstons
idea was to define these expression meanings themselves by means of the meanings of illocutionary actions. If this is the line along which meaning should be
analyzed, then the concept of an illocutionary act is the most fundamental concept
in semantics and, hence in the philosophy of language (Alston 1964: 37 [emphasis
mine]).
If we substitute a communicative act for an illocutionary act, Alstons
speech act approach would seem to coincide with the Intention-Bases Semantics as
sketched in 2.3 above. But notice that even Alston in 1964 subscribed to the two
dogmas of classical speech act theory. However, nearly 40 years later, he has extended his basic idea of meanings as CDIAPs into a full blown PS of its own (Alston 2000). Beside Alston, the most detailed and best developed rule-oriented PS is
Eike von Savignys theory (1983). Whereas the intention-based PS as sketched
above starts with the communicative meaning of concrete utterances (action-

222

Nikola Kompa and Georg Meggle

tokens, i.e. field (4) of the schema in 2.1 above), Savignys rule-based PS starts
with the conventional meaning of utterance types (action-types, i.e.field (3) in the
same schema). According to this approach, differences in meaning are correlated
with differences in the conventional commitments of the persons involved in the
respective utterance-situation. And these commitments in turn are defined by
means of rules as explicated by Hart (1961) with reference to regularities, sanctions and acceptance of these sanctions. Finally, sentence meanings are to be introduced as theoretical constructs in order to systematically explain the salient structures to be found in these regularities.
To systematically compare the IB- versus these and other Rule-Based Approaches to PS with regards to their various pros and cons is one of the biggest desiderata in the whole field of modern philosophy of language.
2.6.

Variety of language games

As far as Wittgensteins meaning = use equation is concerned (see 2.2 above),


the usual distinction between the rule-reading versus the intention-reading of it
should not be the last word. At least not in the sense of an either-or distinction because there is some chance that both readings need not be really exclusive. First, as
noticed in 2.6, the two readings may refer to different explication-levels, the intention-reading to the token-one, the rule-reading to the type-one. Next, as noticed in
2.4, only a special class of perlocutionary intentions is relevant for the definition of
communicative/illocutionary acts: communicative intentions (see 2.3.1). As these
intentions are necessary conditions for some action to be a communicative action
(a directive, for example), they may be looked at as being part of the internal relation, which, according to Wittgenstein ( 21), is constitutive of the kind of
meaning his various language games are meant to be exemplifications of. According to this reconstruction of Wittgensteins meaning as use theory,15 the relevant intention-reading is part of the rule-reading itself.
But there are many other language games (many other illocutions) whose internal relations do not include any reference to perlocutionary intentions. One
might therefore say that meanings (illocutions) do not per se entail perlocutions.
But from this it does not follow that perlocutions (as part of meanings) are per se
excluded. There is more than just one kind of meaning.
2.7.

Evaluations

This difference is of the utmost importance for any evaluation of the various PSapproaches.
What kinds of illocution are the most important ones? The IB-semanticists (as
sketched in 2.3 above) take it that interpersonal communication is at the very basis.
But they do not deny, of course, that languages may have many other functions as

Pragmatics in modern philosophy of language

223

well. Some neo-Griceans, interpreting Grices original starting-point of speakers


meaning as covering much more than communicative intentions, do accordingly
extend this basis and include, e.g., the intention to express ones thoughts or feelings (Davis 2003). Others, like Austin, think it best to start with fully institutionalized or even ritualized kinds of performances. Finally, there are scholars like Brandom (1994), who take the whole PS-project (of starting with primitive acts or
intentions and then constructing something like linguistic meanings out of them) to
be on the wrong track because they claim that language has to be presupposed before attributing even these most primitive elements.
Before we postulate the one PS which embraces all sorts of things language
may be used for, we should keep in mind that we just cannot come up with the complete theory of language and still hope for its consistency.

Notes
1. For an overview of the development of semiotics, see e.g., Scholz 2003, for semiotics
and pragmatics Nth, this volume, and for a general historiographic overview of pragmatics Koyama, this volume.
2. For an account of the relationship between semantics and pragmatics cf. Saeed, this volume.
3. Kaplan also speaks of the character of indexical expressions as of a rule that fully determine[s] the referent for each context. (1989: 491)
4. For this specific conception of context cf. Lewis A context is a location time, place,
and possible world where a sentence is said. (1980: 21), and Kaplan, who adds the
speaker or acting participant as a further constituent, argues that an index representing a
context is a 4-tuple <w, x, p, t>, consisting of a world (w), an actor (x), a place (p) and a
time (t) such that x is in w at t in p (Kaplan 1989: 509).
5. Stalnakers notion of context might be promising here: So I propose to identify a context (at a particular point in a discourse) with the body of information that is presumed,
at that point, to be common to the participants in the discourse. (1998: 98)
6. For more on inferences to the best explanation in general, see, e.g., Lipton (1991).
7. Some context-sensitive sentences may have preferred or default completions, though. In
the absence of any special stage setting, we will interpret the sentence: Sams daughter
got married and became pregnant, for example, as meaning that Sams daughter first
married and then became pregnant. For more on the notion of a preferred or default interpretation see Levinson (2000).
8. See also Searle (1978: 220221) and (1980: 221).
9. Which may be treated as something like the cornerstone of the whole of PS.
10. For formal explications of step (i) see Meggle (1997) and for step (ii) Meggle (2010).
11. Following Max Weber one could also say that Action Understanding in this sense is to
know the actions subjective meaning / subjektiver Sinn.
12. With (BM) being already a reconstructed version of Grices own version (1957): S
meant something by x is (roughly) equivalent to S intended the utterance of x to produce some effect in an audience by means of the recognition of this intention.

224

Nikola Kompa and Georg Meggle

13. Where Common Knowledge in P that p may be easily defined this way: (i) All members
of P know that p; (ii) All members of P know that (i) etc., with this etc to be defined
in formal terms as usual by recursion.
14. See Austin (1962: ch. 9), where perlocutionary effects are by definition restricted to effects going beyond (conventional) illocutionary effects.
15. Very forcefully proposed for the first time by Feyerabend (1955) and heavily relying on
Wittgensteins famous language game, cf. 2 in his Philosophical Investigations.

References
Alston, William P.
1964
Philosophy of Language. Englewood Cliffs, NJ: Prentice-Hall.
Alston, William P.
2000
Illocutionary Acts and Sentence Meaning. Ithaca and London: Cornell University Press.
Austin, John L.
1962
How to do things with Words. Oxford: Oxford University Press.
Austin, John L.
31979a
The Meaning of a Word. In: John Austin, Philosophical Papers, 5575. Oxford: Oxford University Press.
Bach, Kent and Robert M. Harnish
1979
Linguistic Communication and Speech Acts. Cambridge, MA: Harvard University Press.
Bach, Kent
1994
Semantic slack. In: Savas L. Tsohatzidis (ed.), Foundations of speech act
theory: Philosophical and linguistic perspectives, 267291. London/New
York: Routledge.
Bach, Kent and Anne Bezuidenhout
2002
Distinguishing Semantics and Pragmatics. In: Joseph Keim Campbell, Michael
ORourke and David Shier (eds.), Meaning and Truth: Investigations in Philosophical Semantics, 284309. New York/London: Seven Bridges Press.
Barwise, Jon and John Perry
1983
Situations and Attitudes. Cambridge, MA: MIT Press.
Bennett, Jonathan
1976
Linguistic Behaviour. Cambridge: Cambridge University Press.
Bezuidenhout, Anne
2002
Truth-Conditional Pragmatics. Philosophical Perspectives 16: 105134.
Bloom, Paul
2000
How Children Learn the Meanings of Words. Cambridge, MA: MIT Press.
Borg, Emma
2004a
Formal Semantics and Intentional States. Analysis 64: 215223.
Borg, Emma
2004b
Minimal Semantics. Oxford: Oxford University Press.
Borg, Emma
2007
Minimalism versus Contextualism in Semantics. In: Gerhard Preyer and Georg
Peter (eds.), Context-Sensitivity and Semantic Minimalism: New Essays on
Semantics and Pragmatics, 339359. Oxford: Oxford University Press.

Pragmatics in modern philosophy of language

225

Brandom, Robert B.
1994
Making it explicit. Reasoning, Representing and Discursive Commitment.
Cambridge, MA: Harvard University Press.
Brandom, Robert, B.
2002
Pragmatics and Pragmatisms. In: James Conant and Urszula M. Zeglen (eds.),
Hilary Putnam: Pragmatism and Realism, 4059. London: Routledge.
Clark, Eve V.
1993
The lexicon in acquisition. Cambridge Textbooks in Linguistics, Cambridge:
Cambridge University Press.
Carnap, Rudolf
1956
Meaning and Necessity: A Study in Semantics and Modal Logic. 2nd ed. Chicago: The University of Chicago Press.
Carston, Robyn
2002
Thoughts and Utterances: The Pragmatics of Explicit Communication. Oxford: Blackwell.
Cohen, Jonathan L.
1985
A Problem about Ambiguity in Truth-Theoretical Semantics, Analysis,
Vol. 45/3: 129134.
Croft, William and Cruse, Alan D.
2004
Cognitive Linguistics. Cambridge Textbooks in Linguistics, Cambridge: Cambridge University Press.
Davis, Wayne A.
2003
Meaning, Expression, and Thought. Cambridge: Cambridge University Press.
Feyerabend, Paul
1955
Wittgensteins Philosophical Investigations. Philosophical Review: 449483.
Frege, Gottlob
1891
Funktion und Begriff. Lecture held 1891 at a meeting of the Jenaischen Gesellschaft fr Medizin und Naturwissenschaft. Reprinted in Gottlob Frege
1986, 6th ed., 1839.
Frege, Gottlob
1892a
ber Sinn und Bedeutung. Zeitschrift fr Philosophie und philosophische Kritik 100, 2550. Reprinted in Gottlob Frege 1986, 6th ed., 4065.
Frege, Gottlob
1892b
ber Begriff und Gegenstand. Vierteljahrsschrift fr wissenschaftliche Philosophie XVI, 192205. Reprinted in Gottlob Frege 1986,6th ed., 6680.
Frege Gottlob
1986
Funktion, Begriff, Bedeutung. Edited by Gnther Patzig, 6th ed. Vandenhoeck
und Ruprecht, Gttingen.
Goodman, Nelson
1976
Languages of Art. An Approach to a Theory of Symbols. 2nd ed. Indianapolis:
Hackett.
Grice, H. Paul
1957
Meaning. The Philosophical Review 66: 377388. Reprinted in Grice 1989.
Grice, H. Paul
1989
Studies in the Way of Words. Cambridge, MA: Harvard University Press.
Hart, Herbert L.A.
1961
The Concept of Law. Oxford: Clarendon Press.

226

Nikola Kompa and Georg Meggle

Jeffrey, Richard C.
1983
The Logic of Decision. 2nd ed. Chicago: University of Chicago Press.
Kaplan, David 1989
Demonstratives. An Essay on the Semantics, Logic, Metaphysics, and Epistemology of
Demonstratives and Other Indexicals. In: Joseph Almog, John Perry, and Howard Wettstein (eds.), Themes from Kaplan, 481563. Oxford: Oxford University Press.
Keller, Rudi and Ilja Kirschbaum
2003
Bedeutungswandel: Eine Einfhrung. Berlin: de Gruyter.
Koons, Robert
2009
Defeasible Reasoning. In: Edward N. Zalta (ed.), The Stanford Encyclopedia
of Philosophy (Winter 2009 Edition). <http://plato.stanford.edu/archives/
win2009/entries/reasoning-defeasible/>.
Kutschera, Franz von
1975
Conventions of Language and Intensional Semantics. Theoretical Linguistics
Vol 2/3: 255283.
Kutschera, Franz von
1976
Einfhrung in die intensionale Semantik. Berlin: de Gruyter.
Lakoff, George and Mark Johnson
1980
Metaphors We Live By. Chicago: University of Chicago Press.
Levinson, Stephen C.
2000
Presumptive Meanings: The Theory of Generalized Conversational Implicature. Cambridge, MA: MIT Press.
Lewis, David
1969
Convention. Cambridge, MA: Harvard University Press.
Lewis, David
1975
Languages and Language. In: Keith Gunderson (ed.), Minnesota Studies in the
Philosophy of Science, Vol. VII, 335. University of Minnesota Press. Reprinted in David Lewis 1983: Philosophical Papers. Volume 1, 163188. Oxford: Oxford University Press.
Lewis, David
1980
Index, Context, and Content, in Stig Kangar and Sven hman (eds.), Philosophy and Grammar, 79100. Dordrecht: Reidel.
Lipton, Peter
1991
Inference to the Best Explanation. London: Routledge.
Meggle, Georg
1997
Grundbegriffe der Kommunikation. Berlin: de Gruyter.
Meggle, Georg
2010
Handlungstheoretische Semantik. (Habilitation thesis 1984). Berlin: de
Gruyter.
Montague, Richard
1970
English as a Formal Language. In: Bruno Vicentini et al. (eds.), Linguaggi
nella societ e nella tecnica, 189224. Mailand: Edizioni di Comunit. Reprinted in Montague 1974, 188221.
Montague, Richard
1970b
Universal grammar. Theoria 36: 373398. Reprinted in Montague 1974,
222246.

Pragmatics in modern philosophy of language

227

Montague, Richard
1973
The proper treatment of quantification in ordinary English. In: Jaakko J. Hintikka, Julius M. Moravcsik and Patrick Suppes (eds.), Approaches to natural Language, 221242. Dordrecht: Reidel. Reprinted in Montague 1974,
1734.
Montague, Richard
1974
Formal Philosophy: Selected Papers of Richard Montague. Edited by Richmond H. Thomason. New Haven/London: Yale University Press.
Moravcsik, Julius M.
1998
Meaning, Creativity, and the Partial Inscrutabilty of the Human Mind. CSLI
Publications, Stanford.
Manor, Ruth
2001
On the Overlap of Pragmatics and Semantics. Synthese 128: 6373.
Morris, Charles
1938
Foundations of the Theory of Signs. In: Otto Neurath, Rudolf Carnap and
Charles Morris (eds.), International Encyclopedia of Unified Science. Vol. 1,
Nr.2, 77138. Chicago: University of Chicago Press. Reprinted in Charles
Morris, 1971, Writings on the General Theory of Signs, 1771. Mouton: The
Hague/Paris, quotations are to Morris 1971.
Partee, Barbara
1973
Some transformational extensions of Montague grammar. Journal of Philosophical Logic 2: 509534.
Partee, Barbara
1975
Montague grammar and transformational grammar. Linguistic Inquiry 6/2:
203300.
Portner, Paul and Barbara H. Partee
2002
Introduction. In: Paul Portner and Barbara H. Partee (eds.), Formal Semantics:
The Essential Reading, 116. Oxford: Blackwell.
Recanati, Franois
2004
Literal Meaning. Cambridge; Cambridge University Press.
Savigny, Eike von
1983
Zum Begriff der Sprache. Konvention, Bedeutung, Zeichen. Stuttgart: Reclam.
Schiffer, Stephen
1972
Meaning. Oxford: Oxford University Press.
Scholz, Oliver R.
2003
Semiotik und Hermeneutik. In: Roland Posner, Klaus Robering and Thomas A.
Sebeok (eds.), Semiotik/Semiotics: Ein Handbuch zu den zeichentheoretischen
Grundlagen von Natur und Kultur / A Handbook on the Sign-Theoretic Foundations of Nature and Culture, Vol. 3, 25112561. Berlin: de Gruyter.
Searle, John R.
1969
Speech Acts. Cambridge: Cambridge University Press.
Searle, John R.
1978
Literal meaning. Erkenntnis 13: 20724.
Searle, John R.
1980
Background of Meaning. In: John Searle, Ferenc Kiefer, Manfred Bierwisch
(eds.), Speech Act Theory and Pragmatics, 221232. Dordrecht: Reidel.

228

Nikola Kompa and Georg Meggle

Stalnaker, Robert C.
1998
On the Representation of Context. Journal of Logic, Language, and Information 7. Reprinted in Stalnaker 1999 Context and Content. 96113. Oxford: Oxford University Press.
Stanley, Jason
2002
Making it Articulated. Mind and Language 17: 149168.
Stanley, Jason
2005
Semantics in Context. In: Gerhard Preyer and Georg Peter (eds.), Contextualism in Philosophy: Knowledge, Meaning, and Truth, 221253. Oxford: Oxford
University Press,
Tomasello, Michael
2003
Constructing a Language: A Usage-Based Theory of Language Acquisition.
Cambridge, MA: Harvard University Press.
Travis, Charles
1985
On what is strictly speaking true. Canadian Journal of Philosophy 15:
187229.
Travis, Charles
1996
Meanings Role in Truth. Mind 105: 451466.
Travis, Charles
1997
Pragmatics. In: Bob Hale and Crispin Wright (eds.), A Companion to the Philosophy of Language, 87107. Oxford: Blackwell.
Wittgenstein, Ludwig
1953
Philosophische Untersuchungen. Ed. by G.E.M. Anscombe and Rush Rhees.
Oxford: Basil Blackwell.

8.

Foundations of pragmatics in functional


linguistics
Saskia Daalder and Andreas Musolff

1.

Introduction

The relatively young discipline of linguistic pragmatics has had forbears in several
older branches of linguistics. This chapter focuses on important prefigurations of
parts of linguistic pragmatics from the late nineteenth century to around 1970
which were informed by a functional view of language. We view functional linguistics as the study of language that starts its research with a focus on the functions of language in social life: the effects of language use, differentiated as to
types of communicators, types of contexts and types of language uses. This includes the analysis of the primarily intended goals of the speech participants in
their speaking and reacting as well as the study of the long-time effects of language
use in the life of the individual and in the history of the language.
One could name quite a few functionally inclined linguists of that period whose
theories are to some extent congenial to modern pragmatics, starting with the
theorists Dwight Whitney (18271894), Michel Bral (18321915), Georg von
der Gabelentz (18401893) and Henry Sweet (18451912) and ending with, e.g.,
Andr Martinet (19081999) and Michael Halliday (*1925). However, in this article we will confine ourselves to results of functional linguistics that have actually
been transmitted to linguistic pragmatics and may be counted among its foundations. Such results are found foremost in the works of four theorists: the German
psychologist Karl Bhler (18791963), the Czech scholar of English Vilm Mathesius (18821945), the Russian general linguist and scholar of Slavonic languages
Roman Jakobson (18961982) and the British general linguist Michael Halliday.
Of these, we will discuss the relevant theories of the first three. Bhler proposed an
influential model of three fundamental language functions, which Jakobson
came to extend to the model with six such functions that has had a strong impact
on pragmatics, the ethnography of communication and social semiotics.1 Furthermore, both Bhler and Jakobson analysed the functioning of the deictic elements of language, emphasising their role in the situated nature of language
use. Mathesius, who was more specifically concerned with the comparison of living languages as to their possibilities of expression, provided the foundations of
the functional analysis of utterances in terms of the notions theme and rheme
(topiccomment), which has become one of the stock items of pragmatic
analysis. For their ideas in this domain, both Bhler and Mathesius were indebted
to the work of the older and less well-known psychologically oriented German lin-

230

Saskia Daalder and Andreas Musolff

guist Philipp Wegener (18481916), who held views that also show a strong affinity to those of modern linguistic pragmatics. In our sections 25, we will in turn
discuss the contributions by Wegener, Bhler, Mathesius and Jakobson. In the concluding section 6, we consider the questions of whether and how their insights into
the functionality of language can still be of value for linguistic pragmatics. In our
view, these insights still provide valid reasons to integrate the description of elements of the situational context into linguistic analysis itself and thus to question
the wisdom of completely separating research into the system of language from
pragmatics, the study of its use.

2.

Philipp Wegener

Philipp Wegener (18481916), a classics teacher, grammar school headmaster and


psychological linguist from Northern Germany, developed in a relatively small
body of work a remarkably modern, communication-oriented analysis of language.2 It resulted from his opposition against a particular claim of contemporary
historical linguistics. The Neogrammarian theorists had acknowledged the relevance of psychological factors for processes such as the borrowing of linguistic
elements by one language or dialect from another, but not for so-called internal,
primarily phonetic language change, which was thought to be caused purely by
physiological mechanisms. Motivated by his study of developments in the local
dialects, Wegener wished to demonstrate the concrete psychological and communicative background also of internal changes, and this forms the first topic in
his 1885 publication Investigations into the Fundamental Questions on the Use of
Language (Untersuchungen ber die Grundfragen des Sprachlebens).3 Wegener
uses the key concept of automatisation: changed phonetic realisations may appear in hindsight not to fulfil any special goals, but they are in fact the results of
conscious choices by speakers, made at one point in time, of alternative possibilities with special communicative effects, which subsequently have become automatised through habit. A communicative basis can also be seen in the ontogeny of
language. In time, the child learns to use sound patterns with meanings of a general
nature, but this develops only after a stage in which a particular vocal expression is
a purposive action, essentially connected to the childs momentary needs and
meant to appeal to the carers for help (Wegener [1885] 1991: 419).
Wegener goes on to claim that the general purpose of speaking is indeed to influence the actions, volition or thoughts of another person in a way the speaker
thinks worthwhile (Wegener 1991: 6467, 180). He has no use for the well-known
subjectivist picture of the individual speaker who, in speaking, merely mirrors
his own thoughts; such a view would make the idea of communication as accommodation between different minds all but void (Wegener 1902: 402). Focusing
rather on the role of the hearer, Wegener emphasises that the hearer is the true locus

Foundations of pragmatics in functional linguistics

231

of meaning, because he constructs meaning first and foremost not from linguistic
cues but according to his knowledge of the world and of the purposes of social action, helped further by his inferential abilities. Within this dialogical theory of
meaning, the function of the words and grammatical elements the speaker presents
is to help organise the hearers expectations (Wegener 1991: 157158, 180182).
The importance of the asymmetrical interaction between speaker and hearer is
highlighted in the detailed analysis of many grammatical constructions that Wegener provides in his book. As the general functional structure of utterances, Wegener proposes a bipartition consisting of exposition (Exposition) and predicate (Prdicat).4 The predicate carries the core function of the utterance, it
indicates the new and important material for the situation of utterance, and whatever effect is achieved in the hearer is connected primarily to this part. The exposition is everything in the utterance that the speaker has deemed necessary for
the hearer to situate the information contained in the predicate in the way it is
meant to be situated (Wegener 1911; 1991: 1922).
Elements of the situation itself may provide enough information to let the
predicate do its work, as Wegener explains. Any aspects of the hearers momentary
attention, such as a context of actions that have just been carried out (or are in the
process of being carried out), a topic from the social or institutional situation at
hand or an element from the general cultural situation, can be sufficient for him to
understand an unaccompanied predicate Well done as uttered by the speaker.
Language thus need not always be congruous to what the mind understands (Wegener 1991: 2127).5 However, an exposition of linguistic material is often
added to help achieve clarity as to the situated fact in relation to which the predicate has to be understood. Note that according to Wegener, the exposition need not
always precede the predicate; there may be many reasons for a speaker to choose
the reverse ordering. Intonation, but also grammar, can fulfil the task of indicating
to the hearer just what part of the utterance has which role.
Wegener (1991: 4760) also relates important phenomena of word semantics to
his communicative bipartition of utterances.6 It is not the same for a word to appear
in the exposition of an utterance or to appear in the predicate. The words of the exposition are used by the hearer to determine the real-world referent, and their
meanings should therefore be relatively congruous to the relevant thought. By
contrast, words chosen for the predicate may well have a variable, only loosely-fitting relation to their real-world referent and often have a metaphorical character.
For his precise understanding, the hearer can rely on inferences that are drawn
from elements of the situation and from expositional parts of the utterance. When
such inferences become automatised through repeated use, the word meaning loses
its flexibility, increases in its congruousness to thought and thereby becomes appropriate to fulfil a role also in the expositional parts of utterances. To give a modern example: it is often said, figuratively, that the family is the cornerstone of society. The phrase has become so well-known that the original predicate part (the

232

Saskia Daalder and Andreas Musolff

cornerstone of society) can now also be used in a role of exposition and still be
understood to have the family as referent: the cornerstone of society is in trouble.
But this does not yet hold unrestrictedly for cornerstone on its own; of course,
this could well be a further development. The process as sketched amounts to semantic change and, more specifically, the well-attested, continuous fading (abblassen) of metaphors (Wegener 1991: 5157).
It is an interesting fact that in this picture, pragmatic analysis can explain linguistic phenomena that are usually considered to be semantic or grammatical in nature. Pragmatics could thus be conceived of as an overarching discipline encompassing linguistics the full potential of such a conception does not seem to be
exhausted by modern pragmaticists.7
Although Wegeners approach was appreciated and taken into account by such
functionalist theorists as the important Polish anthropologist Bronisaw Malinowski (18841942) and the British linguists Alan Gardiner (18791963) and Michael Halliday,8 it never gained real notoriety in general linguistics and was not
used by the founders of modern linguistic pragmatics. Nor did the revival of Wegeners pioneering ideas in works by the German communication theorists Gerold
Ungeheuer (19301982) and Johann Juchem (19392003)9 succeed in making
them known beyond a rather small circle of pragmaticists. However, some of Wegeners proposals did reach modern pragmatics through an intermediary. Karl
Bhler helped spreading Wegeners analysis of the interweaving, in each utterance, of the speakers expression of thoughts, the intended effect on the hearer and
the representation of a part of reality. This tripartite analysis had a major influence
on the development of Bhlers theory of the functioning of language, as can
be seen already in one of his earliest linguistic publications (Bhler 1909b:
119123).10 It eventually led to Bhlers model of three language functions, the
organon model, to be discussed in section 3.1, which in turn was further differentiated by Roman Jakobson and became part of linguistic pragmatics in that extended version (see section 5.1).

3.

Karl Bhler

3.1.

Bhlers axiomatic view of language functions

In 1934, Karl Bhler (18791963), Ordinarius professor of psychology at the


University of Vienna, published a monograph entitled Theory of Language. The
Representational Function of Language (Sprachtheorie. Die Darstellungsfunktion
der Sprache). Its introductory First Part deals with The Principles of Language
Studies, which include four basic axioms. The first of these axioms, the organon-model, is introduced as the maxim of the three semantic functions (Leitsatz
von den drei Sinnfunktionen) of all linguistic structures (Bhler 1990: 28; 1934:

Foundations of pragmatics in functional linguistics

233

22).11 Bhler cites Platos definition in Kratylos of language as an organum


(means/tool) for a speaker to inform the other of something about the things
(Bhler 1990: 30; 1934: 24) to motivate the term organon-model. It was this
1934 version of the model of language functions that inspired a number of later
scholars to develop the notion of Functional Linguistics with a strong emphasis
on pragmatic as well as sociolinguistic aspects, and it is quoted as such in textbooks to this day.12 Before we look at its systematic theoretical status and further
adaptation, however, we need to analyse its development in Bhlers own work.
1934 did not mark the starting point of Bhlers work on meaning functions
by any means; rather, it was its climax and almost also its ending due to the interruption of Bhlers work in 1938 when Nazi Germany annexed Austria.13 Building on Edmund Husserls (18591938), Anton Martys (18471914) and Wilhelm Wundts (18321920) theories of meaning,14 as well as on his own work on
thought psychology in the context of the Wrzburg school around Oswald Klpe
(18621915),15 Bhler had published an article on the Critical assessment of new
sentence theories (Kritische Musterung der neueren Theorien des Satzes) as early
as 1918, in which he proposed a triadic model of language functions (Leistungen). It is based on the three main foundations of any meaning-constitutive
interaction, i.e.,
1) the speaker whose feelings and attitudes are given expression (Kundgabe),
2) the hearer for whom it provides a stimulus that elicits reactions (Auslsung),
3) objects or states of affairs that form the referents for representation (Darstellung) (Bhler 1918: 16).
After moving to the Chair of Psychology at Vienna University in 1922, Bhler
went on to develop this triadic model further in two major psychological studies. In
The Crisis of Psychology (Die Krise der Psychologie) of 1927, the model serves as
the basis for a critical assessment of contemporary psychology. The two basic semantic functions of expression (Kundgabe, which, as Bhler emphasises, also includes its receptive counterpart, Kundnahme) and guidance (which also involves
mutuality) are present even in animal communication; the ability to produce symbolic representations (of objects/states of affairs), on the other hand, is characteristic of human communication and helps to adapt the first two functions to a higher
level, so that all three of them are only fully activated in human communication
(Bhler 1927: 3233, 37, 5051). The three functions model thus forms a framework for conceiving of psychology as the integrated study of subjective experience, social behaviour and structures of objective sense (Bhler 1927: 29,
5762). The enormous conceptual range of this function-concept is evident also
in the further monograph on Theory of expression (Ausdruckstheorie) of 1933,
where Bhler refers to it as the basis to analyse the relationship between miming,
gesture and language (Bhler 1933a: iiiiv).

234

Saskia Daalder and Andreas Musolff

With specific respect to linguistics, Bhler explicitly claimed axiomatic


status for the functions model in his two articles The whole of language theory, its
system and its parts (Das Ganze der Sprachtheorie, ihr Aufbau und ihre Teile) and
The axiomatics of the language sciences (Die Axiomatik der Sprachwissenschaften), published in 1932 and 1933, respectively. He now called the functions
expression (Ausdruck), appeal (Appell), and representation (Darstellung)
(Bhler 1932: 106; 1933a: 7490) and incorporated them into a set of four principles, which prefigure the axiomatic introductory part of Sprachtheorie (although they appear in different order).16 Compared with the earlier versions of the
three functions-model, both articles share with the Sprachtheorie the detailed discussion and acknowledgement of Ferdinand de Saussures (18571913) Cours de
linguistique gnrale (published posthumously in 1916; German translation published in 1931).
Saussures famous lectures had motivated Bhler to reconceptualise language
theory from a semiological (Saussure), or as Bhler calls it, sematological
viewpoint (1990: 33, 42; 1934: 27, 3435). However, Bhlers interpretation of the
Cours differed in significant ways from contemporary structuralist readings.17
Whilst acknowledging the epoch-making impact of the Cours (Bhler 1990: 78;
1934: 7), he criticises its presentation of the speech circuit as a relapse into 19th
century psycho-physics and he insists that linguistic communication comprises
different types of signs (1990: 3134, 4243; 1934: 2528, 3437). Most important
in our context is his discussion of Saussures distinction of a linguistique de la
langue from a linguistique de la parole. Bhler puts the emphasis not on the langueaspect, but instead praises Saussure for having shown what would have to be discovered in order to be really able to initiate a linguistique de la parole (1990: 8;
1934: 7). This foregrounding of the speech/action aspect of language was at odds
with the predominant reading of the Cours as an endorsement of a system-oriented
core linguistics. It thus provides a first indication of Bhlers interest in the pragmatic orientation as an essential methodological perspective for linguistic research.
Besides the pragmatically interpreted Cours, Bhler cites three further works
as coming closest to providing a functionalist grounding of linguistics as an urgently needed complement to the old grammar (1990: 27; 1934: 23):
a) Philipp Wegeners Investigations into the Fundamental Questions of the
Use of Language ([1885] 1991) (discussed in our section 2) had taught Bhler
about the listeners role in the social action dimension of speech;18
b) Karl Brugmanns (18491919) essay on Indo-European pronoun systems
(cf. 1904) had provided him with a theory of demonstrative deixis as a framework
to analyse the speech situation;
c) Alan Gardiners Theory of Speech and Language (1932), which owed its
publication partly to Bhlers encouragement.19 Bhler praised it as the most interesting attempt to carry out a project similar to his own but adds the reservation
that Gardiner strives exclusively for a situational theory of language, whereas

Foundations of pragmatics in functional linguistics

235

he, Bhler, wants to account also for types of language use that are removed from
the situation (Bhler 1990: 28; 1934: 3435).20
Supported by these three authorities (as well as by the parole-oriented reading
of Saussures Cours), the organon-model appears in the Sprachtheorie as the
first of four axioms; it postulates three semantic functions (semantische Funktionen) for every instance of a linguistic sign (Sprachzeichen):
Es ist Symbol kraft seiner Zuordnung zu Gegenstnden und Sachverhalten, Symptom
(Anzeichen, Indicium) kraft seiner Abhngigkeit vom Sender, dessen Innerlichkeit es
ausdrckt, und Signal kraft seines Appells an den Hrer, dessen ueres und inneres
Verhalten es steuert wie andere Verkehrszeichen. (Bhler 1934: 28).
It is a symbol by virtue of its co-ordination to objects and states of affairs, a symptom
(Anzeichen, indicium: index) by virtue of its dependence on the sender, whose inner
state it expresses, and a signal by virtue of its appeal to the hearer, whose inner and outer
behaviour it directs as do other communicative signs. (Bhler 1990: 35).

It is in this version that the organon-model has entered pragmatic-functionalist


textbooks as a well-known historical reference (see above). However, its presentation in the Sprachtheorie as the first axiom (possibly meant to underline its preeminence) is less convincing than its introduction in the Axiomatics essay from a
year earlier. In that essay it appears as the last axiom and forms the conclusion of
the previous argumentation.21 In the Sprachtheorie, the model is followed first by
illustrative comments on various styles and registers in which either of the three
functional dimensions appears dominant (e.g. poetry, rhetoric, scientific discourse)
and a hint that its main thesis will be verified when all three books [on language]
that the organon model requires have been written (Bhler 1990: 39; 1934: 33).
This hint underlines the point of the subtitle of Sprachtheorie, The Representational Function of Language (Die Darstellungsfunktion der Sprache): this work
does not contain the complete Theory of Language whose necessity is implied
by the organon model; it still focuses mainly on the one function that had been the
object of traditional language philosophy and semantics.
This discrepancy, although openly announced, is regrettable from a pragmatic
point of view, for historical reasons (as the three promised books were never completed: The Crisis of Psychology and Expression Theory contain valuable linguistic
insights but have a much wider, psychologically defined remit) as well as in theoretical respects. As a largely programmatic formulation, the organon model lacks an
explication by Bhler that would have clarified its methodological implications. In
the Sprachtheorie itself, the model is only intermittently and cursorily referred to
and does not provide a systematic structuring principle of the book. The following
remarks therefore highlight aspects of the Sprachtheorie that can be most closely
linked to the three functions model, without claiming to give a systematic overview of Bhlers opus magnum.
Within the Axiomatics-part of the Sprachtheorie, the second axiom is the signcharacter of language, which Bhler takes over not just from Saussure but from

236

Saskia Daalder and Andreas Musolff

the sign-theoretical tradition going back to the medieval scholastic debates on representation and which he relates in a bold move to the contemporary phonemephone distinction. Bhlers formula for the principle of the semantic character of
signs is abstractive relevance (Bhler 1990: 4056; 1934: 3348). Building on
the organon- and sign-axioms, he attempts in axiom III to achieve an integration of
functional and systemic language aspects by way of a cross-classification based on
the parole-langue and ergon-energeia distinctions by Saussure and Humboldt.
This yields four perspectives on language as an object of scientific investigation,
i.e. language as (1) speech action (Sprechhandlung), (2) language work
(Sprachwerk), (3) speech act (Sprechakt) and language structure (Sprachgebilde) (Bhler 1990: 5780; 1934: 4869).
This schema has been criticised for being terminologically and methodologically inconsistent.22 Its English translation in the 1990 edition adds further problems as it suggests false friends such as Sprechakt and speech act. Bhlers
understanding of this aspect is far removed from Austinian, Searlean or Gricean
concepts; rather, it rests on the philosophical notion of sense-conferring acts
(sinnverleihende Akte), which Bhler had adopted from Husserl. Bhler praises
Husserls Logische Untersuchungen for highlighting sense-conferring acts as a
necessary object of language theory but he criticises the implication of an exclusively subject-centred perspective on meaning constitution, against which he insists
on the inter-subjective side of all use of signs (1990: 7374; 1934: 6365; see also
Nerlich and Clarke 1996: 194198, 211215).
The most interesting aspect of Bhlers schema from a pragmatic perspective is
the category of the speech action, which encapsulates the insight that all concrete speech is in vital union with the rest of a persons meaningful behaviour; it is
among actions and is itself an action (Bhler 1990: 61; 1934: 52). In the discussion in the Sprachtheorie Bhler gives only a relatively brief illustration by way
of famous quotations that embody a whole situation (e.g. Caesars alea jacta est)
and refers to developmental psychology; he also highlights the field-character of
every speech action and the importance of the act(ion) history, which includes the
language acquisition process (1990: 6566; 1934: 56). In the Axiomatics essay,
however, Bhler gives a more detailed explanation, drawing strongly on the action
theory developed by the fellow Wrzburg School psychologist Abraham Anton
Grnbaum (18851932), who had given a paper on Language as action (Sprache
als Handlung) at a workshop organised by Bhler at a psychology congress in
Hamburg in 1931 (Bhler 1933a: 5051). For Grnbaum, every action is a dynamic system of organism and environment (System: Organismus Umwelt); its
constitutive aspects are the agentive organisms need (Bedrfnis) and intention (Intention), the active completion (Erledigung) of the action and the agents reaction
to the completion (Erledigungsreaktion) (Grnbaum 1932: 165173). For the application to the speech action, Bhler re-labels the reaction to the completion as
fulfilment (Erfllung) and describes it further as consisting of the speakers con-

Foundations of pragmatics in functional linguistics

237

scious realisation that the process of meaning constitution has been concluded
(Bhler 1933a: 5051).23
In the Sprachtheorie, Grnbaums theory is not explicitly mentioned but
Bhler uses Grnbaums terminology in some parts and refers for further reading
to his own Axiomatics essay as well as to disputes among the schools of Gestalt
theory, where he takes a polemical stance against the supposed monism of the
Berlin School of Gestalt psychology around Max Wertheimer, Wolfgang Khler
and Kurt Koffka (Bhler 1990: 66; 1934: 56). In this allusion-laden discussion the
contours of Bhlers concept of speech as action are sometimes barely discernible.
There is just one hint that the topic is taken up again in chapter 10, which deals
with the integration of speaking into other meaningful behaviour, and more specifically into different types of surrounding fields (Umfelder) (Bhler 1990: 61;
1934: 52). This distance or detour of six chapters is motivated by Bhlers more
immediate focus on the concept of language fields, which overlaps with the notion of surrounding fields, thus leading to a potentially confusing double-terminology.
Within the Axiomatics-part of Sprachtheorie, the field concept comes into
focus in the fourth axiom, which deals with the syntax-semantics interface under
the heading Word and Sentence: the S[ymbol]-F[ield]-system of the type language (Bhler 1990: 8190; 1934: 6978), but the concept also dominates the remainder of the monograph. It serves Bhler to gain an overview of language phenomena that ranges from phonology and morphology via syntax and case theory to
text theory and stylistics. The principle unifying this theoretical and methodological tour de force is the distinction between, on the one hand, those aspects of linguistic signs that receive their precise meaning in the deictic field (Zeigfeld), i.e.
the spatio-temporal surroundings, the relationship of speaker and listener, their
shared knowledge and action goals, and on the other hand, those aspects of linguistic signs that are fully integrated in syntagms and are detachable from the immediate situation, thus forming a symbol field (Symbolfeld).24 As a programmatic outline, this two-field theory provides, as Garvin (1994) has pointed out,
an alternative to any rigid syntax-semantics-pragmatics distinction and has
proved to be a constant source of fresh insights. In terms of a theoretical focus,
however, Bhlers wide-ranging discussion covers too many data and theoretical
aspects and the important insights from the organon model and the speech action concept are taken up only intermittently in the field-theoretical chapters. In
the following section, we highlight these pragmatically relevant theory aspects
without claiming to provide a comprehensive reconstruction of Bhlers general
field-theoretical framework.

238
3.2.

Saskia Daalder and Andreas Musolff

Linguistic deixis and action

The two main concepts in the Sprachtheorie that are of relevance for linguistic
pragmatics are those of the deictic field and the surrounding fields, i.e. the context in which speech is meaningful as social action. Terminologically, both categories are evidently strongly integrated in Bhlers overarching field-theory,
but conceptually they also depend on the three-functions model. In both cases, the
multi-functionality of language and the fundamental importance of the speakerhearer relationship, which derive from the organon model, are constitutive of
Bhlers argumentation.
The origin of linguistic indexicality/deixis lies in the I-now-here-position of
the speaker, which serves as the starting point for spatial, temporal and social
orientation and co-ordination between the communication partners (Bhler 1990:
117136; 1934: 102120). The interlocutors shared horizon can then be extended
through the quasi-figurative use of the co-ordinates of the demonstratio ad oculos
(visual [or other sensory] demonstration) for the purpose of imaginationoriented deixis. When referring to objects that are not immediately given in the
discourse situation, speaker and hearer must either project them linguistically into
that situational context (e.g. by putting a problem before themselves) or, vice
versa, project their own situation co-ordinates onto an imagined context, or use a
mixture of both. Such imagination-oriented deixis is present already in everyday
language use and marks the first step in the gradual emancipation of language from
the immediate situational context (Bhler 1990: 158166; 1934: 140148). The
highest developmental stage of deictic language use is reached when indexical
language signs are used anaphorically and cataphorically as joints of speech, e.g.
as relative pronouns, text-deictic prepositions, prepositional adverbs or conjunctions (Bhler 1990: 438452; 1934: 385397). Here, the symbol-field is reconstructed along deictic co-ordinates; at the same time, deictic signs are used to
demonstrate relationships within a representational context.25
The second main pragmatically interesting discussion in the Sprachtheorie,
which Bhler had already flagged up in the context of the Axiomatics section when
dealing with the speech action, is chapter 10. Its title provides an additional typology of surrounding fields: The sympractical, the symphysical and the synsemantic field of Language Signs (1990: 175; 1934: 154). Again, as in the case of
the deixis-symbolising continuum, Bhler conceptualises the field-series as a cline
of abstraction. Linguistic signs can be tied closely to a specific situation or specific
objects (= the sympractical and symphysical fields) and so remain situation-bound.
The more strongly they are integrated into the linguistic con-text (= the synsemantic field) however, the more liberated they are from the specific circumstances
of the speech situation. In some genres, e.g. algebraic formula systems, the liberation from concrete, situation-related content is almost taken to the maximum (cf.
chapters 11, 25 of Sprachtheorie). The overlap between the synsemantic field

Foundations of pragmatics in functional linguistics

239

and the symbol field (Bhler 1990: 179189; 1934: 159168) blurs the boundaries
between the two field-theoretical concepts to some extent,26 which may be indicative of Bhlers difficulty to combine his focus on functional aspects of language with the perceived necessity of integrating these insights into the predominant contemporary tendencies of explaining language mainly as a system of
representation.
In the context of the surrounding fields, Bhler comes closest to a pragmatic
interpretation of language data in the passages dealing with the sympractical
field, which presuppose his three-functions axiom and Grnbaums action field
theory (see above). The sympractical field includes what would be called in modern terminology the situational context (as opposed to co-text), which communication partners need to know (and share) in order to be able to disambiguate
situation-bound utterances such as short commands, requests, greetings etc. Such
utterances had been viewed traditionally as deficient, on account of not being
syntactically well-formed sentences (Bhler 1990: 176179, 187189; 1934:
156159, 166167).27 Through examples such as that of a patron in a coffee house
asking for a particular type of coffee by way of uttering something like an espresso, please28 and of a passenger in a tram car choosing between types of
tickets, Bhler explains that an ellipsis interpretation is mistaken. In such cases,
the speaker does not have to produce semantically explicit or grammatically wellformed utterances, because the interlocutor, on account of shared knowledge of the
situation (the Umfeld), understands motives and plans for the action (in modern
terminology: the action schema or script) and uses the language sign only as a
diacritic. The situational context provides a sufficient fulfilment of all meaning
functions in the action field: the utterance is not elliptical but rather, in Bhlers
terminology, empractical (1990: 177179; 1934: 155158).
Generalizing from Bhlers hints, one can argue that language use is empractical in principle, insofar as it is always aimed at resolving ambiguities of intersubjective co-operation and based on some degree of shared background knowledge
(or at least, background assumptions). Whilst Bhler never formulated any communicative co-operation or relevance principles in the Sprachtheorie, his notion of the sympractical field as the basic action-structure in which language functions operate can be viewed at least as compatible with more recent pragmatic and
relevance-theoretical approaches.29
4.

Vilm Mathesius and the Prague Linguistic Circle

4.1.

Introduction

From 1930 onward, Bhler had scattered connections with members of the Prague
Linguistic Circle, the avant-garde group of linguists that propagated a view of language as a functional system with structural laws (Prague Linguistic Circle

240

Saskia Daalder and Andreas Musolff

[1929] 1983: 7779). Founded by the Czech linguist and scholar of English Vilm
Mathesius (18821945) in 1926, the Circle united under Mathesiuss presidency
some of his Czechoslovak colleagues and linguists of other nationalities who lived
in Prague or nearby, notably a number of Russian and Ukrainian migrs. Whereas
the functional point of view was the original contribution of Mathesius himself,
Roman Jakobson, the Russian vice-president of the Circle, added a structuralist,
much more system-oriented approach. Not surprisingly, Mathesiuss and Jakobsons legacies to pragmatics vary. The following section will detail Mathesiuss
functionalist findings that are relevant for pragmatics; in section 5, we will turn to
foundations of pragmatics deriving from Jakobsons work.
4.2.

Mathesiuss functional view of language

Since his earliest linguistic writings in the first decade of the century, Mathesius
had argued for a view of language as a means to an end.30 In the years of his university education in Prague, his general methodological outlook was formed by the
call for concreteness from the Czech philosopher Toms Masaryk (18501937),
who coupled this appeal with the desirability of democratising scientific and philosophical education (Masaryk 1887). Science had to give priority to the study of
close and accessible phenomena, including their comparative analysis; this
would prepare for historical research at a later stage.31 Accordingly, Mathesius
sought his linguistic inspirations in the works of opponents of Neogrammarian
dogmatism, such as Sweet, Otto Jespersen (18601943), von der Gabelentz and
Wegener (Mathesius [1936] 1966: 137), and he became an advocate of synchronic
language study.
The ideal of concreteness took shape in the emphasis which Mathesius laid
on the inherent synchronic variability of linguistic phenomena among individuals
of a language community as well as in the speech of a single individual; he spoke
of the static oscillation in speech and called this aspect of language its potentiality (Mathesius [1911] 1983a: 3). He stressed that the seeming homogeneity of
language is a product of the method of analysis and that without the phonetic phenomenon of synchronic oscillation, languages could not change (Mathesius
[1911] 1983a: 4, 32). These claims sound very modern and they were indeed approvingly cited in the influential paper on variability and language change by
Weinreich, Labov and Herzog (1968: 167169), although these authors also criticised Mathesius for not having pointed to the sociolinguistic regularities that are
inherent in variation.
In his ideas on synchronic oscillation in the domain of semantics, Mathesius
was at first strongly influenced by the stylistics of the Geneva linguist Charles
Bally (18651947) as well as by the idealistic approach of the German scholar of
Romance languages Karl Vossler (18721949). This stance implied an emphatic
interest in the expression of the speaking individual rather than in the social char-

Foundations of pragmatics in functional linguistics

241

acter of language, which he saw as having been overstated by nineteenth-century


linguistics (Mathesius [1911] 1983a: 2931, 3536; [1926] 1983b: 5253). In his
later work, Mathesius redressed the balance to some extent by way of an acknowledgement of the dominance, at least for language change, of the communicative
function of language vis--vis the original expressive function (Mathesius
[1926] 1983b: 5455). However, on account of his continued interest in stylistics
he kept a predilection for the expressive function and defined his functional
point of view as one that takes the meaning or function as its starting point and
tries to find out by which means it is expressed (Mathesius [1926] 1983b: 5658;
also [1929] 1983c: 122125). Mathesius thus came to neglect some of the necessary focus on the hearer as the key player in the constitution of meaning, something
which Wegener and Bhler had already reached in their work. This has to be
viewed as a general restriction of Praguian functional linguistics in its foundational
role for pragmatics.
Mathesiuss emphasis on the expressive function is understandable in the light
of his long-standing interest in the comparative study of word order patterns,
which was first kindled by the work of the French scholar of Ancient Greek Henri
Weil (18181909) (Weil [1844] 1887). Confronted with differences in word order
between Czech and English, Mathesius saw languages of different character.
Where a speaker of Czech usually filled the slot of the grammatical subject with the
agent of the action expressed by the verb, an English speaker appeared to prefer the
theme (Thema) of the utterance for this role. The Czech speaker therefore was
relatively free to order his sentence parts according to their contextual status of
theme or nucleus (rheme). For the English speaker to have a similar possibility readily at hand, the language had developed grammatical means for alternative choices of the grammatical subject, i.e. several types of passive constructions, which could be observed to make out a considerable proportion of English
verbal constructions (Mathesius 1929: 202203; [1929] 1983c: 126128, 133
134; 1930). The character of a language was thought to explain the links between
such descriptive facts of different types.
Mathesiuss characterology (or: language typology) thus relied on what he
called the actual articulation32 of utterances, which was to be distinguished from
its logico-grammatical articulation. The notion of actual articulation pointed to
the inherent situatedness of the structure in a specified context, and to its concrete quality; in short: it was a pragmatic structuring. Specifically, it consisted of
a basis or theme on the one hand and a nucleus, enunciation, or rheme
on the other.33 In a formulation from Mathesius (1975a: 81): The element about
which something is stated may be said to be the basis of the utterance or the theme,
and what is stated about the basis is the nucleus of the utterance or the rheme.
To an extent, Mathesiuss proposal of a functional sentence perspective resembles Wegeners ideas on the bipartition of utterances in exposition and
predicate that we discussed in section 2. Still, we note the less pronounced com-

242

Saskia Daalder and Andreas Musolff

municative emphasis with Mathesius: the hearer is absent in his formulation, and
even the speaker is no more than implied. Furthermore, Mathesiuss insight in the
role of the situation is much less developed. In any case, there appear to be no references by Mathesius to Wegeners theories on this subject, at least not from the
period in which Mathesius developed his notion of functional sentence structure
(the late 1920s). We have to decide that the relation is probably not a direct one.
It is certain however that Mathesiuss notions found favour in many quarters of
(pragmatic) linguistics and have become, under some of their original names or as
topic and comment, classics of pragmatic analysis (see, e.g., Lyons 1977:
501512; Levinson 1983: 8889).34

5.

Roman Jakobson

5.1.

Jakobsons model of linguistic functions

The Russian general linguist and scholar of Slavonic languages Roman Jakobson
(18961982) was intrigued by the seemingly autonomous functionality (teleology, self-movement) of language change all his life (Liberman 1987). For a
long time, Jakobson hoped to be able to account for it with the help of a Hegelian
logic of system change (Toman 1995: 171176). But in the milieu of the Prague
Linguistic Circle, he gradually came to recognise that the stages of a change in
progress had in fact to be conceived as different simultaneous facts, or styles,
within the language system. Especially after Jakobsons emigration to the United
States in 1941, his teleological point of view shifted to a focus on the less mysterious functions in a means-ends model of a synchronic system of systems (Jakobson 1963: 107).
In his new homeland, Jakobson found the study of language to have been, as he
remarked at a conference of anthropologists and linguists in 1952, strongly bulwarked by the impressive achievement of two conjoined disciplines the mathematical theory of communication and information theory (Jakobson 1953: 12).
This led him to discuss the factors of the communication process that were treated,
e.g., in the well-known publication by Shannon and Weaver (1949). In his paper,
Jakobson enumerates the four factors connected to the message in the information
theorists terminology (sender, receiver, topic of the message, code) and points to
the variability of their relative importance for a message. Apart from being associated with the factor of topic (information), a message may also be oriented toward the sender or the receiver, leading to a function of the message different
from a purely informative one, or, as the case may be, a hierarchical bundle of several functions (Jakobson 1953: 13). These three basic function varieties can be
considered equivalent to Bhlers (1990, 1934) functions of representation, expression and appeal, in the three-functions model that Jakobson subsequently

Foundations of pragmatics in functional linguistics

243

refers to as the traditional model of language (Jakobson 1960: 355; [1976] 1985:
115).35 It should be kept in mind however that Jakobson views the language functions as systemic facts throughout (facts of a system of systems). This distinguishes his theory from Bhlers approach (recall Bhlers championing of parole)
and even more from Wegeners; indeed, we are not aware of any references to Wegener in Jakobsons work.
Jakobson had come to linguistics via the study of Russian verse and of poetics
in general (cf. Rudy 1978; Toman 1995; Waugh and Rudy 1998: 22622265). Together with his compatriot, the literary scholar Jurij Tynjanov (18941943), he had
made a case for a parallel development of language, literature and other cultural
systems by structural necessity; there was to be no separation of the interconnected
codes of language and literature (Tynjanov and Jakobson ([1928] 1972). In a section of the 1929 Theses of the Prague Linguistic Circle, Jakobson and the Czech
literary scholar Jan Mukarovsky (18911975)36 had sought to contrast poetic language with communicative language. They had stated that poetic expression is
directed towards the way of expression itself and that the linguistic means of different levels, which in communicative speech tend to become automatised, in
poetic utterances tend to become foregrounded themselves (Prague Linguistic
Circle [1929] 1983: 94; cf. also 89). It is this idea that Jakobson takes up when he,
in his paper from 1953, adds a fourth, poetic function of language to Bhlers set
of three. For a message that is partly loosened from its referential meaning by
being oriented towards its own self, towards the palpability of its signs, the
poetic function of language is at work, be it as the predominant function of a literary fragment or as a subordinate, additional function of an otherwise prosaic
message (Jakobson 1953: 14).
Jakobson mentions also a fifth possibility: an orientation of the message toward
the factor of code, interesting because here the code (i.e., the langue) is not just an
abstraction but interacts with language use. The orientational set (explained as
Einstellung) for the code is witnessed in non-specialist discourse about such linguistic subjects as clarification of code elements, code switching, bilingualism,
areal diffusion, language variation and language change (Jakobson 1953: 1418).
This function is described with more clarity and termed the metalingual function
in a publication based on Jakobsons presidential address at a meeting of the Linguistic Society of America in 1956 (cf. Jakobson 1985: 116121). A sixth function
is mentioned in that address as well: Jakobson analyses anthropologist Malinowskis phatic communion function37 as being attached to messages with a set for the
factor of contact, or channel, as when interlocutors, without much information
transfer, check or revive the physical communication channel, or keep it going
(Jakobson 1985: 115).38
Jakobson (1960: 353357) sums up the resulting model of six language functions (with some terminological changes): the referential (denotative, cognitive)
function, the emotive (expressive) function, the conative39 function, the poetic

244

Saskia Daalder and Andreas Musolff

function, the metalingual (or glossing) function and the phatic function. They
are explained by the orientation of the message toward one of the six constitutive
factors of speech events, to wit the context referred to (the referent), the addresser,
the addressee, the message, the code, and the contact channel. It is in this version
that Jakobsons model of language functions has been widely cited by linguists of
several persuasions, including theoreticians of linguistic pragmatics (see, e.g.,
Hymes [1962] 1968: 99, 110124; Hymes 1983: 334; Lyons 1977: 5256; Levinson 1983: 41; Schiffrin 1994: 3233; Duranti 1997: 284287). It is clear, however,
that the model is focused on language as an undivided phenomenon, in which, just
as in Bhlers approach, the many facets of a pragmatic nature remain inclusive,
and therefore partly implicit. In order to gain an unimpeded view of pragmatic
facts, a systematic linguistic pragmatics had still to be drawn out.
5.2.

Jakobsons treatment of deixis

Among the further aspects of Jakobsons functionalism that are relevant for the development of linguistic pragmatics, his view of deictic phenomena is the most important. In a publication of 1957, which goes back to a paper delivered in Geneva
in 1950 (cf. Jakobson 1950), he treats the differentia specifica of indexical items
such as the personal pronouns I and you (Jakobson [1957] 1971). Like Bhler, Jakobson points to the fact that their precise meaning in context needs to take into account aspects of the situation, always unique, in which they happen to be used;
adopting Otto Jespersens (18601943) (1922: 123) term, he calls them shifters
(Jakobson 1971: 131). Jakobsons main interest however is focused on the question
of the status of these shifters as elements of the linguistic code, and he insists that
their symbolic meaning (for I, the addresser of the pronoun instance, and for you,
the addressee) should not be neglected either. Using Charles Sanders Peirces
(18391914) classification of sign modes and specifically following its interpretation by the American mathematician and philosopher Arthur Burks (19152008),
Jakobson argues in favour of according them the status of indexical symbols.40
First and second person pronouns in use are associated with the object they represent by an existential relation (i.e., pronoun token and object exist in the same
situation), therefore they indeed function as indices, just like acts of pointing
with the index finger would. However, on account of their conventional relation
with the role of speaker or hearer, which in different language codes holds for different forms (Latin ego, German ich, Russian ja), they partake just as well of the
symbolic mode of signs (Jakobson 1971: 132).41
Whereas the sober-minded Jespersen (1922: 123), in reaction to a remark by
the German idealist philosopher of subjectivity Johann Gottlieb Fichte
(17621814), had belittled attempts to derive deep philosophical conclusions from
the little linguistic trick involved in using the pronoun I correctly, Jakobson goes
on to develop its theoretical significance further. Considering the constitutive fac-

Foundations of pragmatics in functional linguistics

245

tors of speech situations, he notices for the shifters an overlapping of the factors
Code and Message: their general meaning, which is part of the Code, contains a
reference to the very Message of which the shifter in an instance of its use happens
to form a part (Jakobson 1971: 132). Jakobson explains this possibility by asserting
that both Code and Message can in principle refer as well as be an object of reference; in the specific case of the shifters, the Code has the function of referring,
while the Message is being referred to.42
This explanation takes up a similar remark by the French linguist mile Benveniste (19021976) in an article about pronouns that was published a year earlier
in the Festschrift For Roman Jakobson (Benveniste [1956] 1966a: 252). Benveniste, likewise anxious to find an aspect of the Saussurean langue (the Code) within
the typical discourse (Message) reality of first and second person pronouns, goes
further than Jakobson in postulating that it is precisely the use of elements symbolizing the roles of utterer-conceptualiser (nonciateur) and allocutee of a discourse event which leads to language users consciousness of their role as subjects. Those symbolic features of the language system then get parcelled out into
psychological reality. Benveniste also mentions that the process is strengthened by
various verbal forms that align with the perspective of the utterer of the discourse
(Benveniste 1966a: 254255; also [1958] 1966b: 258263). In his 1957 article,
Jakobson indeed uses the theory of the shifters to set up a new classification of verbal categories, in particular those of Russian. Like first and second person pronouns, verbal shifter categories (several forms in the paradigms of person, mood,
tense and evidentiality) have an encoded meaning in which the speech event (the
message) and/or one of its participants play a decisive role (Jakobson 1971:
133147).43
The relevance of the idea of the interweaving of Code (langue) and Message
(instance of discourse) to linguistic pragmatics becomes evident when the phenomenon of the shifters is described as a fact about speech events rather than the
linguistic code. It can be said then that by means of various indexical elements
of the language, speech events often encode, or index, important components of
the communitys social life. It was the American anthropologist Dell Hymes
(19272009) who realised that on this basis, Jakobsons speech functions could be
extended and seen to be differently organised for different communities, opening
up an additional realm of structure (Hymes 1983: 335). This was to result in the
ethnography of speaking and more broadly, the ethnography of communication
(Hymes [1962] 1968: 110129; 1964; 1972).
6.

Conclusion

Philipp Wegener, Karl Bhler, Vilm Mathesius and Roman Jakobson all of them
characteristically modernist theorists were fully aware of the pervasive relevance of language in countless domains of human life and culture. They projected

246

Saskia Daalder and Andreas Musolff

the pragmatic aspects which they had identified, such as the multifunctionality of
language, the functional structure of the sentence and the nature of deixis, onto a
rich picture of language that did not separate semantic from pragmatic phenomena
and often contained no fixed answer to the question as to what is language itself
and what is its use. Seen from a present-day perspective, their works may appear to
be lacking in systematic treatment of pragmatic facts. By the same token, however,
they have had and continue to have potential not for just one but many branches
of linguistic pragmatics. Wegeners ideas for the linguistic analysis of utterances
together with their embedding situations may still be used to further loosen the grip
of the received wisdom about the structure of linguistic utterances as phenomenal
entities per se. Similarly, Bhlers and Jakobsons theories of the functions of language, the associated models of the communication situation and the deictic signs/
shifters that derive their specific meaning from it offer the chance to review the
immunity of the theory of the language system to results of the analysis of language use.

Notes
1. Outside the discipline of linguistics, Bhlers model has been taken up by Jrgen Habermas, who in his theory of communicative action (cf. 1981) uses the three functional dimensions as reference points for a model of social communication. According to Habermas, every speaker, when making a meaningful utterance, intends it to be understood by
the hearer as legitimate (within the social context), truthful (i.e. as an expression of the
speakers own convictions) and true (in relation to objective states of affairs). Even
though these assumptions are often disappointed in actual communication, they act as
regulating norms for meaningful social exchange (Habermas 1981, vol. 1: 372377,
412). See further the article by Cooke in this volume.
2. For information on Wegener and his work, see especially Knobloch (1991); also Juchem
(1984, 1986), Knobloch (1988: 292297), Nerlich (1990: 153192; 1992: 8187), Graffi
(1991: 8789), Elffers (1993), Nerlich and Clarke (1996: 177183), Morpurgo Davies
(1998: 318320), Grimm-Vogel (1998), Tenchini (2008).
3. An English translation of Wegeners 1885 book appeared as part of Abse (1971); note
that Knobloch (1991: xliv) expresses a caveat about details of the translation.
4. In his publication from 1885, Wegener uses the term logical predicate; in later works he
switches to psychological predicate and statement (Mitteilung).
5. Knobloch (1991: xvi, xxxii) mentions Lazarus ([18561857] 1884) as a possible source
of Wegeners ideas in this area. About Moritz Lazarus (18241903), see Knobloch (1988:
268273, 411422) and Nerlich and Clarke (1996: 165168).
6. For details, see Hlzer (1987), Knobloch (1988: 296297; 1991: xviii-xix), Nerlich
(1990: 181185; 1992: 8687).
7. This is the opinion also of Knobloch (1988: 297) and Nerlich (1992: 87).
8. See the references to Wegeners work in Malinowski ([1923] 1946: 297), Gardiner (1932:
passim) and Halliday (1978: 109), particularly in relation to the notion of situation.

Foundations of pragmatics in functional linguistics

247

9. See Ungeheuer (1987, 1990) and Juchem (1984, 1986).


10. For details see Knobloch (1991: xxxix-xli).
11. Quotations from Bhlers works are referenced, where possible, to both the original
German publications and their English translation, i.e., in the case of the Sprachtheorie,
1934 for the German volume, 1990 for its translation.
12. See e.g. Schlieben-Lange (1979: 13); Levinson (1983: 41, 61); Brown and Yule (1983:
1): Coupland (2007: 12).
13. After the Sprachtheorie (1934), Bhler published several other linguistic articles up
until 1938 (Bhler 1935, 1936a, 1936b, 1938). Immediately after the occupation of
Vienna by German troops, Bhler was imprisoned by the Gestapo. His wife and academic collaborator, Charlotte Bhler, managed to effect his release and to organise their
emigration to the USA, but the continuation of their careers proved difficult: they only
gained posts as visiting or assistant professors and did not return to Europe after the war
(see Garvin 1966, Ch. Bhler 1984, Musolff 1997). The 1960 article Von den Sinnfunktionen der Sprachgebilde (Of the Semantic functions of linguistic structures) is an
excerpt from Sprachtheorie and of dubious editorial status (e.g. it cites the Organonmodell as Organmodell).
14 Husserl (19001901); Wundt (1900); Marty (1908). Bhler reviewed Martys work with
specific emphasis on the functional typology, see Bhler (1909a: 964946).
15. For Bhlers experiments on sentence comprehension in the context of Thought Psychology, see Bhler (1908), Wundt (1908), Wettersten (1988).
16. For the changes in the functions-terminology and the status and ordering of the
axioms, see Wunderlich (1969); Kamp (1977: passim); Beck (1980: 169192); Busse
(1975: 213215, 222229); Camhy (1980a: 9294); Swiggers (1981: 5455); Innis
(1982: 310); Eschbach (1984b: 9193); Henzler (1987: 354356); Knobloch (1988:
424430); Graumann (1988); Musolff (1990: 2631); Vonk (1992: passim); Elffers
(2005; 2008: 2124).
17. For the contemporary context of the Saussure-reception in relation to Bhler see
Scheerer (1980: 3640), Koerner (1984), Musolff (1990: 4043) and Ehlich (1996).
18. Details of what Bhler had learnt from Wegener can be found in Bhler (1909b:
119123).
19. See Gardiner (1932: viii).
20. For detailed discussion of these and other pragmatic-functional influences cf. Graumann (1984), Herrmann (1984), Knobloch and Schallenberger (1993), Musolff (1993)
and Nerlich and Clarke (1996: 182183, 226236, 341343).
21. See Bhler (1933b: 90); in Strkers edition (1976: 116117).
22. See Laziczius (1939); Lohmann (19421943); Camhy (1980b); Ortner (1986).
23. For detailed discussion of Grnbaums influence on Bhler see Knobloch and Schallenberger (1993: 8590).
24. For Bhlers field-theoretical approach (as opposed to the contemporary notion of
semantic fields) cf. Wolf (1932); Heger (1984); Musolff (1990: 61120); Garvin
(1994).
25. For Bhlers influence on modern pragmatically oriented studies of deixis see Hrmann
(1978: 394424); Weissenborn (1988); Conte et al. (1989); Innis (1992: 556557);
Ehlich (1996: 959).
26. Garvin (1994: 6064) interprets the relationship as a superordination of the more general surrounding field over the two subcategories of symbolic and deictic field; he

248

27.
28.
29.

30.

31.
32.

33.

34.
35.

36.
37.
38.

39.
40.

Saskia Daalder and Andreas Musolff

bases this reading on an excellent comparison with the contemporary gestalt-psychological theories, which Bhler himself highlighted (1990: 175176, 1934: 154155) as
well as with early semiotics (Charles Morris). However, in the Sprachtheorie this classification is not explicated nor consistently followed it constitutes a further theoretical
(and highly plausible) development, the implications of which are still to be worked out
in detail.
For critical assessments see Klein (1984), Ortner (1983).
Living in Vienna at the time, Bhler of course chose an appropriate local variety (einen
schwarzen a black one).
But compare Nerlich and Clarkes (1996: 236237) critical comment that Bhler, despite being one of the most important figures among language researchers engaged in
pragmatics avant la lettre, never crossed the last hurdle on his way to a fully developed theory of linguistic pragmatics, especially on account of his [clinging] too much
to his three functions of language, overlooking to some extent the polyfunctionality of
speech acts, rediscovered by Wittgenstein.
Toman (1995: 7779). About Mathesiuss work, see further Trnka ([1946] 1966), Vachek (1982), Danes (1987, 2003), Fronek (1988), Graffi (1991: 179183, 249250),
Toman (1995: 7186, 97101 and passim), Leska (1995), Nekula (1999).
See Toman (1995: 8486, 97101), also Fronek (1988).
This notion features in the original title of the Czech article Mathesius ([1939] 1975b);
see the translators remark in Mathesius (1975b: 479). It came to be translated as information-bearing structure of the sentence, shortened to information structure and
alternating with functional sentence perspective (FSP).
The terminology differs; see Mathesius ([1929] 1983c: 126128; 1929: 202; 1930: 58).
The term rheme came into use only later, through the writings of the Brno scholar of
English Jan Firbas (19212000). See the explanation by Vachek in Mathesius (1975a:
185 n71), further Vachek (1982: 123124) and Danes (1987: 2326; 2003: 4041).
The Prague version of the information structure of the utterance has remained a distinctive one. See a short description in Hajicov (1994).
Jakobson referred to Bhlers three functions already in 1932 (Jakobson [1932] 1966:
2728). In its Theses of again some years earlier, the Prague Linguistic Circle had distinguished, under the heading On the functions of langugage, intellectual speech
from two types of emotional speech and had mentioned the different hierarchy of
functions for each utterance (Prague Linguistic Circle [1929] 1983: 8889). Jakobson
probably wrote (part of) this section (nr. 3a) of the Theses (cf. Toman 1995: 289 n13).
For the attribution to Jakobson and Mukarovsky, see again Toman (1995: 289 n13).
See Malinowski ([1923] 1946: 315316). Cf. Nekula and Ehlers (1996: 190, 192) on the
at first marginal place of the phatic function of language in Jakobsons model.
On Jakobsons model of language functions, see Holenstein (1976: 153164); Waugh
(1976: 2526); Waugh and Monville-Burston (1990: 1516); Nekula and Ehlers (1996);
Nerlich and Clarke (1996: 284286); Waugh and Rudy (1998: 22602261).
Conative: with the notion of striving to effect some act by the addressee; cf. Jakobson ([1932] 1966: 22, 27).
See Burks (1949). Among the signs having the property of indexicality, Burks views the
indexical symbol as the most fundamental type; a pure index cannot in effect point
to a well-defined object and is therefore a less fundamental type of sign (Burks 1949:
678688).

Foundations of pragmatics in functional linguistics

249

41. Jakobsons (1971: 132) statement to the effect that Bhler acknowledged the indexical
side of personal pronouns only seems not fully justified. It is when Bhler introduces the
indexical field (Zeigfeld) that he contrasts indexical and appellative words as to the
field in which, specifically, their meaning fulfilment (Bedeutungsprzision, Bedeutungserfllung, see above, 2.2) is located: in the indexical field and the symbol-field of
an utterance, respectively (Bhler 1990: 9194; 1934: 7881). This does not imply that
personal pronouns would not have a symbolic meaning aspect to begin with, such as
the sender of the word in question (Sender des aktuellen Wortes) (Bhler 1990: 91;
1934: 78); indeed, the notion of meaning fulfilment presupposes the opposite idea.
42. Considering the three other possibilities of combining the choices of Code/Message and
referring/referred-to item, Jakobson theorises further in the following way. When the
functions are turned round and it is rather the Message that refers and the Code that is referred to, we have to do with the elucidation of words and sentences (what would be
named the metalingual function of language a few years later); when a Message refers
to a Message, the result is the well-known phenomenon of reported speech; and lastly,
with Code referring to Code we could describe the general nature of the meaning of
proper names, Jakobson holds (Jakobson 1971: 130131).
43. On Jakobsons notion of shifters, see Holenstein (1976: 158, 162163); Waugh (1976:
2425, 32, 48, 52); Waugh and Monville-Burston (1990: 1719); Nerlich and Clarke
(1996: 282284).

References
Abse, D. Wilfred
1971
Speech and Reason: Language Disorder in Mental Disease. With a Translation
of The Life of Speech by Philipp Wegener. Charlottesville, VA: University
of Virginia Press.
Beck, Gtz
1980
Sprechakte und Sprachfunktionen: Untersuchungen zur Handlungsstruktur
der Sprache und ihren Grenzen. Tbingen: Niemeyer.
Benveniste, mile
1966a
La nature des pronoms. In: mile Benveniste, Problmes de linguistique gnrale, tome 1, 251257. Paris: Gallimard. First published The Hague: Mouton
[1956].
Benveniste, mile
1966b
De la subjectivit dans le langage. In: mile Benveniste, Problmes de linguistique gnrale, tome 1, 258266. Paris: Gallimard. First published Paris: PUF
[1958].
Brown, Gillian and George Yule
1983
Discourse Analysis. Cambridge: Cambridge University Press.
Brugmann, Karl
1904
Die Demonstrativpronomina der indogermanischen Sprachen: Eine bedeutungsgeschichtliche Untersuchung. In: Berichte ber die Verhandlungen der
Kniglichen schsischen Gesellschaft der Wissenschaften zu Leipzig, Philologisch-Historische Classe, No. 6: 3151.

250

Saskia Daalder and Andreas Musolff

Bhler, Charlotte
1984
Karl Bhler: Eine biographische Skizze. Transl. by Achim Eschbach. In:
Achim Eschbach (ed.), Bhler-Studien, Vol. 1, 2530. Frankfurt am Main:
Suhrkamp.
Bhler, Karl
1908
Tatsachen und Probleme zu einer Psychologie der Denkvorgnge: II) ber
Gedankenzusammenhnge. Archiv fr die gesamte Psychologie 12: 123; III)
ber Gedankenerinnerungen. Archiv fr die gesamte Psychologie 12: 2492.
Bhler, Karl
1909a
Review of Anton Marty, Untersuchungen zur Grundlegung der allgemeinen
Grammatik und Sprachphilosophie (1908). Gttingische gelehrte Anzeigen
171: 947979.
Bhler, Karl
1909b
ber das Sprachverstndnis vom Standpunkt der Normalpsychologie aus. In:
Bericht ber den 3. Kongre fr experimentelle Psychologie (Frankfurt 1908),
94130. Leipzig: Wilhelm Engelmann.
Bhler, Karl
1918
Kritische Musterung der neuern Theorien des Satzes. Indogermanisches Jahrbuch 6: 120.
Bhler, Karl
1927
Die Krise der Psychologie. Jena: J.A. Barth.
Bhler, Karl
1931
Phonetik und Phonologie. Travaux du Cercle Linguistique de Prague 4: 2253.
Bhler, Karl
1932
Das Ganze der Sprachtheorie, ihr Aufbau und ihre Teile. In: Gustav Kafka
(ed.), Bericht ber den 12. Kongre der Deutschen Gesellschaft fr Psychologie (Hamburg 1931), 95122. Jena: G. Fischer.
Bhler, Karl
1933a
Ausdruckstheorie: Das System an der Geschichte aufgezeigt. Jena: G. Fischer.
Bhler, Karl
1933b
Die Axiomatik der Sprachwissenschaften. Kant-Studien 38: 1990. (Reprinted
1969 and 1976 under the same title in a truncated version without the introductory chapter, edited by Elisabeth Strker, 25117). Frankfurt am Main: Klostermann.
Bhler, Karl
1934
Sprachtheorie: Die Darstellungsfunktion der Sprache. Jena: G. Fischer.
Bhler, Karl
1935
Forschungen zur Sprachtheorie: Einleitung. Archiv fr die gesamte Psychologie 94: 401412.
Bhler, Karl
1936a
Psychologie der Phoneme. In: Daniel Jones and Dennis B. Fry (eds.), Proceedings of the Second International Congress of Phonetic Sciences, 162169.
Cambridge: Cambridge University Press.
Bhler, Karl
1936b
Das Strukturmodell der Sprache. Travaux du Cercle Linguistique de Prague 6:
312.

Foundations of pragmatics in functional linguistics

251

Bhler, Karl
1938
Der dritte Hauptsatz der Sprachtheorie: Anschauung und Begriff im Sprechverkehr. In: Henri Piron and Emile Meyerson (eds.), 11e Congrs International de Psychologie (Paris 1937). Paris: Rapports et Comptes Rendus,
196203.
Bhler, Karl
1960
Von den Sinnfunktionen der Sprachgebilde. In: Richard Wisser (ed.), Sinn und
Sein: Ein philosophisches Symposion; 95112. Tbingen: Niemeyer:
Bhler, Karl
1990
Theory of Language: The Representational Function of Language. Translation
by Donald Fraser Goodwin. Amsterdam: Benjamins.
Burks, Arthur W.
1949
Icon, index, and symbol. Philosophy and Phenomenological Research 9:
673689.
Busse, Winfried
1975
Funktionen und Funktion der Sprache. In: Brigitte Schlieben-Lange (ed.),
Sprachtheorie, 207240. Hamburg: Hoffmann and Campe:
Camhy, Daniela G.
1980a
Karl Bhlers Sprachtheorie. PhD. Diss. Graz: University of Graz.
Camhy, Daniela G.
1980b
Sprechereignis als Handlung: Bhlers Beitrag zu handlungstheoretischen Anstzen in der Sprachphilosophie. In: Rudolf Haller and Wolfgang Grassl (eds.),
Sprache, Logik und Philosophie, 401405. Wien: Hlder-Pichler-Temsky.
Conte, Maria Elisabeth, Jnos Snder Petfi and Emel Szer (eds.)
1989
Text and Discourse Connectedness. Amsterdam: Benjamins.
Coupland, Nikolas
2007
Style: Language, Variation and Identity. Cambridge: Cambridge University
Press.
Danes, Frantisek
1987
On Prague School functionalism in linguistics. In: Ren Dirven and Vilm
Fried (eds.), Functionalism in Linguistics, 338. (Linguistic and Literary
Studies in Eastern Europe 20.). Amsterdam: Benjamins.
Danes, Frantisek
2003
Vilm Mathesius Konzeption der funktionalen Linguistik. In: Marek Nekula
(ed.), Prager Strukturalismus: Methodologische Grundlagen, 3147. (Slavica:
Monographien, Hand-, Lehr- und Wrterbcher 3.). Heidelberg: Universittsverlag Winter.
Duranti, Alessandro
1997
Linguistic Anthropology. (Cambridge Textbooks in Linguistics.) Cambridge:
Cambridge University Press.
Ehlich, Konrad
1996
Sprache als System versus Sprache als Handlung. In: Marcelo Dascal, Dietfried Gerhardus, Kuno Lorenz, Georg Meggle (eds.), Sprachphilosophie: Ein
internationales Handbuch zeitgenssischer Forschung, vol. 1, 952963.
(Handbcher zur Sprach- und Kommunikationswissenschaft 7.1). Berlin/New
York: de Gruyter.

252

Saskia Daalder and Andreas Musolff

Elffers, Els
1993
Philipp Wegener as proto-speech act theorist. Linguistics in the Netherlands
1993: 4959. Amsterdam: Benjamins.
Elffers, Els
2005
Reichling and de Groot: Two Dutch reactions to Bhlers organon-model. Historiographia Linguistica 32: 87116.
Elffers, Els
2008
Interjections and the language functions debate. The Henry Sweet Society
Bulletin 50: 1729.
Eschbach, Achim (ed.)
1984a
Bhler-Studien. 2 vols. Frankfurt am Main: Suhrkamp.
Eschbach, Achim
1984b
Karl Bhlers Axiomatik und das Axiomensystem der Zeichentheorie. In: Carl
F. Graumann and Theo Herrmann (eds.), Karl Bhlers Axiomatik: Fnfzig
Jahre Axiomatik der Sprachwissenschaften, 53104. Frankfurt am Main:
Klostermann.
Fronek, Josef
1988 T.G. Masaryk and Prague School linguistic structuralism. In: Josef Novk (ed.), On
Masaryk: Texts in English and German, 8196. (Studien zur sterreichischen
Philosophie, 13). Amsterdam: Rodopi.
Gabelentz, Georg von der
1875
Weiteres zur vergleichenden Syntax: Wort- und Satzstellung. Zeitschrift fr
Vlkerpsychologie und Sprachwissenschaft 8: 129165, 300338.
Gardiner, Sir Alan Henderson
1932
The Theory of Speech and Language. Oxford: Clarendon Press. (2nd ed. with
additions 1951.)
Garvin, Paul L.
1966
Karl Bhlers contributions to the theory of linguistics. The Journal of General
Psychology 75: 212215.
Garvin, Paul L.
1994
Karl Bhlers Field Theory in the Light of the Current Interest in Pragmatics.
In: Svetla Cmejrkov and Frantisek Stcha (eds.), The Syntax of Sentence and
Text: A Festschrift for Frantisek Danes, 5966. Amsterdam: Benjamins.
Graffi, Giorgio
1991
200 Years of Syntax: A Critical Survey. (Studies in the History of the Language
Sciences 98.) Amsterdam: Benjamins.
Graumann, Carl F.
1984
Wundt Mead Bhler: Zur Sozialitt und Sprachlichkeit menschlichen Handelns. In: Carl F. Graumann and Theo Herrmann (eds.), Karl Bhlers Axiomatik: Fnfzig Jahre Axiomatik der Sprachwissenschaften, 217247. Frankfurt
am Main: Klostermann.
Graumann, Carl F.
1988
Aspektmodell und Organonmodell: Die Problematik des Verhltnisses zwischen
Sprachwissenschaft und Psychologie bei Karl Bhler. In: Achim Eschbach
(ed.), Karl Bhlers Theory of Language, 107124. Amsterdam: Benjamins.
Grimm-Vogel, Irmingard Hildburg
1998
Philipp Wegener 18481916: Wesen Wirken Wege. Dissertation Rheinischen Friedrich-Wilhelms-Universitt zu Bonn.

Foundations of pragmatics in functional linguistics

253

Grnbaum, Abraham Anton


1932
Sprache als Handlung. In: Gustav Kafka (ed.), Bericht ber den 12. Kongre
der Deutschen Gesellschaft fr Psychologie (Hamburg 1931), 164176. Jena:
G. Fischer.
Habermas, Jrgen
1981
Theorie des kommunikativen Handelns. 2 vols. Frankfurt am Main: Suhrkamp.
Hajicov, Eva
1994
Topic/Focus and related research. In: Philip A. Luelsdorff (ed.), The Prague
School of Structural and Functional Linguistics: A Short Introduction,
245275. (Linguistic and Literary Studies in Eastern Europe 41.). Amsterdam:
Benjamins.
Halliday, Michael A.K.
1978
Language as Social Semiotic: The Social Interpretation of Language and
Meaning. London: Edward Arnold.
Heger, Klaus
1984
Zeigfeld und Symbolfeld. In: Carl F. Graumann and Theo Herrmann (eds.),
Karl Bhlers Axiomatik: Fnfzig Jahre Axiomatik der Sprachwissenschaften, 105116. Frankfurt am Main: Klostermann.
Henzler, Stefan
1987
Die Funktionen der Sprache im Gebrauch: Zu Konsistenz und Funktionalismus von Karl Bhlers Sprachtheorie. In: Werner Abraham and Ritva rhammar (eds.), Linguistik in Deutschland, 353365. Tbingen: Niemeyer.
Herrmann, Theo
1984
Sprachstrukturen und ihre aktionale Realisation: Psychologische Erwgungen
zu Karl Bhlers Axiomatik der Sprachwissenschaften. In: Carl F. Graumann
and Theo Herrmann (eds.), Karl Bhlers Axiomatik: Fnfzig Jahre Axiomatik
der Sprachwissenschaften, 143172. Frankfurt am Main: Klostermann.
Holenstein, Elmar
1976
Roman Jakobsons Approach to Language: Phenomenological Structuralism.
Transl. by Catherine Schelbert and Tarcisius Schelbert. Bloomington: Indiana
University Press.
Husserl, Edmund
19001901 Logische Untersuchungen. Vol. I: Prologomena zur reinen Logik. Vol. II/1:
Untersuchungen zur Phnomenologie und Theorie der Erkenntnis. Vol. II/2:
Elemente einer phnomenologischen Aufklrung der Erkenntnis. Halle: Niemeyer.
Hrmann, Hans
1978
Meinen und Verstehen: Grundzge einer psychologischen Semantik. Frankfurt
am Main: Suhrkamp.
Hlzer, Heike
1987
Die Metapher: Kommunikationssemantische berlegungen zu einer rhetorischen Kategorie. Mnster: Nodus.
Hymes, Dell
1964
Introduction: Toward ethnographies of communication. In: John J. Gumperz
and Dell Hymes (eds.), The Ethnography of Communication, 134. (Special
issue of American Anthropologist 66, 6, part 2.)

254

Saskia Daalder and Andreas Musolff

Hymes, Dell H.
1968
The ethnography of speaking. In: Joshua A. Fishman (ed.), Readings in
the Sociology of Language, 99138. The Hague: Mouton. First published
Washington, D.C. [1962].
Hymes, Dell
1972
Models of the interaction of language and social life. In: John J. Gumperz
and Dell H. Hymes (eds.), Directions in Sociolinguistics: The Ethnography of
Communication, 3571. Oxford: Blackwell.
Hymes, Dell
1983
The pre-war Prague school and post-war American anthropological linguistics.
In: Dell Hymes, Essays in the history of linguistic anthropology, 331345.
Amsterdam: Benjamins.
Innis, Robert E.
1982
Karl Bhler: Semiotic Foundations of Language Theory. New York/London:
Plenum Press.
Innis, Robert E.
1992
Karl Bhler (18791963). In: Marcelo Dascal, Dietfried Gerhardus, Kuno Lorenz, Georg Meggle (eds.), Sprachphilosophie: Ein internationales Handbuch
zeitgenssischer Forschung. Vol. 1, 550562. (Handbcher zur Sprach- und
Kommunikationswissenschaft 7.1.) Berlin/New York: de Gruyter.
Jakobson, Roman
1950
Les catgories verbales. [Summary]. Cahiers Ferdinand de Saussure 9: 6.
Jakobson, Roman
1953
[From the point of view of linguistics.] In: Results of the Conference of Anthropologists and Linguists, 1121. (Indiana University Publications in Anthropology and Linguistics; International Journal of American Linguistics, Memoir 8.)
Baltimore: Waverly Press.
Jakobson, Roman
1960
Closing statement: Linguistics and poetics. In: Thomas A. Sebeok (ed.), Style
in Language, 350377. Cambridge, MA: MIT Press.
Jakobson, Roman
1963
Efforts toward a means-ends model of language in inter-war continental linguistics. In: Christine Mohrmann, F. Norman and Alf Sommerfelt (eds), Trends
in modern linguistics, 104108. Utrecht/Antwerp: Spectrum.
Jakobson, Roman
1966
Zur Struktur des russischen Verbums. In: Eric P. Hamp, Fred W. Householder
and Robert Austerlitz (eds.), Readings in Linguistics II, 15. Chicago/London:
The University of Chicago Press. First published Prague: Prague Linguistic
Circle [1932].
Jakobson, Roman
1971
Shifters, verbal categories, and the Russian verb. In: Roman Jakobson, Selected Writings, vol. 2: Word and Language, 130147. The Hague/Paris: Mouton. First published Cambridge, MA: Harvard University [1957].
Jakobson, Roman
1985
Metalanguage as a linguistic problem. In: Roman Jakobson, Selected Writings,
vol. 7: Contributions to Comparative Mythology: Studies in Linguistics and
Philology, 19721982, ed. by Stephen Rudy, 113121. Berlin: Mouton de
Gruyter. First published [1976].

Foundations of pragmatics in functional linguistics

255

Jespersen, Otto
1922
Language: Its Nature, Development and Origin. London: George Allen and
Unwin; New York: Holt.
Juchem, Johann G.
1984
Die Konstruktion des Sprechens: Kommunikationssemantische Betrachtungen
zu Philipp Wegener. Zeitschrift fr Sprachwissenschaft 3: 318.
Juchem, Johann G.
1986
Wegener und Wundt. Kodikas/Code 9: 155166.
Juchem, Johann G.
1987
Einleitung. In: Gerold Ungeheuer, Kommunikationstheoretische Schriften,
vol. 1: Sprechen, mitteilen, verstehen. Ed. by Johannes G. Juchem, 117. (Aachener Studien zur Semiotik und Kommunikationsforschung 14.). Aachen:
Rader Verlag.
Kamp, Rudolf
1977
Axiomatische Sprachtheorie: Wissenschaftstheoretische Untersuchungen zum
Konstitutionsproblem der Einzelwissenschaften am Beispiel der Sprachwissenschaftstheorie Karl Bhlers. Berlin: Duncker and Humblot.
Klein, Wolfgang
1984
Bhler Ellipse. In: Carl F. Graumann and Theo Herrmann (eds.), Karl Bhlers
Axiomatik: Fnfzig Jahre Axiomatik der Sprachwissenschaften, 117141.
Frankfurt am Main: Klostermann.
Knobloch, Clemens
1988
Geschichte der psychologischen Sprachauffassung in Deutschland von 1850
bis 1920. (Reihe Germanistische Linguistik 86.) Tbingen: Niemeyer.
Knobloch, Clemens
1991
Introduction. In: Philipp Wegener, Untersuchungen ber die Grundfragen des
Sprachlebens. New edition by Konrad Koerner, with an introduction by Clemens Knobloch, xili. (Classics in Psycholinguistics 5.) Amsterdam: Benjamins.
Knobloch, Clemens and Stefan Schallenberger
1993
Sprechhandlung und Sprachbedeutung in der Sprachpsychologie um 1930.
Histoire pistmologie Langage 15/1: 81109.
Koerner, E. F. Konrad
1984
Karl Bhlers theory of language and Ferdinand de Saussures Cours. Lingua 62: 324.
Lazarus, Moritz
1884
Geist und Sprache: Eine psychologische Monographie. 3rd ed. Berlin:
Dmmler. First published Berlin [18561857].
Laziczius, Julius von
1939
Das sog. dritte Axiom der Sprachwissenschaft. Acta Linguistica 1: 162167.
Leska, Oldrich
1995
Vilm Mathesius (18821945). [In Czech.] Slovo a Slovesnost 56: 8197.
Levinson, Stephen C.
1983
Pragmatics. Cambridge: Cambridge University Press.
Liberman, Anatoly
1987
Roman Jakobson and his contemporaries on change in language and literature
(the teleological criterion). In: Krystyna Pomorska, Elzbieta Chodakowska,

256

Saskia Daalder and Andreas Musolff

Hugh McLean and Brent Vine (eds.), Language, Poetry and Poetics: The
Generation of the 1890s: Jakobson, Trubetzkoy, Majakovskij, 143155. Berlin:
Mouton de Gruyter.
Lohmann, Johannes
19421943 Karl Bhlers Drittes Axiom. Acta Linguistica 3: 516.
Lyons, John
1977
Semantics. 2 vols. Cambridge: Cambridge University Press.
Malinowski, Bronislaw
1946
The problem of meaning in primitive languages. In: Charles K. Ogden and Ivor
A. Richards, The Meaning of Meaning: A Study of the Influence of Language
upon Thought and of the Science of Symbolism. 8th ed, 296336. London:
Kegan Paul. First published London: Kegan Paul [1923].
Marty, Anton
1908
Untersuchungen zur Grundlegung der allgemeinen Grammatik und Sprachphilosophie. Vol. 1. Halle: Niemeyer.
Masaryk, Toms Garrigue
1887
Versuch einer concreten Logik: Classification und Organisation der Wissenschaften. Wien: C. Konegen.
Mathesius, Vilm
1929
Zur Satzperspektive im modernen Englisch. Archiv fr das Studium der
neueren Sprachen und Literaturen 155: 202210.
Mathesius, Vilm
1930
On linguistic characterology with illustrations from Modern English. In: Actes
du premier Congrs International de Linguistes la Haye, du 1015 avril
1928, 5663. Leiden: Sijthoff.
Mathesius, Vilm
1966
Ten years of the Prague Linguistic Circle. Transl. by Josef Vachek. In: Josef
Vachek (ed.), The Linguistic School of Prague: An Introduction to its Theory
and Practice, 137151. (Indiana University Studies in the History and Theory
of Linguistics.) Bloomington/London: Indiana University Press. First published in Czech, Prague [1936].
Mathesius, Vilm
1975a
A Functional Analysis of Present Day English on a General Linguistic Basis.
Ed. by Josef Vachek, transl. by L. Duskov. (Janua Linguarum, Series Practica
208). The Hague/Paris: Mouton; Prague: Academia. First published in Czech,
Prague [1961].
Mathesius, Vilm
1975b
On information-bearing structure of the sentence. Transl. by Tsuneko Olga Yokoyama. Harvard Studies in Syntax and Semantics 1: 467480. First published
in Czech, Prague [1939].
Mathesius, Vilm
1983a
On the potentiality of the phenomena of language. Transl. by Josef Vachek. In:
Josef Vachek and Libuse Duskov (eds.), Praguiana: Some basic and less
known aspects of the Prague Linguistic School, 343. (Linguistic and Literary
Studies in Eastern Europe 12.). Amsterdam: Benjamins. First published in
Czech, Prague [1911].

Foundations of pragmatics in functional linguistics

257

Mathesius, Vilm
1983b
New currents and tendencies in linguistic research. In: Josef Vachek and
Libuse Duskov (eds.), Praguiana: Some basic and less known aspects of the
Prague Linguistic School, 4563. (Linguistic and Literary Studies in Eastern
Europe 12.). Amsterdam: Benjamins. First published Prague [1926].
Mathesius, Vilm
1983c
Functional linguistics. Transl. by Libuse Duskov. In: Josef Vachek and
Libuse Duskov (eds.), Praguiana: Some basic and less known aspects of the
Prague Linguistic School, 121142. (Linguistic and Literary Studies in Eastern Europe 12.). Amsterdam: Benjamins. First published in Czech, Prague
[1929].
Morpurgo Davies, Anna
1998
History of Linguistics. Vol. 4: Nineteenth-Century Linguistics. Ed. by Giulio
Lepschy. (Longman Linguistics Library.) London/New York: Longman. First
published in Italian, Bologna [1992].
Musolff, Andreas
1990
Kommunikative Kreativitt: Karl Bhlers Zweifelderlehre als Ansatz zu einer
Theorie innovativen Sprachgebrauchs. Aachen: Alano.
Musolff, Andreas
1993
Karl Bhlers and Alan Gardiners concepts of metaphor in the context of their
theories of speech and language. Beitrge zur Geschichte der Sprachwissenschaft 3: 255272.
Musolff, Andreas
1997
Karl Bhler. In: Jef Verschueren, Jan-Ola stman, Jan Blommaert, Chris Bulcaen (eds.), Handbook of Pragmatics 1997, 115. Amsterdam: Benjamins.
Nekula, Marek
1999
Vilm Mathesius. In: Jef Verschueren, Jan-Ola stman, Jan Blommaert, Chris
Bulcaen (eds.), Handbook of Pragmatics 1999, 114. Amsterdam: Benjamins.
Nekula, Marek, and Klaas-Hinrich Ehlers
1996
Jakobsons Bilanz des Strukturalismus: Ein Vortrag, der jahrzehntelang nicht
gedruckt werden konnte. Beitrge zur Geschichte der Sprachwissenschaft 6:
187199.
Nerlich, Brigitte
1990
Change in Language: Whitney, Bral, and Wegener. (Routledge History of
Linguistic Thought Series.) London/New York: Routledge.
Nerlich, Brigitte
1992
Semantic Theories in Europe 18301930: From Etymology to Contextuality.
(Studies in the History of the Language Sciences 59.) Amsterdam: Benjamins.
Nerlich, Brigitte and David D. Clarke
1996
Language, Action, and Context: The Early History of Pragmatics in Europe
and America, 17801930. Amsterdam: Benjamins.
Ortner, Hanspeter
1983
Karl Bhlers Stellung in der Ellipsenforschung. Zeitschrift fr Germanistische
Linguistik 11: 149165.
Ortner, Hanspeter
1986
Bhlers Vierfelderschema (Das dritte Axiom der Sprachtheorie) Grundgedanken und Rezeption. Kodikas/Code, Ars Semeiotica 9: 211226.

258

Saskia Daalder and Andreas Musolff

Prague Linguistic Circle


1983
Theses presented to the first Congress of Slavists held in Prague in 1929.
Transl. from Czech by Josef Vachek. In: Josef Vachek and Libuse Duskov
(eds.), Praguiana: Some basic and less known Aspects of the Prague Linguistic
School, 77120. (Linguistic and Literary Studies in Eastern Europe 12.). Amsterdam: Benjamins. First distributed as loose sheets in Czech, Prague [1929].
Rudy, Stephen
1978
Jakobsonian Poetics of the Moscow and Prague Periods. Doctoral dissertation
Yale University.
Saussure, Ferdinand de
1931
Grundfragen der allgemeinen Sprachwissenschaft. Berlin: W. de Gruyter.
Transl. by Hermann Lommel of: Cours de linguistique gnrale, ed. by
Charles Bally and Albert Sechehaye, 1916.
Scheerer, Thomas M.
1980
Ferdinand de Saussure: Rezeption und Kritik. (Ertrge der Forschung 133.)
Darmstadt: Wissenschaftliche Buchgesellschaft.
Schiffrin, Deborah
1994
Approaches to Discourse. Oxford: Blackwell.
Schlieben-Lange, Brigitte
1979
Linguistische Pragmatik. Stuttgart: W. Kohlhammer.
Shannon, Claude E. and Warren Weaver
1949
The Mathematical Theory of Communication. Urbana: University of Illinois
Press.
Swiggers, Pierre
1981
Two dimensions in Karl Bhlers sign theory. Kodikas/Code, Ars Semeiotica 4:
5356.
Tenchini, Maria Paola
2008
Aspetti funzionali e pragmatici nel pensiero di Philipp Wegener. Con la traduzione antologica di Philipp Wegener, Untersuchungen ber die Grundfragen
des Sprachlebens. (Quaderni del Centro di Linguistica dellUniversit Cattolica 11.) Brescia: Editrice La Scuola.
Toman, Jindrich
1995
The Magic of a Common Language: Jakobson, Mathesius, Trubetzkoy, and the
Prague Linguistic Circle. Cambridge, MA: MIT Press.
Trnka, Bohumil
1966
Vilm Mathesius. Transl. by Vladimir Honsa. In: Thomas A. Sebeok (ed.),
Portraits of Linguists: A Biographical Source Book for the History of Western
Linguistics, 17461963, vol. 2, 474489. Bloomington: Indiana University
Press. First published in Czech, Prague [1946].
Tynjanov, Jurij, and Roman Jakobson
1972
Probleme der Literatur- und Sprachforschung. Translated by Rolf Fieguth und
Inge Paulmann. In: Texte der russischen Formalisten, vol. 2: 386391. First
published [1928].
Ungeheuer, Gerold
1987
Kommunikationstheoretische Schriften, vol. 1: Sprechen, mitteilen, verstehen.
Ed. by Johannes G. Juchem. Aachen: Rader Verlag.

Foundations of pragmatics in functional linguistics

259

Ungeheuer, Gerold
1990
Kommunikationstheoretische Schriften, vol. 2: Symbolische Erkenntnis und
Kommunikation. Ed. by H. Walter Schmitz. Aachen: Alano Verlag.
Vachek, Josef
1982
Vilm Mathesius living heritage to world linguistics. Philologica Pragensia
25:121127.
Vonk, Frank
1992
Gestaltprinzip und abstraktive Relevanz: Eine wissenschaftshistorische Untersuchung zur Sprachaxiomatik Karl Bhlers. Mnster: Nodus.
Waugh, Linda R.
1976
Roman Jakobsons Science of Language. Lisse: Peter de Ridder Press.
Waugh, Linda R., and Monique Monville-Burston
1990
Introduction: the life, work, and influence of Roman Jakobson. In: Roman
Jakobson, On Language. Ed. by Linda R. Waugh and Monique MonvilleBurston, 145. Cambridge, MA: Harvard University Press.
Waugh, Linda R., and Stephen Rudy
1998
Jakobson and structuralism. In: Roland Posner, Klaus Robering and Thomas
A. Sebeok (eds.), Semiotik: Ein Handbuch zu den zeichentheoretischen Grundlagen von Natur und Kultur, vol. 2, 22562271. (Handbcher zur Sprach- und
Kommunikationswissenschaft 13.2.). Berlin/New York: de Gruyter.
Wegener, Philipp
1902
Review of Berthold Delbrck, Grundfragen der Sprachforschung (Strassburg:
Trbner, 1901). Literarisches Centralblatt 12 (22. Mrz 1902): 401410.
Wegener, Philipp
1911
Exposition und Mitteilung: Ein Beitrag zu den Grundfragen des Sprachlebens.
In: Festschrift zur Feier des 350 jhrigen Bestehens des Gymnasiums zu
Greifswald, Bd. 2: Wissenschaftliche Aufstze, 121. Greifswald: Kunike.
Wegener, Philipp
1991
Untersuchungen ber die Grundfragen des Sprachlebens. New edition by
Konrad Koerner, with an introduction by Clemens Knobloch. (Classics in Psycholinguistics 5.) Amsterdam: Benjamins. First published Halle: Niemeyer,
[1885].
Weil, Henri
1887
The Order of Words in the Ancient Languages compared with that of the
Modern Languages. Transl. by Charles W. Super. Boston: Ginn. First published in French, Paris: Joubert [1844].
Weinreich, Uriel, William Labov and Marvin I. Herzog
1968
Empirical foundations for a theory of language change. In: Winfred P. Lehmann and Yakov Malkiel (eds.), Directions for Historical Linguistics: A Symposium, 95195. Austin/London: University of Texas Press.
Weissenborn, Jrgen
1988
Von der demonstratio ad oculos zur Deixis am Phantasma: Die Entwicklung
der lokalen Referenz bei Kindern. In: Achim Eschbach (ed.), Karl Bhlers
Theory of Language, 257276. Amsterdam: Benjamins.
Wettersten, John
1988
Klpe, Bhler, Popper. In: Achim Eschbach (ed.), Karl Bhlers Theory of
Language, 327347. Amsterdam: Benjamins.

260

Saskia Daalder and Andreas Musolff

Wolf, Kthe
1932
Darstellungsfelder in der Sprache. In: Gustav Kafka (ed.), Bericht ber den
12. Kongre der Deutschen Gesellschaft fr Psychologie (Hamburg 1931),
449453. Jena: G. Fischer.
Wunderlich, Dieter
1969
Karl Bhlers Grundprinzipien der Sprachtheorie. Muttersprache 79: 5262.
Wundt, Wilhem
1900
Die Sprache. 2 Parts (= Vol. 1 of Vlkerpsychologie). Leipzig: Engelmann.
Wundt, Wilhem
1908
Kritische Nachlese zur Ausfragemethode. Archiv fr die gesamte Psychologie
11: 445459.

9.

Foundations: ethnomethodology and


Erving Goffman
Christine Domke and Werner Holly

1.

Genesis and foundations of ethnomethodology

1.1.

Harold Garfinkels perspective on peoples everyday life activities

The most crucial assumption of ethnomethodology is that social reality exists only
through the everyday activities of a given societys members. Ethnomethodologists thus focus on the question of how social order arises in everyday life and is recognized as such. Answering this question entails identifying the actual practices
used by members of a society to create order. The goal of ethnomethodological
studies is to identify and reconstruct the methods used by individuals in everyday
life to produce meaningful and differentiable activities and to indicate how they interpret the activities of others.
Ethnomethodology was founded in the 1960s by Harold Garfinkel, who received his doctorate in 1952 for a dissertation entitled The perception of the other:
A study in social order under the direction of the sociologist Talcott Parsons (cf.
Heritage [1984] 1992; Schegloff 1992). While Garfinkel viewed Parsons studies
(alongside those of Alfred Schtz and Edmund Husserl) as crucial in the development of his own perspective on the world of everyday activities (Garfinkel
[1967] 1996: ix), he also insisted on distancing himself from his former mentors
theoretical assumptions about the crucial question of how social order arises. In
Parsons structural functionalism, social order is made possible by the orientation
of the individual to intersubjectively valid norms. According to Parsons, these cultural values and rules serve as a framework for the actions of the individual, who
perceives and interprets social reality on the basis of these shared norms. Garfinkel, however, rejected the notion that the individual is geared to given norms and
asked instead how he goes about interpreting situations. Garfinkel thus shifted the
focal point away from collectively held norms as constitutive of social order to
identify the continuous interpretation and attribution of meaning as central to the
genesis of social order.
Hence in his own analysis of social order, Garfinkel concentrated on the assessments made by ordinary people and the activities they produce which can be perceived by others (Heritage 1992: 736). Garfinkel understood ethnomethodology
in contrast to the then prevalent assumption of an objective reality. In the preface
to his book Studies in Ethnomethodology (1996), he remarks that the objective
reality of social facts should not be seen as given, but rather as the result of an ongoing accomplishment of the members of a society. Social order thus arises out of

262

Christine Domke and Werner Holly

the interactive activities of the participants. These activities also serve as accounts to explain how the activities of the participants should be understood.
On the basis on these reflections, everyday life became a legitimate subject of
sociological analyses. Moreover, the new focus directed attention at examining
how the order of everyday life, including the world of work, is organized. For as
Garfinkel (1996: 3575) pointed out in the late 1960s, although the familiar common sense world of everyday life occupied a central place as a topic of sociological inquiry, most scholarly literature included little data and few methods which
could be used to investigate the features of familiar scenes. His main concern
was thus to affirm and to (re)discover common sense activities in a societys everyday life as a relevant object of study.
1.2.

Early ethnomethodological investigations

One of the first ethnomethodological studies (see e.g. Turner 1974b; Sudnow
1972) examines the activities which jurors as non-professional adjudicators engage in so that others will recognize them as jurors and take their actions seriously
(Garfinkel 1974: 16). Using tape recordings and interviews with jurors as a basis,
this study focuses on becoming a juror (Garfinkel 1996: 110). This process includes, for example, using rules of the official line instead of the rules of daily
life (Garfinkel 1996: 110113). When describing their actions, the jurors present
their decisions in correspondence with the official rules as right, as based on law
and evidence, not personal preferences or interests. Moreover, they justify their
decisions by stressing the integrative features of the deliberations, thereby
neglecting the anomic or random ones (Garfinkel 1996: 113). These examples
demonstrate how a continuous perspective, that of the jurors, is successively generated.
Another early study about suicidal deaths examines the question of how the
involved parties, in this case coroners, arrive at the diagnosis suicide in cases of
sudden, unnatural death (Garfinkel 1996: 11). This work analyzes the process of
inquiry through which investigators establish what really happened as they provide an account of how a particular person died (Garfinkel 1996: 15). Ethnomethodological studies are thus concerned with how activities of everyday life are organized and how this organization is accomplished through peoples everyday life
activities.
The approach as well as the name ethnomethodology can be traced back to the
study on jurors (see above) which Garfinkel carried out together with Saul Mendlovitz in Chicago during the 1950s before going on to the University of California
at Los Angeles (UCLA) in 1954. When he wrote up the study about the tapes and
interviews, he was looking for a suitable name for the doing methodology as observed in the jurors (Garfinkel 1974: 16). The prefix ethno- is intended to refer to
a societys body of general knowledge that serves the individual as a point of orien-

Foundations: ethnomethodology and Erving Goffman

263

tation. Hence the term ethnomethodology designates the procedures and practices,
that is, ethno-methods, which are valid for the members within a particular society
and which produce structure. While the approach of ethnomethodology is analogous to, for example, ethnobotany, it is not intended to stand for a specific knowledge domain such as, for example, ethno-physics (Garfinkel 1974: 1617). The
genesis of the term highlights one of the main thoughts behind this sociological research approach as laid out in Garfinkels 1967 volume Studies in Ethnomethodology: the search for the everyday methods through which a societys members
structure their activities and make them comprehensible and visible as such should
be oriented towards situations and conditions of everyday life (1996: vii). This direct orientation to the accomplishment of everyday activities was achieved with
the help of tape recordings and ethnographical observations from the field. Basing
their work on common sense activities and daily situations, Garfinkel and other
ethnomethodologists of his day vehemently denounced all forms of standardization of ethnomethodological research and stressed what they were not intending
to do: neither theories, nor advice, nor recipes, but rather the practices and features
of everyday life would be the focus of their investigations (Garfinkel 1996: viii; see
also Turner 1974a).
This avoidance of solely theoretical assumptions to focus on everyday activities represents one of Garfinkels central ideas, which can be linked to classical
American pragmatism: The relevance of real lived experiences, habitual practices,
and language with its indexically and socially constitutive characteristics had already been established in the work of Peirce, Mead, Dewey, and others. Seen from
this perspective, the questions arises as to whether certain elements of Garfinkel
and ethnomethodology can be seen as the fulfillment of the unfinished business
of pragmatism (Emirbayer and Maynard 2006).
In Studies in Ethnomethodology (1996) Garfinkel presents various studies
from the preceding twelve years, thereby demonstrating the possibilities for his
form of analysis of everyday activities. In addition to introductory articles and the
study of the everyday practices of jurors mentioned above, the volume also includes two of Garfinkels best-known areas of investigation: one about a psychiatric clinic and the methods used there to identify patients for treatment, and another known as Agnes. Agnes is a pseudonym for a young woman who was
born and raised as a male person (Garfinkel 1996: 120). At the age of 17, Agnes assumed a female identity by altering her appearance and dressing as a woman; three
years later she underwent a sex-change operation. Garfinkels study uses Agnes
story to illustrate that, contrary to commonly held notions about the fundamental
difference between the sexes, the perception of normal sexuality is actually the
result of a constant situational production which, in Agnes case, was constituted
by the daily management of herself as natural female (1996: 184).

264
1.3.

Christine Domke and Werner Holly

Basic assumptions of ethnomethodology: making activities accountable

While the establishment of a full-fledged research program was not a concern of


the first ethnomethodologists (Turner 1974a), several of their assumptions can be
seen as central.
As mentioned above, ethnomethodology assumes that the members of a society
provide accounts of their own actions. The main focus of ethnomethodological
studies is to show how people make their activities observable, reportable and visible so that social settings become accountable for their participants. Another basic
thought behind ethnomethology is that the actions themselves indicate how they
should be understood as they are being carried out. Because they constitute both a
part of the situation in progress and a means of interpreting it, these human actions
can be characterized by a basic reflexivity. Hence the context in which an activity
should be understood is locally produced in situ by the activities themselves. This
reflective character of the accounts makes it possible to assign sense and meaning
in spite of the indexicality of the expressions and the vagueness of language. Another central concept in the work of Garfinkel known as indexicality emphasizes
the contextuality of language. Because indexical expressions are so common and
in spite of the efforts of many scholars who are ironically mentioned by Garfinkel
irreplaceable, their properties provide important clues for understanding everyday
actions. Concerned only with objective expressions or context-free language, the
exact sciences describe indexical, that is, contextual expressions as simply
awkward (Garfinkel 1996: 5) in spite of their frequency of occurrence and their
utility; from this perspective, indexical expressions appear to be irremediable nuisances (Garfinkel 1996: 7). Garfinkel, however, is most interested in exploring the
rational properties (1996: 911) of indexical expressions and thus the following
central question: How is it possible to gain understanding in spite of the recognizable and irremediable contexuality of language and human action?
In order to illustrate this contextuality and reflexivity of language and human
activity, Garfinkel had his students conduct experiments in familiar scenes of
everyday affairs known as breaching experiments. He wanted his students to
examine ordinary conversations and episodes to test the degree to which everyday
events are characterized by seen but unnoticed features (Garfinkel 1996: 3649).
Garfinkel borrowed the term seen but unnoticed from the phenomenologist
Alfred Schtz, whose work about the world of everyday life influenced Garfinkel
profoundly (Heritage 1992: 3774). Schtz referred to the background expectancies which the members of a society use to understand and interpret human actions as the attitude of daily life. Garfinkels aim was thus to find out how commonplace scenes become familiar for the members of a society. The experiments
were designed to discover the mutual expectations and assumptions which people
share as a kind of basis. Garfinkels method was to examine what happens when
people depart from the usual everyday order by breaking social rules or norms, or

Foundations: ethnomethodology and Erving Goffman

265

in his words: to start with familiar scenes and ask what can be done to make
trouble (Garfinkel 1996: 37). Garfinkels students were asked to engage someone
in an ordinary conversation and to then respond to indexical phrases which are
normally not further scrutinized by asking the person to clarify their remarks. For
example when asked How are you?, they would respond: How am I in regard to
what? My health, my finances, my school work []? (Garfinkel 1996: 44). Unaware of the experiment, most people reacted to this unexpected change in the normal course of conversation by showing confusion or annoyance or by changing the
subject. It also became clear that one explanation tended to result in another, indicating that indexicality cannot be suspended or healed. Reflexivity, which is, as
mentioned above, constitutive of all actions, normally prevents these kinds of deviations in everyday life and assures that the participants in a conversation alternately receive hints to help them deal with vagueness as a matter of course.
2.

Conversation Analysis

2.1.

Harvey Sacks Lectures on Conversation

These basic assumptions about the production of order through everyday activities
have also been fundamental for conversation analysis (CA), which grew out of ethnomethodology from the mid to late 1960s to became an independent area of research (see e.g. Sudnow 1972; Turner 1974b). Harvey Sacks, who, like Garfinkel,
was at UCLA from 1963, is known as the founder of CA. Through his work, he
paved the way for a growing interest in everyday conversations as an object of
study. Garfinkel and Sacks were engaged in close scientific exchange since they
first met in a seminar by Talcott Parsons at Harvard in 1954 (Schegloff 1992: xiii;
see also Bergmann 2000a). Known as the cornerstone of CA, Lectures on Conversation is composed of lectures which Sacks held between 19641972. Following his early death in 1975, Sacks former student Gail Jefferson edited the lectures
(Sacks 1989, 1992 Vol. I & II). Sacks used recorded conversations (including telephone recordings from an emergency psychiatric hospital) as a basis for intensely
analysing the structural characteristics of verbal interactions. In the course of his
work, he introduced and advanced the position that with respect to both verbal and
nonverbal activities, the generation of social order in everyday life is a legitimate
subject of sociology (Sacks 1972). Topics covered in his extensive analyses include lies to be expected in conversations (e.g. as an answer to the question How
are you?, Sacks 1992, Part V: 8, 9), the techniques used to choose a conversations
next speaker, and the procedures used by members to categorize someone or an
event. These techniques are used automatically by the participants in a conversation and assure, as a rule, two things: that conversations proceed not chaotically,
but according to a certain order, and that each party is able to assign meaning to the
utterances of the other. These order providing mechanisms are the focus of CA.

266

Christine Domke and Werner Holly

Sacks dealt extensively with questions about the everyday use of categories of
membership (Sacks 1992, inter alii Part I: 6, Part V: 11, 12). In his view, the procedures and devices which members use to categorize others constitute a central
machinery of social organization. Sacks developed an apparatus called MIR device (M for membership, I for inference-rich, R for representative) (1992:
41) which includes three categories of questions: questions about the chosen category sets (e.g. race, sex, age, occupation); questions about the conclusions which
are made possible by the categories used, e.g. baby and mommy in the wellknown example The baby cried. The mommy picked it up (see Sacks 1992,
Part III: 12; Sacks [1972] 1974); and questions about the activities ascribed to
particular members as category bound, e.g. crying to babies and picking up to
mothers.
2.2.

The sequential organization of verbal interaction

Sacks thus took what Garfinkel (and others) discussed about the relevance of
everyday activities for the production of social order and applied it to verbal interactions. CA empirically examines how order is generated in conversations. Proceeding from the basic principle of ongoing accomplishment (see above), the
question of the successive generation of various ordering structures comes into
focus. The processual creation of ordering structures during interactions is conceptualized as doing (see Sacks 1984 on doing being ordinary, Garfinkels analysis
of Agnes from 1996 on doing gender, Aya 2004 on doing interviewing).
A concentration on sequential organization constitutes a central and pivotal
point of analyses in the work of Sacks and within CA. The ever-present question
about the localisation of individual elements of conversation puts the focus on the
direct context of single utterances. By analysing the sequence of utterances generated locally within the situation, it is possible to determine what an utterance refers
to, what problem a particular turn solves, and which turn can be expected to follow.
Sequential organization is the core of the methodical approach of CA, which tries
to identify the processes which give order to conversation. In conjunction with
Emanuel Schegloff, a former fellow student who later became a colleague, and
Gail Jefferson, a former student, Sacks developed various research methods for
studying sequential organization. These methods constitute the centrepiece of CA
and prominently demonstrate how CA examines interactions within the context of
their own order, which is produced turn by turn. These assumptions about the selfreferential and reflexive processes involved in the production of a conversation
make it possible to describe the autonomous nature of interactions and to designate
them as systems (see below).
The first basic structure describes how people take turns in ordinary discourse
according to a turn-taking-machinery which functions automatically during conversation. Analysing this machinery indicates how the simplest systematic for

Foundations: ethnomethodology and Erving Goffman

267

the organization of turn-taking in conversation functions and how the latent problem of choosing the next speaker at a transition relevance place can be solved,
e.g. through self-selection or other selection (Sacks, Schegloff, and Jefferson
[1974] 1978). The local, sequential organization of a conversation underscores
Garfinkels assumptions about the reflexivity of all actions in everyday life: understanding in verbal interaction is possible because the utterance of one speaker reflects and makes accountable his understanding of what was said before. One of the
central methods used by speakers to organize their conversation in the course of
conversation are so-called utterance pairs or adjacency pairs (Schegloff and
Sacks 1973; Schegloff 1972). Because some utterances provoke certain responses,
they are said to operate according to a property called conditional relevance
(inter alii Schegloff 1968), which compels the next speaker to respond in an expectable way in terms of both content and form (as a rule, questions are followed
by answers and greetings are followed by return greetings). Understanding is made
possible not only in the sequence of utterances and the structural order (question/
answer, accusation/defence), but also by virtue of the fact that utterances are retrospectively repaired. The sequential repair of utterances constitutes a special feature
of conversation: an utterance can be readily revised or repaired with respect to
formulations or details in a second turn. As shown in various studies (Schegloff,
Jefferson, and Sacks 1977; Schegloff 1979; Jefferson 1974), repairs can be differentiated according to who initiates them and who actually carries them out. The
analyses indicated that participants prefer self-initiated self-repair (as opposed to
other-repair), which means that the speaker of the utterance requiring repair makes
the correction without prompting from another. This idea has also been developed
in further studies of conversation under the name of a general preference organization and can be described with reference to common adjacency pairs such as
agreement/disagreement (inter alii Pomerantz 1984). In addition to these three central areas (turn-taking, adjacency pairs, repairs), a number of structural characteristics and specific tasks of verbal interaction (such as recipient design, laughter,
opening and ending of conversation, stories) as well as conversation types (everyday conversations, consultations, trouble talks, e.g. Jefferson 1984) have been analyzed since the beginnings of CA (see Schenkein 1978; Psathas 1979; Atkinson
and Heritage 1984; Hutchby and Wooffitt 1998). The initial work of Sacks, Sudnow, Schegloff, and Jefferson in Berkeley and at UCLA placed a sustained focus
on the regularities of everyday speech (also in working life) as well as on the
(ethno)-methods by means of which this verbal order is generated.
2.3.

The analysis of empirical data

Because talk-in-interaction (Schegloff 1987) had not been the subject of such
minute analysis up until this point, another achievement of the first CAs was to develop and promote appropriate methods for archiving and preserving conver-

268

Christine Domke and Werner Holly

sations (Bergmann 1981, 1985). Using recordings of natural data (initially telephone conversations) as a basis, transcripts were made which attempted to record
entire exchanges word for word. Hence word interruptions, reformulations and
ums were not edited out, but rather became a part of the transcript. Both the recordings and the transcripts provide the data used in the analysis of authentic conversations (Hutchby and Woofitt 1998; ten Have 2007). Both then and now, these
natural data stand in strong contrast to data which have been contrived or generated
in experimental settings (inter alii Heritage 2001). With the help of sound recordings and, more recently, video recordings (on this topic see Heath and Luff 2000;
Goodwin 2000), CA preserves interactions by recording them for future use. These
data transcripts then form the basis for exploring the central issue of ordering structures. Seen from this perspective, reconstructive methods (e.g. narrative interviews, memorized observations) commonly used prior to CA (and even today) do
not constitute data which have been recorded verbatim from an actual situation.
Already tainted by the perspective of the one reporting it, data collected in this way
cannot provide a basis for analyzing the genuine course of interactions (Bergmann
1985).
The strictly empirical approach of CA starts from the data themselves
(Schegloff and Sacks 1973: 291) in order to reconstruct the mechanisms und
methods through which the participants of verbal interaction generate turn-by-turn
order. This focus on the course of the interactions themselves disaffirms the practice of arbitrarily applying a methodological or normative apparatus which is always supposed to be universally valid. Just as ethnomethodologists reject theoretical claims and programmatic assertions (see Turner 1974b; Garfinkel 1996), there
is a consensus within CA about the inappropriateness of these kinds of standardizations. In the view of CA, verbal interactions should be studied using methods
that arise out of the analysis of the data: to use hypotheses or theoretical links as a
starting point would be to narrow ones perspective on the order of a given set of
interactions. As it seeks to identify the specific features of a particular interaction
using turn-by-turn analysis of the data, this position can also be understood as a
kind of analytic mentality (Schenkein 1978: 6).
This absolute concentration on the order of interaction as recorded by the transcript (which now includes nonverbal elements as well) emphasizes the special
empiric understanding of CA and its perspective on interactions. The fact that the
methods of the participants are concretely reconstructed from the data themselves not only makes it possible to study authentic processes by reconstructing
the creation of order and the sequential structure of talk turn by turn. This method
of analysis with its step by step procedure also highlights the autonomy of interaction. In other words, the particular empiric understanding points to certain assumptions about the object, the interaction itself. With each successive turn, the order of
interaction shows which options emerge for the participants, what the interaction
requires, and what interactively appears as relevant. From this perspective, self-

Foundations: ethnomethodology and Erving Goffman

269

referential and reflexive processes of interaction move to the centre of CA. This
focus on the autonomy of interaction makes it possible to connect CA with more
recent system theory (Luhmann 1984, 1997), which presupposes the self-referential and autopoietic organization of a system (see Hausendorf 1992; Schneider
1997; Domke 2006). These interconnections correspond to CAs fundamental interest in interaction and open up empirical links to system theorys strictly theoretically based thoughts on the formation of systems.
One of the leading achievements and characteristics of CA is the way in which
it unlike mechanistic communication models points out the internal dynamics
of interactions. However, this focus on interaction as an autonomous system also
comprises a great challenge: the context of the interaction and the participants
must be taken into account appropriately and methodically examined. It is in these
points that CA differs from Erving Goffmans perspective on interactions.

3.

Erving Goffmans Interaction Order

Alongside Garfinkels ethnomethodology, Erving Goffmans interaction analysis


paved the way for numerous studies on the subject of everyday interactions and
analyses of how social order is generated by these interactions. Goffman taught at
the University of California at Berkeley from 19571968 before taking a position at
the University of Pennsylvania. His first book, The Presentation of Self in Everyday
Life was published in 1959 and is considered to be one of his best-known works.
Goffman had a great influence on Harvey Sacks, who studied with Goffman in Berkeley before going on to UCLA. In his introductions to Sacks lectures, Schegloff
emphasises that both sociologists influenced each other, an aspect which needs to
be explored more closely (Schegloff 1992: xxii). The special nature of their relationship is underlined by Goffmans reported answer to the question of whether
Sacks studied under him: What do you mean; I was his student! (Schegloff 1992:
xxiii). In spite of numerous ties and points of common ground (Drew and Wootton
1988; Widmer 1991) Goffmans work can neither be subsumed under the name of
ethnomethodology nor that of CA, but rather occupies an independent area of empirical social research with its own form of analysing everyday activities. However,
this does not mean that a Goffman School or methodology actually developed
out of his studies. It was primarily by pointing out the crucial role and function of
the interaction order as an institution in its own right (Goffman 1983b) that Goffman came to have such an important influence on conversation analysts like Sacks
and Schegloff (Heritage 2001; see also Bergmann 1991, 2000b).
According to Goffman, the features of traffic rules for pedestrians can be
studied in a crowded kitchen as well as anywhere else, and distinguishing between public and private settings is not crucial for understanding social interactions (Goffman 1983b). In his view, solely an understanding of face-to-face inter-

270

Christine Domke and Werner Holly

action as an independent domain with independent structures is necessary for the


analysis of everyday situations in the presence of others. However, this does not
mean in reverse that it can be effective to analyse the structure and rules of conversations out of context. On the contrary: in his studies on interaction order (initially
Interaction Ritual, 1967), Goffman focuses on the rituals and processes (such as
deference and demeanor, see also 1967) which participants adhere to in their
everyday presentations of self in order to create face and engage in moral enterprises (Drew and Wootton 1988: 67). From this perspective, the importance of
interactions for the formation of socially relevant structures (such as face, role) becomes clear. In order to differentiate social situations, Goffman uses, among
others, the term encounter to describe participants engaged in copresence-based,
mutually ratifying interaction, and the term gathering for unfocused situations
(such as waiting for a train) (Goffman 1964, 1983b).
Goffmans analyses work under the assumption that the individual self is created and shaped in every everyday face-to-face interactions. In his early essays
On Cooling the Mark Out (Goffman 1952) and On Face-work (Goffman
1955), Goffman introduces the concept of face-work as an essential process of
presenting the self and interacting with others. He developed the concept further in
later studies. For example in Stigma (1963b), he explores how persons who deviate
from normality (such as the handicapped or homosexuals) try to come to terms
with their threatened identity in everyday life. The exclusion of the individual in
public places, e.g. via the claiming of territories (such as seats or phone boxes) is
the focus of Behaviour in Public Places (Goffman 1963a). Goffman uses central
concepts from the theater to describe the production of self (Goffman 1959). Like
actors in a play, people constantly engage in impression management in order to
embody a certain role for their audience. This everyday dramaturgy includes
various methods and attributes such as the front stage, where the performers play
their roles in full view of the audience, and the backstage, a less accessible area
which is not available for public viewing.
The key points outlined above indicate a fundamental difference between Goffmans ideas and the core of CA. In CA, the focus is on the sequential structure of
interaction and the empirical analysis of interaction structures. This approach
brings the autonomy of interaction to the fore, which allows for an understanding
of interactions as systems as described in 2.3. Goffman, by contrast, is more interested in the individual and the effects of situations on his activities. The pertinent
concepts of face and ritual put the focus on the individual in relation to his surroundings. For Goffman, interaction within the structures of these surroundings is
less relevant than the individual participant in his emerging social identity. Hence,
by looking at the pertinence of interaction for both CA and Goffman, significant
differences can be identified. In light of the mutual influence between CA and
Goffman, these differences have prompted numerous attempts to explain and distinguish the two approaches (see Bergmann 1991).

Foundations: ethnomethodology and Erving Goffman

271

Goffman addresses linguistic systems and dialogic structures more concretely


(see Bergmann 1991) in his later works. The essays collected in Forms of Talk
(Goffman 1981) identify the various possibilities for differentiating social situations and encounters and discuss the role of the participants. Of particular relevance are his thoughts on participation status and participation framework (Goffman 1981: 137), concepts which stand in contrast to the rigid ascriptions of a
speaker/hearer model (see Auer 1999: 155163; Bergmann 1991: 314317). In one
of the essays entitled Footing, which first appeared in 1979 in the magazine Semiotica, Goffman describes a range of different types of speakers and hearers
(Goffman 1981: 124157). The participant who is engaged in talking is at that moment merely the sounding box in use (Goffman 1981: 144); he or she functions
as an animator. The animator and the author of what is said are not necessarily
identical and must be differentiated. A third function is that of the principal, the
one who is responsible for what is said. Taken together, these ascriptions reveal
what Goffman refers to as the production format of an utterance (Goffman 1981:
145). Hearers can also be differentiated into different modes (Goffman 1981:
131137). There are two kinds of ratified hearers: those who are directly addressed
by the speaker and those who are unaddressed. Non-ratified participants may be
listening intentionally, which Goffman calls eavesdropping, or unintentionally,
which he refers to as overhearing. All of these elements constitute the footing of an
interaction. Displayed interactively, footing is characterized by constant change.
The footing of an interaction also acts as a kind of orientation aid similar to the
more general frame of events, a concept which Goffman developed in Frame
Analysis (Goffman 1974) with reference to the earlier work of Bateson. For Goffman, frames involves definitions of situations and cognitive structures which
guide participants and indicate what might be expected of them. Frame analysis
(e.g. Goffman 1974) includes Goffmans analyses of response cries, a term which
was also the title of an article (Goffman 1978) later published in Forms of Talk.
Speakers produce exclamatory utterances like Oops! or Brr! in everyday life
as self talk without directly addressing these interjections to a particular person.
Using his distinction between social situation and encounter (see above), Goffman
describes these response cries as creatures of social situation, not states of talk
(Goffman 1981: 121) and thus identifies them as conventionalised expressions and
not as primitive, unsocialised (Goffman 1981: 120. The open state of talk is yet
another of Goffmans important designations which is useful for describing certain
communicative situations. An open state designates a conversation in which the
participants have the right but not the obligation to talk, as is the case when individuals are involved in a joint physical task or a common meal (Goffman 1981:
134135; also 1983a).

272
4.

Christine Domke and Werner Holly

Recent developments

Taking the fundamental studies described here as a starting point, the analysis of
verbal interaction has developed further in many respects. In the following, a few of
the most prominent out of a large number of works will be mentioned, starting with
the studies of work which grew out of ethnomethodology. In the introduction to a
volume of studies (1986) which he edited, Garfinkel emphasises that the analysis of
work processes (e.g. within a jazz ensemble studied by Harvey Sacks) makes it
clear that there exists a locally produced order of works things (Garfinkel 1986:
vii) which has largely been ignored. The specific features of individual occupational activities have become the focus of studies on everyday practices in the working world including both verbal and nonverbal activities (see Bergmann 2000b).
So-called workplace studies, which concentrate primarily on the relationship
between society and technology, have also been influenced by studies of work
(Knoblauch and Heath 1999). This area of research focuses on computer supported
work environments and the consequences of the mechanisation of various processes (inter alii Heath and Luff 2000). The next step will be to extend the methods
used in these studies to carry out large-scale document and video analyses; this endeavour will provide adequate material for future works of conversation analysis.
Other important studies which were motivated by conversation analysis analyse the characteristics of work realms or specific forms of interactions by focusing
on the verbal component (see Drew and Heritage 1992 on Talk at work; Clayman
and Heritage 2002 on News Interview). In more recent studies, this concentration on speech has given way to a more holistic approach (Dausendschn-Gay
and Krafft 2002), which tries to incorporate facial expressions, gestures and the environment of interactions into each analysis. Conversation analytical work with
video recordings has led to a discussion of classical CA focus areas (such as turntaking, conversational openings) and elaborated the interplay and necessary coordination between speech, gesture and objects for various areas including medicine,
everyday conversations and the media (inter alii Goodwin 2000; Aya 2004; Kissmann 2009; Schmitt 2007; Streeck 2009).

5.

Influence of conversation analysis on linguistic pragmatics

Conversation analysis is one of the two cornerstones of pragmalinguistics. In retrospect, it appears to be the less prevalent of the two, but in the long run it may be the
more sustainable. In simplistic terms, one could say that linguistic pragmatics has
essentially followed two major lines in its reorientation. On the one hand, it has
taken its impulse from the philosophy of language, incorporating and integrating
not only ordinary language philosophy along with the speech act theory of Austin
and Searle, but also the concept of meaning and understanding as explicated by

Foundations: ethnomethodology and Erving Goffman

273

their attributed intellectual father Ludwig Wittgenstein and further developed by


Grice. On the other hand, linguistic pragmatics has been influenced by sociological
impulses which were mediated chiefly through conversation analysis and its historical backdrop, including names ranging from Harold Garfinkel and Alfred
Schtz to Max Weber. In addition, the microsociological studies of Erving Goffman with their direct relevance to every kind of interaction have impacted the field
to a lesser degree.
No linguist working in the field of pragmatics has been able to avoid coming to
grips with speech act theory and its string of tradition because speech act is the
central concept that has defined not only the whole linguistic subdiscipline, but
also the refounding of the entire field of linguistics. The study of conversation
analysis (and especially Goffman), however, has been less obligatory for two reasons. First, because it applies only to verbal interaction, CA was thought to represent a limited subject area. Secondly, CA was in competition with other approaches which also attempted to analyze spoken interaction or which did not
make a rigorous distinction between (spoken) interactions and (written) texts. Another possible explanation for the resistance to CA may be the challenges it posed
to adherents of speech act theory from the philosophy of language who were accustomed to working with constructed and isolated example sentences. The consistent empirical approach to collecting data and diligently transcribing them
required a fundamentally different kind of methodology than was required by
grammar-oriented research. Finally, many were put off by what they saw as a kind
of unorthodox leftist movement, a reputation which CA earned with its wild
1960s California image, its antitheoretical mistrust of academic omniscience, and
its uniquely different perspective: the adherents of CA were searching for the
methods of the members of society themselves. Others, however, were attracted
by this unconventionality. For example, Bergmann (1981: 38) admires das Unsystematische und Unabgeschlossene, das Ungeschtzte und Riskant-Artistische, das
Fragmentarische, Labyrinthhafte und Antiakademische [the unsystematic and inconclusive, the unprotected and risky-artistic, the fragmentary, labyrinthian and
anti-academic]. In any case, it cannot be denied that the actual subject matter of CA
was paradoxically just what the more theoretical and less disputed branch of pragmatics actually had in mind without analyzing it empirically: ordinary language.
While CA researchers repeatedly denied having a primary or special interest in language and conversation (Schegloff and Sacks 1973, cited in Coulthard 1977: 52;
see also Sacks comment as quoted in Auer 1999: 137), they nonetheless were
compelled to and wanted to attract interest by designating themselves as conversation analysts, especially in light of the new pragmatic orientation of the field
of linguistics and the relevance and central importance of their subject matter, as
highlighted here by Levinson: It is not hard to see why one should look to conversation for insight into pragmatic phenomena, for conversation is clearly the
prototypical kind of language usage [] (1983: 284).

274

Christine Domke and Werner Holly

The gradual triumph of CA, particularly in the field of linguistics, was thus inevitable to a certain extent. The appearance of important studies by sociological researchers in linguistic journals (Sacks, Schegloff, and Jefferson 1978; Schegloff,
Jefferson, and Sacks 1977 in Language) or in a socioliguistic context (e.g. Sacks
1974, first in Gumperz and Hymes 1972) was no accident. The proximity and
kinship to newly developing sociolinguistic approaches (e.g. the ethnography of
speaking) created a situation in which the various communication-based approaches benefited each other in an area of overlap between sociology, anthropology, and linguistics; this interconnectedness also strengthened interest in works on
conversation analysis. Recognition of CA expanded beyond a small circle of interested scholars to the entire (Anglo-Saxon) field of linguistics after the publication
of the two chapters The ethnography of speaking and Conversation analysis
(following the chapter entitled Speech acts) in An Introduction to Discourse
Analysis by Malcolm Coulthard (1977). Coulthard founded the Birmingham
School together with John Sinclair on the basis of their study of classroom interaction (Sinclair and Coulthard 1975).
Within the field of German linguistics, two research structures in Bielefeld and
Freiburg were primarily responsible for the spread of studies in conversation
analysis. As early as 1973, researchers in Bielefeld published two volumes of texts
by American sociologists, including studies by Garfinkel, Sacks und Psathas on
ethnomethodology and by Hymes on the ethnography of speaking (Arbeitsgruppe
Bielefelder Soziologen 1973). The work of the group in Bielefeld also led to the
publication of a highly regarded article by the linguist Werner Kallmeyer and the
sociologist Fritz Schtze on Conversation Analysis (Kallmeyer and Schtze
1976). The Freiburg project Dialogstrukturen [dialogue structures] from 1974
under the direction of Hugo Steger investigated spoken language, thereby incorporating CA research in its investigation of spoken language (Berens et al. 1976; for
details see Schwitalla 2001a). Thanks to this early engagement in the field, the
1980 Annual Conference of the Institute for German Language on the topic of dialogue research included an overview article on ethnomethodological conversation
analysis (Bergmann 1981). This topic was also covered in later publications, including German guides to dialogue analysis and conversation linguistics (Bergmann 1994, 2001) as well as an introduction to linguistic interaction by Auer
(1999) on the basis of several classic authors in the field.
The research paradigm conversation analysis was effectively linguistically
canonized in the 1980s with the publication of the chapter on conversational structure in Pragmatics by Stephen Levinson (1983), the internationally best-known
book on the topic. Levinson deals extensively with the methods and results of CA
in this chapter, and similar introductions published later follow his example (e.g.
Green 1989; Mey 1993; Grundy 1995; Meibauer 1999). Levinsons work also
served as a model for other writers because he discusses CA in contrast to other approaches, which are combined under the label discourse analysis (DA).

Foundations: ethnomethodology and Erving Goffman

6.

275

Conversation analysis vs. other approaches in the linguistics


of spoken interaction

While Levinson advocates CA, others have taken a position between reservation
and rejection. As a whole, the field of discourse analysis is quite diverse, especially
since it sometimes includes analyses of both written texts (not considered here) and
larger units such as discourse (in the sense of Foucault). This discussion, however,
will continue to focus on spoken language. Alongside the previously mentioned
approaches (Dell Hymes ethnography of speaking, Sinclair and Coulthards Birmingham School), John Gumperz developed his interactional sociolinguistics;
Levinson also cites Van Dijk (1972) and Labov and Fanshel (1977) as typical
representatives of discourse analysis; Bublitz (1991) and Ventola (2001) provide a
general overview of conversation and discourse analysis, and Hausendorf discusses the situation in German-speaking regions (2001). Some approaches are
based more strongly on speech acts: in Germany, for example dialogue analysis
and dialogue grammar (e.g. Hundsnurscher 1995, 2001; Hindelang 1994). The
volume Handbuch der Dialoganalyse [Handbook of Dialogue Analysis] (Fritz and
Hundsnurscher 1994) includes a discussion of the Geneva model of discourse
analysis (Moeschler 1994, 2001) and an action theoretically based concept of
linguistic communication analysis (Gloning 1994), which is associated with
Heringers practical semantics. Konrad Ehlich and Jochen Rehbein take their
own separate direction with their functional pragmatics in Diskursanalyse
(Rehbein 2001). Yet another example of DA is the critical discourse analysis
(Kritische Diskursanalyse) of the Wiener Schule under Ruth Wodak (e.g.
Wodak et al. 1990), an approach which is based on Faircloughs Critical Discourse Analysis (CDA) (e.g. Fairclough 1995).
Writing from the perspective of CA, Levinson (1983: 286294) targets text
grammatical or speech act theoretical approaches of DA with his criticisms of
methods and theoretical tools (Levinson 1983: 294). Claiming that DA sometimes falls back on constructed examples, he criticizes a lack of empirical focus as
well as a fundamentally intuitive approach which ultimately cannot tolerate the
possibility of falsifiability. He also takes issue with DAs acceptance of the familiar
theoretical principles of well-formedness (see also Kohrt 1986: 7475), a viewpoint which sees discourses as nothing more than strings of interlinked sentences.
In the eyes of DAs critics, this oversimplifying grasp (Kohrt 1986: 80) can
easily lead to the construction of rigid or even normative structures (Bublitz 1991:
271). According to Levinson, the practice of assigning individual utterances to one
particular speech act and then constructing structurally definable sequence mechanisms (as in sentences) is bound to fail because it cannot do justice to the complex
reality of the unpredictable dynamic interactions which are themselves embedded
in complex contexts (Levinson 1983: 288294). Proponents of CA defend their
method by pointing out that it is strictly empirical and inductive, with as little ap-

276

Christine Domke and Werner Holly

peal as possible to intuitive judgements (Levinson 1983: 287), and that it does not
immediately categorize individual cases, but rather attempts to compile functionally similar cases from the data in order to prove the validity of an analysis (Bergmann 2001: 923). The most effective argument for CA is definitely its indisputable
success in the recognition of the structures which organize speech.
However, CA has also had its critics. For example, Levinson (1983: 287)
makes reference to Labov and Fanshel (1977: 25) and Coulthard and Brazil (1979)
as he points out: DA theorists can accuse CA practitioners of being inexplicit, or
worse, plain muddled, about the theories and conceptual categories they are actually employing in analysis. It is indeed questionable whether all existing order
can be deduced solely from the object of investigation. In other words, it is doubtful that utterances are each reflexive and indexical enough to be understood without being categorized by the interpreter from the outside. In fact, it is commonly
recognized that by drawing on his or her analytical mentality, the ethnomethodologist is forced to draw repeatedly on the intuitive understanding which he or she
possesses as a competent member of a linguistic community (Bergmann 1981: 23).
At the same time, this demonstrative refusal to draw on additional outside knowledge in an attempt to immunize CA against possible allegations of interpretive arbitrariness has not only led several critics to label the field as anti-psychological
and individual-psychological (Redder 1990: 7), but it has also caused some to
cross the line to (unjustifiably) accuse CA of secret positivism (Flader and Trotha 1988). A further criticism cites an alleged failure to take social background into
account (e.g. Fairclough 1989: 12; see also Bublitz 1991: 271), but this assessment
is not particularly convincing in view of CAs sociological and sociocritical origins. Moreover, more recently, a growing number of studies in CA have focused on
institutional and work contexts as well as political communication.
On the whole, the field of linguistic pragmatics has not really seen an intense
dispute over theoretical and methodological issues surrounding CA; the predominant pattern has been rather a more or less reflected and more or less selective appropriation of methods and results. Evidence of these integrative tendencies can be
found in many textbooks and introductions, but they arise above all out of the experiences and needs of concrete research. Hence at the end of the 1990s, Hausendorf (2001: 977) came to the following conclusion with respect to the situation in
the German-speaking realm: To the extent to which the methodological praxis of
conversation analysis consolidated itself and the findings and conclusions of empirically sophisticated i.e. complex and long-term projects were presented in
the course of the 1990s, the methodological struggle between the various approaches has faded away.

Foundations: ethnomethodology and Erving Goffman

7.

277

Influence of Erving Goffman on linguistic pragmatics

While CA quickly attained prominence within linguistic pragmatics on account of


its direct relevance for the study of spoken language, the linguistic importance of
Goffmans research initially appeared to be limited to marginal questions. However, his work has been widely recognized and discussed within many other disciplines because of its broad spectrum of topics (self-portrayal, asylums, interaction
rituals, role distance, gender advertising, frame analysis), so that gradually, his
studies have also entered the field of linguistics indirectly via other routes. At least
four strands of Goffmans work have been adopted by pragmatics in one form or
another over the course of time.
First, Goffmans concept of face was further developed by Brown and Levinson ([1978] 1987) into a politeness model with universal applicability. This model
was adopted in numerous studies on the topic of language and politeness, but it
was also vehemently criticized for its weak empirical basis of single utterances as
well as for the questionable way in which the model was universalized (for summary see Fraser 2001; Bublitz 2009: 270). However, the complexity of Goffmans
concept (see Tracy 1990) comprises at its core something that goes beyond mere
aspects of politeness to encompass the ritual negotiation of identities and relationships between interaction partners. This second Goffmanian strand was received
by conversation analysis quite early in a few instances (e.g. Holly 1979, 2001),
but it has taken more time to develop the concept fully for linguistics with respect
to routines (Coulmas 1981), rituals (e.g. Rauch 1992; Werlen 2001), conflict
settlement (e.g. Schank and Schwitalla 1987; Schwitalla 2001b) and further issues. The third important strand constitutes Goffmans frame analysis, which has
been applied more or less prominently in many pragmalinguistic studies (e.g.
Tannen 1984). Finally, the fourth strand involves Goffmans explicit investigations of conversations (Forms of Talk), in particular the deconstruction of
participant roles. This deconstructive aspect of Goffmans work was adopted by
Levinson (1988) and formed the basis of a new conceptualization of the idealized
and simplistic speaker-hearer-constellation (see Schwitalla 2001c), e.g. to analyse multiple addressing (Khn 1995) or testimonials in a television commercial
(Holly 2007). Moreover Goffmans analysis of different intensities of focus on
talk generated the useful concept of open state of talk, which is connected to
Bhlers observations on action-accompanying talk and is also essential for the
analysis of certain uses of speech (see e.g. on the reception of television texts
Holly, Pschel, and Bergmann 2001). The fact that four articles in the Handbook
of Linguistics of Text and Conversation focus primarily on Goffmanian concepts
(Werlen 2001; Schwitalla 2001b, 2001c; Holly 2001) is an index of how widely
Goffmans research has been received in the field of pragmalinguistics. Nonetheless, his importance for the analysis of spoken language continues to be underestimated in pragmalinguistics, probably because of his independence as a scholar,

278

Christine Domke and Werner Holly

which fostered a mentality against the formation of schools, and his conceptual
flexibility.

8.

The convergence of various approaches

Although ethnomethodological conversation analysis constitutes a markedly independent and distinctive school of thought in the study of spoken language, there is
a gradual tendency toward increasing convergence with other related approaches
which are anchored more firmly in the fields of anthropology, sociology or linguistics; Heritage (2001) combines the three under the label ethno-sciences.
These ethno-sciences share in common not only work with empirical data, but also
a series of core ideas which he characterizes in the following:
This research embodies the core notions that (i) communicative meaning is inherently
contextual in character, (ii) social context is unavoidably dynamic and is managed
through the participants actions, (iii) the specific contextual significance of actions is
structurally achieved by means of rules and practices of conduct which are systematically related and organized as systems, (iv) the contextual significance of action also involves a web of inferences which are inescapable, very often involve personal, moral
and social accountability, and thus connect interaction with culture, social structure and
personality, and (v) all this is managed through the integrated significance of talk, paralanguage and body movement. (Heritage 2001: 916)

With these kinds of common convictions as a basis, infighting, which also has seen
adherents of CA criticize a less than rigorous handling of methodological questions (Streeck 1983), could relax. At the same time, this convergence is being promoted in the field of linguistics by an increasing return to the original competencies, resulting in a common focus on linkages between grammatical and interactive
structures (Uhmann 1997; Ochs, Schegloff, and Thompson 1997), in particular
with reference to prosodic-intonational structures as foreseen in the concept of
interactional linguistics by Elisabeth Couper-Kuhlen und Margret Selting. This
approach sees itself as an interface between linguistics und interaction theory
(above all functional linguistics, conversation analysis, contextualization theory,
anthropological linguistics) (Selting and Couper-Kuhlen 2000: 76). Other efforts
concentrate more closely on the connection between grammar and interaction (e.g.
Deppermann, Fiehler, and Spranz-Fogasy 2006), so that the discipline of conversation linguistics reencounters many older orientations in the investigation of
spoken language (see Schwitalla 2001a), be it on the basis of Ehlich und Rehbeins
Functional Pragmatics which was originally intended to focus on grammar as
well, or on the basis of construction grammar (Gnthner and Imo 2006; Deppermann 2007).

Foundations: ethnomethodology and Erving Goffman

279

References
Arbeitsgruppe Bielefelder Soziologen (ed.)
1973
Alltagswissen, Interaktion und gesellschaftliche Wirklichkeit. 2 vols. Reinbek:
Rowohlt.
Atkinson, John Maxwell and John Heritage (eds.)
1984
Structures of Social Action: Studies in Conversation Analysis. Cambridge:
Cambridge University Press.
Auer, Peter
1999
Sprachliche Interaktion. Eine Einfhrung anhand von 22 Klassikern. Tbingen: Niemeyer.
Aya, Ruth
2004
Konversationsanalytische Medienforschung. Medien und Kommunikationswissenschaft, 52 (1): 529.
Berens, Franz-Josef, Karl-Heinz Jger, Gerd Schank and Johannes Schwitalla
1976
Projekt Dialogstrukturen. Ein Arbeitsbericht. Mnchen: Hueber.
Bergmann, Jrg R.
1981
Ethnomethodologische Konversationsanalyse. In: Peter Schrder and Hugo
Steger (eds.), Dialogforschung. (Jahrbuch 1980 des Instituts fr deutsche
Sprache), 953. Dsseldorf: Schwann.
Bergmann, Jrg R.
1985
Flchtigkeit und methodische Fixierung sozialer Wirklichkeit. Aufzeichnungen als Daten der interpretativen Soziologie. In: Wolfgang Bon and Heinz
Hartmann (eds.), Entzauberte Wissenschaft. Zur Relativitt und Geltung soziologischer Forschung, 299321. Gttingen: Schwartz.
Bergmann, Jrg R.
1991
ber Erving Goffmans Soziologie des Gesprchs und seine ambivalente Beziehung zur Konversationsanalyse. In: Robert Hettlage and Karl Lenz (eds.),
Erving Goffman ein soziologischer Klassiker der zweiten Generation, 301
326. Bern/Stuttgart: Haupt.
Bergmann, Jrg R.
1994
Ethnomethodologische Konversationsanalyse. In: Gerd Fritz and Franz
Hundsnurscher (eds.), Handbuch der Dialoganalyse, 316. Tbingen: Niemeyer.
Bergmann, Jrg R.
2000a
Harold Garfinkel und Harvey Sacks. In: Uwe Flick, Ernst von Kardorff and
Ines Steinke (eds.), Qualitative Forschung. Ein Handbuch, 5162. Reinbek:
Rowohlt.
Bergmann, Jrg R.
2000b
Ethnomethodologie. In: Uwe Flick, Ernst von Kardoff and Ines Steinke (eds.),
Qualitativeforschung. Ein Handbuch, 118135. Reinbek: Rowohlt.
Bergmann, Jrg R.
2001
Das Konzept der Konversationsanalyse. In: Brinker et al. (eds.), 919927.
Brinker, Klaus, Gerd Antos, Wolfgang Heinemann and Sven Frederik Sager (eds.)
2001
Text- und Gesprchslinguistik. 2nd vol. (Handbcher zur Sprach- und Kommunikationswissenschaft 16.) Berlin/New York: de Gruyter.

280

Christine Domke and Werner Holly

Brown, Penelope and Stephen C. Levinson


1987
Politeness. Some Universals in Language Usage. Cambridge: Cambridge University Press. First published in: Esther N. Goody (ed.), Questions and Politeness, 56289. Cambridge: Cambridge University Press 1978.
Bublitz, Wolfram
1991
Discourse Analysis: The State of the Art. In: Claus Uhlig and Rdiger
Zimmermann (eds.), Anglistentag 1990 Marburg. Proceedings, 259284. Tbingen: Niemeyer.
Bublitz, Wolfram
2009
Englische Pragmatik. Eine Einfhrung. 2nd ed. Berlin: Erich Schmidt Verlag.
Clayman, Steve and John Heritage
2002
The News Interview. Journalists and Public Figures on the Air. Cambridge:
Cambridge University Press.
Coulmas, Florian
1981
Routine im Gesprch. Zur pragmatischen Fundierung der Idiomatik. Wiesbaden: Athenaion.
Coulthard, Malcolm
1977
An Introduction to Discourse Analysis. London: Longman.
Coulthard, Malcolm and David Brazil
1979
Exchange Structure. Birmingham: Birmingham University.
Dausendschn-Gay, Ulrich and Ulrich Krafft
2002
Text und Krpergesten: Beobachtung zur holistischen Organisation der Kommunikation. Psychologie und Sozialwissenschaft. Zeitschrift fr Qualitative
Forschung. Themenheft: Sprechen vom Krper Sprechen mit dem Krper 4:
3060.
Deppermann, Arnulf
2007
Grammatik und Semantik aus gesprchsanalytischer Sicht. Berlin/New York:
de Gruyter.
Deppermann, Arnulf, Reinhard Fiehler and Thomas Spranz-Fogasy (eds.)
2006
Grammatik und Interaktion. Untersuchungen zum Zusammenhang zwischen
grammatischen Strukturen und Interaktionsprozessen. Radolfzell: Verlag fr
Gesprchsforschung.
Domke, Christine
2006
Besprechungen als organisationale Entscheidungskommunikation. Berlin/
New York: de Gruyter.
Drew, Paul and John Heritage (eds.)
1992
Talk at work: interaction in institutional settings. Cambridge: Cambridge University Press.
Drew, Paul and Anthony Wootton (eds.)
1988
Erving Goffman. Exploring the interaction order. Cambridge: Polity Press. Reprinted Cambridge: Polity Press [1996].
Emirbayer, Mustafa and Douglas Maynard
2006
Pragmatism and Ethnomethodology. Paper presented at the annual meeting of
the American Sociological Association, Montreal Convention Center. Montreal, Canada: Online <PDF>. 2009-05-24 from http://www.allacademic.com/
meta/p95497_index.html.

Foundations: ethnomethodology and Erving Goffman

281

Fairclough, Norman
1989
Language and Power. London: Addison Wesley.
Fairclough, Norman
1995
Critical Discourse Analysis: The critical study of language. London: Longman.
Flader, Dieter and Thilo von Trotha
1988
ber den geheimen Positivismus und andere Eigentmlichkeiten der ethnomethodologischen Konversationsanalyse. Zeitschrift fr Sprachwissenschaft 7:
92115.
Fraser, Bruce
2001
The form and function of politeness in conversation. In: Brinker et al. (eds.),
14061425.
Fritz, Gerd and Franz Hundsnurscher (eds.)
1994
Handbuch der Dialoganalyse. Tbingen: Niemeyer.
Garfinkel, Harold
1952
The Perception of the Other: A Study in Social Order. Ph.D. Dissertation. Boston: Harvard University.
Garfinkel, Harold
1974
The Origins of the Term Etnomethodology. In: Turner, Roy (ed.), Ethnomethodology. Selected Readings, 1519. Harmondsworth: Penguin Education.
Garfinkel, Harold (ed.)
1986
Ethnomethodological studies of work. London: Routledge and Kegan Paul.
Garfinkel, Harold
1996
Studies in ethnomethodology. Cambridge: Polity Press. First published Englewood Cliffs, NJ: Prentice Hall [1967].
Gloning, Thomas
1994
Praktische Semantik und Linguistische Kommunikationsanalyse. In: Fritz et
al. (eds.), 113129.
Goffman, Erving
1952
On Cooling the Mark Out: Some Aspects of Adaption to Failure. Psychiatry 15
(4), 451463. Reprinted in: Charles Lemert and Ann Branaman (eds.), The
Goffman Reader, 420. Oxford: Blackwell [1997].
Goffman, Erving
1955
On Face-work: An Analysis of Ritual Elements in Social Interaction. Psychiatry 18 (3), 213231.
Goffman, Erving
1959
The Presentation of Self in Everyday Life. Garden City, NY: Doubleday, Anchor Books.
Goffman, Erving
1963a
Behaviour in Public Places: Notes on the Social Organization of Gatherings.
New York: The Free Press.
Goffman, Erving
1963b
Stigma: Notes on the Management of Spoiled Identity. Engelwood Cliffs, NJ:
Prentice Hall.
Goffman, Erving
1964
The Neglected Situation. American Anthropologist 66: 133136. Reprinted in
Charles Lemert and Ann Branaman (eds.), The Goffman Reader, 229233. Oxford: Blackwell [1997].

282

Christine Domke and Werner Holly

Goffman, Erving
1967
Interactional Ritual: Essays on Face-to-Face Behaviour. Garden City, NY:
Doubleday, Anchor Books.
Goffman, Erving
1974
Frame Analyses. An Essay on the Organization of Experience. London: Harper
and Row.
Goffman, Erving
1978
Response Cries. Language 54 (4): 787815. Reprinted in Erving Goffman,
Forms of talk, 78122. Philadelphia: University of Pennsylvania [1981].
Goffman, Erving
1979
Footing. Semiotica 25 (1/2): 129. Reprinted in Erving Goffman, Forms of
talk, 124159. Philadelphia: University of Pennsylvania [1981].
Goffman, Erving
1981
Forms of talk. Philadelphia: University of Pennsylvania Press.
Goffman, Erving
1983a
Felicitys Condition. American Journal of Sociology 89 (1): 151. Reprinted in
Charles Lemert and Ann Branaman (eds.), The Goffman Reader, 167192. Oxford: Blackwell [1997].
Goffman, Erving
1983b
The Interaction Order. American Sociological Review 48: 117. Reprinted in
Charles Lemert and Ann Branaman (eds.), The Goffman Reader, 233261. Oxford: Blackwell [1997].
Goodwin, Charles
2000
Practices of seeing: visual analysis an ethnomethodological approach. In:
Theo van Leeuwen and Carey Jewitt (eds.), Handbook of visual analysis,
157182. London: Sage.
Green, Georgia M.
1989
Pragmatics and Natural Language Understanding. Hillsdale, NJ: Lawrence
Erlbaum.
Grundy, Peter
1995
Doing Pragmatics. London: Arnold.
Gumperz, John and Dell Hymes (eds.)
1972
Directions in Sociolinguistics: The Ethnography of Communication. New
York/London: Holt, Rinehart and Winston.
Gnthner, Susanne and Wolfgang Imo (eds.)
2006
Konstruktionen in der Interaktion. Berlin/New York: de Gruyter.
Hausendorf, Heiko
1992
Gesprch als System. Linguistische Aspekte einer Soziologie der Interaktion.
Opladen: Westdeutscher Verlag.
Hausendorf, Heiko
2001
Gesprchsanalyse im deutschsprachigen Raum. In: Brinker et al. (eds.), 971
979.
Have, Paul ten
2007
Doing Conversation Analysis. A Practical Guide. 2nd ed. Los Angeles: Sage.
Heath, Christian and Paul Luff
2000
Technology in action. Cambridge: Cambridge University Press.

Foundations: ethnomethodology and Erving Goffman

283

Heritage, John
1992
Garfinkel and ethnomethodology. Cambridge: Polity Press. First published
Cambridge: Polity Press [1984].
Heritage, John
2001
Ethno-sciences and their significance for conversation linguistics. In: Brinker
et al. (eds), 908919.
Hindelang, Gtz
1994
Sprechakttheoretische Dialoganalyse. In: Fritz et al. (eds.), 95112.
Holly, Werner
1979
Imagearbeit in Gesprchen. Zur linguistischen Beschreibung des Beziehungsaspekts. Tbingen: Niemeyer
Holly, Werner
2001
Beziehungsmanagement und Imagearbeit. In: Brinker et al. (eds.), 13821393.
Holly, Werner
2007
Audiovisuelle Hermeneutik. Am Beispiel des TV-Spots der Kampagne Du
bist Deutschland. In: Fritz Hermanns and Werner Holly (eds.), Linguistische
Hermeneutik, 389428. Tbingen: Niemeyer.
Holly, Werner, Ulrich Pschel and Jrg Bergmann (eds.)
2001
Der sprechende Zuschauer. Wie wir uns Fernsehen kommunikativ aneignen.
Wiesbaden: Westdeutscher Verlag.
Hundsnurscher, Franz
1995
Some Remarks on the Development of Dialogue Analysis. In: Franz Hundsnurscher and Edda Weigand (eds.), Future Perspectives of Dialogue Analysis,
7993. Tbingen, Niemeyer.
Hundsnurscher, Franz
2001
Das Konzept der Dialoggrammatik. In: Brinker et al. (eds.), 945952.
Hutchby, Ian and Robin Wooffitt
1998
Conversation Analysis: Principles, Practices and Applications. Cambridge:
Polity Press.
Jefferson, Gail
1974
Error corrections as an international resource. Language in Society 2:
181199.
Jefferson, Gail
1984
On the organization of laughter in talk about troubles. In: John Maxwell Atkinson and John Hertitage (eds.), Structures of social action: studies in conversation analysis, 346369. Cambridge: Cambridge University Press.
Kallmeyer, Werner and Fritz Schtze
1976
Konversationsanalyse. Studium Linguistik 1: 128.
Kissmann, Ulrike Tikvah (ed.)
2009
Video Interaction analysis. Methods and Methodology. Frankfurt am Main:
Lang
Knoblauch, Hubert and Christian Heath
1999
Technologie, Interaktion und Organisation: Die Workplace Studies. Schweizerische Zeitschrift fr Soziologie 25 (2): 163181.
Kohrt, Manfred
1986
Dialoggrammatik und/oder Konversationsanalyse. In: Franz Hundsnurscher
and Edda Weigand (eds.), Dialoganalyse. Referate der 1. Arbeitstagung
Mnster 1986, 6982. Tbingen: Niemeyer.

284

Christine Domke and Werner Holly

Khn, Peter
1995
Mehrfachadressierung. Untersuchungen zur adressatenspezifischen Polyvalenz sprachlichen Handelns. Tbingen: Niemeyer.
Labov, William and David Fanshel
1977
Therapeutic Discourse: Psychotherapy as Conversation. New York: Academic Press.
Lemert, Charles and Ann Branaman (eds.)
1997
The Goffman reader. Ed. with introductory essays by Charles Lemert and Ann
Branaman. Oxford: Blackwell.
Levinson, Stephen C.
1983
Pragmatics. Cambridge: Cambridge University Press.
Levinson, Stephen C.
1988
Putting Linguistics on a Proper Footing. In: Paul Drew and Anthony Wootton
(eds.), Erving Goffman. Exploring the interaction order, 161227. Cambridge:
Polity Press.
Luhmann, Niklas
1984
Soziale Systeme. Grundriss einer allgemeinen Theorie. Frankfurt am Main:
Suhrkamp.
Luhmann, Niklas
1997
Die Gesellschaft der Gesellschaft, Vols 1 and 2. Frankfurt am Main: Suhrkamp.
Meibauer, Jrg
1999
Pragmatik. Eine Einfhrung. Tbingen: Stauffenburg.
Mey, Jacob L.
1993
Pragmatics. An Introduction. Oxford: Blackwell.
Moeschler, Jacques
1994
Das Genfer Modell der Gesprchsanalyse. In: Fritz et al. (eds.), 6994.
Moeschler, Jacques
2001
The Geneva School. In: Brinker et al. (eds.), 952957.
Ochs, Elinor, Emanuel A. Schegloff and Sandra Annear Thompson (eds.)
1997
Interaction and grammar. Cambridge: Cambridge University Press.
Pomerantz, Anita
1984
Agreeing and disagreeing with assessments: some features of preferred/dispreffered turn shapes. In: John Maxwell Atkinson and John Heritage (eds.),
Structures of Social Action: Studies in Conversation Analysis, 57101. Cambridge: Cambridge University Press.
Psathas, George (ed.)
1979
Everyday language: studies in ethnomethodology. New York: Irvington.
Rauch, Elisabeth
1992
Sprachrituale in institutionellen und institutionalisierten Text- und Gesprchssorten. Frankfurt am Main: Lang.
Redder, Angelika
1990
Grammatiktheorie und sprachliches Handeln: denn und da. Tbingen:
Niemeyer.
Rehbein, Jochen
2001
Das Konzept der Diskursanalyse. In: Brinker et al. (eds.), 927945.

Foundations: ethnomethodology and Erving Goffman

285

Sacks, Harvey
1972
An Initial Investigation of the Usability of Conversational Data for Doing
Sociology. In: David Sudnow (ed.), Studies in social interaction, 3174. New
York: The Free Press.
Sacks, Harvey
1974 [1972] On the Analysability of Stories by Children. In: Roy Turner (ed.), Ethnomethodology. Selected Readings, 216232. Harmondsworth: Penguin Education. First published in John Gumperz and Dell Hymes (eds.), Directions in
sociolinguistics: The ethnography of communication. New York: Holt, Rinehart, & Winston.
Sacks, Harvey
1984
On doing being ordinary. In: John Maxwell Atkinson and John Heritage
(eds.), Structures of Social Action: Studies in Conversation Analysis, 413429.
Cambridge: Cambridge University Press.
Sacks, Harvey
1989
Lectures 19641965. Ed. by Gail Jefferson. Dordrecht: Kluwer Academic Publishers.
Sacks, Harvey
1992
Lectures on Conversation. Vols I and II. Ed. by Gail Jefferson. With an introduction by Emanuel A. Schegloff. Oxford: Basil Blackwell.
Sacks, Harvey, Emanuel A. Schegloff and Gail Jefferson
1978
A simplest systematics for the conversation of turn taking for conversation. In:
Jim N. Schenkein (ed.), Studies in the organization of conversational interaction, 755. New York: Academic Press. First published Language 50: 696735
[1974].
Schank, Gerd and Johannes Schwitalla (eds.)
1987
Konflikte in Gesprchen. Tbingen: Narr.
Schegloff, Emanuel A.
1968 [1972] Sequencing in conversational openings. American Anthropologist 70:
10751095. Reprinted in John Gumperz and Dell Hymes (eds.), Directions
in sociolinguistics: the ethnography of communication, 346380. New York:
Rinehart & Winston.
Schegloff, Emanuel A.
1972
Notes on a Conversational Practice. Formulating Place. In: David Sudnow
(ed.), Studies in social interaction, 75119. New York: The Free Press.
Schegloff, Emanuel A.
1979
The Relevance of Repair to Syntax-for-conversation. In: Talmy Givn (ed.),
Syntax and Semantics 12: Discourse and Syntax, 261286. New York: Acadamic Press.
Schegloff, Emanuel A.
1987
Some sources of misunderstanding in talk-in-interaction. Linguistics 25:
201218.
Schegloff, Emanuel A.
1992
Introduction. In: Harvey Sacks: Lectures on Conversation. Vols I and II. Ed. by
Gail Jefferson, ixxii. Oxford: Basil Blackwell.
Schegloff, Emanuel A., Gail Jefferson and Harvey Sacks
1977
The Preference for Self-Correction in the Organization of Repair in Conversation. Language 53 (2): 361382.

286

Christine Domke and Werner Holly

Schegloff, Emanuel A. and Harvey Sacks


1973
Opening up closings. Semiotica 8: 289327.
Schenkein, Jim N. (ed.)
1978
Studies in the organization of conversational interaction. New York: Academic Press.
Schmitt, Reinhold (ed.)
2007
Koordination. Analysen zur multimodalen Interaktion. Tbingen: Narr.
Schneider, Wolfgang Ludwig
1997
Die Analyse von Struktursicherungsoperationen als Kooperationsfeld von
Konversationsanalyse, objektiver Hermeneutik und Systemtheorie. In: Tilmann Sutter (ed.), Beobachtung verstehen, Verstehen beobachten. Perspektiven einer konstruktivistischen Hermeneutik, 164227. Opladen: Westdeutscher Verlag,
Schwitalla, Johannes
2001a
Gesprochene-Sprache-Forschung und ihre Entwicklung zu einer Gesprchsanalyse. In: Brinker et al. (eds.), 896903.
Schwitalla, Johannes
2001b
Konflikte und Verfahren ihrer Bearbeitung. In: Brinker et al. (eds.), 1374
1382.
Schwitalla, Johannes
2001c
Beteiligungsrollen im Gesprch. In: Brinker et al. (eds.), 13551361.
Selting, Margret and Elizabeth Couper-Kuhlen
2000
Argumente fr die Entwicklung einer interaktionalen Linguistik. Gesprchsforschung Online-Zeitschrift zur verbalen Interaktion 1: 7695
Sinclair, John and Malcolm Coulthard
1975
Towards an analysis of discourse. London: Oxford University Press.
Streeck, Jrgen
1983
Konversationsanalyse. Ein Reparaturversuch. Zeitschrift fr Sprachwissenschaft 2: 72104.
Streeck, Jrgen
2009
Gesturecraft. The manufacture of meaning. Amsterdam: Benjamis.
Sudnow, David (ed.)
1972
Studies in social interaction. New York: The Free Press.
Tannen, Deborah
1984
Conversational Style: Analyzing Talk Among Friends. Norwood, NJ: Ablex.
Tracy, Karen
1990
The many faces of facework. In: Howard Giles and W. Peter Robinson (eds.),
Handbook of language and social psychology, 209226. Chichester: John
Wiley and Sons.
Turner, Roy
1974a
Introduction. In: Roy Turner (ed.), Ethnomethodology. Selected Readings,
712. Harmondsworth: Penguin Education.
Turner, Roy (ed.)
1974b
Ethnomethodology. Selected Readings. Harmondsworth: Penguin Education.
Uhmann, Susanne
1997
Grammatische Regeln und konversationelle Strategien. Fallstudien aus Syntax
und Phonologie. Tbingen: Niemeyer.

Foundations: ethnomethodology and Erving Goffman

287

Van Dijk, Teun


1972
Some Aspects of Text Grammars. The Hague: Mouton.
Ventola, Eija
2001
Discourse studies in the English-speaking countries. In: Brinker et al. (eds.),
979994.
Werlen, Iwar
2001
Rituelle Muster in Gesprchen. In: Brinker et al. (eds.), 12631278.
Widmer, Jean
1991
Goffman und die Ethnomethodologie. In: Robert Hettlage and Karl Lenz
(eds.), Erving Goffman ein soziologischer Klassiker der zweiten Generation,
211242. Bern/Stuttgart: Haupt.
Wodak, Ruth, Johanna Pelikan, Peter Nowak, Helmut Gruber, Rudolf de Cillia and Richard
Mitten
1990
Wir sind alle unschuldige Tter!. Diskurshistorische Studien zum Nachkriegsantisemitismus. Frankfurt am Main: Suhrkamp.

10.

Pragmatics in Habermas Critical Social Theory


Maeve Cooke

1.

Introduction

Sociology, as the scientific or systematic study of society, emerged as an independent academic discipline only towards the end of the nineteenth century. From the
beginning it has been marked by a dispute as to the method of inquiry appropriate
to it. A central question is whether its methodology should model that of the natural sciences or that of humanistic and cultural studies. The first position is often attributed to Auguste Comte and nineteenth and twentieth century positivism; it
holds that social phenomena can be subjected to observable natural laws. The second position is associated with Wilhelm Dilthey and nineteenth and twentieth century historicism and hermeneutics; it asserts an essential difference between the
methodology of the natural sciences and humanistic and cultural studies, emphasizing the interpretative component of social scientific inquiry.
The first position adopts an external perspective on the production of social
order, viewing it as a quasi-natural process that can be observed in its empirical
regularities and explained by way of nomological hypotheses. The second position
adopts an internal perspective on the production of social order, viewing it as a
meaningfully structured reality for social agents that should be interpreted in terms
of intentional actions.
To be sure, the lines of division are by no means clear cut. Max Weber, for
example, defines sociology as a science that attempts the interpretative understanding of social action in order thereby to arrive at a causal explanation of its course
and effects (Weber 1969). In line with this, he holds that rational interpretation, in
the sense of objective or impartial appraisal, is possible, when the agent and her interpreter equally accept the standards of judgement as valid (Habermas 1988).
Moreover, Weber is ambivalent on the question of whether the methods of the
social sciences should model those of the natural sciences. In this regard we find
striking differences in attitude and emphasis in his writings (McCarthy 1978:
141145).
We cannot explore this debate further here. For our present purposes, it is sufficient to note that, today, there is widespread, if not universal, agreement that the
methods of the natural sciences cannot simply be mapped onto sociology. Furthermore, among those who accept this, a significant number agree that the difference is due in large part to the intentional component of human action.1 By this is
meant, first, that the concept of society is internally linked with the concept of
human action and, second, that in the case of such action it is always in principle

290

Maeve Cooke

possible to ask what the agent(s) intended meant when carrying it out. The implication here is that a description of a particular human action that failed to consider the intentions of the agent would lack an essential ingredient.2 This implies in
turn that intention, understood as subjective meaning, is a sociological primitive:
an irreducible element of sociological analysis. It should also be noted that even if
we accept the reference to subjective meaning as constituting an essential methodological difference between the natural sciences and sociology, a number of important questions remain open, for example, the questions of whether causal explanations of social phenomena are possible and the degree to which objectivity is
achievable in sociology. Thus, accepting the irreducibility of subjective meaning
does not commit the sociologist to a purely internal perspective on social phenomena and require her to abstain from empirical-analytic analyses and causal explanations. However, it presents her with the challenge of developing a theoretical
framework, and corresponding methodology, that negotiates the tensions between
the external and internal perspectives.
In the early 1970s, as part of an endeavour to meet this challenge, Jrgen Habermas, a philosopher and sociologist in the Frankfurt School tradition of critical
social theory, turned to pragmatics.3 The importance of pragmatics in contemporary
sociology is largely due to Habermas contribution to the study of society. This contribution is so significant that no sociologist today can ignore his social theory and,
by extension, the programme of pragmatics underpinning it. While Habermas owes
a significant debt to his colleague Karl-Otto Apel, who introduced him to pragmatics in the late 1960s, Apels work has been more influential in philosophy than
social theory (Apel 1980). Furthermore, unlike Habermas, Apel did not develop his
reflections on pragmatics into a systematic programme. For these reasons, we will
take Habermas as our reference point in the remainder of our discussion.
In turning to pragmatics, Habermas concerns were not only methodological.
He sought, in addition, to provide an answer to central questions of sociology, such
as how we should conceptualize social action, how social order is possible and how
social order is reproduced. A further, equally important, aim was to provide a basis
for the rational critique of society without appealing to metaphysical notions of
truth, justice or the good. The Theory of Communicative Action, first published in
two volumes in 1981, attempts to meet these aims. In the following, we will examine the role of pragmatics, first, from the point of view of method, where we will
consider the conceptual strategy he deems most appropriate for sociological inquiry; then, from the point of view of the interconnected questions of social action,
social order and social reproduction; finally, from the point of view of social
critique, where we will look at the role pragmatics plays in critique of social pathologies and in developing a postmetaphysical conception of reason, which is the
normative reference point for this critique. In the concluding section we will look
briefly at some challenges to Habermas appeal to pragmatics.

Pragmatics in Habermas' critical social theory

2.

291

Method

In the course of the 1970s Habermas develops a theory of pragmatics that, as indicated, is supposed to help him answer central sociological questions relating to social action, social order and social reproduction as well as provide a normative
basis for his critical theory of society. In a preliminary essay based on lectures delivered at Princeton University in 1971, he focuses specifically on methodology,
identifying a number of metatheoretical issues as centrally important in the endeavour to develop a conceptual strategy appropriate for sociology.4 In each case,
in line with his concern to avoid a simple decision for either explanation or interpretation, Habermas adopts a position that seeks to retain the most valuable elements in the apparently conflicting approaches. For example, in his reflections on
the debate between methodological individualism and holism, he does not straightforwardly endorse either of these approaches but attributes to them complementary
strengths and weaknesses that make each useful for different purposes and in different contexts (Habermas 1971: 1314). Most relevant in the present instance is
the methodological decision he makes in favour of subjective meaning as a sociological primitive (Habermas 1971: 310). This decision leads him to view society
as a continuous process of generating a meaningfully structured reality, whereby
meanings are produced according to an underlying system of abstract rules and
have a sociologically relevant, intentional component. Having generally endorsed
theories that take subjective meaning as an irreducible element of sociological
analysis, Habermas now makes a distinction within this set between constitutive
and communicative theories.5 It is at this point that pragmatics enters the scene.
In his account, constitutive theories base the process of the generation of meaning on a transcendental subject. Some, taking their lead from Husserls account of
the constitution of the everyday world of lived experience (the lifeworld), conceive of this transcendental subject as an intelligible ego modelled on the empirical
individual subject. Others, taking their lead from Hegel and Marx, conceive of it as
a species-subject that constitutes itself in history. In each case, they encounter the
problem of how to move from the level of experience to the level of society. More
precisely, they run up against the problem of how adequately to conceptualize intersubjective meanings within a subject-object model of cognition and action. In
models of this kind, knowledge and action are conceived monologically, as the result of a relationship between a subject and an object that is distinct from it. However, if meaning is conceived of essentially as the product of a relationship
between a subject and some distinct object, commonly shared, intersubjective
meanings appear to have a secondary status. From a social-theoretical point of
view this gives rise to considerable difficulties. One difficulty is that it makes it
hard to explain how society, as a process of generating intersubjective meanings,
can be granted the primordial status it has in phenomenological accounts. This
weakness of constitutive theories, together with other, related ones, prompts Ha-

292

Maeve Cooke

bermas to look to communicative theories, which construe meaning as intersubjectively generated from the beginning.
In this preliminary essay, Habermas finds himself in the unwelcome situation
of being unable to refer to an existing body of work that would support the kind of
communicative theory he is seeking; he admits that the theoretical approach that he
deems methodologically most appropriate for social scientific inquiry has not yet
been adequately developed. However, he is clear as to the direction it is necessary
to take. What is required in his view is a sociological theory that would start from
the premise that everyday communication, as a process of generating shared meanings, is the basis for social action (and socialization); the use of ordinary language
would be seen as constituting the intersubjective relations within which action
takes place (as well as the personality structures of the subjects of action). His
name for this project is universal pragmatics.
Habermas programmatic essay What is Universal Pragmatics? first appeared in German in 1976. The adjective universal indicates the difference between his reconstructive approach to the study of linguistic behaviour and empirically oriented analyses. In his account, empirical-pragmatic approaches set
themselves the task of describing the speech acts typical of a certain milieu, which
may then be analyzed from sociological, ethnological and psychological points of
view. Habermas favours a broader perspective. He aligns himself with general
pragmatic theory, which is concerned not with specific instances of language use,
but rather with formally reconstructing the rule system that underlies the ability of
a subject to utter sentences in any relevant situation whatsoever (Habermas 1998:
54). Accordingly, he characterizes the task of universal pragmatics as being to
identify and reconstruct universal conditions of possible Verstndigung, which we
can translate here as mutual understanding6 (Habermas 1998: 21). Soon after the
publication of this essay, however, he distanced himself from the name universal
pragmatics. In a footnote to the English translation, published three years later in
1979, he writes that he is no longer happy with his original terminology, suggesting
that his project instead be named formal pragmatics. In opting for the word formal he wishes to emphasize the continuity between his approach and formal semantics, notwithstanding his focus on the use of language in speech acts or utterances as opposed to the semanticists concern with the properties of isolated
sentences.7 His choice of the word formal also makes sense in the context of his
adoption of the procedure of rational reconstruction which, as indicated, is the
method he attributes to his universally oriented, pragmatic approach. By rational
reconstruction he means the systematic reconstruction of the intuitive knowledge
of competent subjects (Habermas 1998: 2941). Reconstructions relate to pretheoretical knowledge of a general sort, to universal capabilities and not merely to the
competences of particular groups or individuals. Thus, what begins as an explication of meaning ultimately aims at the reconstruction of species competences.
Habermas sees it as the great merit of Noam Chomsky to have developed this idea

Pragmatics in Habermas' critical social theory

293

in the case of grammatical theory (Chomsky 1969). He also commends Jean Piaget
and Lawrence Kohlberg for, respectively, their theories of cognitive and moral development, which, like Chomskys, rationally reconstruct the pretheoretical and
implicit knowledge and competences of speaking and acting subjects in general.
In distinguishing his formal-pragmatic approach from an empirical-pragmatic
one, Habermas does not want to deny the importance of empirical validation. As
indicated, reconstructive theories belong to the category of general theory in the
sense that they seek to reconstruct the rule system that underlies the ability of a
subject to generate well-formed sentences in any language whatsoever. At the
same time, they belong to the category of empirical theories. Their explications of
general structures and universal conditions have a merely hypothetical status and
must be subjected to the usual methods of testing. Thus, Chomskys and Habermas reconstructions of linguistic competence are open to checking against speakers intuitions, scattered across as broad a sociological spectrum as possible (Habermas 1984:138). As is appropriate for scientific investigations of this sort, the
checking procedure is circular: it moves in a circle between theory formation and a
more precise rendering of the object domain (Habermas 1998: 40).
From this it is clear that Habermas places reconstruction, as an explicative exercise, on the side of interpretation rather than explanation, thereby connecting it
with an internal rather than external perspective on social phenomena. The point of
reconstructions is neither to describe behaviour nor to systematize their findings in
law-like explanations, but rather to bring to consciousness implicit knowledge
through a maieutic method of interrogation, which involves the choice of suitable
examples and counterexamples, the use of contrast and similarity relations, translation, paraphrase and so on (Habermas 1998: 40). This is not to deny that the procedures employed in constructing and testing hypotheses, in appraising competing
reconstructive proposals and in gathering and selecting data, are not in many ways
like the procedures used in the nomological sciences (Habermas 1998: 46). Like
such sciences, reconstructive theories usually operate at a high level of abstraction,
seeking to explicate basic conceptual frameworks and to formalize assumptions
built into everyday behaviour. The key difference, to recall our earlier discussion,
is that reconstructive social theories take subjective meaning as an irreducible element of social scientific analysis whereas theories that adopt a purely external
perspective do not.
We have seen that Habermas turns to pragmatics in order to find a method of
sociological analysis that takes subjective meaning as a sociological primitive,
while avoiding the weaknesses of constitutive models which attribute the process
of generating a meaningfully structured reality to some sort of transcendental subject. His aim is to develop a communicative model that replaces the subject-object
paradigm underlying constitutive theories with an intersubjective paradigm in
which the generation of shared meanings is seen as fundamental in the constitution
of social reality. He sees the use theory of meaning developed by Ludwig Wittgen-

294

Maeve Cooke

stein in his later work as making an important contribution towards developing


such a model (Wittgenstein 2001). But for reasons that will become clearer in the
next section, he finds speech act theory as developed by Austin and by Searle (in
his early work) more fruitful for his purposes.
Speech act theory shares with reconstructive language analysis a concern to explicate the rules that a competent speaker must master in order to form grammatical sentences and to utter them in an acceptable way. However, by contrast with
Chomskian linguistics, instead of starting from the postulate of an implicit ability
to produce sentences, speech act theory starts from the assumption of an implicit
ability to employ sentences in speech acts (utterances). Thus, it is concerned with a
competent speakers ability to communicate as opposed to produce a grammatical
sentence. Accordingly, a general theory of speech acts explicates the system of
rules that competent speakers master when they fulfil the conditions for the successful employment of sentences in utterances, no matter to which particular languages the sentences may belong and in which random contexts the sentences may
be uttered (Habermas 1998: 47). The importance of speech act theory in the development of Habermas conceptual strategy is twofold. First, by virtue of its focus on
utterances as communicative actions, it enables him to show how the meaning of
speech acts is produced by way of an action involving speaker and hearer(s); this
provides support for his shift in paradigm from a subject-object based, monological model of meaning production to an intersubjective one. Second, it enables him
to show how the production of meaning is internally connected with a number of
different kinds of claims to validity. The connection between meaning and validity
in a number of dimensions emerges from speech act theorys analysis of the successful employment of sentences in utterances. Habermas contends that for an utterance or speech act to be successful it must count as true for the speaker and hearers, in the sense of representing something in the world, must count as normatively
right, in the sense of conforming to socially recognized expectations and must
count as truthful, in the sense of expressing something intended by the speaker. As
we shall now see, the connection between meaning and validity claims is of crucial
importance in Habermas accounts of social action, social order and social reproduction and in his proposal for a postmetaphysical conception of reason that permits social critique.

3.

Social action, social order and social reproduction

Formal pragmatics forms the basis for the kind of communicative theory that Habermas thinks necessary in order to avoid the weakness of constitutive social theories and those that share their subjectivist orientation. In his view, such theories
not only have difficulties adequately conceptualizing meanings shared in common;
they are also not well suited to the task of adequately conceptualizing social action,

Pragmatics in Habermas' critical social theory

295

social order and social reproduction. This is because they tend to privilege instrumental action, understood as action undertaken as a means to achieving an end that
is specified in advance of the undertaking. Accounts of social order and reproduction that take instrumental action as the fundamental mode of social action run up
against the problem of how to explain long-term social cohesion, when the preservation and reproduction of the social order in question does not evidently equally
enable everyone to obtain their particular ends (cf. Habermas 1998: 234). A central
aim of Habermas theory of communicative action is to show that instrumental action is not the primary mechanism in social cohesion and social reproduction. He
draws on formal-pragmatic analyses of everyday language in order to achieve this
aim.
As the background to our discussion in this section and the next, it is important
to note that in The Theory of Communicative Action Habermas offers an account of
the developmental path of modern societies in terms of two categorially distinct
modes of societal integration. In what he calls the lifeworld, co-ordination of action takes place primarily by way of communicative action, which relies on intentionally acting human subjects who establish validity-based relationships with one
another. In the system, by contrast, co-ordination of action takes place primarily
by way of the functional interconnection of action consequences, bypassing the intentions of the agents concerned. For Habermas, rationalization processes, by
which he means the development of the internal logic of a particular mode of societal integration, are possible and desirable in each case. The rationalization of the
system refers to increasing complexity and increasing capacity to take on steering
functions such as material reproduction and administration. By contrast, the
rationalization of the lifeworld is a matter of communicative rationalization: the
development of the internal logic of communicative action. It takes place in the domains of cultural tradition, social integration and socialization and entails the increasing independence of processes of justification from traditional normative contexts of validity, together with increasing reliance on action oriented to achieving
understanding (Verstndigung).
Communicative action is a form of social interaction in which the plans of various agents are co-ordinated through an exchange of speech acts. Drawing on the
speech act theory of Austin and the early Searle, Habermas maintains that the success of a speech act is not just a matter of comprehending the sentence embedded in
the utterance, but of responding in the appropriate manner to what the speaker does
in performing the action. In addition to commending these speech act theorists for
emphasizing the action component of linguistic understanding, he praises them for
directing attention to modes of language use other than the assertoric and descriptive modes. In this respect, too, they take their lead from the later Wittgenstein.
However, Habermas finds speech act theory more promising than Wittgensteins
use theory of language, due to its assertion of a connection between meaning and
validity in a context-transcending sense. For Habermas, truth and truth-analogous

296

Maeve Cooke

concepts such as moral rightness have a context-transcending aspect in the sense


that they assert a validity that transcends particular places and times. This is one
reason why he refers to them as universal validity claims.8 The connection between
meaning and universal validity disappears from view in Wittgensteins theory, since
validity is held to be relative to particular, historically specific, language games. To
be sure, Habermas finds Austins and Searles approaches to universal validity
claims unsatisfactory. He criticizes Austin for blurring the distinction between clear
cut universal validity claims such as truth and normative rightness and a host of
other evaluative criteria (Habermas 1998: 290291). He admonishes Searle for acknowledging only the representational function, together with the claim to truth
corresponding to it, as the only clear cut universal validity claim (Habermas 1998:
291292). Against Austin and Searle Habermas seeks to demonstrate that everyday
language use is connected with three clear cut universal validity claims.
This is one of the two main respects in which Habermas moves beyond speech
act theory. The other is his emphasis on the relationship of obligation into which
the speaker and hearer(s) enter with the performance of a speech act. Taking his initial lead from Austin, he finds it instructive to consider what a speaker does in saying something. He begins by discussing Austins distinction between locutions and
illocutions (Austin 1962). Initially Austin reserved the term locution for the
propositional content of what is said and the term illocution for the performative
act of uttering sentences with a propositional content. He later reformulated the
distinction as one between constatives and performatives, defining constatives as
speech acts in which we concentrate on the truth or falsity of what is said and performatives as speech acts in which we concentrate on the illocutionary force of the
utterance (Habermas 1998: 6681). Habermas finds this way of distinguishing between constative and performative speech acts unhelpful, proposing instead that
we distinguish between utterances on the basis of the kinds of validity claims they
raise. In his reformulated version of Austins classificatory schema, constatives involve the cognitive use of language and raise claims to truth (examples are assertions and descriptions); here we thematize the content of the utterance as a statement about what is or could be the case. Regulatives involve the interactive use of
language and raise claims to normative rightness (examples are promises, requests
and warnings); here we thematize the relationship of obligation into which speaker
and hearer enter with the speech act. Habermas adds a third category, expressives,
to these revisions of Austins schema; as the name indicates, these involve the expressive use of language and raise claims to truthfulness (examples are confessions
and disclosures); here we thematize the truthfulness or sincerity of the speakers
intentions. He proposes, further, that we understand the illocutionary force of the
utterance what the speaker does in performing any speech act as a rationally
binding force (bindende Kraft), in the twin sense of binding and compelling (Habermas 1998: 8188). This rationally binding force is held to be operative in all
three modes of language use: the constative, regulative and the expressive.

Pragmatics in Habermas' critical social theory

297

This, then, is Habermas answer to the question of what we do in performing a


speech act: we enter into a relationship of rationally-based obligation with the
hearers. The obligation is rationally-based since it arises from the speakers undertaking to support what she says with reasons. He conceptualizes this in terms of a
validity claim raised by the speaker. The speaker claims that what she says is true
or that she is normatively right to say it or that she is sincere in saying it, thus
raising a claim to validity that calls for a yes or no response from the hearer.
The success of the interaction depends on the speakers ability to supply reasons to
support the validity of the claim, if challenged, and on the hearers ability to provide reasons in support of her yes or no. For Habermas, therefore, the illocutionary force of an utterance is not simply an aspect of its meaning conveyed by a
performative verb; it is a co-ordinating power that is based on reciprocal obligations to provide reasons. This is what he means by the rationally binding force of
linguistic interaction.
Habermas holds, furthermore, that every speech act raises all three claims simultaneously. More precisely, he holds that with every speech act, the speaker raises
one of these three kinds of validity claims directly, while the other two are raised
indirectly. Thus, for example, if a speaker states she has no money to pay the rent,
she is raising a claim to truth that calls for a yes or no response from the hearer(s) or for an abstention on grounds of insufficient knowledge of the facts. At the
same time she raises a claim that her utterance is appropriate in the context in question (normatively right) and that she is sincere in uttering it. In this case, a claim
to truth is raised directly and claims to normative rightness and sincerity are raised
indirectly. If a speaker requests a hearer to lend her some money, she raises a claim
to normative rightness directly and claims to truth and truthfulness indirectly (Habermas has clarified that indirect claims to truth should be understood as claims
that certain existential presuppositions are met, 1998: 146). A speaker who confesses that she has spent all her money raises a claim to truthfulness directly and
claims to normative rightness and truth indirectly.
The thesis that every speech act raises three claims to validity simultaneously is
important for Habermas attempt to show that an orientation to a plural multi-dimensional conception of validity is built into everyday language use. As we shall
see, this plural conception of validity is the basis for his non-logocentric conception of communicative rationality, which does not privilege propositional truth but
conceives of reason as an interplay of truth, moral rightness and truthfulness/authenticity. The thesis of three simultaneous validity claims is also important from the
point of view of social integration and reproduction. Its purpose is to show that the
three validity claims raised with every speech act amount to three structural components of speech acts that correspond to the three functions of the communicative
use of language: (a) representing states of affairs (or something in the world that
confronts the speaker), (b) entering into a relationship with a hearer and (c) giving
expression to the speakers intentions or experiences. Referring to the communi-

298

Maeve Cooke

cation-theoretic approach expounded by the German psychologist Karl Bhler


(Bhler 1934), Habermas posits these as three mutually irreducible but internally
connected linguistic functions.9 His thesis is that these linguistic functions, carried
out with every speech act, serve to maintain and reproduce the lifeworld in its three
domains of cultural tradition, social integration and socialization.
For this, however, the thesis of three simultaneous validity claims is not sufficient. Even if successful (see Cooke 1994: 9091), it would show merely that the
communicative use of language fulfils functions of representation, of establishing
a relationship with the hearer and of expression; it would not show that the communicative use of language is of fundamental importance in maintaining and reproducing the lifeworld. Showing the latter is central to Habermas critical project.
We will recall his claim that the colonization of the communicatively regulated domains of the lifeworld by the functional rationality of the system is a deeply troubling pathology of modernity. If he were unable to show that communicative action
is indispensable for the purposes of establishing, maintaining and reproducing
social order, he would lack grounds for referring to the shrinking of the communicatively regulated domains of the lifeworld as a pathology and for adopting the
metaphor of colonization to describe the expansion of functional rationality into
the communicatively regulated domains. In order to meet this challenge, Habermas
seeks to demonstrate the primacy of communicative action over instrumental (and,
in particular, strategic) action. Once again, he employs a formal-pragmatic strategy. This entails showing the primacy of the communicative use of language over
other modes, specifically the strategic one.
Just as the communicative use of language amounts to communicative action,
the strategic use of language amounts to strategic action. Strategic actions are instrumental insofar as one agent uses another as a means of achieving his respective
goals; he treats other persons as though they were objects or entities in the physical
world. In contrast to communicative action, in which the success of the interaction
is not at the disposal of the individual agent but depends on all parties involved,
strategically acting agents are not dependent on the co-operation of others (Habermas 1998: 220227). While Habermas acknowledges that strategic action, like
communicative action, is a mode of action within the lifeworld, he insists that it is a
merely secondary mode of social integration. His position, in other words, is that
communicatively regulated interactions must predominate in the lifeworld if this is
to maintain and reproduce itself. He seeks to show the priority of communicative
action by demonstrating that the strategic use of language is parasitic on the communicative one.10
In The Theory of Communicative Action Habermas attempts to show that the
use of language oriented to reached understanding (Verstndigung) is the primary
or original mode of language use by drawing on Austins distinction between locutions and perlocutions (Habermas 1998: 122129) In Austins account, the locutionary act is the act of saying something, the illocutionary act is the action per-

Pragmatics in Habermas' critical social theory

299

formed by saying something and the perlocutionary act is the effect the speaker
produces upon the hearer. Initially, Habermas took this to mean that perlocutionary
acts are an indication of the integration of speech acts into contexts of strategic action (Habermas 1998: 126). This interpretation neglects the importance of distinguishing between different types of perlocutions. It is important, above all, to distinguish between the contingent, often unintended effects on the hearer(s) of
particular speech acts and the intended, but concealed effects of particular speech
acts. An example of the former is the pleasure a speaker gives to his hearers husband by lending him money for the rent (for the act might have given him displeasure); an example of the latter is the speakers intention to use his loan of
money to blackmail the hearer. Habermas has subsequently modified his position,
clarifying that only perlocutionary effects of the latter kind are relevant for his argument (Habermas 1998: 329333). He sees them as examples of latently strategic
action. As illustrated by the case where the speaker intends to blackmail the hearer
by lending her money, the success of latently strategic action depends on the speakers success in hiding his intentions from the hearer; he must successfully pretend
that he is using language with an orientation towards mutual understanding (Verstndigung), as opposed to strategically, using the interaction with the hearer as a
means to furthering his own particular ends.
The latently strategic mode of language use is by definition parasitic on the
communicative use (Cooke 1994: 24).11 Thus, demonstrating its secondary character will not be sufficient for Habermas purposes. The more interesting and difficult case is the manifestly strategic use of language. In this mode of language use,
the presupposition of an orientation towards reaching understanding is overtly suspended (Habermas 1998: 225226). A speaker who backs up his demand that a
hearer lend him some money with the threat of sanction (for example, by threatening to harm her daughter) illustrates such language use. In order to make good his
claim that the communicative use of language is the primary mode of language use,
Habermas has to show that this manifestly strategic use of language, too, is parasitic on the communicative one. This part of Habermas argument is underdeveloped (Cooke 1994: 2326).
In this section we have considered the importance of formal pragmatics in Habermas accounts of social action, social order and social reproduction. We have
seen that its role is threefold: to demonstrate that everyday linguistic communication has an in-built rationally binding force, to show that this binding force is operative in three dimensions corresponding to the three linguistic functions of representation, establishing a relationship with the hearer(s), and expression, and to
demonstrate that the communicative use of language is the primary mechanism in
establishing, maintaining and reproducing the social order. In the next section, we
will look at the importance of formal pragmatics for Habermas project of social
critique.

300
4.

Maeve Cooke

Social critique

Habermas is first and foremost a critical social theorist. His sociological concerns
are intimately bound up with his concern to diagnose and criticize the pathologies
of modernity. We have already mentioned the colonization of the lifeworld as one
of these pathologies. Habermas, drawing on Weber, sometimes refers to this as the
modern loss of freedom (Habermas 1987: 301302). The second principal pathology he diagnoses is a loss of meaning (here, too, he draws on Webers analysis of
modernity). By this he means the cultural impoverishment of the lifeworld that results from the differentiation of the cultural value spheres of science, law/morality,
and art under conditions of modernity; these value spheres become fields for the
specialized treatment of questions of truth, normative rightness and beauty/authenticity by experts. While these specialized discourses promote a growth in cognitive-instrumental, moral-practical and aesthetic-expressive knowledge, thus giving rise to significant learning processes, they are also a potential source of
pathology. This is because with progressive modernization, the gap between the
expert cultures and the general public becomes ever wider. Learning processes
within the expert cultures do not automatically flow back into the communicative
practices of the lifeworld, contributing to their renewal and regeneration, but remain cut off from them. Lacking revitalizing impulses from the specialist cultures
(or, indeed, from other sources12), everyday communicative practices threaten to
become impoverished.
His critique of these pathologies is guided by a corresponding normative vision. At the core of this vision is a conceptual triad that has defined his work from
his earliest major publication The Structural Transformation of the Public Sphere
(Habermas 1989): the concepts of the public sphere, discourse and reason. The interplay of these concepts yields a utopian vision of a communicatively rationalized
lifeworld (Habermas 1991: 6469). In the projected society, cultural traditions
would be reproduced through processes of intersubjective evaluation of validity
claims, legitimate orders would be dependent on critical and open argumentative
practices for justifying laws/norms and for making political decisions, and individual identities would be self-regulated through processes of critical reflection. We
must assume, furthermore, that in such a communicatively rationalized lifeworld
there would be ongoing processes of semantic renewal in the form of continuous
flows of knowledge from the specialized, expert cultures into the communicative
practices of everyday life.13 This utopian projection of a rationalized lifeworld is in
turn part of a vision of society in which the development of systems rationality and
communicative rationality would be evenly balanced: systems rationality would
no longer encroach into the communicatively regulated domains of the lifeworld.
Formal pragmatics contributes to Habermas critical project in two main respects. His formal-pragmatic investigations are intended, first, to provide a normative reference point for the critique of the two principal pathologies of modern

Pragmatics in Habermas' critical social theory

301

societies: the colonization of the communicatively regulated domains of the lifeworld by systems rationality and its cultural impoverishment due to the emergence
of specialized discourses among experts in the areas of science, law/morality and
art. They are intended, second, to provide a normative reference point for criticism
of the conduct and outcomes of actual practices of deliberation in everyday life. In
each case, the normative reference point is given the name communicative
rationality.
As we have seen, Habermas thesis of the colonization of the lifeworld makes
the claim that the uneven developmental path of modern societies is pathological.
What is required, instead, is an evenly balanced pattern of societal rationalization.
Habermas expresses the idea of a non-pathological, evenly balanced process of societal rationalization in terms of the harmonious, free interplay of the three cultural
value spheres of science, morality/law and art, in which no one sphere would predominate at the expense of the other (Habermas 1984: 240). His formal-pragmatic
investigations are used as a normative underpinning for this idea.
The relevant formal-pragmatic finding is that everyday linguistic communication fulfils three mutually irreducible functions: representation, establishing a relationship between speaker and hearer(s), and expression; as we have seen, this is
the conclusion Habermas draws from his thesis that the speaker raises three validity
claims simultaneously with every speech act. He asserts, in addition, that these
three linguistic functions can be correlated with the kind of knowledge defining
each of the three cultural value spheres: scientific knowledge, moral-practical
knowledge and aesthetic-expressive knowledge. He then makes a further correlation between the three cultural value spheres and the three moments of a once substantively unified concept of reason, which have become separated from one another under conditions of modernity;14 the image of harmonious, free interplay is
extended to these three moments of reason, evoking an idea of reason whose unity is
defined purely formally as opposed to substantively, and that is non-logocentric, expanding the traditional philosophical focus on theoretical reasoning and truth to include other dimensions of validity (such as justice and beauty) on an equal footing.
If the colonization thesis suggests as its opposite evenly balanced processes of
functional rationalization and communicative rationalization, the thesis of cultural
impoverishment suggests as its opposite the continuous feedback back into everyday communicative practices of the knowledge produced by experts in their various fields. However, whereas Habermas evidently wishes to use his formal-pragmatic analyses to provide empirically based, normative support for the idea of
evenly balanced processes of societal rationalization, he gives no indication of any
formal-pragmatic finding that would lend support to the idea of the free flow of
knowledge between expert cultures and the lifeworld. Nor is it easy to see which
part of his analyses could help him in this regard.
In addition to these two principal pathologies he diagnoses and attributes to
deficiencies in modern processes of societal rationalization, Habermas identifies a

302

Maeve Cooke

number of further potential pathologies (Habermas 1987: 142144); by contrast


with the pathology of colonization, which threatens it from the outside, these potential disturbances arise within the lifeworld (in this they are similar to the pathology of cultural impoverishment). In addition to the loss of meaning that results
from the encapsulation of specialist cultures, he mentions the anomie that results
from disturbances in the process of social integration and the mental illnesses (psychopathologies) that result from disturbances in the process of individual socialization. At least initially, Habermas held that formal pragmatics could play a role in
the critique of these disturbances. In The Theory of Communicative Action and in
the preliminary studies that led up to it, he suggested that formal-pragmatic analyses could help to establish a normative basis for criticizing the phenomena in
question as pathologies. The specific strategy he envisaged was a formal-pragmatic
account of systematically distorted communication (Habermas 1974). Despite occasional remarks in his writings up to the middle of the 1980s that indicate that he
had not yet abandoned this strategy, he has not since then made any serious attempt
to develop such an account. His key intuition seemed to be that systematically distorted communication violates what formal pragmatics shows to be the internal organization of speech, by undermining the three linguistic functions of representation, establishing a relationship between speaker and hearer(s) and expression.
The violation, which takes the form of disconnecting meaning from validity, speaking from acting and/or meaning from intention, is due to the overwhelming pressure exerted on the internal organization of speech by its external organization; by
the external organization of speech he means the regulation of the normative context in which the communicative interaction takes place: for example, the determination of who is allowed to take part in which discussion, who can initiate topics,
who can bring the discussion to a close, who can contribute and in which order, how
the topics are ordered, and so on. To be sure, in order to develop a convincing account of systematically distorted communication, Habermas would have had to
embed it within his critical account of societal rationalization as colonization (at
the time of publication of his 1974 essay, he had not yet provided such an account);
specifically, he would have had to show how the infiltration of functional rationality into the communicatively regulated domains of social life affects the external
organization of speech, contributing to a violation of its internal organization. Formal-pragmatic arguments would not be helpful here. Nor would they be able to
ground the normative force of the thesis implicit in Habermas account of systematically distorted communication: the thesis that the regulation of the context in
which speech takes place should conform to demanding normative standards of inclusivity, fairness and openness. In short, even a well-developed formal-pragmatic
account of systematically distorted communication would show only what the internal organization of speech and its disruption would look like; on its own it would
not be sufficient to provide the sought-for normative basis for critique of disturbances such as anomie and psychopathologies (Cooke 1994: 148150).

Pragmatics in Habermas' critical social theory

303

As indicated, the normative reference point for Habermas diagnosis and


critique of social pathologies is the concept of communicative rationality. Habermas sets himself the task of elaborating a concept of reason that responds to two
(closely connected) challenges facing contemporary critical social theorists:
maintaining a dialectics of immanence and transcendence and maintaining concepts of validity that transcend times and spaces while acknowledging the historical situatedness of all actual conceptions of context-transcending validity. The
first is specific to theorists in the Frankfurt School tradition who take seriously the
Hegelian thesis that reason is embedded within history itself; the second arises for
any social theorists today who seek to uphold an idea of reason while taking on
board the post-Enlightenment suspicion of transcendental ideas such as truth, justice and the good. Formal pragmatics plays a role in meeting each of these challenges.
The term Left-Hegelian is sometimes used to characterize modes of critical
social theory that, taking their lead from Marx and Lukcs, are concerned to re-articulate Hegelian speculative thinking in a materialist way. Most critical social theorists in the Frankfurt School tradition fall into this category. The aspect of Hegels
thinking most relevant here is his emphasis on reasons simultaneous immanence
and transcendence. Hegel endeavours to show a) that reason is immanent to human
history in the sense that the latter is a process in which reason actualizes itself by
way of the progressive overcoming of contradictions and b) that reason always
transcends its particular historical self-actualizations.15 The attempt to maintain a
dialectical relationship between immanence and transcendence is taken up explicitly by Theodor W. Adorno, Max Horkheimer, Herbert Marcuse and other first
generation critical social theorists (see, for example, Adorno 1981) and continues
to play a central role today in the writings of Axel Honneth, the most prominent
member of the third generation of critical social theorists (Honneth 1994, 2003).
Habermas, the most prominent member of the second generation is less inclined to
describe his project in explicitly Left-Hegelian terms; nonetheless, a concern to
sustain a dialectics of immanence and transcendence is certainly also one of its
most important elements (see, for example, Habermas 1996:125). In Habermas
work, this amounts to the elaboration of a conception of reason that is at once context-transcending, in the sense of transcending times and places, thus extending to
human beings in general, and immanent, in the sense of being anchored in practices
within everyday historical practices, specifically in everyday practices of using
language communicatively.
Habermas theory of communicative action is of crucial importance to his attempt to elaborate a conception of reason that is at once immanent and transcendent. We have seen that he uses a formal-pragmatic analysis of everyday language
use as the basis for an account of social action, order and reproduction that has a
normative as well as an empirical content. It is empirical in the sense that it uses
empirical evidence to reconstruct the relevant structural features of everyday lan-

304

Maeve Cooke

guage use; it is normative in the sense that the resulting account of language is used
as the basis for a normative idea of a communicatively rationalized lifeworld that
co-exists harmoniously with a well-functioning system. An analogous interplay of
the empirical and normative characterizes the concept of communicative rationality. This concept expresses the rational potential implicit in immanent to everyday communicative action. As a practice of raising and responding to validity
claims, communicative action is conceptually tied to more or less rudimentary
practices of argumentation. Habermas maintains that the validity claims raised in
communicative action have a context-transcending, critical force that derives from
idealizations built into the concept of argumentation (see Cooke 2006: 4753).
One set of idealizations refers to the procedure of argumentation: to the normative
promise contained in the general understanding of how argumentation should be
conducted. Examples here are the idealizing presuppositions that all participants
are motivated by a concern to find the single right answer, that no force is exerted
except that of the better argument, that all relevant arguments are considered, that
no one should be excluded from the deliberation on grounds such as race, class and
gender and that everyones voice should be given an equal hearing. Another set
refers to the outcome of argumentation: to the normative promise contained in the
general understanding of the aim of argumentation. An example here is the idealizing presupposition that a discursively reached agreement warrants the truth of
propositions and constitutes the justice of norms. In both cases, the tension between the normative promise contained in these idealizations and what actually
happens in everyday communicative practices provides a basis for criticism: in the
one case, they permit criticism of the ways in which the outcomes of argumentation are reached; in the other case, they permit criticism of the outcomes from the
point of view of an idea of context-transcending validity (e.g. truth or justice). The
critical power of communicative action resides primarily in the tension between an
idealized notion of argumentation and what happens in actual empirical practices.
It should be noted that this critical power is not restricted to a particular socio-cultural context. Since it is grounded in universal features of language use, it expresses a critical perspective with context-transcending force, in the sense that its
validity would have to be accepted by everyone, everywhere, irrespective of sociocultural context. It should be noted, too, that the critical power of the concept of
communicative rationality refers to an idealized practice of communication that
has the status of a methodological fiction. Habermas warns against essentialist
misunderstandings of the ideal speech situation, emphasising that it is merely an
idealizing projection and best understood as enabling a fruitful thought experiment
(Habermas 1996: 322323). In this way, his formal-pragmatic analyses of language enable Habermas to assert an idea of reason that is at once immanent to history, in the sense that it is built into everyday practices of linguistic communication, and transcendent, in the sense that it provides a reference point for critique
of all actually existing communicative practices.

Pragmatics in Habermas' critical social theory

305

As part of his engagement with the post-Enlightenment critique of reason, from


the Romantics, through Nietzsche and historicism, up to contemporary post-structuralism, Habermas emphasizes the postmetaphysical character of the concept of
communicative rationality. The role played by formal pragmatics here is very similar to its contribution to sustaining a dialectics of immanence and transcendence.
By postmetaphysical he means, principally, that the concept of communicative
rationality breaks with traditional views of the substantive unity of reason; refrains
from specifying the content of truth, justice and the good life; moves from a subject-object model of cognition to an intersubjective one; gives up the traditional
fixation on truth and theoretical reason; and renounces all dependency on otherworldly projections such as God or the Good. Our discussion in the foregoing has
cast light on how his formal-pragmatic investigations help him to achieve these
aims. The reference to reason built into everyday language use is a formal as opposed to substantive conception of reason, for it amounts only to a procedure for
determining what is true or right and leaves open the question of the respective
content of these concepts; it is non-subjectivist in the sense that it starts from interaction in communicative practices rather than an individual subjects relation to
an object independent of it; it is non-logocentric in the sense that it does not assert
the primacy of theoretical or propositional truth but comprises a plurality of dimensions of validity that interact with each other and relate to each other on an
equal footing; it is historically rooted in the sense that it is contained within actual
practices of using language; and it is innerworldly in the sense that the ideal reference point to which it appeals is not some otherworldly idea of God or the
Good but an idealized projection of a speech situation in which the idealizing presuppositions relating to the conduct of intersubjective deliberation would be realized in practice and in which the outcomes of such deliberation would contribute
constructively to validity in a context-transcending sense.
In this section we have considered the ways in which formal pragmatics enables Habermas to develop a concept of communicative rationality that is the indispensable normative reference point for his critical social theory, both with regard to the critique of social pathologies and with regard to criticism of the
shortcomings and distortions of actual practices of intersubjective deliberation. In
my final remarks I will outline the principal lines of objection that have been raised
against Habermas use of formal pragmatics as the normative foundation for his
critical social theory.

5.

Objections

There are three principal lines of objection to Habermas use of formal pragmatics
as the underlining for his critical project. The first casts doubt on the ability of formal pragmatics to provide a foundation for social critique that is sufficiently robust

306

Maeve Cooke

for his purposes. The second criticizes the specific shape of the conception of reason that emerges from his formal-pragmatic analyses. The thi