Philosophical Perspectives For Pragmatics

Philosophical Perspectives for Pragmatics
Handbook of Pragmatics Highlights (HoPH)
The ten volumes of Handbook of Pragmatics Highlights focus on the

most salient topics in the field of pragmatics, thus dividing its wide
interdisciplinary spectrum in a transparent and manageable way.
EachÂ€volume starts with an up-to-date overview of its field of interest
and brings together some 12–20 entries on its most pertinent aspects.
Since 1995 the Handbook of Pragmatics (HoP) and the HoP Online (in
conjunction with the Bibliography of Pragmatics Online) have provided
continuously updated state-of-the-art information for students and
researchers interested in the science of language in use. Their value as a
basic reference tool is now enhanced with the publication of a topically
organized series of paperbacks presenting HoP Highlights.
Whether your interests are predominantly philosophical, cognitive,
grammatical, social, cultural, variational, interactional, or discursive, the
HoP Highlights volumes make sure you always have the most relevant
encyclopedic articles at your fingertips.
Editors
Jef Verschueren Jan-Ola Östman
University of Antwerp University of Helsinki
Volume 10
Philosophical Perspectives for Pragmatics
Edited by Marina Sbisà, Jan-Ola Östman and Jef Verschueren
Philosophical Perspectives
for Pragmatics
Edited by
Marina Sbisà
Università di Trieste
Jan-Ola Östman
University of Helsinki
Jef Verschueren
University of Antwerp
John Benjamins Publishing Company

Amsterdamâ•›/â•›Philadelphia
TM
The paper used in this publication meets the minimum requirements
8
of American National Standard for Information Sciences – Permanence
of Paper for Printed Library Materials, ansi z39.48-1984.
Library of Congress Cataloging-in-Publication Data
Philosophical perspectives for pragmatics / edited by Marina Sbisà, Jan-Ola Östman and Jef Ver-
schueren.
p. cm. (Handbook of Pragmatics Highlights, issn 1877-654X ; v. 10)
Includes bibliographical references and index.
1. Pragmatics. 2. Linguistics (Philosophy) I. Sbisà, Marina. II. Östman, Jan-Ola. III. Verschueren, Jef.
P99.4.P72P55â•…â•… 2011
302.44--dc22 2010051884
isbn 978 90 272 0787 6 (Pb ; alk. paper)
isbn 978 90 272 8913 1 (Eb)
© 2011 – John Benjamins B.V.

No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any other
means, without written permission from the publisher.
John Benjamins Publishing Co. · P.O. Box 36224 · 1020 me Amsterdam · The Netherlands
John Benjamins North America · P.O. Box 27519 · Philadelphia pa 19118-0519 · usa
Table of contents
Preface to the series xiii
Acknowledgements xv
Introduction 1
Marina Sbisà
1. Pragmatics and philosophyâ•… 1
2. Conceptions of meaningâ•… 2
3. Speech as actionâ•… 4
4. Mind and selfâ•… 6
5. Doing pragmatics, doing philosophyâ•… 8
Analytical philosophy – Ordinary language philosophy 11

Marina Sbisà
1. Philosophy as analysisâ•… 11
1.1 The ‘linguistic turn’â•… 12
1.2 The influence of Fregeâ•… 12
1.3 Analysis in G.E. Moore and B. Russellâ•… 13
2. Analysis and the ideal of scientific languageâ•… 14
2.1 Wittgenstein’s Tractatusâ•… 14
2.2 Rudolf Carnap and the encyclopedia of unified scienceâ•… 14
3. Analysis and ordinary languageâ•… 15
3.1 The evolution of Wittgenstein’s thoughtâ•… 16
3.2 Wittgenstein’s influence and ordinary
language philosophyâ•… 16
3.3 Some Oxford philosophersâ•… 16
3.3.1 J.L. Austinâ•… 16
3.3.2 P.F. Strawsonâ•… 17
3.3.3 H.P. Griceâ•… 17
4. Further developments of analytical philosophyâ•… 18
4.1 W.V.O. Quine: From analysis to naturalizationâ•… 18
4.2 From intensional semantics to discourse
representation theoryâ•… 19
VI Philosophical Perspectives for Pragmatics
4.3 Meaning and understandingâ•… 19

4.4 Philosophy of mindâ•… 20
5. Analytical philosophy and pragmaticsâ•… 21
John L. Austin 26
Marina Sbisà
1. J.L. Austin and his approach to philosophyâ•… 26
1.1 Austin’s philosophical methodâ•… 26
1.2 “Linguistic phenomenology”â•… 28
1.3 General tendenciesâ•… 28
2. Epistemologyâ•… 29
2.1 Knowledge and beliefâ•… 29
2.2 Perceptionâ•… 30
3. Philosophy of languageâ•… 31
3.1 Meaningâ•… 31
3.2 Performative utterancesâ•… 31
3.3 Assertion and truthâ•… 32
3.4 The speech actâ•… 32
4. Philosophy of actionâ•… 33
4.1 Actionâ•… 34
4.2 Freedom and responsibilityâ•… 34
5. Austin and pragmaticsâ•… 35
Mikhail Bakhtin 38
Martina Björklund
1. Biographical sketchâ•… 38
2. The ‘Bakhtin industry’â•… 40
3. Bakhtin’s view of languageâ•… 41
3.1 Dialogueâ•… 42
3.2 Heteroglossiaâ•… 43
3.3 Polyphonyâ•… 43
3.4 Metalinguisticsâ•… 44
3.5 Speech genresâ•… 47
3.6 Chronotopeâ•… 49
3.7 Carnivalâ•… 49
4. Conclusionâ•… 50
Contextualism 53
Claudia Bianchi
1. Two perspectivesâ•… 53
2. Semantic minimalismâ•… 54
Table of contents VII
3. Indexicalismâ•… 55
4. Radical contextualismâ•… 57
4.1 Overviewâ•… 57
4.2 Wittgenstein, Austin, Searle, and Travisâ•… 58
4.3 Motivations for radical contextualismâ•… 60
4.4 Objections to radical contextualismâ•… 62
5. Nonindexical contextualismâ•… 64
Deconstruction 71
Tony Schirato
1. Introductionâ•… 71
2. Historical backgroundâ•… 71
3. Basic tenetsâ•… 73
4. Deconstruction in literature and linguisticsâ•… 76
5. Against deconstructionâ•… 77
Epistemology 79
Filip Buekens
Epistemology of testimony 82
Paul Faulkner
Michel Foucault 85
Luisa Martín Rojo & Angel Gabilondo Pujol
2. Foucault and the discursive turnâ•… 86
2.1 Discourse as a practiceâ•… 89
2.2 Discourse, knowledge and powerâ•… 92
3. The order of discourseâ•… 95
4. Rethinking the analytical practiceâ•… 97
5. Conclusionsâ•… 100
H.P. Grice 104

Frank Brisard
1. Lifeâ•… 105
2. Languageâ•… 108
2.1 Meaningâ•… 108
2.2 The conversationalist hypothesisâ•… 112
2.3 Rationalityâ•… 114
VIII Philosophical Perspectives for Pragmatics
3. alue and the new metaphysicsâ•… 117

V
3.1 Creature constructionâ•… 117
3.2 Absolute value (Kantotle)â•… 119
4. Concluding remarksâ•… 120
5. Further readingâ•… 122
Hermeneutics 125
Piet Van de Craen
2. The origins of hermeneutic thinkingâ•… 125
3. Some aspects of the evolution of hermeneutic thinkingâ•… 126
4. The nature of the hermeneutic enterprise and the
hermeneutical circleâ•… 127
5. Linguistics and hermeneuticsâ•… 127
5.1 Structuralismâ•… 128
5.2 Linguistic anthropologyâ•… 128
5.3 Cognitive linguisticsâ•… 128
5.4 Conversation analysisâ•… 129
Indexicals and demonstratives 131

Eros Corazza
2. Indexical expressionsâ•… 132
3. Demonstratives, pure indexicals, and essential indexicalsâ•… 134
4. Indexicals as singular termsâ•… 136
5. Indexicals and anaphorsâ•… 137
6. Indexicals and contextsâ•… 148
Intensional logic 153

Paul Gochet
1. The distinction between intension and extensionâ•… 153
2. The principle of extensionality and its failuresâ•… 153
3. The Frege-Carnap treatment of intensional contextsâ•… 154
4. The problem of hybrid contextsâ•… 155
5. Intensional constructions in natural language:
The montagovian paradigmâ•… 156
6. Inadequacies of the standard semantics of intensional logicâ•… 158
7. Hyperintensionalityâ•… 159
Table of contents IX
8. Propositional attitudes and pragmaticsâ•… 160

9. Intension, compositionality and context-dependenceâ•… 161
Modal logic 163

Paul Gochet
1. The development of modal logicâ•… 163
2. Irving Lewis’ contributionâ•… 164
3. What is modal logic all about?â•… 165
4. Quantified modal logicâ•… 166
5. Tense logicâ•… 167
6. From tense logic to pragmaticsâ•… 168
Model-theoretic semantics 171

Paul Gochet
1. The meeting of two different approaches to semanticsâ•… 171
2. The basic notions of Tarski’s semanticsâ•… 172
3. The scope and limits of Tarski’s semanticsâ•… 174
4. Generalized quantifiersâ•… 175
5. The layers of contextsâ•… 176
6. A model-theory for contextsâ•… 177
Charles Morris 180

Susan Petrilli
1. Morris’s behavioristics and pragmaticsâ•… 180
2. Morris’s pragmatics and Peirce’s pragmaticism:
Towards a ‘behavioral semiotic’â•… 185
3. Pragmatic philosophy in the United Statesâ•… 192
4. Pragmatics, signs and valuesâ•… 195
Notation in formal semantics 202

Walter De Mulder
1. Objectivesâ•… 202
2. Principlesâ•… 202
3. The basis: Predicate logic and model theoryâ•… 203
4. Semantic typesâ•… 205
5. Lambda abstraction and lambda conversionâ•… 207
6. Quantifiersâ•… 208
7. Intensionalityâ•… 209
8. Contextsâ•… 212
9. List of some frequently used symbolsâ•… 214
X Philosophical Perspectives for Pragmatics
Phenomenology 217
Peter Reynaert & Jef Verschueren
1. The study of ‘phenomena’â•… 217
2. History and basic tenets of the phenomenological movementâ•… 217
3. Phenomenology, linguistics, and the social sciencesâ•… 219
4. Implications for pragmaticsâ•… 221
Philosophy of action 222

Filip Buekens
Philosophy of language 228

Asa Kasher
2. Natural languageâ•… 230
3. Certain aspects of natural languageâ•… 231
4. Programmatic theoriesâ•… 231
Philosophy of mind 235

Stefaan E. Cuypers
1. Philosophy of mind naturalizedâ•… 235
2. The mind–body problemâ•… 235
3. The problem of intentionality or the status of folk psychologyâ•… 237
4. Psychosemanticsâ•… 239
5. Cognitive pragmaticsâ•… 240
Possible worlds semantics 244

Paul Gochet
1. Some logical problemsâ•… 244
2. The emergence of possible worlds semanticsâ•… 245
3. Key concepts of possible worlds semanticsâ•… 247
4. From possible worlds semantics to pragmaticsâ•… 248
Reference and descriptions 253

Andrea Bianchi
1. Do proper names describe?â•… 253
1.1 Gottlob Fregeâ•… 253
1.2 Bertrand Russellâ•… 256
1.3 John Searleâ•… 258
Table of contents XI
1.4 Taking stockâ•… 260

1.5 Saul Kripkeâ•… 261
2. Do definite descriptions refer?â•… 265
2.1 Russell againâ•… 265
2.2 Peter Strawsonâ•… 269
2.3 Keith Donnellanâ•… 271
3. One further issue: Semantics versus pragmaticsâ•… 274
Truth-conditional semantics 280

Robyn Carston
1. The basicsâ•… 280
2. Short history and key namesâ•… 282
3. Brief comparison with other approaches to linguistic meaningâ•… 283
4. Truth-conditional semantics and pragmaticsâ•… 284
4.1 Indexicalityâ•… 285
4.2 Non-truth-conditional aspects of semanticsâ•… 286
Universal and transcendental pragmatics 289

Joachim Leilich
1. Originsâ•… 289
2. Habermas’ universal pragmaticsâ•… 290
3. Apel’s transcendental pragmaticsâ•… 292
Ludwig Wittgenstein 297

Joachim Leilich
1. Biographical sketchâ•… 297
2. General characteristics of Wittgenstein’s conception of philosophyâ•… 298
3. The Tractatus and the picture-theory of meaningâ•… 299
4. Logical constants and the doctrine of saying and showingâ•… 301
5. Wittgenstein’s later philosophyâ•… 302
6. Influenceâ•… 306
Index 309
Preface to the series
In 1995, the first installments of the Handbook of Pragmatics (HoP) were

published. The HoP was to be one of the major tools of the International Pragmatics
Association (IPrA) to achieve its goals (i) of disseminating knowledge about prag-
matic aspects of language, (ii) of stimulating various fields of application by making
this knowledge accessible to an interdisciplinary community of scholars approaching
the same general subject area from different points of view and with different methodolo-
gies, and (iii) of finding, in the process, a significant degree of theoretical coherence.
The HoP approaches pragmatics as the cognitive, social, and cultural science of
language and communication. Its ambition is to provide a practical and theoretical
tool for achieving coherence in the discipline, for achieving cross-disciplinary intel-
ligibility in a necessarily diversified field of scholarship. It was therefore designed
to provide easy access for scholars with widely divergent backgrounds but with
converging interests in the use and functioning of language, in the topics, traditions,
and methods which, together, make up the broadly conceived field of pragmatics. As
it was also meant to provide a state-of-the-art report, a flexible publishing format was
needed. This is why the print version took the form of a background manual followed
by annual loose-leaf installments, enabling the creation of a continuously updatable
and expandable reference work. The flexibility of this format vastly increased with
the introduction of an online version, the Handbook of Pragmatics Online (see
www.benjamins.com/online).
While the HoP and the HoP-online continue to provide state-of–the-art informa-
tion for students and researchers interested in the science of language use, this new
series of Handbook of Pragmatics Highlights focuses on the most salient topics in the
field of pragmatics, thus dividing its wide interdisciplinary spectrum in a transparent
and manageable way. The series contains a total of ten volumes around the following
themes:
– Key notions for pragmatics

– Pragmatics and philosophy
– Grammar, meaning and pragmatics
– Cognition and pragmatics
– Society and language use
– Culture and language use
– The pragmatics of variation and change
XIV Philosophical Perspectives for Pragmatics
– The pragmatics of interaction

– Discursive pragmatics
– Pragmatics in practice
This topically organized series of paperbacks, each starting with an up-to-date over-
view of its field of interest, each brings together some 12-20 of the most pertinent HoP
entries. They are intended to make sure that students and researchers alike, whether
their interests are predominantly philosophical, cognitive, grammatical, social, cul-
tural, variational, interactional, or discursive, can always have the most relevant ency-
clopedic articles at their fingertips. Affordability, topical organization and selectivity
also turn these books into practical teaching tools which can be used as reading mate-
rials for a wide range of pragmatics-related linguistics courses.
With this endeavor, we hope to make a further contribution to the goals underly-
ing the HoP project when it was first conceived in the early 1990’s.
Jan-Ola Östman (University of Helsinki) &

Jef Verschueren (University of Antwerp)
Acknowledgements
A project of the HoP type cannot be successfully started, let alone completed, Â�without
the help of dozens, even hundreds of scholars. First of all, there are the authors
themselves, who sometimes had to work under extreme conditions of time Â�pressure.
Further, most members of the IPrA Consultation Board have occasionally, and
some repeatedly, been called upon to review contribuÂ�tions. Innumerable additional
Â�scholars were thanked in the initial versions of handbook entries. All this makes
the Handbook of Pragmatics a truly joint endeavor by the pragmatics community
world-wide. We are greatly indebted to you all.
We do want to specifically mention the important contributions over the years
of three scholars: the co-editors of the Manual and the first eight annual install-
ments, Jan Blommaert and Chris Bulcaen were central to the realization of the
project, and so was our editorial collaborator over the last four years, Eline Ver-
sluys. Our sincerest thanks to all of them.
The Handbook of Pragmatics project is being carried out in the framework of the
research program of the IPrA Research Center at the University of Antwerp. We are
indebted to the university for providing an environment that facilitates and nurtures
our work.
Jan-Ola Östman (University of Helsinki) &

Jef Verschueren (University of Antwerp)
Introduction
Marina Sbisà
University of Trieste
1.â•… Pragmatics and philosophy
It is well known that pragmatics – like many branches of the social and even natural
Â�sciences – has its roots in philosophy. Pragmatics is (comparatively) young and its
roots are recent; moreover, like other disciplines concerned with language and com-
munication, its separation from these philosophical roots is incomplete, that is, the
discipline is still at least in part philosophical in character. By this, I mean that prob-
lems in pragmatics are often not only a matter of research within one accepted para-
digm, but also a matter of defining the conceptual framework itself in terms of which
research should be conducted. All this makes it relevant for pragmatic research both
to be aware of the philosophical ideas that have fostered its beginnings and to keep
track of the developments in the philosophical context. In this awareness, we have
selected for publication in this collection, from the various sections of the Handbook
of Pragmatics, a number of articles that provide the reader with materials for pos-
sible routes of reflection at the interface between pragmatics (as a interdisciplinary
field) and philosophy. The Â�collection provides information about some of the philo-
sophical sources of pragmatics (Austin, Grice, Morris) or other authors whose work
is relevant to pragmatic issues (Bakhtin, Foucault, Wittgenstein), about some trends
or areas of research that are in various ways relevant for the definition of the main
concepts of pragmatics (Philosophy of language, Philosophy of action, Philosophy of
mind, Epistemology) and the characterization of the cultural context in which it has
developed (Analytical philosophy, Phenomenology, Hermeneutics, Deconstruction),
about some theories or debates within the philosophy of language and neighbouring
fields that are concerned with pragmatic aspects (Indexical and demonstratives, Ref-
erence and descriptions, Contextualism, Universal and Transcendental Pragmatics,
Epistemology of testimony), and, last but not least, about Truth-conditional semantics
and some main branches of formal semantics (Model-Theoretic Semantics, Possible
worlds semantics, Intensional logic, Modal logic; see also: Notations in formal seman-
tics). If the collection does not comprise articles dealing with such classic topics of
philosophical pragmatics as speech acts, presupposition, implicature, and the seman-
tics vs pragmatics interface, this is precisely because the central role of these issues in
pragmatics made the articles dealing with them indispensable in other volumes of the
same Highlights series.
2 Marina Sbisà
It should be noted that most references of the selected articles concern analytic
philosophy, the brand of philosophy programmatically attentive to clarity in the use of
words, to the details and the logical monitoring of argumentation, and to syntony with
the Â�development of both natural and cognitive sciences, which is most widespread today
in the English-speaking world. Indeed, it is within this trend of philosophical research
that pragmatics was invented and launched. This does not mean that other philosophical
traditions are not relevant to pragmatic research: they are (and our Â�volume is giving vari-
ous hints in this direction, since it comprises e.g. articles on Bakhtin, Deconstruction,
Â�Foucault, Hermeneutics, Phenomenology, Universal and Transcendental Pragmatics).
But the contributions to pragmatic issues coming from these areas of philosophy are
made less visible by their being spread out among different schools of thought, national
traditions, and languages: the common label “Continental philosophy”, by which
they all are sometimes referred to, is even more misleading than the label “Analytical
Â�philosophy” with which it is contrasted (against such geo-linguistic categorizations, see
Glock 2004: 61–88). Moreover, differences in philosophical jargon make it difficult even
to Â�philosophers to build bridges between different so-called “continental” trends, as well
as between the issues they discuss and issues in the analytic tradition.
Of course, our selection from the Handbook of Pragmatics could not aim to be
Â�exhaustive as regards the philosophical trends, methods and doctrines that have been
or may be of interest to scholars doing pragmatics. There would be lots of interesting
things to say about, for example, pragmatics and pragmatism; about especially Â�relevant
authors in the phenomenological tradition, such as Reinach or Schutz; about the philo-
sophical problem of other minds, skepticism about other minds and arguments against
it (a field directly connected with the nowadays popular cognitive-Â�science theme of
the “theory of mind”); on social epistemology; on John Searle’s notion of “collective
intentionality”; on the current main trends in ethics (including meta-ethics and virtue
ethics) and aesthetics (theories of aesthetic judgment, emotions and passions). One
might want to add feminist thought (which tackles a type of differences that are con-
stantly relevant to everyday interaction). Although such extensions are desirable, our
collection as it stands already contains materials enough for getting reflections started
about numerous aspects of the interface between pragmatics and philosophy. I will
now try to outline three main directions of reflection, selected from among the ones
that the articles included in the volume may prompt.
2.â•… Conceptions of meaning
Philosophy has elaborated various conceptions of meaning. These are relevant for
Â�pragmatic research for two main reasons. First, because one of the objects of pragmatic
research is human communicative activity, in its various manifestations, but study-
ing such an activity requires taking for granted that the texts by means of which it is
Introduction 3
performed are meaningful. So whatever meaning may be, pragmatics presupposes its
existence and Â�cannot avoid accepting, explicitly or implicitly, one or other of its defini-
tions. Second, because it has been argued on various occasions that defining meaning
is not a matter of semantics alone, but also of pragmatics.
In the context of analytic philosophy, the mainstream view takes meaning to be
truth-conditional. That is, the meaning of a sentence amounts to its truth conditions,
and the meaning of a sub-sentential phrase to the contribution it makes to the truth
conditions of the entire sentence in which it occurs. Such a view of meaning comes
from Frege’s way of connecting thought and truth, and from Wittgenstein’s Tractatus.
In its neo-positivist, epistemic version, it gave rise to verificationism. Later on, it has
been reinforced by Davidson in his Tarski-inspired view of how we come to understand
a language. It has also been developed by formal semantics into rules for the analysis
of many types of sentence. The later Wittgenstein and ordinary language Â�philosophers
were critical of this view, since they could not believe that it exhausts everything that
has to be said about meaning. In the Wittgensteinian perpective, meaning is the role
played by a linguistic expression in a language game, that is, it is embodied in use. One
could say it consists of the rules for the use of the linguistic expression in question. Such
a perspective, though, tends to confound meaning and force (the kind of function that
the utterance performs). Later on, it was from a philosopher formed in the environment
of ordinary language philosophy, Paul Grice, that there came the proposal to consider
meaning as primarily a matter of a certain kind of speaker intention.
Meaning as speaker intention is not logically incompatible with meaning as truth
Â�conditions. Grice himself thought that what a speaker says with an utterance amounts
to its truth-conditional meaning (while utterances usually imply more than that): his
distinction between saying and implicating allowed him to deal with what is said as
included in speaker meaning. Still, speaker meaning is what a speaker means with
an utterance and does not coincide with sentence meaning. Its relationship with sen-
tence meaning is ambiguous: it is one of contextual application and adaptation if
sentence meaning is felt to be prior; it is a foundational relationship if one tries to
use speaker meaning to explicate sentence meaning. Be that as it may, in the long
run the co-existence within the analytic philosophy of language of the two concep-
tions (speaker meaning and meaning as truth conditions) has triggered the develop-
ment of new, “contextualist” perspectives, according to which it is only in their use as
utterances in a context that sentences have complete truth conditions (and, therefore,
acquire their meaning).
Truth-conditional meaning, meaning as use, and meaning as speaker intention
do not exhaust the conceptions of meaning that may be at issue. Both pragmatism
and structuralism have proposed their own conceptions, the former in terms of the
“interpretant” of a sign (according to Peirce, the reaction of the interpreter to the sign),
including complex dispositions to act, the latter in terms of the (arbitrarily set) role of
the sign in the system of langue. The structuralist conception, that has some aspects in
4 Marina Sbisà
common with the later Wittgenstein’s idea of meaning as role in a language game (in
both cases the systematic aspect of language becomes prominent), has further evolved
into post-structuralist and deconstructionist conceptions in which systematic relations
among meanings turn into meaning shift and deferral, and in which the aim is not so
much to fix or calculate meanings, but rather to destabilize them and let them slide
across chains of signifiers. Here too the original purely semantic association of signifier
and signified turns into something more dynamic and complex, involving enuncia-
tion, intentions of both author and audience, interpretation processes involving both
bottom-up and top-down procedures, all of which count for us as pragmatic factors.
3.â•… Speech as action
There have been in philosophy different ways to realize that words do things. This
theme is obviously relevant to pragmatics, in Morris’s sense because doing something
pertains to an agent, but more generally because doing something with words pertains
to our use of words (in context).
Famously, Wittgenstein introduced a picture of speech activity as consisting of
Â�language games. In this view, the meaning of our words depends on the role they
play in a language game and, therefore, on the rules in accordance to which they are
used. Language games are taken to be (potentially) infinite, that is, there is no fixed
limit to the rules that may be adopted by a speech community for the use of words in
the context of one or more of its forms of life. And forms of life themselves may vary,
although it should be recognized that they all are grounded in more basic ways of
behaving that are common to all mankind. Thus, talk of language games suggests an
approach to speech activity highlighting the variety of uses of language, the public and
rule-governed character of language, and the relationship between ways of speaking
(the “grammar” of a language game) and practical aims within forms of life.
Another well-known way of approaching what we do with words was proposed by
Austin, putting emphasis on those uses of language that amount to the performance of
actions. Austin contrasted action descriptions or reports (“I run”) with utterances that
perform an action or make it explicit what action our utterance should be taken to perform
(“I promise”, “I warn”). The idea that utterances may be, or count as, performances of
actions, once highlighted by the notion of the “performative utterance”, was extended
to speech in general, to be considered (according to Austin) not only as involving acts
of saying something, but also acts performed in and by saying something, i.e. illo-
cutionary and perlocutionary acts. Austin maintained that illocutionary acts can be
successfully performed only if the audience’s uptake is secured, and that their perfor-
mance involves conventional effects. Perlocutionary acts correspond to the speaker’s
responsibility for further effects or consequences on the audience. In Austin’s view,
Introduction 5
while it is correct to trace meaning back to use, it is necessary to make distinctions

among various senses in which we may speak of the uses of language: in particular,
meaning should be distinguished from Â�illocutionary force, while using language to
achieve extralinguistic, non-conventional goals amounts to perlocution. The subse-
quent development of speech act theory owes much to John Searle, who combined
Austin’s view of speech as action with the post-Wittgensteinian view of language as
a rule-governed activity and with greater attention to the speaker’s intentions. Paul
Grice, who together with Austin had participated in the Oxford ordinary language
philosophy movement, focused attention on what a speaker intends to do in uttering
words (basically, making the audience think or intend to do something) and on the
peculiar, complex structure of communicative intentions.
Outside the analytic philosophical environment, other authors have tackled
approximately the same topics. Bühler, Benveniste, Jakobson have in various ways
taken into account the act of uttering or énonciation (seen not as the mere issuing of
an utterance, but as the act of uttering meaningful words) as opposed to the content
of the uttered sentence (énoncé). In this volume, an article on Bakhtin illustrates his
developments of this theme, particularly his view of language as intrinsically dialogic
(that is, involving more agents than the speaker alone) and his analysis of polyphony.
An article on Universal and transcendental pragmatics summarizes how Apel and
Habermas have re-elaborated themes from speech act theory (J. Searle in particular) in
the framework of a broader interest in the normative and ethical aspects of social rela-
tionships. An article on Michel Foucault illustrates his conception of discourse and his
approach to its analysis. Foucault, whose thought has its roots in French structuralism,
assumes the arbitrariness of the word-world relationship and the systematic character
of both language and society. He also detects forms of systematic organization in our
speech activities (discourse formations, but beyond these, the order of discourse), and
investigates how discursive practices relate to social practices, shedding light on the
constraints and transformations to which discourse is subject through and by power
relationships. At the same time, he is aware that discourse itself produces power, and
that while it may reinforce it, it may also reveal its fragilities or contribute to ques-
tioning it. Thus speech, in Foucault’s perspective, enters in a double relationship with
social action: it is shaped by the dynamics of society but also contributes to reinforcing
or reshaping it.
Differences in theoretical framework between those approaches to what words
can do and how they do it, belonging to the tradition of analytic philosophy, and those
closer to the (European) structuralist tradition, are an interesting issue to explore.
Exploring similarities and synergies could be even more fruitful, particularly in rela-
tion to Foucault and Austin: illocution viewed as producing changes in what partici-
pants can or have to do (see Sbisà 1984, 2006: 157–159; 2009: 43–46) basically involves
matters of power.
6 Marina Sbisà
Among these similarities, I would like to underscore one theme: the connection
between knowledge and action (with its background of abilities, competencies, entitle-
ments, or to say it in one word, of pouvoir – the Â�nominalized modal verb usually trans-
lated in this context with the noun “power”). Knowledge and abilities are assimilated
by the later Wittgenstein, while in Austin the notion of knowledge emerges from the
interplay of competence and commitment in making assertions. In Foucault, we find
a triad Â�discourse-knowledge-power in which the connection at issue becomes explicit,
and which may lend support to the Austinian refusal to admit a category of utter-
ances that “only” say (i.e. exert a purely cognitive function, without thereby doing
anything). Recent debates in the epistemology of testimony come close to this same
issue, when they investigate whether and how, in communication, we inherit each
other’s knowledge.
Philosophical questions as to what is action, what is agency, what is responsibility,
always underlie the investigation of speech as action. The replies that one can get from
a philosophy of action largely preoccupied with aligning oneself with naturalism and
materialism, and disregarding its possible connections with the socio-anthropological
study of human interaction, often fail to be helpful.
4.â•… Mind and self
Both philosophy and linguistics in the 20th century have moved back and forth between
mentalism and behaviorism. Morris attempted to construct a behaviorist semiotics (on
a pragmatist basis); Wittgenstein famously argued against the possibility of any “pri-
vate” language in which meaning would be fixed by the powers of mind alone; Quine
assumed a behaviorist conception of language. Grice was perhaps the first to reverse
the anti-mentalist trend by connecting meaning with the speaker’s intentional states
(specifically, his or her communicative intentions): in doing so, he even appeared to
reduce semantics to psychology. Later on, under Chomsky’s influence, mentalism was
rehabilitated not only in linguistics, but also in analytic philosophy, as is testified by
semantic internalism, Fodor’s “language of thought” hypothesis, and (in general) the
increasing attention paid to the philosophy of mind in the last decades of the 20th cen-
tury. In this connection it has been observed that the “linguistic turn” characteristic of
philosophy (particularly, analytic) in the first half of the 20th century was then replaced
by a “representational turn”, that is, a focus on (mental) “representation” (Williamson
2004: 107–108). These transformations had their repercussions, or analogues, in prag-
matics. Those trends of pragmatic research that investigate verbal interaction, relation-
ship between discourse and society, human social and communicative behavior, are
keen to focus on phenomena belonging to the public, intersubjective sphere and may be
concerned with the self as a projection or perhaps internalization of what occurs there.
Introduction 7
Other trends of research, particularly so-called cognitive pragmatics, are mainly con-
cerned with the explanation of facts about the use of language or (in general) human
communication in terms of cognitive processes, internal to the mind and operating on
mental representations. In these studies, societal, intersubjective aspects of the use of
language appear sometimes to be neglected or underrated as mere applications of the
discoveries concerning the cognitive structure of individual minds.
At present in analytic philosophy, mentalism is paired with naturalism, an attitude
towards the mind and its functions that seeks explanations of them in conformity to
the natural sciences. Such an attitude, applied to matters of language and meaning,
tends to trace them back to psychological states and processes. Thus, the effort to get
a clear conception of language is replaced by the search for a good model of utterance
processing. Still, a philosophical perspective on utterance processing is not reducible
to a scientific, psychological one: rather, in the latter there often remain hidden philo-
sophical presuppositions, which may require overt philosophical discussion. What
philosophy can say about utterance processing, moreover, hardly counts as a descrip-
tion of actual processes and is more likely to hold as the rational reconstruction of
how a certain understanding may be arrived at and justified. Indeed, this is the way in
which Grice’s theory of conversational implicature is best understood.
In contrast with analytic philosophy, phenomenological and post-Â�
phenomenological approaches to philosophy are not so much involved in debates
about the naturalization of mind, as in questioning the notion of self (or, perhaps
more broadly, of “subject”), which does no longer appear as something primitive and
epistemically transparent to introspection. Pragmatics, for the most part under the
influence of analytic philosophy (and therefore of its mainstream empiricism and nat-
uralism), tends to assume that speaker and hearer (as individuals, endowed with indi-
vidual minds) are all there is to be examined in order to account for the functions of
utterer (énonciateur), addressee, agent, observer. But these functions do not so simply
and one-to-one correspond to individuals physically present on the scene of the inter-
actional event. They may be shifted to fictional, virtual or institutional domains and
in connection to this, greater awareness of the problematic character of the notion of
subject would be called for. Acquaintance with the ways in which the subject has been
discussed in post-structuralist and post-phenomenological thought may contribute to
the understanding of this complexity. This holds also for an interesting contribution to
the problematization of the subject, which is intermediate between analytic and post-
phenomenological philosophy and is not represented in this volume: Stanley Cavell’s
analysis of interpersonal acknowledgement, showing how it is up to us to acknowl-
edge other subjects and their minds in our everyday interaction or let ourselves be
Â�overwhelmed by skepticism and tragedy.
A theme concerning the mind that cuts across received divisions among approaches
to philosophy concerns the ways in which we acquire knowledge. Â�Philosophers may
8 Marina Sbisà
want to claim that at least part of our knowledge of reality is “direct”, that is, attains
reality itself and is not, for example, merely inferred from our acquaintance with inter-
mediate mental entities; this is the case with Austin and, in some respects at least,
with Wittgenstein. This directness is sometimes extended to meaning (as in the case
of direct reference) and to the way in which we understand it. But many philoso-
phers prefer to assume some kind of mediation and therefore indirectness in cogni-
tive processes as well as in understanding: so, it is often held that knowledge (even at
the perceptual level) necessarily involves inferences, and that understanding is not
just intuitive, or automatically “conventional”, but requires interpretation. Readers will
recognize assumptions of this kind in various philosophical theories reviewed or dis-
cussed in this volume. This theme is relevant for pragmatics in two ways: for those who
believe that it is only inferentially retrieved meaning that pertains to pragmatics, the
issue is how far pragmatics extends; for those who do not make this assumption, the
issue is how much a pragmatically oriented view of language, speech, social interac-
tion, and even human cognition is bound to side with inferential theories of meaning,
force, social cognition and perceptual knowledge.
5.â•… Doing pragmatics, doing philosophy
Taking Wittgenstein’s way of doing philosophy a bit too literally (philosophy as an

activity of clarification!) and assuming, with Austin, some kind of contiguity between
philosophy and the sciences of language, I have often experienced a peculiar ambigu-
ity between research in pragmatics and philosophical reflection. When the pragmatic
investigation of a text, or of a given interactional event, succeeds in clarifying the inter-
subjective dynamics that underlie it, the identities of the participants involved, their
common ground (or failure to share one), their actions, and the like, why shouldn’t it
count as an activity of clarification, and why shouldn’t it be recognized (in this respect
at least) as philosophical? Insofar as its results clearly display, for example, how our
use of language is articulated in a certain area, or how the “grammar” of certain words
was applied and understood on a certain occasion, shouldn’t they be seen as close or
even internal to some kind of philosophical activity? And once it is conceded that phi-
losophy is no longer a matter wholly separated from actual, everyday life (or perhaps
that it never really was), why should its tasks fail to include methodological advice for
the clarification of everyday situations and events, which would amount to putting
forward methodological claims for pragmatics?
Even leaving aside such questions (which might be idiosyncratic of my own way of
understanding both pragmatics and philosophy), there are other senses in which doing
pragmatics comes close to doing philosophy. In pragmatics, we not only pursue clari-
fication (of social and interpersonal relations, of interaction, of implicit Â�meaning…),
Introduction 9
but also do so by means of reasoning and argumentation. The rationality that deploys
itself in the inferential retrieval of implicatures (but also of free enrichment or of com-
mon ground) is perhaps best understood not as the property of a cognitive system,
but as the satisfaction due to the normative requirement that one’s moves be justi-
fied (cf. Grice 1991: 79–83). Moreover, pragmatics’ highlighting interpersonal rela-
tions (speaker-hearer, agent-observer…) and the bottom-up emerging of social bonds
can be seen as a manifestation of the same preoccupation that, in philosophy, has
prompted phenomenological reflections on intersubjectivity as well as Wittgenstein’s
remarks on our attitude towards other minds, Grice’s Cooperative Principle or Searle’s
collective intentionality.
A more traditional way of viewing the relationship between pragmatics and phi-
losophy would be to say that pragmatics needs philosophy in order to clarify its own
core notions. A crucial example is the notion of context. It is still open to discussion
whether (or when) we should adopt a cognitive and internalist conception of context
or a situational, externalist one. Moreover, it is certainly a philosophical problem to
argue for the viability of a conception of context that avoids the inconsistency deriving
from the attempt to speak of the very context in which one is speaking. Indeed, in a
sense what we think and speak about is always and only our context (what is within
reach of our senses and our mind, or what is selected as relevant by the goals of the
activity in which we are engaged). But in another sense, explicit talk of one’s context
can only be done by stepping outside.
Finally, the relationship between pragmatics and philosophy can also be consiÂ�
dered from the other side: indeed, philosophy might sometimes itself profit from
pragmatic research and thus be led to embody bits and results of pragmatic inves-
tigations. By this I mean, for example, that the philosophy of language might find
advantage in testing, sometimes at least, the descriptive and explanatory potential
of its proposals, not against constructed examples but against natural data, using to
this end data and even corpora collected in pragmatic research. And ethics could
exploit pragmatic research on, for example, service encounters (or for that matter,
family dinners) to study the actual functioning of morally significant relationships,
thus avoiding the risks of aprioricity and implausibility, and enhancing its fidelity to
real life (cf. Railton 2004). No trace of this kind of collaborations can yet be found in
this volume, but bringing pragmatics and philosophy closer to each other (or making
them more aware of their vicinity) might indirectly contribute to such developments.
References
Glock, H.-J. (2004). What is analytic philosophy? Cambridge University Press.

Grice, P. (1991). The conception of value. Oxford University Press.
10 Marina Sbisà
Railton, P. (2004). Toward an ethics that inhabits the world. In B. Leiter (ed.) The future of philosophy:
265–284. Oxford University Press.
Sbisà, M. (1984). On illocutionary types. Journal of Pragmatics 8: 93–112.
Sbisà, M. (2006). Communicating citizenship in verbal interaction. Principles of a speech actÂ�
oriented discourse analysis. In: H. Hausendorf, A. Bora (eds), Analysing citizenship talk: 151–180.
John Benjamins.
Sbisà, M. (2009). Uptake and conventionality in illocution. Lodz Papers in Pragmatics 5: 33–52.
Williamson, T. (2004). Past the linguistic turn? In B. Leiter (ed.) The future of philosophy: 106–128.
Oxford University Press.
Analytical philosophy
Ordinary language philosophy
Marina Sbisà
1.â•… Philosophy as analysis
The analysis of concepts has always been an important part of the work of philosophers.
We find brilliant samples of this in Plato and Aristotle, or in Medieval philosophy (e.g.
in the modists), or in modern philosophy (Descartes, Hobbes, the British empiricists,
to name a few). But the analytical activity was not explicitly theorized as central to
philosophy until the rise of the trend in contemporary philosophy, called analytical
philosophy.
The label ‘analytical philosophy’ (nowadays often, more shortly, ‘analytic philoÂ�
sophy’) usually refers to a certain range of philosophical inquiries and doctrines which
have developed from the last decade of the 19th century to present times. These do
not share a fixed number of features, but rather display some ‘family resemblances’.
Most of them belong to the British and American philosophical environments, but
there are remarkable exceptions among the founders: Gottlob Frege was German
and Â�Wittgenstein was Austrian; moreover, the analytical philosophical style has now
spread into many different countries. Most of them share an effort towards a pre-
cise and detailed formulation of philosophical questions as well as towards a careful
argumentative style in philosophical answers. Many of them are connected with the
empiricist tradition, and many of them are committed to the idea that the only propo-
sitions that we know with certainty as true are analytic ones. Many of them conceive of
philosophy as an analytical activity and/or envisage the study of language as necessary
for tackling or perhaps solving philosophical problems.
The last two features are perhaps most important for an understanding of the ori-
gins of the label ‘analytical philosophy’. Modern philosophy has most often presented
itself as building up views of man and of the world, and of the history of these views.
Analytical philosophers have focused on our understanding of what we say and do,
and on the chances and the means we have for controlling our own moves, when we
do philosophy. Thus, they have been engaged in the search for clarity and in careful
argumentation; they have often been attentive to meta-philosophical issues (what is
philosophy? can we ever solve philosophical problems and how?), and have assigned
a prominent role to the study of language.
12 Marina Sbisà
An overview of the main authors and problems of analytical philosophy in the

Â�Twentieth century, in the light of present-day philosophical debates, is provided by
Soames (2003). Two introductions, very different in style but both attentive to the
historical Â�dimension and to the differences between analytical philosophy and other
contemporary trends of philosophical research, are Engel (1997) and Glock (2004).
1.1â•… The ‘linguistic turn’

The beginnings of analytical philosophy have often been identified with a change in the
way of doing philosophy that occurred in the 20th century called the ‘linguistic turn’. As
is observed by Rorty (1967: 15), many philosophers, facing the enormous progresses of
those sciences in which there is cooperation among scientists, who accept each others’
conclusions when they are proved to be true, have started to search for criteria for ratio-
nal agreement in philosophy. With such a goal in mind, they have focused attention
not on man, things in the world, and so on, but on the language we use to speak about
all this. Either an ideal language, or an ideal understanding of our ordinary language,
could avoid misunderstandings among philosophers and promote rational agreement.
This move involves a reconsideration of most traditional philosophical problems
as problems of language. For example, the fundamental ethical question ‘What is
good?’ becomes a question about the meaning of ‘good’, or the way in which ‘good’ can
have meaning. The ontological question ‘What is there?’ becomes a question about
when and how we can refer to objects.
Also other contemporary trends in philosophy, e.g. hermeneutics, conceive of
themselves as stemming from a ‘linguistic turn’. However, in so far as it is related to
analytical philosophy, the linguistic turn can also be characterized as a conception of
thought and of its relation to language. According to Dummett (1988), what distin-
guishes analytical philosophy from other schools is the belief, first, that a philosophical
account of thought can be attained through a philosophical account of language, and,
secondly, that a comprehensive account of thought can only be so attained.
1.2â•… The influence of Frege

Gottlob Frege gave fundamental contributions to formal logic and to the philosophy
of mathematics, in attempting to reduce arithmetic to logic. His conception of logic
lead him to investigate also some notions that are basic for the study of language, such
as the notion of meaning. With respect to meaning, he proposed a distinction between
Sinn (sense), that is, the cognitive content associated with a sign, and Â�Bedeutung
(meaning; usually translated into English, in the Fregean technical sense, as reference),
the object which the sign refers to. He also anticipated the notion of force, the notion
of presupposition, and the problems of the analysis of sentences expressing proposi-
tional attitudes.
Nowadays he is generally held to have laid the foundations for analytical philoso-
phy. At the basis of the analytical Linguistic Turn lies a fundamental conception that
Frege shared with other philosophical writers who used the German language (Bol-
zano, Meinong, Â�Husserl), which Dummett (1975, 1976) has called ‘the extrusion of
thoughts from the mind’. In this conception, a thought is the sense of a sentence and is
grasped by a mental act, but is not itself a mental content, since senses, as opposed to
representations (which are subjective), are communicable and therefore objective. This
doctrine gave a non-psychological direction to the analysis of concepts and proposi-
tions. Among analytical philosophers, this direction was soon to become a linguistic
direction. A decomposition of the sentence has a corresponding decomposition of the
thought it expresses, and therefore, to speak of the structure of a thought is to speak of
the semantic interrelation of the parts of the corresponding sentence.
1.3â•… Analysis in G.E. Moore and B. Russell

At the end of the 19th century, the most widespread kind of philosophy in Britain was
neo-Hegelian Absolute Idealism. G. E. Moore and B. Russell refused to follow it and
took a different path, opposing an “analytic” way of doing philosophy to the idealists’
“synthetic” one and thus contributing to the birth of analytical philosophy (cf. Urmson
1956: 1–5; Hacker 2007: 125 ff.).
Both Moore and Russell, each in his own way, were interested in finding proposi-
tions, the truth of which could be held to be known with certainty. Moore defended
common sense knowledge about matters such as the existence of the external world;
Russell was interested in the propositions of mathematics and logic, and in scientific
knowledge. Both linked their inquiries to the analysis of propositions, the truth of
which they were interested in. Moore thought he could demonstrate that some propo-
sitions of common sense are true with certainty, but he thought also that the analy-
sis of their meaning was a difficult philosophical question. Russell, inspired by the
work of Frege, tackled the analysis of mathematical languages (Russell 1903; Russell &
Â�Whitehead 1913) as well as many logically relevant aspects of language, for example,
the analysis of sentences containing denotative phrases (Russell 1905), which has been
considered as a ‘paradigm of philosophy’ in the analytical sense.
Russell developed his conception of philosophical analysis in connection with his
doctrine of ‘logical atomism’ (Russell 1924). According to this doctrine, sentences are
to be analyzed as built up of parts, some of which name objects in the world. By analyÂ�
zing language, we can find the fundamental constituents of facts, i.e. the logical atoms.
G.E. Moore did not formulate an analytical philosophical program, but rather
practised philosophy as analysis, in a continuous effort to philosophize in a careful and
detailedly argumentative way. In his view analysis is not to be identified with the whole
of philosophy, but is nevertheless a fundamental philosophical activity. In his Lectures
14 Marina Sbisà
on Â�philosophy of the years 1933–34 (Moore 1966), he distinguished two main kinds
of philosophical problems, those regarding what we mean by some particular words,
phrases or forms of expressions in common use, which are a matter of analysis, and
those Â�regarding the world as a whole, some of which can be tackled by philosophical
analysis (i.e. by Â�proposing an analysis of an answer we already take to be true), while
others need other argumentative procedures.
2.â•… Analysis and the ideal of scientific language
Most analytical philosophers in the decades preceding World War II focused attention
on matters such as the relations between philosophical and scientific languages and the
logic underlying the languages of the sciences. Philosophical analysis took as its main
goal the discovery of misleading features in our everyday uses of language as well as
in its philosophical use, and the substitution of supposedly unreliable languages with
logically sound ones.
2.1â•… Wittgenstein’s Tractatus

Ludwig Wittgenstein came to Britain to study engineering, but soon turned to phi-
losophy, under the tutorship of Russell. His peculiar philosophical genius made him
become influential on Russell’s thought (in particular, on his conception of logical
atomism) at least as much as Russell was influential on his. According to Wittgenstein’s
Tractatus logico-philosophicus (1922), propositions can describe facts in so far as they
share their logical form: such a logical form cannot itself be described, it can only be
shown by the proposition itself. Ordinary language often does not allow us to see the
logical forms of its complex propositions. The analysis of complex propositions, how-
ever, yields elementary propositions, the logical form of which is manifest.
Wittgenstein also claimed that the only propositions which state something
meaningful are those of the natural sciences. In his view, the propositions of logic
and mathematics are always true, but do not state anything, because they are tautolo-
gies. By contrast, philosophical propositions are nonsensical, since they try to say what
Â�cannot be subject to description; thus they violate the limits of language.
2.2â•… Rudolf Carnap and the encyclopedia of unified science

In Vienna some philosophers of an empiricist orientation, who were interested in
Â�logical and scientific thought, gathered together with some scientists and mathemati-
cians, and in 1928 formed the Vienna Circle. The group was influenced by thinkers
such as Frege and Ernst Mach, and included as members Moritz Schlick, Rudolph
Â�Carnap and Otto Neurath. They cooperated with an analogous group in Berlin
Â�
(including Hans Reichenbach), and had some cooperation with Wittgenstein too,
drawing some Â�inspiration from his Tractatus, especially with regard to the dichotomy
between descriptive, synthetic propositions and tautological, analytical ones. Thus a
philosophical tendency called logical empiricism or logical positivism was generated.
Logical empiricists formulated a criterion for meaning called the principle of
Â�verifiability, according to which the meaning of a proposition lies in its method of
verification. This principle involved the rejection of metaphysics and a restriction
of analysis, often conceived of as translation or even construction according to explicit
rules, to the language of science. Philosophy became the logic of science, i.e. the logical
analysis of its sentences, terms, concepts, and theories.
In this spirit, a project of an International encyclopedia of unified science was
outlined, which was to be carried out at least in part in the late 1930s in the USA,
after the diaspora (caused by nazism) of the German and Austrian logical empiri-
cists in the English speaking world, and with the cooperation of American pragmatist
Â�philoÂ�sophers such as John Dewey and Charles Morris.
Among logical empiricists, Rudolph Carnap was the most influential on the
Â�subsequent developments of analytical philosophy. In his philosophy of language, he
at first focused attention on logical syntax (1934), characterizing it as a purely formal
theory of language; later on, under the influence of the semantic theory developed
by the Polish logician Alfred Tarski, he tackled semantic problems (1947), propos-
ing a reformulation of Frege’s distinction between sense and reference as a distinction
between the intension and the extension of a linguistic expression, and sketching the
outlines of a modal logic.
The attention to science that was typical of logical empiricism has continued in
Quine’s appeal for the use of ‘canonical’ language (first order logic) in both scientific
and philosophical discourse (see below, §4.1.) and in his proposal of a naturalized
epistemology, in the debates on the mind-body problem and on the naturalization
of intentionality (see below, 4.3.), and in the recent, ongoing collaboration between
Â�philosophers belonging to the analytical tradition and cognitive scientists.
3.â•… Analysis and ordinary language
During the 1930s, in the context of British universities such as Cambridge and Oxford,
a different conception of analysis began to emerge (some have even claimed that it
should no longer be called analysis). The philosophers’ concern began to shift from
reduction, reformulation, and translation to description and elucidation. Correspond-
ingly, the object of analysis shifted from the language of science to ordinary language,
the polemics against metaphysics softened, and more attention began to be paid to
Â�non-assertive uses of language.
16 Marina Sbisà
3.1â•… The evolution of Wittgenstein’s thought

Wittgenstein, teaching in Cambridge since 1929, began changing his philosophical
perspective and criticizing his own former way of doing philosophy. He claimed that
philosophical theories are the effect of “the bewitchment of our intelligence by means
of language”, superstitions produced by misinterpreting the grammar of the language
we use (1953: 47). But, even though ordinary language can be misleading, we do
not have to translate it into another, ideal one; we have to look into its workings and
Â�understand its grammar correctly.
Wittgenstein did not think that philosophers should investigate ordinary language
for its own sake, but as a therapy for the inclination towards philosophizing. Never-
theless, a few notions he formulated are of interest for students of language and of the
relations between language and social life. In particular, he proposed to view Â�language
as consisting of ‘language games’, i.e. rule-governed linguistic activities, rooted in
Â�culturally bound social activities and attitudes or ‘forms of life’ (1953, 1969).
3.2â•… Wittgenstein’s influence and ordinary language philosophy

Wittgenstein’s new ideas, circulating in manuscript form or as lecture notes, immedi-
ately influenced his Cambridge students as well as other philosophers. In Cambridge,
John Â�Wisdom focused attention on a restatement of the nature of philosophical activ-
ity, emphasizing its similarity to an illness and its misleading, although illuminating,
character (Wisdom 1953). In Oxford, Gilbert Ryle searched for the origin of meta-
physical theories in the misinterpretation of the way in which we ordinarily use certain
linguistic expressions (Ryle 1931, 1949). Their influence, together with Wittgenstein’s,
largely contributed to the rise of the philosophical movement known as ordinary lan-
guage philosophy, which Â�flourished in the years after the Second World War.
3.3â•… Some Oxford philosophers

Analysis as elucidation of concepts and attention for ordinary linguistic usage were
already present in the philosophical environment of Oxford University as an effect
of its tradition in classical and Aristotelian studies (Passmore 1957: 449). Like
Â�Wittgenstein, Oxford philosophers linked philosophical enquiries to the investiga-
tion of the Â�workings of ordinary language, but they intended the latter, when correctly
carried out, as giving positive contributions to the former. Therefore, they were inter-
ested in describing the structure and the functioning of language and some of them
gave Â�positive contributions to the philosophy of language and to linguistics.
3.3.1â•… J.L. Austin

John L. Austin can be considered as the most influential representative of postwar
Oxford ordinary language philosophy. His philosophical method, which he once
labeled ‘linguistic phenomenology’ (but others labeled also ‘linguistic botanizing’:

Grice 1989: 376), drew largely on linguistic analysis, to be practised with the help of
such tools as dictionaries. He maintained that language is the instrument of the phi-
losopher and that philosophers ought to check carefully what is ordinarily meant and
implied by the words they use. Moreover, he argued that philosophers can find useful
suggestions (the ‘first word’ in philosophy, although not the last one) in the concep-
tual distinctions that are embodied in ordinary linguistic usage, since these are the
product of a selection which has been developed for centuries by the community who
was using that language and are therefore likely to have a point, while conceptual dis-
tinctions just invented by philosophers are often arbitrary and misleading. In order to
analyze ordinary linguistic usage, he used to ask ‘what you would say when’, namely,
to imagine contextualized discourse sequences in which the use of a certain word or
construction is appropriate (Austin 1956). He believed that philosophy, if practised on
such bases, could become a cooperative enterprise. He viewed philosophy and linguis-
tics as distinct, but as able to stimulate each other’s development (Austin 1958).
Among his main contributions to the philosophy of language is the analysis of
performative utterances (Austin 1946) and its subsequent development into a general
conception of language as action, which gave rise to speech act theory (Austin 1962).
3.3.2â•… P.F. Strawson

Peter F. Strawson tackled major problems in logical theory from an ordinary language
perspective. He argued that formal logic is useful in appraising context-free discourse,
but needs to be supplemented by a logic of everyday discourse, since it is incapable of
coping with the complexities of ordinary speech (1952).
He maintained that a distinction has to be drawn between linguistic expressions
as such and their use, e.g. between sentences and their use in making statements, and
emphasized that a sentence can have a variety of uses. In this framework, he elabo-
rated the notion of presupposition, already sketched by Frege, according to which the
actual existence of the objects to which ordinary denotative expressions refer is not
part of what is asserted by the statement containing the denotative expressions, but
constitutes a presupposition of that statement, so that the statement can be assigned a
truth-value only if the presupposition is satisfied (1950).
Later on, Strawson argued that reliance on a close examination of ordinary lan-
guage was not enough for doing philosophy and undertook the construction of a
‘descriptive metaphysics’, with the aim not of revising the structure of our thought, but
of laying bare its most general features (1959, 1992).
3.3.3â•… H.P. Grice

H. Paul Grice participated in the debates of the Oxford ordinary language philoso-
phers already in the 1940s, but his most relevant contributions to philosophy were to
18 Marina Sbisà
come later. In 1957 he put forward an analysis of the notion of meaning in which he
argued for the primacy of the notion of speaker meaning (what is meant by a speaker
by an utterance on a given occasion) over those of the meaning of an utterance, a
sentence, or a word. According to him, speaker meaning consists in the intention by
the speaker to produce an effect in the hearer by means of the hearer’s recognition of
the intention to produce that effect (Grice 1957). In 1967 he delivered his lectures on
Logic and conversation, in which he Â�propounded something that was going to count as
a new conception of the relations between semantics and pragmatics: the distinction
between what is said by an utterance (conceived of as truth-conditional and depending
on the conventions of language) and what is conversationally implicated (conceived of
as depending on speakers’ and hearers’ expectations and inferences) (Grice 1975).
Throughout the evolution of his thought, Grice never lost sight of the spirit of
Â�ordinary language philosophy in that he kept on searching to elucidate everyday
notions and trying to capture what characterizes ordinary language or which values
are embodied in the philosophical investigation of it (Grice 1989: 376–85).
4.â•… Further developments of analytical philosophy
Not all analytical philosophers accepted the ordinary language approach and, anyway,
after 1960 ordinary language philosophy rapidly declined. In the meantime, the atten-
tion towards the construction of artificial languages had continued, which came to be
seen not only as a device for analyzing the language of the sciences, but as a means
for giving a formal basis to any philosophical investigation about language. Moreover,
since the 1960s, the sharp opposition between constructed formal languages and natu-
ral languages softened, and the tendency arose (perhaps corroborated by the parallel
development of generative grammar) to consider the former as capable to deal with
the latter.
4.1â•… W.V.O. Quine: From analysis to naturalization

The American philosopher Willard V.O. Quine received some main inspirations and
themes of discussion from logical empiricism, but greatly changed the general frame-
work in which they were collocated, by challenging the dichotomy between analytic
and Â�synthetic propositions (Quine 1953). According to him, synonymity as an inten-
sional notion, which was involved in the received definition of analyticity, was too
vague and uncontrollable. He outlined, therefore, a view of language both behavioris-
tic and holistic: meaning assignment (beyond the identification of the perceptual stim-
ulus to which the speaker is reacting) is made to a whole language and even so, suffers
from indeterminacy (Quine 1960). Also verification occurs not for each Â�sentence,
Â�considered in isolation, but for a whole theory of the world.
In this framework, philosophical analysis is no longer possible either as elucida-

tion of concepts or as translation of sentences into synonymous ones. Also, the separa-
tion between philosophy (as analysis) and science (concerned with substantive claims)
is softened or overcome: philosophy purports to accept the results of natural science
and attempts to adopt its methods. Quine continued in his effort to give a clear system-
atization to the language of science, purifying it from intensional language, and con-
sidered a simple formal language (first order predicate calculus) as capable to account
for its essential Â�structure. Ordinary language can be reformulated and regimented in
such a canonical notation, without any claim to preserving sense, but rather with the
aim of achieving an advance.
4.2â•… From intensional semantics to discourse representation theory

Since Carnap’s transition from logical syntax to formal semantics (1947), and inde-
pendently from Quine’s rejection of intensional notions, research on formal languages
paid more and more attention to semantics and took into consideration not only
the extensions of linguistic expressions, but also their intensions. In this framework,
modal logic developed into possible worlds semantics. Important contributions to
this area were given by Saul Kripke, to whom a theory of reference for proper names
is also due, which is opposed to the Fregean distinction between sense and reference
(Kripke 1972).
Another important step in formal logic was taken by Richard Montague, who
attempted to deal with at least some parts of a natural language such as English in a
formalized way. He also proposed to shift the attention from semantics to pragmat-
ics by taking into consideration pragmatic languages, namely, languages which
are to be interpreted by taking into account relevant features of the contexts of use
Â�(Montague 1974). Consideration of intersentential relations has subsequently led to
discourse semantics such as Hans Kamp’s Discourse Representation Theory (Kamp &
Reyle 1993), which differs from received Montague Grammar also because it aims at
psychological plausibility.
4.3â•… Meaning and understanding

From the 1960s on, many philosophers have worked in a Quinean framework. Among
the most influential ones is Donald Davidson, who has combined Quine’s holism
with a reconsideration of Tarski’s semantics. He put forward a conception of meaning
which focuses on the requirements for a ‘theory of meaning’ for a given language. Such
a theory has to allow us to understand any utterance of the language and to be control-
lable without reference to the notion of meaning itself. He argued that it is possible to
construct such a theory for a natural language by drawing on a theory of truth inspired
by Tarski’s theory of truth for formal languages. However, the interpretation of any
20 Marina Sbisà
individual utterance of a language depends on the whole theory of meaning for it, but
since the adequacy of a theory of meaning can be controlled only by reference to the
assent or dissent of speakers to individual utterances, it seems to become impossible
for anybody to understand an utterance in a language or idiolect different from his/
her own. Therefore, in order to account for communication, Davidson developed the
idea of a ‘principle of charity’, which enables us to interpret other people’s utterances
on the basis of the assumption that they hold and express true beliefs (Davidson 1984).
Some criticisms of the Quine-Davidson philosophical perspective on language
have been formulated by Michael Dummett, who has stressed the relationship betweenÂ�
meaning, understanding, and the manifestation of the latter. He argued that a theory
of Â�meaning has to account for our understanding of language, namely, for the implicit
knowledge which enables us to use linguistic expressions correctly; and it has to include
a theory of reference, a theory of sense and a theory of force. He also proposed to link
meaning not to the truth-conditions of sentences, but to their assertibility conditions,
i.e. the conditions that justify the speaker’s assertion of them (Dummett 1975, 1976).
Since understanding an utterance involves grasping its inferential grounds and
consequences, meaning has also been considered in terms of inference rules, giving
rise to finely articulated projects of inferential semantics (see e.g. Brandom 1994).
4.4â•… Philosophy of mind

Since Chomsky showed, against behaviorism, that the mind can be an object of
study, not only as a site where external events are recorded, but as having contents
and structures of its own, also philosophers belonging to the analytical tradition
shifted the focus of their interest from language, its logical form and its public use to
the mind. As early as 1975, Jerry Fodor argued that natural languages are grounded
in a language of thought (Fodor 1975). With the language of thought hypothesis,
meaning became as much an issue in the philosophy of mind as it used to be one in
the philosophy of language. Another example of the shift from language to the mind
is John R. Searle’s discussion of intentionality (1983), following his earlier contribu-
tions to speech act theory (Searle 1969, 1979).
In the area of the philosophy of mind, philosophers in the analytical tradition
have been concerned with the mind-body problem (considering e.g. type identity or
token identity between mental states and brain states) and have discussed the nature
of mental states or attitudes (often in functionalist terms) and of their salient feature,
intentionality (i.e. aboutness, grounded in the directedness of consciousness). Func-
tionalism, assimilating the mind to a program to be run on the human brain, has
been widely debated and criticized (e.g. by John Searle in his Chinese room thought
experiment, purporting to show that a functionalist analysis, being merely syntactic,
leaves the semantic dimension of mental content untouched; see Searle 1984). About
intentionality, it remains open whether it is a fact (intentional realism) or a delusion

(eliminativism) or the result of our interpretive strategies (interpretationism, seeÂ�
Dennett 1987); whether (if a fact at all) it can be exhaustively studied by a naturalistic
approach; and whether the content of intentional states (hence, also the content of
linguistic utterances) is to be identified with internal structures of the mind (as inter-
nalists have it) or is grounded in independent features of the natural world or human
society (as externalists claim, see e.g. Putnam 1975). Principles and processes guaran-
teeing the connection between external states of affairs and their mental representa-
tions have also been widely debated (see Fodor 1987 for a causal theory, Millikan 1993
for a teleological one).
5.â•… Analytical philosophy and pragmatics
The subdivision of semiotics into syntax (regarding the relations among signs), seman-
tics (regarding the relations between signs and what they designate), and pragmatics
(regarding the relations between signs and their users), was theorized by the American
pragmatist philosopher Charles Morris within the project of the International encyclo-
pedia of unified science (Morris 1938; see above, 2.3.). The subdivision was accepted
by Carnap, who claimed that the whole study of natural languages should belong to
pragmatics (1942: 8–12), and that the development of a theoretical pragmatics would
help the student of formal languages in dealing with such concepts as intension and
belief (1955).
A few philosophers and logicians contributed to the study of deixis, which clearly
belongs to pragmatics (according to Morris’s definition, since deixis draws on the
relationship between an utterance and its context, which always comprises an user):
among them, there are Hans Reichenbach, who proposed an analysis of tense (1947)
and Yehoshua Bar-Hillel, who investigated indexical expression (1954). Since the
recent formal investigations of natural languages (see above, 4.2.) cannot neglect the
relation between utterances and contexts, the study of deixis has remained a field in
which logic and pragmatics meet. A remarkable contribution as regards both indexi-
cals and demonstratives is due to David Kaplan, who has distinguished between their
content, depending on the context of utterance, and their semantic value, depending
on the circumstance or possible world with respect to which the sentence uttered is to
be evaluated (Kaplan 1989).
The primacy assigned by logical empiricism to scientific, and thus assertive,
Â�language made it difficult to understand how non-assertive utterances (such as
ethical judgements) can have meaning at all; this suggested to some philosophers
that one should investigate how they function (Stevenson 1944; Hare 1952). Later
on, the emphasis put by Wittgenstein (1953) and by the Oxford ordinary language
22 Marina Sbisà
Â�
philosophers on the many different uses of language, prepared the ground for speech
act theory and, beyond it, for all kinds of applied investigations of the communica-
tive, interpersonal functions of language.
Oxford ordinary language philosophy especially contributed to the development
of pragmatics. The notion of appropriateness owes much to the idea that being false
or even contradictory are not the only faults that an utterance can have and that
there are conditions associated with linguistic forms, which are to be satisfied by
their context of utterance in order for their use to be appropriate. Both the felic-
ity conditions for speech acts (Austin 1962: 12–44) and the notion of presupposi-
tion (Strawson 1950) stemmed from considerations of this kind. Also Grice (1975)
can be considered as attempting to account for appropriateness without giving up
the bivalence of logic (as was suggested by Strawson) (cf. Grice 1989: 3–21). Grice,
moreover, contributed to pragmatics both a speaker-related definition of meaning
and the notion of conversational implicature, which has been applied to a wide range
of linguistic phenomena (from the use of scalar expressions to Â�rhetorical figures).
Robert Stalnaker has taken up Gricean themes in his philosophical work on prag-
matic presupposition (1999).
In the present ‘state of the art’ of analytical philosophy, its identity as a trend of
thought has become somewhat fuzzy. The primacy of philosophy of language over
other areas of philosophy has been shaken. At present, few authors (see e.g. Jackson
1998) maintain that language-based conceptual analysis is a privileged tool for doing
philosophy. Moreover, it is argued that philosophical questions are about the world,
not about language or thought, even if in order to tackle them we have to think about
thought or language (Williamson 2004: 127). In the context of these transformations,
it is difficult to tell what the relations of analytical philosophy to the universe of prag-
matics are going to be. One salient fact is that philosophy has been influenced by cog-
nitive pragmatics, particularly Relevance Theory (Sperber & Wilson 1986; Carston
2002). Another salient fact is that several philosophers have recently focused atten-
tion on the role of context with respect to meaning, communication, and truth, giving
rise to various forms of contextualism. Contextualism in the philosophy of language
ranges from the claim that our language contains far more indexical expressions than
is overtly recognized, to the claim that the propositions we express by uttering sen-
tences often contain elements not spelled out in words but necessary to the truth-
evaluability of the espressed proposition (“unarticulated” constituents, see e.g. Perry
1993), to more radical claims of context-dependency as to the proposition expressed
or the truth of the asserted utterance (see e.g.; Travis 2000; Recanati 2004; MacFarlane
2009) and, finally, to the proposal of a context logic in which the value that conclu-
sions inherit from premises in valid inferences is assertibility in a context rather than
truth (Gauker 2003; it should be noted that Gauker defines assertibility with respect to
situational context as opposed to cognitive context). Philosophers who maintain that
criteria for the attribution of knowledge change with context have argued for contex-
tualism in epistemology too (see de Rose 1999). However, I would like to recall that
independently of their taking sides for one or other philosophical claim about lan-
guage or mind, philosophers in the analytical tradition keep discussing such problems
as reference, force, attitude attribution, interpersonal or cross-cultural understanding,
all of which are at the same time objects of pragmatic research. This will most likely
keep on creating opportunities for fruitful encounters between analytical philosophy
and pragmatics.
References
Almog, J., J. Perry & H. Wettstein (eds.) (1989). Themes from Kaplan. Oxford University Press.
Austin, J.L. (1946). Other minds. Proceedings of the Aristotelian Society (Suppl.) 20: 148–87.
——— (1956a). A plea for excuses. Proceedings of the Aristotelian Society 57: 1–30.
——— (1956b). Ifs and cans. Proceedings of the British Academy 42: 109–32.
——— (1961). Philosophical papers. Oxford University Press.
——— (1962). How to do things with words. Oxford University Press.
Bar-Hillel, Y. (1954). Indexical expressions. Mind 63: 359–79.
Brandom, R. (1994). Making it explicit. Harvard University Press.
Carnap, R. (1934). Logische Syntax der Sprache. Wien.
——— (1942). Introduction to semantics. Harvard University Press.
——— (1947). Meaning and necessity. University of Chicago Press.
——— (1955). On some concepts of pragmatics. Philosophical Studies 7: 89–91.
Carston, R. (2002). Thoughts and Utterances. Blackwell.
Davidson, D. (1967). Truth and meaning. Synthese 17: 304–23.
——— (1973). Radical interpretation. Dialectica 27: 313–28.
——— (1984). Truth and interpretation. Oxford University Press.
de Rose, K. (1999). Contextualism: an explanation and defense. In J. Greco & E. Sosa (eds.): 187–205.
Dennett, D. (1987). The Intentional Stance. The MIT Press.
Dummett, M. (1975). What is a theory of meaning? (Part 1). In S. Guttenplan (ed.) Mind and
Â�language: 97–138. Oxford University Press.
——— (1976). What is a theory of meaning? (Part 2). In G. Evans & J. McDowell (eds.) Truth and
meaning: 67–137. Oxford University Press.
——— (1988). The origins of analytical philosophy. Lingua e Stile 23: 3–49, 171–210.
Engel, P. (1997). La dispute. Une introduction à la philosophie analytique. Minuit.
Fodor, J.A. (1975). The Language of Thought. Cromwell.
——— (1987) Psychosemantics. The MIT Press.
Frege, G. (1879) Begriffsschrift. Nebert.
——— (1892) Über Sinn und Bedeutung. Zeitschrift für Philosophie und philosophisches Kritik
100: 25–50.
——— (1918). Der Gedanke. Beiträge zur Philosophie des deutschen Idealismus 1(2): 58–77.
Gauker, C. (2003). Words without Meaning. The MIT Press.
Glock, H.-J. (2004). What is analytic philosophy? Cambridge Univiersity Press.
Greco, J. & E. Sosa (eds.) (1999). The Blackwell Guide to Epistemology. Blackwell.
24 Marina Sbisà
Grice, H.P. (1957). Meaning. The Philosophical Review 66: 377–388.

——— (1975). Logic and Conversation. In P. Cole & J.L. Morgan (eds.) Syntax and Semantics, vol. 3:
41–58. Academic Press.
——— (1989). Studies in the way of words. Harvard University Press.
Hacker, P.M.S. (2007). Analytic philosophy: beyond the linguistic turn and back again. In M. Beaney
(ed.), The analytic turn. Analysis in early analytic philosophy and phenomenology: 125–141.
Routledge.
Hare, R.M. (1952). The language of morals. Oxford University Press.
Jackson, F. (1998). From Metaphysics to Ethics. Oxford University Press.
Kamp, H. & U. Reyle (1993). From Discourse to Logic. Kluwer.
Kaplan, D. (1989). Demonstratives. In J. Almog, J. Perry & H. Wettstein (eds.): 481–563.
Kripke, S. (1972). Naming and necessity. In D. Davidson & G. Harman (eds.) Semantics of natural
language (2nd ed.): 253–355. Reidel.
——— (1980). Naming and necessity. Blackwell.
McFarlane, J. (2009). Nonindexical contextualism. Synthese 166: 231–50.
Millikan, R.G. (1993). White Queen Psychology and Other Essays for Alice. The MIT Press.
Montague, R. (1974). Formal philosophy. Yale University Press.
Moore, G.E. (1925). A defence of common sense. In J.H. Muirhead (ed.) Contemporary British
Â�philosophy (Series 2): 193–223. Allen & Unwin.
——— (1959). Philosophical papers. Allen & Unwin.
——— (1966). Lectures on philosophy. Allen & Unwin.
Morris, C.L. (1938). Foundations of the theory of signs. University of Chicago Press.
Neurath, O., R. Carnap & C.L. Morris (eds.) (1938). Foundations of the unity of science. University
of Chicago Press.
Passmore, J. (1957). A hundred years of philosophy. Duckworth.
Perry, J. (1993). The Problem of the Essential Indexical and Other Essays. Oxford University Press.
Putnam, H. (1975). Mind, language and reality. Cambridge University Press.
Quine, W.V.O. (1953). Two dogmas of empiricism. In W.V.O. Quine, From a logical point of view.
Harvard University Press.
——— (1960). Word and object. MIT Press.
Recanati, F. (2004). Literal Meaning. Cambridge University Press.
Reichenbach, H. (1947). Elements of symbolic logic. Macmillan.
Rorty, R. (ed.) (1967). The linguistic turn. University of Chicago Press.
Russell, B. (1903). The principles of mathematics. Cambridge University Press.
——— (1905). On denoting. Mind 14: 479–93.
——— (1924). Logical atomism. In J.H. Muirhead (ed.) Contemporary British philosophy (Series 2).
Allen & Unwin.
——— (1956). Logic and knowledge. Allen & Unwin.
Russell, B. & A.N. Whitehead (1913). Principia mathematica. Cambridge University Press.
Ryle, G. (1931). Systematically misleading expressions. Proceedings of the Aristotelian Society 32:
139–70.
——— (1949). The concept of mind. Hutchinson.
Searle, J.R. (1969). Speech acts. Cambridge University Press.
——— (1979). Expression and meaning. Cambridge University Press.
——— (1983). Intentionality. Harvard University Press.
——— (1984). Minds, brains and science. Harvard University Press.
Soames, S. (2003). Philosophical Analysis in the Twentieth Century. Vol. 1: The Dawn of Analysis.
Vol. 2: The Age of Meaning. Princeton University Press.
Sperber, D. & D. Wilson (1986). Relevance. Blackwell. [2nd edition 1995].
Stalnaker, R. (1999). Context and Content. Oxford University Press.
Stevenson, C.L. (1944). Ethics and language. Yale University Press.
Strawson, P.F. (1950). On referring. Mind 59: 320–44.
——— (1952). Introduction to logical theory. Methuen.
——— (1959). Individuals. Methuen.
——— (1971). Logico-linguistic papers. Methuen.
——— (1992). Analysis and metaphysics. Oxford University Press.
Tarski, A. (1936). Der Wahrheitsbegriff in den formalisierten Sprachen. Studia Philosophica 1: 261–405.
——— (1956). Logic, semantics, metamathematics. Oxford University Press.
Travis, C. (2000). Unshadowed Thought. Harvard University Press.
Urmson, J.O. (1956). Philosophical analysis. Oxford University Press.
Williamson, T. (2004). Past the linguistic turn? In B. Leiter (ed.) The future of philosophy: 106–128.
Wisdom, J. (1953). Philosophy and psycho-analysis. Blackwell.
Wittgenstein, L. (1922). Tractatus logico–philosophicus. Routledge & Kegan Paul.
——— (1953). Philosophische Untersuchungen/Philosophical investigations. Blackwell.
——— (1969). Über Gewissheit/On certainty. Blackwell.
John L. Austin
Marina Sbisà
1.â•… J.L. Austin and his approach to philosophy
J.L. Austin (1911–1960) turned to philosophy after a training in classics. From 1933
he worked at the University of Oxford, first as a research Fellow, then as a Fellow
and tutor, and later on as White’s Professor of Moral Philosophy. During the Second
World War he served in the Intelligence Corps of the British Army, making remark-
able contributions to the organization of D-Day (Warnock 1969). After the war and
until his death in 1960, he greatly influenced the philosophical debate in Oxford by
his personal approach to ordinary language philosophy, both in his academic activi-
ties and by promoting informal discussions in which his method was experimented
with (Warnock 1973a). He published only a few papers during his lifetime; he also
translated Frege’s Grundlagen der Arithmetik into English (Frege 1950). The collection
of his papers (Austin 1979) and two volumes containing two series of lectures (Austin
1962; Austin 1975) were published posthumously.
1.1â•… Austin’s philosophical method

Austin was partly influenced by the echoes of Wittgenstein’s teaching in Cambridge
(he must have read the manuscript of Wittgenstein’s Blue Book, 1933–34), and by the
kind of philosophical analysis practised by G.E. Moore. He was also reacting to neo-
positivism, particularly to the version of it proposed in Oxford by Alfred J. Ayer, and
re-elaborating themes discussed by Oxford philosophers of the previous generation,
such as H.A. Prichard. Also, the Oxford tradition in classic and Aristotelian studies,
which involved a practice of analysis as elucidation of concepts and required attention
to ordinary linguistic usage, contributed to shaping his approach to philosophy.
Like Wittgenstein (in the later phases of his philosophy), Austin linked philo-
sophical investigations not to the construction of artificial languages but to the
consideration of the workings of ordinary language, and did not restrict his consider-
ations to the scientific or descriptive language, but was interested in the whole range
of language uses. However, unlike Wittgenstein, he believed that the investigation
of ordinary language could not merely correct philosophical mistakes (e.g. prevent
pseudo-problems from arising), but also provide substantial contributions to the
discussion and possibly the solution of philosophical issues.
John L. Austin 27
Austin’s philosophical method proposed to analyze ordinary linguistic usage by asking

“what we should say when, and so why and what we should mean by it” (1979: 181).
The procedure required imagining contextualized discourse sequences in order to see
whether the use in them of a certain word or construction is appropriate or not. The
method had to be practised with the help of such tools as dictionaries: the lexical
distinctions involved were sometimes so subtle as to earn for the method the label of
“linguistic botanizing” (Grice 1989: 376). Austin attempted to justify his philosophical
method along the following lines. He argued that language is the instrument of the
philosopher and that philosophers ought to carefully check what is ordinarily meant
or implied by the words they use. Moreover, he argued that philosophers can find use-
ful suggestions in the conceptual distinctions embodied in ordinary linguistic usage,
since these are the product of a selection which has been made for centuries by the
community using that language and are therefore likely to have a point, while con-
ceptual distinctions simply invented by philosophers are often arbitrary and mislead-
ing. According to him, therefore, ordinary language should be considered as the “first
word” in philosophy, though not the last one.
Against this proposal, charges of conservativism and of irrelevance have been lev-
elled (Gellner 1959; Graham 1977). Austin’s method seems conservative because it
often leads to preferring the plain man’s views, which are embodied in ordinary lan-
guage, to the philosopher’s conceptual innovations. Moreover, it is hard to say whether
all the usages Austin proposed to investigate were philosophically relevant, and his
method itself does not provide the means for making this distinction. Both charges
stem from a misunderstanding. Austin aims at making certain ideas explicit, which
are implicit in language as a system, and to which we are bound simply because we use
that language; he does not thereby accept the whole range of common sense beliefs
or prejudices typical of his social group, which have many other sources than lan-
guage, and are not so much embodied in language, as formulated by means of it. (It is
therefore no surprise that sometimes Austin defended philosophical theses running
contrary to common sense, as for example in his sharp distinction between knowl-
edge and belief, or, in his philosophy of language, in the analysis of assertion as one
kind of conventional action). As to irrelevance, Austin deserves to be reproached for
not caring enough about explaining the relevance of his investigations of ordinary
usage, but this does not mean that these are irrelevant. Most of the distinctions he
draws are implicitly connected with philosophical issues or with the analysis of some
philosophically interesting notion (some examples will be found in Sections 2 and 4
below). A polemic remark of his which bears on this issue (“I am not sure importance
is important: truth is”, 1979: 271) is not meant as a defence of irrelevance, but mani-
fests his intention to bring philosophy to say things that can be deemed true, much
like sciences do, avoiding the temptation to raise claims simply because of their
Â�relevance with respect to some existential or ideological interest. In fact, he believed that
28 Marina Sbisà
philosophy, practised according to his method, could become a cooperative enterprise

and lead to discoveries.
1.2â•… “Linguistic phenomenology”

A fuller understanding of Austin’s method would require a reassessment of its phenoÂ�
menological character. He once proposed to label it “linguistic phenomenology”
(1979: 182), but this proposal has been largely disregarded: it did not seem credible
that a philosopher so deeply rooted in the British tradition would borrow method-
ological instruments from continental philosophy. However, it would be consistent
with most of Austin’s work to take him as proposing a phenomenological reduction
or epoché, not identical but parallel to Husserl’s (1928). Just as Husserl aimed at iso-
lating consciousness, by suspending any judgement about external reality, so Austin
aimed at isolating language. He wanted to look at concepts through the way in which
they manifest themselves in language, rather like a phenomenologist who looks at
things through the way in which they manifest themselves to us in consciousness.
Thus he says that when practising his philosophical method “we are using a sharpened
awareness of words to sharpen our perception of (…) the phenomena” (1979: 182; see
Di Giovanna 1989). In fact, even when Austin speaks of the relationship between
words and world, he is speaking of the ways in which the words-world relationship
is represented in language (1979: 134). It is interesting to notice that, like Austin and
unlike Wittgenstein, Husserl granted informative value to philosophical discourse.
1.3â•… General tendencies

Austin’s thought displays also some recurrent tendencies, which are connected to his
linguistic approach to philosophy but which do not directly depend on it.
i. He refuses oversimplifications. While philosophy usually reduces the complex-

ity of reality by resorting to unifying abstraction, by operating reductions, or
introducing symmetrical, supposedly exhaustive dichotomies, Austin is suspi-
cious of “logical constructions” (among which the notions of truth and of state-
ment), and takes sides against reductionism (Austin 1970: 48 objects in this way
to both materialism and dualism: “Why, if there are nineteen of any thing, is it
not philosophy?”). When concerned with the discussion of dichotomies, such as
“sense-datum” and “material thing”, or “performative” and “constative” (see below
Sections 2.2. and 3.2.), his strategy constantly consists of trying to show that philo-
sophical oversimplification has vitiated both terms and that the dichotomy has to
be reformulated as a bundle of gradual distinctions.
ii. He takes structured wholes, not atomic elements, as primitive. Thus he maintains
that the object of his philosophy of language is “the total speech act in the total
John L. Austin 29
speech situation” (1975: 52), he analyzes reference as primarily belonging not to

names but to the whole statement containing them (1979: 122), and proposes a
view of perception as regarding not isolated objects, but objects within contexts
(1962: 53).
iii. Finally, Austin typically focuses on marginal or non-standard cases in order to
throw light on standard ones. He begins his investigation of speech acts from
the study of infelicities, and proposes to study action through the ways in which
responsibility for it can be disclaimed or aggravated. Thus he exploits the sharp-
ened awareness of a situation produced when it is not smooth so as to bring out those
features of similar situations, which go unnoticed in smooth cases. This feature of his
thought has been mistakenly disregarded by Derrida (1977) in his criticism of speech
act theory (see also Cavell 1995).
2.â•… Epistemology
Austin provided an analysis of the notion of knowledge which contrasts it with belief,
and a discussion of perception grounded in the way it is talked about in ordinary
language.
2.1â•… Knowledge and belief

In order to distinguish knowledge from belief, Austin distinguishes between different
kinds of challenges to an assertion. If the assertion is considered as issued on the basis
of knowledge of how things are, it can be challenged by asking “How do you know?”,
i.e. by asking the speaker to specify how he or she has come to be in a position to
know about that matter, or on the basis of what criteria he or she formulates his or
her judgement. If the assertion is considered as implying belief, it can be challenged
by asking “Why do you believe?”, i.e. what are the reasons which lead the speaker to
hold that belief (1979: 76–81). If the speaker’s reply is unsatisfactory, the former chal-
lenge leads to denying that he or she knows, the latter leads to saying that he or she
ought not to believe, but not to denying that he or she holds that belief. Therefore, Austin
views belief as a mental state to be expressed or described, knowledge as the subject’s
being entitled to issue a certain claim. Claiming to know something equals giving others
one’s word that things are so, on the basis of one’s own entitlement to say that things are
so (1979: 97–103). In this view the acceptance of the inherent fallibility of the human
intellect and senses does not lead to scepticism, but to emphasizing the responsibility of
speakers for their entitlement to issue assertions and raise claims to knowledge.
This view of Austin’s has been criticized because it has been taken as disregarding
the difference between knowing something and saying “I know” (Graham 1977: 130 ff.;
30 Marina Sbisà
Warnock 1989: 39). In the light of the interpretation of Austin’s theses given above, as
well as of the phenomenological reading of his linguistic method (cf. Section 1.2.), such
a criticism appears beside the point.
2.2â•… Perception
As to perception, Austin takes sides against those theories which view it as consist-
ing of the entertaining of sensations from which we conclude by inference that there
are material objects, and he takes sides in particular against Ayer’s approach (1940).
Â�Austin does not maintain that we do perceive material things all the time, but contends
that the dichotomy which opposes sense-data to material things is spurious, since
there are many different kinds of things we may be said to “perceive” and perceiv-
ing something which is not to be called a “material thing” (such as for example a
rainbow) does not always amount to being deceived by the senses (1962: 8–9). He
analyzes arguments in favour of the sense-data theory of perception, paying atten-
tion to the language in which they are cast and observing that this language involves
philosophical usages of everyday terms or expressions which play a misleading role.
He argues that cases of delusion often quoted by sense-data theorists are not in prin-
ciple indistinguishable from cases of veridical perception, as is alleged, since nothing
is perceived in isolation, and the context of an experience usually makes discrimina-
tion possible: when we see the stick as bent (while it is straight), we also see the water
in which it is immersed (1962: 53). He claims that there is no dichotomy between
one sense of “perceive” by which we say we perceive material objects and another by
which we say we perceive sense-data, but that we normally describe, identify, and
classify what we perceive in lots of different ways. Consistent with what we have
described above as his refusal of reductionism (cf. 1.3.), he claims that utterances
such as “I see a silvery speck” and “I see a huge star” can both be true, since the sil-
very speck is a huge star (1962: 98–99).
Austin also disputes the idea that sense-data statements are incorrigible and there-
fore the foundations of knowledge. He argues that it is possible to misdescribe sensa-
tions because of inattention or failure to discriminate, but also that when a subject is
in the very best possible position to make a statement, even about a material object, he
or she is entitled to have complete confidence in it.
Austin’s way of dealing with perception has given rise to debates (Ayer 1969;
Â�Forguson 1969a), has been considered defective because it does not consider causal
theories of perception (theories which represent perception as the causing of Â�sensations
by a material object, such as the one put forward in Grice 1961) (Warnock 1989: 30–31),
and has been reassessed in terms of “natural realism” by Putnam (1994). Soames
(2003) has discussed its anti-skeptical potential (for Austin’s anti-skepticism, see also
Kaplan 2008).
John L. Austin 31
3.â•… Philosophy of language
Several aspects of Austin’s work are relevant to the philosophy of language.
3.1â•… Meaning
In a Wittgensteinian vein, he criticized the question “What is the meaning of a word?”
as being a spurious generalization from the specific questions we ask about the mean-
ings of particular words, which can be answered by explaining the use of each word,
while the general question cannot (1979: 55–62). He claims that there are many dif-
ferent ways in which words have meaning (denoting an object is only one of these)
and that we call different kinds of things by the same name according to a number
of strategies including analogical relations, chains of resemblances and paronymy (a
peculiar way of relating different senses of one word which Austin derives from Aris-
totle, Categories ) (1979: 69–75). A case of paronymy is the word “healthy”, which
may be used to mean a property of a living body, but has also related senses such as
“productive of healthy bodies” (as when it is predicated of food) or “resulting from a
healthy body” (1979: 27). In his analysis of the word “real” (in expressions such as “a
real x”) (1962: 65–76) he made use of the idea that there must not be one quality com-
mon to all the cases to which the same word is legitimately applied and claimed that
the uses of “real” are meant to exclude, in various kinds of situations and for differ-
ent kinds of objects, different ways of not being authentic (being fake, dyed, artificial,
dummy, toy,..).
3.2â•… Performative utterances

One of Austin’s most famous contributions to the philosophy of language is his notion
of the performative utterance. He first discussed the performative utterance “I prom-
ise” in the attempt to compare it to claims to knowledge (in which, he claimed, the
speaker “gives his word” that things are so) (1979: 97–103; see above, §2). He claimed
that utterances such as “I promise” (at the first person singular, present indicative
tense), in the appropriate circumstances, do not describe the action the speaker is
doing, but function as ritual phrases by which the speaker performs the action. Not
being descriptive, they cannot be lies (although they can imply lies), and are therefore
not to be assessed as true or false.
Austin submitted the notion of performative utterance to further scrutiny in his
How to Do Things with Words (1975). Here, it is claimed that utterances of other syn-
tactic forms besides the original performatives can be said to perform an action, pro-
vided they are “reducible, or expandible, or analysable into a form, or reproducible in a
form” (1975: 61–62) which is explicitly performative, by using a verb in the first person
singular, present indicative active, to make explicit the action the speaker performs in
32 Marina Sbisà
issuing them. Some ambiguities in Austin’s view of performatives, among which that
between “doing” something and “making explicit” that something is being done, have
been discussed by Warnock (1973b) and Urmson (1977).
3.3â•… Assertion and truth

Austin was closely connected to Fregean themes as regards meaning (which he intended
as “sense and reference”, 1975: 100), assertion, and truth. His work in this area can
be viewed as consisting of polemic responses to Fregean claims. According to Frege
(1918), the sense of a declarative sentence is a thought (endowed with a truth-value);
asserting amounts to judging a thought to be true, and truth is not a matter of cor-
respondence because it cannot be a matter of degree. Austin (1979: 117–133) argues
for a theory of truth as correspondence which does not involve any structural sim-
ilarity between language and the world, but requires a match between “descriptive
conventions” linking the sentence to a type of situation and “demonstrative conven-
tions” linking the actual statement to a historic situation in the world. According to
this view, saying that a statement is true is saying something about that statement
(and is therefore saying something other than what is said in asserting the statement).
Finally, according to Austin, truth/falsity assessments are (at least to some extent) a
matter of degree, as is shown by the use of other words besides “true” and “false”, such
as “rough”, “general”, “exaggerated”, in assessments about words-world correspondence
(1979: 129–30; 1975: 143–145).
Austin’s views about truth have been criticized by P.F. Strawson and a complex
debate has ensued (Pitcher (ed.) 1964). A perceptive reading of his views is provided
by Travis (2006: 173–198), who argues that they are not bound to positing “truth-
makers” to be found in the world, and emphasizes that, in an Austinian perpective,
questions of truth are questions not of calculation but of merit, to be settled according
to different standards in different occasions.
3.4â•… The speech act

Austin developed his notion of the performative utterance into a general conception of
language as action, giving rise to speech act theory.
In How to Do Things With Words (1975) he starts from a restatement of the con-
trast between performative utterances and assertions (or “constatives”). He discusses
at length the felicity conditions of performatives (those socially accepted conditions,
which speakers and circumstances must satisfy in order for the utterance to succeed
in performing the action it is designed to perform), considers the criteria for recognizing
a performative utterance and various intermediate cases between performatives and con-
statives, and concludes that the dichotomy has to be abandoned. He then proceeds to for-
mulate a new framework for the study of language, based on the notion of speech act.
John L. Austin 33
A few critics have misunderstood Austin’s intents and have taken him as actually seek-
ing to isolate the performative utterance from the constative, and failing to do so (e.g.
Black 1969); speech act theory, in this interpretation, appears as a mere side-effect of
the failure of the pursuit of performatives. More correctly, Warnock (1989: 106–107) has
pointed out that Austin explored the hypothesis of a performative/constative dichotomy in
order to show why the new start on the problem, i.e. speech act theory, was needed. Austin
chose to examine performatives as a special case in which the operative nature of language
is particularly apparent, in order to show that all language is action and to use the analy-
sis of performatives as a guideline for the analysis of one of the facets of linguistic action
which is common to all kinds of utterances (assertions included), the illocutionary act (for
a Â�reading of How to Do Things with Words so oriented, see Sbisà 2007, 2009).
In discussing the speech act, he distinguished three main facets, that is, three main
ways in which it can be described as a “doing” (1975: 91–120):
i. the act of saying something or locutionary act, to be further analyzed into the act
of uttering certain noises (phonetic act), the act of uttering noises of certain types,
conforming to and as conforming to certain rules (phatic act), and the act of using
the words uttered with a certain meaning (rhetic act);
ii. the act of doing something “in” saying something or illocutionary act, which may
be exemplified by such acts as promising, ordering, warning, asking, thanking,
and stating, all performed, according to Austin, on the basis of conventions and
taking effect in conventional ways;
iii. the act of doing something “by” saying something or perlocutionary act, which
may be exemplified by such acts as persuading, alerting, getting someone to
do something, all consisting of the production of psychological or behavioural
Â�consequences by means of an utterance.
These distinctions have been taken up, but also deeply transformed in meaning, by the
ample literature on speech act theory, starting from Searle (1969).
4.â•… Philosophy of action
Another area of philosophical problems to which Austin contributed is the philosophy

of action. He investigated central notions in this field, such as those of freedom and of
responsibility, as well as the notion of action itself. His investigations did not lead to
final conclusions. His paper on “Excuses” (1979: 175–204), a topic which he tackled
following suggestions from Aristotle’s Nichomachean Ethics (Friggieri 1991: 136–145),
is still largely programmatic and two related papers (1979: 253–287) appear to be just
two small parts of a larger research project. This makes it difficult to give his views on
the matter precisely.
34 Marina Sbisà
4.1â•… Action
Austin’s analyses show that he connected the notion of action with that of what is done
(the effect) and was inclined to identify actions on the basis of their effects, altough
he recognized some role to intentions too. Thus, saying that someone has done some-
thing (has performed a certain action) amounts to ascribing to him or her some kind
of responsibility for what he or she has brought about. Austin’s analysis of the speech
act, too, relies on these ideas. Without taking them into account, a notion such as that
of the perlocutionary act cannot be properly understood. The perlocutionary act can
be an action, as Austin claims (1975: 105–107), only if it is regarded as consisting not
of a gesture by the speaker (there is no such gesture), but of the bringing about (on the
part of the speaker) of the receiver’s reaction to the speech act.
Austin was aware of the complexities of the relationship between action identifica-
tion and action description. Usually, various descriptions are true of the behavior of
a certain person at a certain time: if someone has killed a donkey by shooting it, we
may say that he has moved his finger, pulled the trigger, fired a gun, shot the donkey,
or killed it. It is often taken for granted that he or she performs one action, which can
be described in different ways. But we could also consider each description as cor-
responding to a different action. Austin proposed to rediscuss the criteria for action
identity in the light of the procedures for action description and ascription (1979:201).
The problem of whether the different descriptions quoted above individuate dif-
ferent actions or are merely different descriptions of the same action has subsequently
been tackled by Goldman (1970) (who held that each description identifies a different
action) and by Davidson (1980) (who concluded that the different descriptions are
to be considered as descriptions of the same action, which is to be individuated by
reference to the agent’s movements). According to Friggieri (1991), Austin shares with
Davidson the conviction that the consequences of an action are not part of the action
itself, but this is not enough to conclude (as Friggieri does) that he would endorse
Davidson’s identification of actions with basic actions, that is, physical movements.
Goldman’s proposal seems to be closer to Austin’s refusal of reductionism (§1.3. above)
and to his doubts about the usefulness of referring to bodily movements in order to
identify actions (1979: 179; 1975: 112). Recent discussions of Austin’s perspective on
action, emphasizing its connection with the notion of responsibility and its moral and
legal implications, are Sneddon (2006), Yeager (2006).
4.2â•… Freedom and responsibility

As to freedom, Austin put forward an analysis of past tense usages of “could” and of
the kinds of conditional clauses associated with “can” and “could”, which attempted to
show, against other philosophers among which G.E. Moore, that belief in determinism
cannot be considered as compatible with the way in which we ordinarily speak. He
John L. Austin 35
also maintained that the study of the ways in which we act un-freely, as manifested in
the language of excuses, is a useful approach to the understanding of freedom.
He held that together with “freedom”, “responsibility” is to be considered a key
term in the theory of action. In fact, his investigation of the language of excuses and
aggravations (1979: 189–203, 272–87) is prevailingly concerned with ways of mak-
ing, withdrawing or refining ascriptions of responsibility. The idea underlying these
investigations is that one can throw light on someone’s responsibility for an action, as
well as on the structure of actions (their “stages”: “the intelligence, the appreciation,
the planning, the decision, the execution and so forth”, 1979: 201), by examining rec-
ognizedly “non-normal” cases such as those in which responsibility is represented as
either extenuated or aggravated (1979: 180; Forguson 1969b).
Austin’s remarks about extenuating and aggravating expressions have sometimes
been deemed irrelevant (Graham 1977: 219–228). But let us consider, by way of an
example, his distinction between “by mistake” and “by accident” (1979: 185). If one
accepts to reflect about usages seriously, the former expression turns out to be linked
with the idea of something’s going wrong in the agent’s behaviour, the latter with the
idea of some unexpected external circumstance. The agent’s responsibility is clearly
affected in different ways: in the former case the agent may be held responsible for not
monitoring his or her own behaviour appropriately (there can be culpable mistakes),
in the latter he or she can be reproached at most (if at all) for not foreseeing the occur-
rence of the relevant external circumstance. Different components of responsibility
are at stake.
5.â•… Austin and pragmatics
Together with Grice and Searle, Austin is one of the philosophers who has most influ-
enced the birth and development of pragmatics. The aspect of his thought which has
had most influence is his speech act theory, usually considered in isolation from the
rest of his philosophy. But a correct understanding of his speech act theory requires
placing it within the context of his other claims, especially those regarding the philoso-
phy of language and the philosophy of action.
His views on assertion and truth, particularly the idea that statements (as speech
acts) refer to situations and the claim that truth-assessments are context-sensitive,
have influenced situation semantics (Barwise & Perry 1983; Barwise & Etchemendy
1987) and contextualism (e.g. Recanati 2004; Travis 2000). François Recanati (2007)
has employed a notion of Austinian proposition inspired by Barwise’s reading of
Austin’s theory of truth in the framework of his “moderate relativism”.
Finally, it should be recognized that Austin has insightfully explored many notions
and practices which are of interest to pragmaticians: think e.g. of his distinction between
36 Marina Sbisà
excuse and justification (1979: 175–178; see § 4.2), his discussion of the different
implications of using “know” as opposed to “believe” (see § 2.1), his analyses of per-
ceptual reports (see § 2.2) or of pretence (1979: 253–271). So, his work can be a source
of inspiration independent of agreement, or disagreement, about more general
philosophical claims and arguments.
References
Austin, J.L. (1962). Sense and Sensibilia. Oxford University Press.

——— (1970). Intelligent behaviour. A Critical Review of The Concept of Mind. In O.P. Wood &
G. Pitcher (eds.) Ryle. A Collection of Critical Essays: 48–51. Doubleday.
——— (1975) [1962]. How to Do Things with Words. Oxford University Press.
——— (1979) [1961]. Philosophical Papers. Oxford University Press.
Ayer, A.J. (1940). The Foundations of Empirical Knowledge. Macmillan.
——— (1969). Has Austin Refuted Sense-Data? In K.T. Fann (ed.): 284–308.
Barwise, J. & J. Perry (1983). Situations and Attitudes. The MIT Press.
Barwise, J. & J. Etchemendy (1987). The Liar: an essay on truth and circularity. Oxford University Press.
Berlin, I. et al. (1973). Essays on J.L. Austin. Oxford University Press.
Black, M. (1969). Austin on Performatives. In K.T. Fann (ed.): 401–411.
Cavell, S. (1995). What did Derrida want of Austin? In Philosophical Passages: Wittgenstein, Emerson,
Austin, Derrida: 42–65. Blackwell.
Davidson, D. (1980). Essays on Actions and Events. Oxford University Press.
Derrida, J. (1977). Signature Event Context. Glyph 1. In Limited Inc: 172–97. Galilée.
Di Giovanna, J. (1989). Linguistic Phenomenology. Peter Lang.
Fann, K.T. (1969). Symposium on J.L. Austin. Routledge.
Forguson, L.W. (1969a). Has Ayer Vindicated the Sense-Datum Theory? In K.T. Fann (ed.): 309–341.
——— (1969b). Austin’s Philosophy of Action. In K.T. Fann (ed.): 127–147.
Frege, G. (1918). Der Gedanke. Eine logische Untersuchung. Beiträge zur Philosophie des deutschen
Idealismus 1(2): 58–77.
——— (1950). The Foundations of Arithmetic. Blackwell.
Friggieri, J. (1991). Actions and Speech Actions in the Philosophy of J.L. Austin. Mireva.
Gellner, E. (1959). Words and Things. Gollancz.
Goldman, A.I. (1970). A Theory of Human Action. Prentice Hall.
Graham, K. (1977). J.L. Austin. A Critique of Ordinary Language Philosophy. Harvester.
Grice, P. (1961). The Causal Theory of Perception. Proceedings of the Aristotelian Society Suppl.
Vol. 35: 121–68. (also in Grice (1989): 224–247).
——— (1989). Studies in the Way of Words. Cambridge University Press.
Husserl, E. (1928). Ideen zu einer reinen Phänomenologie und phänomenologischen Philosophie. Halle.
Kaplan, M. (2008). Austin’s Way with Skepticism. In J. Greco (ed.), Oxford Handbook of Skepticism.
Oxford: Oxford University Press.
Pitcher, G. (ed.) (1964). Truth. Prentice Hall.
Putnam, H. (1994). The Importance of Being Austin: the Need for a “Second Naiveté”. Journal of
Philosophy 91: 466–487.
Recanati, F. (2004). Literal Meaning. Cambridge University Press.
John L. Austin 37
——— (2007). Perspectival Thought. A plea for (moderate) relativism. Oxford University Press.
Sbisà, M. (2007). How to read Austin, Pragmatics 17: 461–473.
——— (2009). Uptake and conventionality in illocution, Lodz Papers in Pragmatics 5: 33–52.
Searle, J.R. (1969). Speech Acts. Cambridge University Press.
Sneddon, A. (2006). Action and responsibility. Springer.
Soames, S. (2003). Philosophical Analysis in the Twentieth Century. Vol 2: The Age of Meaning.
Â�Princeton University Press.
Travis, C. (2000). Unshadowed Thought. Harvard University Press.
——— (2006). Thought’s footing. Oxford: Oxford University Press.
Urmson, J.O. (1977). Performative Utterances. Midwest Studies in Philosophy 2: 120–127.
Warnock, G.J. (1969). J.L. Austin, a biographical sketch. In K.T. Fann (ed.): 3–21.
——— (1973a). Saturday Mornings. In I. Berlin et al.: 31–45.
——— (1973b). Some types of performative utterances. In I. Berlin et al.: 69–89.
——— (1989). J.L. Austin. Routledge.
Wittgenstein, L. (1958). The Blue and Brown Books. Blackwell.
Yeager, D. (2006). J.L. Austin and the law. Bucknell University Press.
Mikhail Bakhtin
Martina Björklund
Åbo Akademi University
Mikhail Mikhailovich Bakhtin (1895–1975) is sometimes referred to as literary

scholar, sometimes as philologist, semiotician, linguist, culturologist, or philosopher.
In his own view, he was a philosopher – or thinker. In an interview held in the spring
of 1973, Bakhtin agrees with the interviewer that he was more of a philosopher than
a philologist in the 1920s, adding: “And such have I remained until the present day.
I am a philosopher. A thinker.”, (cited from Emerson 1997: 6). I will here concentrate
on Bakhtin’s linguistic thinking, but first a brief sketch of the course of his life (for
bioÂ�graphies, see Clark & Holquist 1984; Konkin & Konkina 1993) and of the whole
industry that has developed around his person and work.
1.â•… Biographical sketch
Bakhtin was born in Orel (south of Moscow), but he grew up in Vilnius and Odessa.
He studied at the Classics Department of the University of Petrograd 1914–1918. Then
he moved first to Nevel in western Russia and in 1920 to nearby Vitebsk. He worked as
a schoolteacher and participated in philosophical study circles, where he met, among
others, Valentin Vološinov and Pavel Medvedev. The only piece of writing that Bakhtin
published during these years was a short article on the philosophy of authorship entitled
‘Iskusstvo i otvetstvennost’, which appeared in 1919 in the Nevel journal Den’ iskusstva,
‘The day of art’, (English translation: ‘Art and answerability’, in Bakhtin 1990).
In 1924 Bakhtin returned to Leningrad. Most of his friends from the time in Nevel
and Vitebsk were also now in Leningrad. According to Clark & Holquist (1984: 100),
Bakhtin emerged more and more as the leader of the circle. More recently, it has, how-
ever, been claimed that Bakhtin’s achievement has been exaggerated in the established
representations of the activity of the so called Bakhtin Circle, and, conversely, that the
thinkers with whom Bakhtin associated in the 1920s have been downgraded. In light
of the assessment of the original contributions made by Vološinov, Medvedev and less
well-known members, the Circle emerges not as a set of followers of one central fig-
ure, but as a dynamic confederation of independent thinkers (Brandist 2002, Brandist,
Shepherd & Tihanov (eds.) 2004).
From 1924 to 1929 several of the books that would later bring Bakhtin fame were
written, including the so called disputed texts, the most famous of which are two books
Mikhail Bakhtin 39
published under the names of Medvedev (Formal’nyi metod v literaturovedenii, 1928;

English translation: The formal method in literary scholarship, 1978) and of Vološinov
(Marksizm i filosofiia iazyka, 1929; English translation: Marxism and the philosophy
of language, 1973). It has been claimed that Bakhtin is practically the sole author of
these books. This claim has struck some scholars as questionable and the question of
the authorship of the texts has been disputed (see, Clark & Holquist 1984: 146–170;
Morson & Emerson 1990: 102–119; Slavic and east European journal 1986). In view
of the controversy, I will only refer to these books in support of ideas that are also
expressed in Bakhtin’s own texts. Under his own name Bakhtin published the book
Problemy tvorchestva Dostoevskogo in 1929 (‘Problems of Dostoevsky’s creative art’,
translation of title by Morson & Emerson). After that no more of Bakhtin’s writing was
to be published until the 1960s.
In 1929, Bakhtin was arrested, and he was eventually sentenced to six years in
internal exile in Kustanai in Kazakhstan. After that he worked for a year in Saransk
(about 400 miles east of Moscow) at the Mordovian Pedagogical Institute. In 1937,
probably to escape possible rearrest in the great purge, he moved to Savelovo (north
of Moscow), where he stayed until the end of the Second World War. During the
1930s and early 1940s, Bakhtin wrote his famous essays on the theory of the novel,
including Slovo v romane, 1934–1935 (English translation: Discourse in the novel
1981), the long essay on the chronotope, 1937–1938 (English translation: Forms of
time and of the chronotope in the novel, 1981) in addition to a dissertation on Rabe-
lais, which he submitted to the Gorky Institute of World Literature in 1940. Because
of the War, the public defense could not take place until 1946. After great contro-
versy, Bakhtin was eventually given the lower degree of Candidate in 1952, but not
that of Doctor.
After the war Bakhtin returned to Saransk. He was head of the Department of
Literature and taught courses in Russian and world literature until his retirement in
1961. The essay ‘Problema rechevykh zhanrov’, 1952–1953 (English translation: ‘The
problem of speech genres’, 1986), and the notes on the problem of the text in linguis-
tics, 1959–1961 (English translation: 1986) were written during this period.
In the early 1960s a group of young scholars at the Gorky Institute, who had come
across Bakhtin’s 1929 Dostoevsky book and the Rabelais dissertation, discovered to
their great surprise that the author was still alive. They visited Bakhtin in Saransk and
helped him to get his works published. A second reworked edition of the Dostoevsky
book Problemy poètiki Dostoevskogo came out in 1963 (English translation: Problems
of Dostoevsky’s poetics, 1973) and the Rabelais dissertation Tvorchestvo Fransua Rable
i narodnaia kul’tura srednevekov’ia i Renessansa in 1965 (English translation: Rabelais
and his world, 1968). Both created a sensation in the Soviet Union. In 1969 Bakhtin
was brought to the Moscow area, but not until 1972 was he permitted to move into
a Moscow apartment. He led a quiet life, saw people, read proofs for new editions
40 Martina Björklund
of earlier works, and wrote new essays. He died on March 7, 1975. Later that year a
Â�collection of his writings spanning the years 1924–1941 was published.
2.â•… The ‘Bakhtin industry’
With the rediscovery of Bakhtin and the publication of his works in the 1960s and
1970s a Bakhtin boom started in the Soviet Union. By the time of his death Bakhtin
was already the object of a cult. In the late 1960s Bakhtin had been introduced to west-
ern scholars, first through French scholars (notably J. Kristeva and T. Todorov) and
French translations. A French Bakhtin boom already flourished while the master was
still alive. After his death, the Bakhtin cult spread from Paris to the United States and
to the rest of Europe. Since the 1980s an international Bakhtin boom has flourished.
Bakhtin’s works have been translated into English and other languages, monographs
and biographies have appeared, meetings of the International Bakhtin Society have
been held, but without Soviet delegates until 1991. Bakhtinian thinking and Bakhtin-
ian concepts have been used, not always successfully, by ‘everybody’, first by literary
scholars, later by linguists, folklorists, and philosophers. The influence of Bakhtinian
thinking has also spread into social theory and psychology (Marková 2003). More-
over, different versions of ‘Bakhtin’ appear in different cultures and research traditions,
due to diverse readings imposing their own grids on Bakhtin’s oeuvre. Thus versions
of, for instance, the American ‘Bakhtin’ or of the French ‘Bakhtine’ may consider-
ably differ from Russian versions; or versions by literary scholars may differ from ver-
sions by semioticians or linguists. To take just one example, American literary scholars
Gary Saul Morson and Caryl Emerson throughout their book (Morson & Emerson
1990) present Bakhtin’s critique of Saussurean linguistics and the view of language
as a system (see, e.g. p. 170) in such terms that their book may produce the impression
that Bakhtin was totally hostile to systemic thinking. Russian semiotician and linguist
Viacheslav Ivanov (1998: 741–742) writes on the same topic in a completely different
tone. First, he commends Bakhtin for being the first to establish the fundamental dis-
tinction between the sign as an element of a system and the sign in a concrete utterance,
and for showing that the former view is not adequate for the study of concrete utterances.
Ivanov then goes on to praise Bakhtin for the concept of speech genres, which, according
to him, unites the two complementary approaches to the study of language.
Until the 1990s more large-scale writing on Bakhtin took place in the West than in
Russia. Another posthumous collection of Bakhtin’s works had appeared in the Soviet
Union in 1979 and the early philosophical text K filosofii postupka was published in
1986 (English translation: Toward a philosophy of the act, 1993). In the 1990s, how-
ever, we witness a veritable Russian ‘Bakhtin industry’. The first full-length biography
in Russian (Konkin & Konkina 1993) appeared. Journals devoted to Bakhtin’s legacy
Mikhail Bakhtin 41
were launched. An abundance of more modest provincial publications, proceedings,

and conference series have been published. Bakhtin’s collected works (M. M. Bakhtin:
sobranie sochinenii) in seven volumes are under preparation (Volumes 1, 2, 4(1), 4(2),
5, and 6 appeared in 1996–2010). Also in journals, previously unpublished material
continues to appear. For a more detailed description of the Russian Bakhtin boom and
a review of the reception of Bakhtin and his work in Russia, from the earliest times to
the centennial jubilee in 1995, see Emerson (1997).
3.â•… Bakhtin’s view of language
Bakhtin was not first and foremost a linguist. Ivanov (1998) characterizes Bakhtin’s
sphere of interest as philosophical anthropology, including all basic forms of human
activity, although it was the study of literature that received most of Bakhtin’s attention.
Thus Bakhtin occupied himself with language only in so far as this was needed for the
theoretical basis of his other studies. When advancing his ideas about language and
language use, Bakhtin was much ahead of his time, and most of his observations are
still relevant to linguists, especially to those interested in pragmatics. In fact, Bakhtin
was engaged in linguistic pragmatics before there was a name for this kind of study.
From the late 1950s, he himself termed the type of linguistic study he was interested in
metalinguistics (cf. below).
Bakhtin’s insistence on the need to consider language as it is manifested in con-
crete whole utterances used in given contexts dates back to 1924, when he wrote an
article entitled ‘Problema soderzhaniia, materiala, i formy v slovesnom khudozhest-
vennom tvorchestve’ (first published in Bakhtin 1975) and translated into English as
‘The problem of content, material, and form in verbal art’ (in Bakhtin 1990). This article
is critical of formalism similarly to the book The formal method in literary scholarship
later published under Medvedev’s name. A major concern of the article is the place of the
language of linguistics (in those days, language as an abstract system in the Saussurean
sense1) in the study of artistic texts. Thus according to Bakhtin, the language of linguis-
tics is no more than a technical moment for poetry. To him, an artistic text, just like any
other text, is an utterance and needs to be viewed as a whole in its context.
A single, concrete utterance is always given in a value-and-meaning context, whether it
be scientific, artistic, poetical, etc., or in the context of a situation from everyday personal
life. Each separate utterance is alive and has meaning only within these contexts: it is
true or false, beautiful or ugly, sincere or deceitful, frank, cynical, authoritative, etc. –
there are no neutral utterances, nor can there be. (Bakhtin 1990: 292)
ïœ±.â•… De Saussure is, however, not mentioned by name in the 1924 article or the 1929
Dostoevsky book.
But linguistics in the 1920s was poorly equipped for the study of utterances. Not only
was it preoccupied with system rather than concrete living utterances; the study of
language had also not advanced further than the sentence. Bakhtin envisaged a whole
programme for further linguistic studies. Purely linguistic analysis should be extended
beyond the sentence to “large verbal entities – long utterances from everyday life, dia-
logue, speech, treatise, novel, and so on” (Bakhtin 1990: 293). Bakhtin further main-
tained that linguistics would only be able to contribute to the study of the aesthetics
of verbal art when it had mastered the whole of its subject;2 and that, conversely, the
achievement of such a goal would enable linguistics to profit from literary studies. The
type of large verbal entity, or, as he would say later, speech genre that Bakhtin was to
occupy himself with most intensely during the next two decades was the novel. It is
remarkable that after almost 90 years, Bakhtin’s programme is not outdated. On the con-
trary, what Bakhtin asked for in the 1920’s is what is now being done by a large number
of linguists; and, moreover, many linguists today profit from Bakhtin’s literary studies.
In what follows, I will outline the main tenets of Bakhtin’s thoughts aboutÂ�
language, discussing in brief the most famous Bakhtinian catchwords. For discussions
of the intellectual sources of Bakhtin and the members of the Bakhtin Circle, see e.g.
Brandist 2002, 2003 and Lähteenmäki 2004.
3.1â•… Dialogue
The dialogic conception of language was central to Bakhtin’s thought. The word
Â�‘dialogue’ is already mentioned in the 1924 article, but only in the sense of one of the
types of large verbal entities that awaited linguistic study. The specifically Bakhtinian
use of dialogue as a term for his conception of language dates from 1929 (both the
Dostoevsky book and Vološinov’s book). Bakhtin was, however, neither very precise
in the definitions, nor consistent in the use of his terms, and dialogue (dialogization,
dialogism, sometimes double-voicing) is employed in at least three senses, which can
be viewed as dialogue at different levels. First, to begin from the widest sense, dialogue
stands for a view of the whole human existence in the world.
To be means to communicate dialogically. When the dialog is finished, all is finished.
[…] One voice alone concludes nothing and decides nothing. Two voices is the
minimum for life, the minimum for existence. (Bakhtin 1973: 213)
Second, and consequently, every utterance is dialogic, not only the speeches in everyday
dialogue, but also long written texts, such as novels (Bakhtin 1981: 280, 1986: 68–69).
This is so because, to begin with, real understanding is responsive in that the listener/
2.â•… Cf. Jakobson’s (1960) contention that poetics can be kept apart from linguistics only when
the field of linguistics is illicitly restricted, for instance, to the sentence as the highest analyzable
construction.
Mikhail Bakhtin 43
reader agrees or disagrees, arguments, etc. with what s/he hears or reads. Such actively
responsive understanding is the initial preparatory stage of a response, which may
follow immediately or emerge later in subsequent speech or behavior (‘understanding
with a delayed reaction’). Moreover, every speaker is her/himself a respondent, pre-
supposing the existence of preceding utterances, which s/he may build on, polemicize
with, or presume as known to the reader. Hence, “[a]ny utterance is a link in a very
complexly organized chain of other utterances” (Bakhtin 1986: 69). A special case of
links in the chain of utterances is formed by the dialogical relations into which utter-
ances on the same theme enter, even if they belong to different people who know
Â�nothing about one another (Bakhtin 1986: 114–115).
The third sense of dialogue concerns the level of individual discourses (or words,
Bakhtin uses the Russian word slovo, which normally means ‘word’, but also ‘discourse’)
and allows some utterances to be dialogic, while others are more or less monologic
(i.e. in this third sense; in the second sense they are always dialogic). The issue is here
whether words (or other linguistic expressions) that are used in the utterance are felt
to be someone else’s words or not. A basic distinction is thus made between single-
voiced discourse (odnogolosoe slovo) and double-voiced discourse (dvugolosoe slovo;
see Bakhtin 1973: 153–168). Double-voiced discourse is also said to be internally diaÂ�
logized (Bakhtin 1981: 326). Such discourse makes use of, among other things, what
Bakhtin calls hybrid constructions:
…a hybrid construction is an utterance that belongs, by its grammatical (syntactic)
and compositional markers, to a single speaker, but that actually contains mixed within
it two utterances, two speech manners, two styles, two ‘languages’, two semantic and
axiological belief systems. (Bakhtin 1981: 304)
3.2â•… Heteroglossia
The hybrid construction containing two ‘languages’ brings us to another Â�Bakhtinian
catchword, namely heteroglossia (raznorechie, literally ‘varied-speechedness’).
By heteroglossia Bakhtin means the stratification of any language into different socio-
ideological ‘languages’ (Bakhtin 1981: 271–272), which are forms for conceptualizing
specific world views (pp. 291–292). An “internally dialogic potential” is said to be
embedded in linguistic heteroglossia (p. 326). Dialogized heteroglossia (p. 273) is the
result if we “regard one language (and the verbal world corresponding to it) through
the eyes of another language” (p. 296). Heteroglossia should thus not be confused with
polyphony (see below).
3.3â•… Polyphony
Polyphony is another Bakhtinian term that has to do with dialogue. According
to Â�Morson & Emerson (1990: 230) polyphony is one of Bakhtin’s most intriguing
Â�
concepts. According to Bakhtin himself in a letter to Vadim Kozhinov (one of the
young scholars that rediscovered Bakhtin), the concept of polyphony has given rise to
much bewilderment (see Bakhtin 1979: 404). One reason for this is that Bakhtin never
explicitly defines polyphony. The notion is introduced in the 1929 Dostoevsky book
to capture the “artistic key” to Dostoevsky’s novel (Bakhtin 1973: 13) and it is used by
Bakhtin exclusively about a certain type of novelistic form. According to Bakhtin, this
form was invented by Dostoevsky, but not limited to his works. Basically, polyphony
has to do with the position of the author in the novel, the relationship between char-
acter and author.
In Dostoevsky’s polyphonic novel the important thing is not the ordinary dialogical
form of unfolding the material within the limits of its monological conception against
the firm background of a unified material world. No, the important thing is the
final dialogicality, i.e. the dialogical nature of the total work. As we have said, the
dramatic work is, in this sense, monological, and Dostoevsky’s novel is dialogical.
It is not constructed as the entirety of a single consciousness which absorbs other
consciousnesses as objects, but rather as the entirety of the interaction of several
consciousnesses, of which no one fully becomes the object of any other one.
(Bakhtin 1973: 14)
In such a novel, by the very construction, “the author speaks not about the hero, but
with him” (Bakhtin 1973: 52). To this end all elements of the novel are dialogized and
the use of double-voiced discourse is predominant. In the linguistic literature, how-
ever, the notion of polyphony has taken on a life of its own and is used, as far as I can
see, to cover different types of double-voicing (see Roulet 1996).
3.4â•… Metalinguistics
To recapitulate, Bakhtin was primarily interested in concrete utterances and the dia-
logic relations into which they enter. Although such a viewpoint fell outside linguistics
in Bakhtin’s days, for a long time he apparently hoped that his outlook on language
could be implanted in linguistics (see Bakhtin 1996: 541–542). Eventually, giving up
this hope, he introduced the term metalingvistika (‘metalinguistics’) in his notes ‘The
problem of the text’ (1959–1961) to cover the study of “concrete forms of texts and
concrete conditions of the life of texts, their interrelations, and their interactions”
(Bakhtin 1986: 114).3
In many texts, Bakhtin discusses the difference between, on one hand, the view
of language as an abstract system (the linguistic view) and, on the other hand, the
3.â•… The term was most probably borrowed from Whorf (although its semantics was modified);
at least the entry ‘Whorf B. Collected Papers on Metalinguistics, Washington 1952’ is included in
a bibliography dating from the time Bakhtin was writing his notes (see, Bakhtin 1996: 641–642).
Mikhail Bakhtin 45
view of language in use by real speakers in concrete contexts (the metalinguistic view).
Bakhtin time and again demonstrates the limited use of the view of language as a
“stable system of normatively identical forms” (Vološinov 1986: 67, emphasis in the
original) for the type of study he was interested in. However, in my reading of Bakhtin,
he did not “lay aside” linguistics as “flawed and misleading” (as argued in Morson &
Emerson 1990: 123). On the contrary, Bakhtin (1973: 150) said that metalinguistics
could not ignore linguistics, and that it should use the results of linguistics (for a simi-
lar interpretation of Bakhtin’s view of linguistics, see Lähteenmäki 2003). I will outline
the main points in the Bakhtinian view on the linguistic versus the metalinguistic
aspects of language.
The most conspicuous point in the Bakhtinian view of language is that he consid-
ers the language system studied by linguists as being of marginal relevance to speakers
of their native language (or of other languages that they master at near-native level).
This point is first extensively argued for in the Vološinov book (1986: 65–71), which
is highly critical of Saussurean linguistics. In a discussion of the sense in which an
abstract system of language could be said to exist at all, the existence of such a system
as an objective fact is denied. Instead it is argued that if language is observed from
above, disregarding any subjective, individual consciousness, “no inert system of self-
identical norms” (p. 66) would be discovered, but rather “the ceaseless generation of
language norms” (ibid.) or “ceaseless flow of becoming” (ibid.). It is further argued that
a synchronic system of language may be said to exist only from the point of view of a
subjective consciousness. Such a system is arrived at by way of abstraction in view of
some theoretical or practical goal, for instance, in view of the study of defunct, alien
languages, or of language instruction.
“The ceaseless generation of language norms” is further elaborated on in Discourse
in the novel, where Bakhtin discusses the tension between what he calls “common
unitary language” and heteroglossia. According to Bakhtin (1981: 269–271), such a
unitary language is a system of linguistic norms which is the result and “theoreti-
cal expression of the historical process of linguistic unification and centralization”
(p. 270). Hence, it is not something given (dan) but something posited (zadan), and it
is constantly opposed to the realities of heteroglossia. Unitary language is present “as
a force for overcoming this heteroglossia, imposing specific limits to it, guaranteeing
a certain maximum of mutual understanding and crystalizing into a real, although
still relative, unity – the unity of the reigning conversational (everyday) and literary
language, ‘correct language’” (ibid.). Thus, according to Bakhtin, language is unitary
“only as an abstract grammatical system of normative forms, taken in isolation from
the concrete, ideological conceptualizations that fill it, and in isolation from the unin-
terrupted process of historical becoming that is characteristic of all living language”
(p. 288). Alongside the forces of centralization and unification, then, the processes of
decentralization and disunification are also constantly at work (p. 272), and thus at
any point in the history of its existence, “language is heteroglot from top to bottom”
(p. 291). Even literary language itself is stratified (p. 288); and, consequently, its unity
“is not a unity of a single, closed language system, but is rather a highly specific unity
of several ‘languages’” (p. 295).
But for the individual consciousness, Bakhtin argues, language is not an abstract
grammatical system of normative forms, “but rather a concrete heteroglot conception
of the world” (p. 293). As argued in Vološinov (1986: 67), for the immediate purposes
of speaking, “[t]he speaker’s subjective consciousness does not in the least operate with
language as a system of normatively identical forms”. In Bakhtin’s view (1986: 87; see
also, Vološinov 1986: 70), the abstract system of language is thus not the place where
speakers normally turn to find words, when constructing utterances. Rather, they rely
on utterances previously encountered in their linguistic experience, i.e. utterances by
other people and their own previous utterances. (This results in dialogic relations of
the kind that are covered by Kristeva’s term intertextuality, defined in the following
way: “any text is constructed as a mosaic of quotations; any text is the absorption and
transformation of another”; see Kristeva 1980: 66.) Hence, for the speaker, linguistic
forms exist only in the context of specific utterances. We do not “say or hear words, we
say and hear what is true or false, good or bad, unimportant and important, pleasant
and unpleasant, and so on” (Vološinov 1986: 70). The criterion of correctness (what
conforms to the language system) is brought to the fore only in abnormal and special
cases, such as instances of conflict (typically associated with writing) and in language
instruction.
This Bakhtinian view of the knowledge of language and of speech production/
reception by actual speakers has been convincingly developed by Gasparov (1996) in
terms of kommunikativnye fragmenty (‘communicative fragments’) which enter into
the vast conglomerate of communicatively loaded linguistic elements that constitutes
the ‘linguistic memory’ of a speaker.
In the notes on the problem of the text, Bakhtin (1986: 105) compares the two
complementary aspects from which language could be studied: the abstract system
studied by linguistics and the concrete utterance which should be studied by metaÂ�
linguistics. He talks about ‘two poles of the text’. The first pole has to do with every-
thing that conforms to the language system, viz. everything “that is repeated and
reproduced, everything repeatable and reproducible, everything that can be given out-
side a given text” (ibid.). The second pole, then, has to do with the text as an utterance,
which is individual, unique, and unrepeatable. This aspect “inheres in the text itself,
but is revealed only in a particular situation and in a chain of texts” (ibid.). It is thus
linked “not with elements (repeatable) in the system of language […], but with other
texts (unrepeatable) by special dialogic […] relations” (ibid.). Bakhtin (ibid.; see, also
Vološinov 1986: 67–68) thus contends that what is important in the production and
reception of actual utterances or texts is not what conforms to the language system,
Mikhail Bakhtin 47
but what is the new and concrete meaning which the linguistic form acquires in its
particular context, and which, in addition, always has some relation to value, viz. to
truth, goodness, beauty, etc. Herein lies the whole significance of the text, the purpose
for which it was created. From this aspect the language system proves to be no more
than underlying “material, a means to an end” (Bakhtin 1986: 105).
The opposition linguistic–metalinguistic also comes to the fore in ‘The problem
of speech genres’, in a lengthy discussion (see, Bakhtin 1986: 67–100) of the difference
between units of language (words and sentences) and units of speech communication
(utterances). Bakhtin argues that an utterance is a finalized whole, guaranteeing the
possibility of a response. It is determined by three factors: semantic exhaustiveness of
the theme, the speaker’s plan or will, and typical compositional and generic forms
of finalization. Further, the utterance is related both to the speaker himself/herself
and to the addressee. In contrast, units of language, such as words and sentences,
have a finality of grammatical form and a finality of meaning that is only abstract:
In themselves, they belong to nobody and are addressed to nobody. They are the
building blocks of utterances and are typically surrounded by a context of the speech
of the same speaker. Thus, we do not exchange sentences but utterances that are built
from language units (words, phrases, and sentences). If an utterance consists of only
one sentence, this sentence “acquires a special semantic fullness of value” (Bakhtin
1986: 74) and it becomes possible to “assume a responsive reaction with respect to
it” (ibid.). According to Bakhtin, it is important to keep the notion of sentence and
that of utterance apart.
3.5â•… Speech genres

Bakhtin introduced the concept of speech genre in 1952–1953 in ‘The problem of
speech genres’ (first published in Russian in 1979; English translation, 1986). By
speech genres Bakhtin (1986: 60, 81) means relatively stable and normative types
of utterances developed in different spheres of communication. In other words, all
“utterances have definite and relatively stable typical forms of construction of the
whole” (Bakhtin 1986: 78, emphasis in the original). Although they are “much more
changeable, flexible, and plastic than language forms are” (p. 80), they have a norma-
tive significance for the speaker – “they are not created by him but are given to him”
(pp. 80–81; in this connection Bakhtin criticizes de Saussure for ignoring “the fact that
in addition to forms of language there are also forms of combinations of these forms”,
viz. speech genres). The ability to cast one’s speech in speech genres is, moreover,
essential for communication.
Speech genres organize our speech in almost the same way as grammatical (syntactical)
forms do. We learn to cast our speech in generic forms and, when hearing other’s
speech, we guess its genre from the very first words; we predict a certain length (that
is, the approximate length of the speech whole) and a certain compositional structure;
we foresee the end; that is, from the very beginning we have a sense of the speech
whole, which is only later differentiated during the speech process. If speech genres
did not exist and we had not mastered them, if we had to originate them during
the speech process and construct each utterance at will for the first time, speech
communication would be almost impossible. (Bakhtin 1986: 78–79).
Yet, although speech genres are relatively stable and normative they do not constitute
a finalized set. On the contrary, Bakhtin (p. 60) points out that there is a boundless
wealth of diverse speech genres (oral and written) which are used and developed for
various spheres of human activity. In fact, the category of speech genres includes
anything from short rejoinders of daily dialogue and the brief standard military com-
mand to the multivolume novel (pp. 60–61). Further, Bakhtin (pp. 61–62) draws atten-
tion to the fundamental difference between primary (simple) and secondary (complex
or ideological) speech genres. Primary genres (for example, rejoinders of everyday
dialogue and personal letters) have taken form in unmediated speech communication
and are immediately related to actual reality and to real utterances of others. Secondary
genres, on the other hand, “arise in more complex and comparatively highly developed
and organized cultural communication (primarily written) that is artistic, scientific,
sociopolitical, and so on” (p. 62). Secondary genres absorb and digest various primary
genres that enter into them. For instance, a letter found in a novel enters into actual
reality only via the novel as a whole. Finally, it is Bakhtin’s contention (p. 62) that both
primary and secondary speech genres should be studied in order to reveal and define
the nature of the utterance.
Bakhtin (p. 65–66) further claims that speech genres and their styles should be
studied in order to explain linguistic change. According to Bakhtin (p. 65) “[t]here is
not a single new phenomenon (phonetic, lexical, or grammatical) that can enter the
system of language without having traversed the long and complicated path of generic-
stylistic testing and modification”.
Linguists that have been inspired by Bakhtin’s category of speech genre are
Wierzbicka (1991), Gasparov (1996), Fairclough and others working within the
framework of critical discourse analysis (see, e.g. Fairclough & Wodak 1997). In
the last fifteen years or so, extensive research on speech genres has seen the light
in Russia (see e.g. the Zhanry rechi (1–3) collections of articles published in 1997,
1999, and 2002).
To conclude this section, I wish to mention two more famous Bakhtinian concepts
that appear in his investigations into the novelistic genre, viz. chronotope and carnival.
These two concepts have, however, attracted the attention of linguists only marginally,
whereas they are widely used in literary and cultural studies; in folklore and narrative
analysis; and in linguistic anthropology (e.g. the work of M. Silverstein, K. Woolard,
C. Briggs, and R. Bauman).
Mikhail Bakhtin 49
3.6â•… Chronotope
The term chronotope (literally time-space) was borrowed from Einsteinian theory of
relativity, but used “almost as a metaphor (almost, but not entirely)” (Bakhtin 1981: 84).
For the purpose at hand, Bakhtin defines chronotope as “the intrinsic connectedness
of temporal and spatial relationships that are expressed in literature” (ibid.). More spe-
cifically, literary chronotopes “are the organizing centers for the fundamental narrative
events of the novel. The chronotope is the place where the knots of narrative are tied and
untied. […] to them belongs the meaning that shapes narrative” (Bakhtin 1981: 250).
For example, the chronotope of the road is described by Bakhtin in the following way:
Encounters in a novel usually take place ‘on the road’. […] On the road (‘the high
road’), the spatial and temporal paths of the most varied people […] intersect at one
spatial and temporal point. People who are normally kept separate by social and
spatial distance can accidentally meet […] the most various fates may collide and
interweave with one another. […] The chronotope of the road is both a point of new
departures and a place for events to find their denouement. Time, as it were, fuses
together with space and flows in it (forming the road); this is the source of the rich
metaphorical expansion on the image of the road as a course: “the course of life,” “to
set out on a new course,” “the course of history,” and so on; varied and multileveled
are the ways in which road is turned into a metaphor, but its fundamental pivot is the
flow of time. (Bakhtin 1981: 243–244).
Bakhtin further raises questions concerning the relations of literary chronotopes to

the chronotope of the narrator and the chronotope of the author (Bakhtin 1986: 134).
Although Bakhtin discusses only literary chronotopes, he admits of chronotopes in
other areas of culture as well (Bakhtin 1981: 84).
The term of chronotope has been used by some linguists, for instance by Zolotova
et al. (1998). In their terms, chronotope stands for the space-time in which speaking,
observing, or acting ‘subjects’ are situated, a sense which, of course, does not exactly
match the Bakhtinian use of the term.
3.7â•… Carnival
The concepts of carnival and carnivalization are elaborated on in the study of Rabe-
lais and in the 1963 Dostoevsky book. They were the first Bakhtinian concepts to be
introduced in the West (the Rabelais book was translated into English as early as in
1968) and they have been extensively used (to the point of being overused) in literary,
folkloristic, and cultural studies, but, to the best of my knowledge, they have not really
found their way into linguistic studies. (There is, however, mention of Bakhtin and of
Rabelaisian laughter in Kress et al. 1997.).
Briefly, according to Bakhtin (1973: 100–101, 133) carnival is a “sense of the world”
which belonged to the whole people especially in antiquity, in the Middle Ages, and in
the Renaissance. For the period of festivals and rituals of the carnival type, this attitude
liberated man from laws, prohibitions and restrictions which determined the order of
normal, i.e. non-carnival, life. Thus, to a degree, carnivalistic life was turned inside out
(monde à l’envers). The four major ‘categories’ of carnival described by Bakhtin are:
free, familiar contact among people (i.e. all hierarchical barriers of non-carnival life
are suspended); eccentricity (the half-real, half-play modus of interrelationship of man
with man in which behavior, gesture, and word are liberated and become eccentric and
inappropriate from the point of view of normal life); carnivalistic mésalliances (com-
bination of “the sacred with the profane, the lofty with the lowly, the great with the
insignificant, the wise with the stupid, etc.”); and, finally, profanation (blasphemies, a
whole system of lowering of status and bringing down to earth, obscenities connected
with the reproductive power of the earth and body, parodies of sacred texts, etc.).
Further, Bakhtin discusses various carnival performances (e.g. the mock crowning)
and the ambivalence of carnivalistic laughter and of carnivalistic images, such as fire.
By carnivalization Bakhtin means the influence of carnival on literature, more spe-
cifically, on literary genre (Bakhtin 1973: 100). But he also talks about the carnivalization
of verbal life, maintaining that the so-called “familiar speech of the street” (famil’iarno-
ploshchadnaia rech’) of the European peoples was permeated with the carnival sense of
the world in the Middle Ages; and further that such speech is still filled with relics of
carnival, especially in profane and derisive expressions (Bakhtin 1973: 107).
4.â•… Conclusion
Bakhtin’s ideas about language were certainly challenging at the time when he
advanced them. But as pointed out by Paducheva (1996: 218), by the end of the 1970s,
a dialogic conception of language had become an axiom in linguistics, and contem-
porary linguistics, enriched with pragmatics, is well equipped to answer Bakhtin’s
challenge. As for Bakhtin’s own works, they offer broad heuristic perspectives for
the study of language in use rather than actual studies. I agree with the classicist
and philosopher Sergei Averintsev, himself a meticulous philologist, when he says:
“His works are not a depository of ready scholarly results that can be mechanically
‘applied,’ but something other [inoe] and greater: a source of mental energy” (cited
from Emerson 1997: 274).
References
Bakhtin, M.M. (1968). Rabelais and his world. [Transl. H. Iswolsky.] MIT Press.
——— (1973). Problems of Dostoevsky’s poetics. [Transl. R.W. Rotsel.] Ardis.
Mikhail Bakhtin 51
——— (1975). Voprosy literatury i èstetiki. Khudozhestvennaia literatura.

——— (1979). Èstetika slovesnogo tvorchestva. [Ed. S.G. Bocharov; notes S.S. Averintsev &
S.G. Bocharov.] Iskusstvo.
——— (1981). The dialogic imagination. [Ed. M. Holquist; transl. C. Emerson & M. Holquist.]
University of Texas Press.
——— (1986). Speech genres and other late essays. [Transl. V.W. McGee; ed. C. Emerson &
M. Holquist.] University of Texas Press.
——— (1990). Art and answerability: Early philosophical essays. [Ed. M. Holquist & V. Liapunov;
transl. and notes V. Liapunov; supplement transl. K. Brostrom.] University of Texas Press.
——— (1993). Toward a philosophy of the act. [Transl. and notes V. Liapunov.] University of
Texas Press.
——— (1996) Sobranie sochinenii. Tom 5. [Ed. by S. G. Bocharov & L. A. Gogotishvili.] Russkie
slovari.
Brandist, C. (2002). The Bakhtin Circle: philosophy, culture and politics. Pluto Press.
——— (2003). The origins of Soviet sociolinguistics. Journal of sociolinguistics 7(2): 213–231.
Brandist, C., D. Shepherd & G. Tihanov (eds.) (2004). The Bakhtin Circle. In the master’s absence.
Manchester University Press.
Clark, L. & M. Holquist (1984). Mikhail Bakhtin. Harvard University Press.
Emerson, C. (1997). The first hundred years of Mikhail Bakhtin. Princeton University Press.
Fairclough, N. & R. Wodak (1997). Critical discourse analysis. In T.A. van Dijk (ed.) Discourse as
social interaction: 258–284. Sage.
Gasparov, B. (1996). Iazyk, pamiat’, obraz: lingvistika iazykovogo sushchestvovaniia. Novoe liter-
aturnoe obozrenie.
Ivanov, V.V. (1998). Izbrannye trudy po semiotike i istorii kul’tury, tom 1. Shkola “Iazyki russkoi
kul’tury”.
Jakobson, R. (1966) [1960]. Closing statement: Linguistics and poetics. In T. A. Seboek (ed.) Style in
language: 350–377. MIT Press.
Konkin, S.S. & L.S. Konkina (1993). Mikhail Bakhtin (Stranitsy zhizni i tvorchestva). Mordovskoe
knizhnoe izdatel’stvo.
Kress, G., R. Leite-García & T. Van Leeuwen (1997). Discourse Semiotics. In T.A. van Dijk (ed.)
Â�Discourse as structure and process: 257–291. Sage.
Kristeva, J. (1980). Desire in language: a semiotic approach to literature and art. Columbia Univer-
sity Press.
Lähteenmäki, M. (2003). On the interpretation of Baxtin’s linguistic ideas: The problem of the texts
from the 1950–60s. Russian linguistics 27: 23–39.
——— (2004). On the origins of Bakhtinian sociolinguistics. In M. Nenonen (ed.) Papers from
the 30th Finnish conference of linguistics, Joensuu, May 15–16, 2003. Studies in Languages
39: 101–106.
Marková, I. (2003). Dialogicality and social representations: the dynamics of mind. Cambridge
University Press.
Medvedev, P.N. & M.M. Bakhtin (1978). The formal method in literary scholarship: a critical introduc-
tion to sociological poetics. [Transl. A.J. Wehrle.] John Hopkins University Press.
Morson, G.S. & C. Emerson (1990). Mikhail Bakhtin: creation of a prosaics. Stanford University Press.
Paducheva, E.V. (1996). Semanticheskie issledovaniia (Semantika vremeni i vida v russkom iazyke;
semantika narrativa). Shkola “Iazyki russkoj kul’tury”.
Roulet, E. (1996). Polyphony. In J. Blommaert, C. Bulcaen, J-O Östman & J. Verschueren (eds.)
Handbook of pragmatics 1996 installment. Also in Handbook of pragmatics online. http://www.
benjamins.com/online/hop/
Forum on Baxtin (1986). Slavic and East European Journal 30: 81–102.
Vološinov, V.N. (1986) [1973]. Marxism and the philosophy of language. [Transl. L. Matejka &
I.R. Titunik.] Harvard University Press.
Wierzbicka, A. (1991). Cross-cultural pragmatics: the semantics of human interaction. Mouton de
Gruyter.
Zhanry Rechi: Sbornik Nauchnykh Statei 1997(1). GosUNTS “Kolledzh”.
Zolotova, G.A., N.K. Onipenko & M. Lu. Sidorova (1998) Kommunikativnaia grammatika russkogo
iazyka. Rossiiskaia akademia nauk.
Contextualism
Claudia Bianchi
University of Milan
1.â•… Two perspectives
Contextualism is a view about meaning, semantic content and truth-conditions,Â�

bearing significant consequences for the characterisation of explicit and implicit
content, the decoding/inferring distinction and the semantics/pragmatics interface.
According to the traditional perspective in semantics (called “literalism” or “semantic
minimalism”), it is possible to attribute truth-conditions to a sentence independently
of any context of utterance, i.e. in virtue of its meaning alone.1 We must then distin-
guish between the proposition literally expressed by a sentence (“what is said” by the
Â�sentence, its literal truth-conditions) and the implicit meaning of the Â�sentence (“what is
implicated” by a speaker uttering the sentence). Over the past forty years, Â�however, an
increasing number of linguists and philosophers have begun to underline the phenom-
enon of semantic underdetermination: the encoded meaning of the Â�sentence employed
by a speaker underdetermines the proposition explicitly expressed by an utterance of
that sentence. According to the extreme version of this Â�perspective – labelled “radi-
cal contextualism”2 – no sentence of a natural language expresses a complete proposi-
tion, or has fixed truth-conditions, even when unambiguous and devoid of indexicals.
A sentence expresses a proposition only when completed and enriched with pragmatic
constituents that do not correspond to any syntactic element of the sentence and yet are
part of its semantic interpretation.
The opposition between minimalism and contextualism can be traced back to a
disagreement – at the very beginning of the contemporary philosophy of language –
between philosophers interested mainly in formal languages on the one hand, and
those interested mainly in natural languages on the other. Philosophers and logicians
like Gottlob Frege, Bertrand Russell, the early Ludwig Wittgenstein, Alfred Tarski, and
Rudolf Carnap aim to create perfect languages for philosophical and scientific com-
munication, artificial languages devoid of all the ambiguities and imperfections that
characterise natural languages; ordinary language philosophers (the later Â�Wittgenstein,
1.â•… If we abstract from ellipsis, ambiguity and indexicality strictly understood: cf. infra, § 2.
2.â•… Récanati 1993: 267n, 2007; this characterisation refers to the radical version of
contextualism – the one defended by Récanati and Relevance theorists: cf. infra § 4.
54 Claudia Bianchi
Friedrich Waismann, John Austin, Paul Grice, Peter Strawson) view Â�natural languages
as autonomous objects of analysis – and their imperfections as signs of richness and
expressive power. Frege and Russell inspire the minimalist perspective in its thinking
that truth-conditions may be ascribed to a sentence independently of any contextual
considerations; the ordinary language philosophers inspire the contextualist perspec-
tive that it is only in context that sentences have complete truth-conditions.
More broadly, “contextualism” may be used to refer to a family of views which
includes moderate contextualism (also called “indexicalism”), radical contextualism
and non-indexical contextualism – and which contrasts with semantic minimalism.
2.â•… Semantic minimalism
Semantic minimalists hold that “what is said” by a sentence (its truth-conditional

content) is closely related to the conventional meaning of the linguistic expressions
employed in it and departs from that meaning only in cases of ellipsis, ambiguity
and indexicality strictly understood (i.e. concerning only a small number of expres-
sions such as true indexicals and demonstratives) – namely only when the conven-
tional meaning of an expression makes a mandatory appeal to context (Borg 2004;
Â�Cappelen & Lepore 2005). The semantic content of a (unambiguous, non indexical)
sentence is the proposition expressed by every utterance of that sentence.
According to minimalism, all context sensitivity can be traced to syntactical
Â�elements of the sentence – the expressions belonging to Cappelen and Lepore’s Basic
Set: true indexicals (like “I”, “here”, “now”) demonstratives (like “this” or “that”), per-
sonal pronouns (“he”, “she”), and expressions like “local” or “enemy” (Cappelen &
Lepore 2005: 2). These expressions have a reference only given a context of utterance:
different occurrences of the same expression-type can have different referents. The
conventional meaning of a sentence like
(1) I am Italian
independently of any context whatsoever, cannot determine the truth-Â�conditions
of the sentence: the referent of “I” must be identified. The truth-Â�conditions of an
indexical sentence are thus determined as a function of the context of utterance of
the sentence. According to Kaplan, a function – the character – is assigned to each
indexical expression as a type: given a context, the character determines the con-
tent (the intension) of the occurrence – which is a function from circumstances of
evaluation to truth-values. The character of an indexical, then, encodes the specific
contextual Â�co-ordinate that is relevant for the determination of its semantic value:
for “I” the Â�relevant parameter is the speaker of the utterance, for “here” the place
of the utterance, for “now”, the time of the utterance, and so on: the designation is
then automatic, “given meaning and public contextual facts” (Kaplan 1989: 595).
Contextualism 55
Hence, minimalists distinguish between a mandatory semantic process (saturation

of the expressions belonging to the Basic Set) – determining the proposition Â�literally
expressed – and optional pragmatic processes (enrichment, transfer, implicatures) –
determining the proposition implicated by the speaker.
Let’s examine a more controversial example, the sentence
(2) It rains.
According to minimalists, (2) expresses a complete proposition (if we put time aside): its
semantic content is It rains.3 (2) is true iff it rains, i.e. if there is an occurrence of rain at
some place or other in the universe. When (2) is uttered in context, the addressee may
derive the conversational implicature It rains in the place relevant for speaker
and addressee. In a similar vein, the sentence
(3) All the bottles are empty
expresses the minimal proposition All the bottles are empty: (3) is true iff all the
bottle are empty, i.e. if all the bottles in the universe are devoid of any content; in con-
text C the addressee may derive the implicature All the bottles Peter just bought
are empty. And
(4) Tom is tall
expresses the minimal proposition Tom is tall: (4) is true iff Tom is tall, i.e.
if Tom has a degree of height above the norm.5 The semantic content of (4) is the
proposition expressed by every utterance of (4): in different contexts the addressee
may derive different implicatures – in context C, for example, the implicature
Tom is tall for a jockey, and, in context C’, the implicature Tom is tall for a
basketball player.
3.â•… Indexicalism
Indexicalism is an intermediate position between minimalism and contextualism.

Indexicalists such as Jason Stanley and Zoltan Szabò do not postulate an alleged
3.â•… Cf. the critique of the Incompleteness Argument in Cappelen & Lepore 2005: 33–38, 59–68.
Propositions are indicated in small caps.
4.â•… This is not the only analysis available to minimalists. According to Cappelen & Lepore 2005,
for example, an utterance of a sentence expresses indefinitely many propositions: it is the thesis
called “Speech Act Pluralism”. The expanded proposition It rains in the place relevant for
speaker and addressee is the one underlying our truth-value judgements about the speech act,
but it is not the proposition Â�semantically expressed.
5.â•… On this example see infra, § 5.
56 Claudia Bianchi
pragmatic contribution to the semantic interpretation of (2)–(4), but posit hidden

indexical elements in the logical form of the sentences. The only process affecting
the truth-conditions of a sentence is the mandatory semantic process of saturation,
triggered by the presence of a syntactic element (explicit or hidden) occurring in the
syntactic structure of the sentence. As in minimalism, all context sensitivity can be
traced to syntactical elements of the sentence – but those elements may be covert or
hidden indexicals at logical form: “all truth-conditional effects of extra-linguistic con-
text can be traced to logical form” (Stanley 2000: 391). Indexicalists, for example, posit
in the logical form of (2) a covert indexical element for the relevant place: (2) is true
iff it rains in the place the speaker refers to. Similarly, the quantificational domain of
“all the bottles” isn’t present in the surface structure of (3): it is a hidden variable in
the logical form of (3). In a given context, (3) expresses (and not merely conveys) the
proposition All the bottles Peter just bought are empty:
there are covert semantic values which play their role in determining the proposition
expressed. The semantic value of the sentence is a proposition that quantifies over the
relevant [bottles]… the value of a contextual parameter somehow contributes to the
semantic value of the whole sentence. (Stanley & Szabò 2000: 234)6
As far as adjectives like “tall” are concerned, indexicalists claim that the comparative
class (tall relative to which set of individuals?) isn’t present in the surface structure
of (4): it is a hidden variable in the logical form of (4). In context C, (4) expresses the
proposition Tom is tall for a jockey, and in context C’ the proposition Tom is tall
for a basketball player.
Indexicalism may be seen as a more liberal version of minimalism: it retains the
central minimalist tenet, namely that the only semantic process is the saturation of
the indexical expressions (even if some of them are “hidden” in the logical form of the
sentence). Stanley himself maintains that his position is “very conservative”:
My own view of the truth-conditional role of context is very conservative. First there
are expressions which are obviously indexicals in the narrow sense of the term, words
such as ‘I’, ‘here’, ‘you’, ‘now’ and their brethren. Secondly, there are expressions which
are obviously demonstratives, such as ‘this’ and ‘that’. Third, there are expressions
that are obviously pronouns, such as ‘he’ and ‘she’. Overt expressions that are in none
of these classes are not context-dependent. If the truth-conditions of constructions
containing them are affected by extra-linguistic context, this context dependence
must be traced to the presence of an obvious indexical, demonstrative, or pronominal
expression at logical form, or to a structural position in logical form that is occupied
by a covert variable. (Stanley 2000: 400)
6.â•… For a discussion of the quantifier domain restriction, see also Gauker 1997; for a critique,
Bianchi 2006.
Contextualism 57
For Cappelen and Lepore, in contrast, indexicalism is a form of moderate

Â�contextualism – an attempt to extend the Basic Set to expressions which are not “obvi-
ously” indexical – namely predicates like “know”, quantified phrases, adjectives like “tall”
or “green”; moreover, moderate contextualism is doomed to collapse into radical con-
textualism. This last claim may be criticised by pointing out that the difference between
moderate and radical contextualism can be framed not in terms of the range of context-
sensitive expressions each perspective allows, but in terms of the kind of mechanisms
they posit. According to both minimalists and indexicalists, we must provide a value for
a context-sensitive expression only when the conventional meaning of that expression
makes a mandatory appeal to context; this is not the opinion of radical contextualists,
who allow context to make a free pragmatic contribution to the semantic interpretation
of a sentence (Bianchi 2003; Borg 2007; Récanati 2004a).
4.â•… Radical contextualism
4.1â•… Overview
According to the radical contextualist perspective, no sentence of a natural language
expresses a complete proposition, or has fixed truth-conditions, even when unambiÂ�
guous and devoid of indexicals. A sentence expresses a proposition only in the con-
text of a speech act, when completed and enriched with pragmatic constituents that
do not correspond to any syntactic element of the sentence (neither an explicit con-
stituent, as in cases of syntactic ellipsis, nor a hidden indexical present at logical
form) and yet are part of its semantic interpretation.
The identification of a single contextualist paradigm is far from obvious; it is
nonetheless possible to identify a general research program, common to various schol-
ars: John Searle and Charles Travis, François Récanati, the Relevance theorists Dan
Â�Sperber, Deirdre Wilson and Robyn Carston. The cases motivating the contextualist
view are well known – although not everyone agrees on all the examples listed below:
(5) Nobody [famous] goes there any more because it’s too crowded;
(6) I have nothing [appropriate to the occasion] to wear tonight;
(7) Some [not all] children got stomach ’flu;
(8) Jill got married and [then] became pregnant;
(9) Bob has [exactly] three cars;
(10) Jack and Jill are engaged [to each other];
(11) Tom hasn’t had breakfast [today];
58 Claudia Bianchi
(12) You’re not going to die [from this cut];

(13) The conference starts at five [or some minutes later];
(14) France is hexagonal [roughly speaking];
(15) I need a Kleenex [or any paper handkerchief].
According to contextualism, these examples show that there is a significant distance

between the level of the conventional meaning of a sentence and the proposition
expressed by uttering that sentence – a distance not imputable only to ambiguity or
indexicality. In this radical perspective, pragmatic processes are required to bridge
the gap between the two levels of meaning: the interpretation of (5)–(15) (the propo-
sitions they express, their truth-conditions) is the result of pragmatic processes of
expansion and contextual enrichment, giving, as a result, the additional linguistic
material in brackets.
In this perspective, (2) expresses (not merely conveys) the proposition It rains
in the place relevant for speaker and addressee, without positing in the logi-
cal form of (2) a covert indexical element for the relevant place. Similarly, the quan-
tificational domain of “all the bottles” isn’t present in the surface structure of (3) or
at logical form: it must be appropriately constrained only at the level of the semantic
interpretation. Following Perry 1986, Récanati argues for the existence of pragmatic
constituents in the proposition expressed by an utterance, i.e. for constituents that
do not correspond to any syntactic or semantic element of the sentence (“unarticu-
lated constituents”): “No proposition could be expressed without some unarticu-
lated constituent being contextually provided” (Récanati 1993: 260)7. “What is said”
is then identified with the proposition completed by primary pragmatic processes
of enrichment and transfer (Récanati 2001, 2004a, 2004b).
4.2â•… Wittgenstein, Austin, Searle, and Travis

Contextualism, we have said, is a view suggested by ordinary language philosophers,
and broadly based on interpretations of the Wittgensteinian motto “Â�meaning is
use”: to understand a word is to know how to use it.8 In order to show that semantic
underdetermination is essential to natural languages, Wittgenstein varies the contexts
7.â•… Cf. Perry 1986, § 1 and Perry 1998. Perry 1986 seems neutral with respect to the issue whether
the presumed unarticulated constituent is only superficially unarticulated, i.e. whether it is at all
present at logical form.
8.â•… Cf. Wittgenstein 1953 § 43: “For a large class of cases – though not for all – in which we employ
the word ‘meaning’, it can be defined thus: the meaning of a word is its use in the language”.
Contextualism 59
of utterance of sentences containing empirical terms, creating unusual occasions of use,

extraordinary or bizarre cases in contrast with our intuitions.9 He writes:
I say “There is a chair”. What if I go up to it, meaning to fetch it, and it suddenly
disappears from sight?… Have you rules ready for such cases – rules saying whether
one may use the word “chair” to include this kind of thing? But do we miss them when
we use the word “chair”; and are we to say that we do not really attach any meaning to
this word, because we are not equipped with rules for every possible application of it?
(Wittgenstein 1953: § 80)
Similar examples are proposed by Waismann: is the sentence “There is a man” true or
false if, when I come closer, the man disappears, or looks like a man, speaks like a man,
behaves like a man, but is only four inch tall? And what about “It’s gold” uttered about
a substance that looks like gold, satisfies all the chemical tests for gold, but emits a new
sort of radiation (Waismann 1940: 120)? The same goes for Austin: what are we to
say about “It’s a goldfinch” uttered about a goldfinch that “does something outÂ�rageous
(explodes, quotes Mrs. Woolf, or what not)” (Austin 1961: 88)? It is in principle impos-
sible to foresee all the possible Â�circumstances which could lead us to modify or retract
an assertion.
John Searle and Charles Travis explicitly take their thought experiments and their
methodology from Austin, Waismann and Wittgenstein (Searle 1979, 1980, 1992;
Â�Travis 1975, 1981, 1985, 1996, 1997). For rather innocent sentences like
(16) The cat is on the mat,
(17) The leaves are green,
(18) Bill cut the grass,
(19) Tom opened the door,
(20) Bob opened his eyes,
(21) The surgeon opened the wound,
(22) Sally opened the can,
Searle and Travis set up anomalous or strange contexts: the cat and the mat Â�travelling
in interstellar space, the russet leaves of a Japanese maple painted green, Bill Â�cutting
grass like a cake, Tom opening the door with a knife. These examples are meant to
9.â•… Cf. Waismann 1940: 118: “The question of the verification arises only when we come across a
new sort of combination of words… when we say ‘The dog thinks’, we create a new context, we step
outside the boundaries of common speech, and then the question arises as to what is meant by such
a word series”. Cf. Wittgenstein 1953: § 250.
60 Claudia Bianchi
show that every sentence has a literal meaning only against a background of con-
textual assumptions fixing its truth-conditions: the background states, for example,
that gravitation is, or is not, effective, or the way people “normally” cut things, and
grass in particular, or open doors, eyes, or cans. What is more, this background is not
unique, stable or fixed once and for all: it may change with different occasions of use.
Consequently, Searle and Travis argue that the semantic properties of an expression
depend on the context of use of the expression: the conventional meaning of a sen-
tence, if taken independently of any context whatsoever, underdetermines its truth-
conditions. In examples (19)–(22), the conventional meaning of “open” does not
change, but its interpretation is different in each utterance: is (19) true if Tom opens
the door with a can opener, or a scalpel? The stable, conventional meaning of “open”
seems to determine a different contribution to the truth-conditions of each utterance.
Following this same line of thought, Â�contextualism criticises the thesis – essential
to semantic minimalism – according to which there are meanings conventionally
Â�associated with linguistic expressions, and sets of truth-conditions conventionally
associated with sentences.
4.3â•… Motivations for radical contextualism

Contextualism is supported by two main arguments: (a) the “Context Shifting
Â�Argument” (CSA) and (b) the “Inappropriateness Argument”.
(a) The CSA is exemplified by sentences (16)–(19):
“Suppose someone suspects that an expression e… is context-Â�sensitive. How

could he go about establishing this? One way that philosophers of language do so
is to think about (or imagine) various utterances of sentences containing e. If they
have intuitions that a semantically relevant feature of those utterances varies from
Â�context to context, then that, it is assumed, is evidence [that] e is Â�context-sensitive.
(Cappelen & Lepore 2005: 17)
Let’s examine more closely example (17), taken from Travis 1997. Let’s assume that
(17) is devoid of indexicals (time isn’t relevant here) and with its semantics fixed:
intuitively, according to Travis, “are green” – given its meaning in English – is a
device for calling things green and “the leaves” purport to speak of some leaves.
Now, consider Pia’s Japanese maple, that has russet leaves: Pia paints them green.
Consider two different utterances of (17). In Context 1, Pia is with her photographer
friend, seeking green subjects for his photos. Pia utters (17): intuitively she speaks
truthfully. In Context 2, Pia is with her botanist friend, seeking green leaves for a
study of green-leaf chemistry. Pia utters (17): intuitively she speaks falsely. Accord-
ing to Travis, in C1 and C2 (17) has the same conventional meaning, but a crucially
different interpretation:
Contextualism 61
If the story is right, then there are two distinguishable things to be said in speaking
[(17)] with the stipulated semantics. One is true; one is false; so each would be true
under different conditions. That semantics is, then, compatible with semantic variety,
and with Â�variety in truth involving properties. So what the words of [(17)] mean is
compatible with various distinct conditions for its truth. (Travis 1997: 89)
The state of affairs referred to with the two utterances of (17) is the same, but their
truth-value changes: it follows, according to Travis, that their truth-conditions are dif-
ferent in C1 and C2. Even though (17) does not contain any of the obvious indexicals
(overt or hidden), it is in some sense context-sensitive: it expresses different propoÂ�
sitions in different contexts.
(b) The second argument is the “Inappropriateness Argument”. Radical contex-
tualists claim that the pragmatic enrichments10 are necessary in order to account for
the intuitions speakers have about the truth-conditions of their utterances. The prag-
matic processes of completion and enrichment are pervasive, and generally uncon-
scious; the interpretation they generate is unproblematic. Nobody takes the speaker
to mean, with (2), that it’s raining in some place or other of the universe; or, when
uttering (3), that all the bottles in the universe are empty: the propositions expressed
by (2) and (3) are enriched quite naturally and without any effort. Moreover, accord-
ing to Relevance theorists, pragmatic processes of strengthening and broadening are
mandatory even in cases of so-called literality: even if, Â�uttering (14), a speaker wants
to express the “literal” proposition France is Â�geometrically hexagonal, this very
interpretation is still the result of a pragmatic process of modulation.11
As we have said, in the minimalist perspective, every contextual contribution is
triggered by a constituent of the sentence – either explicit or implicit, but nevertheless
present as a variable in the syntactic structure of the sentence: such a thesis complies
with the assumption of an isomorphism between syntactic structure and semantic
interpretation.12 By positing unarticulated constituents (such as those bracketed in
examples (5)–(15)), radical contextualism objects to the principle of isomorphism
between syntax and semantics, one of the benchmarks of the traditional view. Depar-
ture from this principle is usually justified in order to preserve the speakers’ semantic
intuitions concerning truth-conditions. Radical contextualists, then, prefer to speak of
“intuitive truth-conditions”: the proposition expressed by an utterance does not corre-
spond to the logical form of the sentence, but is individuated by the truth-conditional
10.â•… Exemplified by (2) and (3) in § 4.1 and represented in (5)–(15) by the bracketed linguistic
material.
11.â•… See the discussion on ad hoc concepts in Carston 2002, ch. 5.
12.â•… See, for example, Grice 1989: 87.
62 Claudia Bianchi
intuitions of the participants in the conversational exchange. Understanding an utter-

ance amounts to knowing which state of affairs makes that utterance true, i.e. under
which circumstances that utterance would be true. Underlying this idea is the Gricean
thesis that “saying” is a variety of non-natural meaning – and must be accessible to the
addressee: non-natural meaning is a matter of intention recognition. This is the sense
of the “Availability Principle”, proposed by Récanati:
In deciding whether a pragmatically determined aspect of utterance meaning is part
of what is said, that is, in making a decision concerning what is said, we should always
try to preserve our pre-theoretic intuitions on the matter. (Récanati 1993: 248)13
4.4â•… Objections to radical contextualism

Radical contextualism is a bold view, criticized by both minimalists and indexiÂ�
calists. Let us now examine some of the objections raised, and some tentative replies.
(1) The radical contextualist view is supported by the elegance, economy and
generality of its proposal. These are unquestionable qualities, compared to the more
local strategies of the indexicalist view, or the counter-intuitive ones of the minimalist
view; but they may become weak points if radical contextualists cannot posit reason-
able constraints on the appeal to context. Linguists such as Stanley and Szabò express
perplexity towards a view suggesting the same kind of solution for a great variety of
phenomena: saturation of indexicals and demonstratives, disambiguation, comple-
tion, quantifier domain restriction, free enrichment. To avoid this kind of objection,
contextualists propose a range of tests (Binding, Optionality, Availability, Embed-
ding Test) in order to provide non-controversial restricting Â�criteria on the elements
Â�legitimately admitted in semantic interpretation.14
(2) Not everybody agrees on the fact that semantics should account for semantic
intuitions, rather than semantic facts – nor that semantic intuitions could be relevant
13.â•… See Récanati 1995, 2004a: 14, 2001: 79–80; cf. Gibbs & Moise 1997. Searle and Travis have
similar views concerning intuitive truth-conditions: cf. Searle 1992 and Travis 1997.
14.â•… The Binding Criterion is proposed by Stanley 2000: 410: “A contextual Â�ingredient in the
interpretation of a sentence S results from saturation if it can be ‘bound’, that is, if it can be made
to vary with the values introduced by some operator prefixed to S”; the Optionality Criterion
is proposed by Récanati 2004a: 101: “Whenever a contextual ingredient of content is provided
through a pragmatic process of the optional variety, we can imagine another possible context of
utterance in which no such ingredient is provided yet the utterance expresses a complete propo-
sition”; the Embedding Test, proposed by Carston, is based on Récanati’s Scope Principle: “A
pragmatically determined aspect of meaning is part of what is said (and, therefore, not a conver-
sional implicature) if – and, perhaps, only if – it falls within the scope of logical operators such
as negation and conditionals” (Carston 2002: 191); for the Availability Principle see supra, § 4.3.
Contextualism 63
in determining what is said by an utterance, instead of, for example, what is commu-
nicated by a speaker. Moreover, intuitions on the propositional content of a sentence
tend to be sensitive to extralinguistic information and are likely to reflect interpreta-
tions that are conveyed by typical, standard, utterances of that sentence.15 Cappelen
and Lepore label as “Mistaken Assumption” the Â�contextualist idea that
a theory of semantic content is adequate just in case it accounts for all or most of the
intuitions speakers have about speech act content, i.e., intuitions about what speakers
say, assert, claim, and state by uttering sentences. (Cappelen & Lepore 2005: 53)
As we have said, for these two authors we must carefully distinguish between semantic
content (the minimal proposition) and speech act content. The expanded propositions
in (5)–(15) are those that underlie our truth-value judgements about the speech act,
but they are not the propositions semantically expressed: “There is no close and imme-
diate connection between semantic content and speech act Â�content” Â�(Cappelen &
Lepore 2005: 58).
According to moderate contextualists, on the contrary, acknowledging that
two utterances of sentences (5)–(15) have different contents is prima facie evidence
that the sentence uttered is context sensitive (Szabò 2006: 33). Moreover, radi-
cal contextualists do not deny that we could define “semantic content” as content
obtained via saturation only; they simply deny this notion any role in a theory of
language and communication. The notion of minimal proposition Â�postulated by
minimalists is too vague and general, does not correspond to speakers’ intuitions
and plays no role in their cognitive life or in the interpretative process. 16 In many
cases (cases of metaphor, or of loose talk, exemplified in (13)–(15)) minimal-
ists are forced to say that speakers express trivial truths or obvious falsehoods:
intuitively the minimal propositions (France is geometrically hexagonal,
for instance) are neither meant by the speaker, nor recognized by the addressee.17
The strongest argument is that addressees can derive implicatures only taking
into account the enriched propositions, and not the minimal ones. Suppose that
Mary invites Peter for a walk, and he utters (2). Only if (2) expresses the enriched
15.â•… Involving what Bach calls “sentence nonliterality”: Bach 2002; cf. Bach 2001: 26, Bach 2004,
Taylor 2001.
16.â•… Cf. Carston 2004a: 640: “the underlying issue is whether there is any psychologically real level of
representation between encoded linguistic semantics and explicature, a level of minimal proposi-
tionality at which saturation processes alone have taken place”; cf. Carston 2004b. For a discussion,
see Bianchi 2009, ch. IV.
17.â•… Cf. Récanati 2004b: 48–49. Relevance theorists, and Carston in particular, are more cautious
than Récanati about the notion of truth-conditional intuitions: cf. Carston 2002: 168–169.
64 Claudia Bianchi
Â�
proposition It rains here can Mary derive the implicature Let’s stay home; Mary
can’t derive the correct implicature taking into account the minimal proposition
It rains (somewhere):
obvious implicatures of the utterance would depend on the enriched proposition;
for instance… [in the example I’ve had a shower] the implicature that the speaker
doesn’t need to take a shower at that time. It is the enriched propositions that are
communicated as explicatures and which function as premises in the derivation of
implicatures: the uninformative, irrelevant, and sometimes truistic or patently false
minimal propositions appear to play no role in the process of utterance understanding.
(Carston 2004a: 639)
(3) Contextualism in its radical form opts for a “deflationary” philosophy of Â�language,
in which conventional meanings are given no central role. Minimalists and indexical-
ists assign meaning to types of sentences, radical contextualists only to Â�occurrences of
sentences.18 As a consequence, the contextualist perspective loses much of its explana-
tory and predictive power – and must account for the stronger role of the contex-
tual information versus the invariant aspects of syntax and semantics. More generally,
contextualism is at risk of undermining systematic theorizing about language and
communication.
Radical contextualists reply that a sentence expresses a content only in the con-
text of a speech act. Therefore the truth-conditional content of an utterance is jointly
determined by semantics and pragmatics: semantics studies linguistic meaning (a
property of expression-types), while truth-conditions are determined by pragmat-
ics, or, better, truth-conditional pragmatics (Récanati 1993, 2001; Carston 2002;
Bezuidenhout 2002).
5.â•… Nonindexical contextualism
In very recent times, some scholars have proposed a different analysis of the cases of
context sensitivity that go beyond meaning-controlled contextuality (i.e. Â�indexicality
broadly understood). Among the major alternatives to minimalism and contextu-
alism we may list MacFarlane’s “nonindexical contextualism” (MacFarlane 2007,
2009), Predelli’s revised version of minimalism (Predelli 2005a, 2005b); Corazza and
Dokic’s “situationalism” (Corazza 2007; Corazza & Dokic 2007), Gauker’s Â�objective
18.â•… Cf. Cappelen & Lepore 2005: 58: “semantics is a discipline that aims to characterize systemati-
cally certain features of linguistic expressions and to do so in a way that captures general truths
about languages, and not just truths about particular speakers in specific contexts”.
Contextualism 65
contextualism (Gauker 2003)19 and Récanati’s “strong moderate relativism” Â�(Récanati

2007, 2008).20 We will focus on MacFarlane’s position.
In a contextualist perspective, different utterances of (4) or (17) say differ-
ent things (express different propositions) in different contexts; in a minimalist
perspective, on the contrary, there is an invariant semantic content: every utter-
ance of (4) or (17) expresses the same minimal proposition (Tom is tall or The
leaves are green, respectively).21 Minimalism must then answer the intension
problem for minimal propositions: at which possible worlds is the minimal propo-
sition Tom is tall (just plain) true? Being tall for a jockey, MacFarlane argues, is a
way of being tall: the minimal proposition Tom is tall will be true at every world
at which the proposition Tom is tall for a jockey is true. If we reiterate the
Â�argument for any comparative class F, we will Â�conclude that the minimal proposi-
tion Tom is tall is true at every world at which the proposition Tom is tall for
an F is true: “we are left with the surprising conclusion that the minimal proposi-
tion that Tom is (just plain) tall is true at every world at which Tom has any degree
of height at all” (MacFarlane 2007: 242).22
According to nonindexical contextualism, indexicalism and contextualism make
the same error: they mistake intuitions about the truth-values of certain Â�utterances
(intuitions revealed by the CSA) for intuitions about the truth-conditions of those very
utterances (or about the propositions they express). For MacFarlane, the variation in
truth-values of the different utterances of (4) or (17) is not caused by a variation in
their content (or truth-conditions), but by a variation in their circumstances of evalu-
ation. The circumstances of evaluation contain a parameter for the possible world and
a parameter “counts-as” – which is a function from properties (like “tall”, “green”, etc.)
to intensions:
The “counts-as” parameter is so called because it fixes what things have to be like
in order to count as having the property of tallness (or any other property) at a
circumstance of evaluation. (MacFarlane 2007: 246)
In other words, sentences like (4) or (17) are context sensitive not in the sense that they
express different propositions in different contexts, but in the sense that the truth, or
falsity, of their occurrences depends on the circumstance in which they are evaluated.
19.â•… Sbisà dubs “evaluational contextualism” the two brands of contextualism proposed by Gauker
and MacFarlane: see Sbisà 2009.
20.â•… For the distinction between moderate and radical relativism, see Récanati 2007.
21.â•… Even if, according to Speech Act Pluralism, in different contexts, it is possible to perform dif-
ferent speech acts having different contents.
22.â•… MacFarlane’s example is “Chiara is tall”.
66 Claudia Bianchi
According to MacFarlane, the “counts-as” parameter is determined by certain

features of the context of utterance, such as the conversation subject matter, or the
speaker’s intentions. The parameter for (4) will change, for instance, if we are discuss-
ing jockeys (C1) or basketball players (C2). The circumstances of evaluation of C1
and C2 may well be different, even if they are situated at the same possible world: two
occurrences of (4) may then have different truth-values even if they express the same
proposition.
In a similar vein, Predelli offers an interesting alternative defence of the tradi-
tional view, indicating a form of contextuality intervening “at the post-composi-
tional level”, i.e. not at the level of “what is said” (the level of the truth-conditions),
but at the level of the evaluation of “what is said” (the level of the truth-values).
Arguing in favour of traditional compositional modules, Predelli tries to provide a
semantic account of sentence-index pairs that avoids the semantic opulence of the
indexicalist view and the pragmatic opulence of the contextualists, who postulate
pragmatic processes (such as free enrichment or strengthening) at the semantic level
(Predelli 2005a, 2005b).
MacFarlane’s position has been labelled “relativism” by some – for the truth-Â�values
of (4) and (17) are thought to be relative to a possible world and an Â�additional param-
eter. Consider the sentences:
(23) Pitt is more handsome than Depp
(24) Tom knows that the bank is open
(25) Greece might win the world cup.
In the same line of thought as MacFarlane’s, some argue that the truth of (4), (17) and
(23)–(25) must be relativised not only to a parameter for a possible world, but also to
a “counts-as” parameter (for (4) and (17)), or to a standard of taste (for (23)), or to
practical interests (for (24)), or to a state of knowledge (for (25)).23
23.â•… Cf. Kölbel 2008: 4. Récanati argues that his moderate relativism is equivalent to MacFar-
lane’s nonindexical contextualism – contrary to MacFarlane’s own opinion: Récanati 2008: 10n,
MacFarlane 2005: 325. MacFarlane claims that his nonindexical contextualism is similar to the
position held by Predelli 2005b: MacFarlane 2009: 246n. Note that merely adding an extra param-
eter (“counts-as”) is relativistic in the moderate sense of Kölbel and Récanati, but not in MacFar-
lane’s sense. According to MacFarlane 2009: 248, in non-indexical contextualism truth-evaluation
depends on the context of the utterance to be assessed, while in relativism it depends on the context
of assessment: “Whereas nonindexical contextualism lets the epistemic standard Â�paraÂ�meter be ini-
tialized by the context of use, relativism lets it be initialized by the context of assessment”.
Contextualism 67
Several issues are still open. For one thing, some clarification on the notion of seman-
tic intuitions would surely be welcome, along with more extensive reflection on the
role that intuitions play – or must play – in semantic theories. Moreover, contextual-
ism (even in its moderate or nonindexical versions) faces the problem of determining
to what degree contextual information should enrich the minimal proposition. More
generally, the contextualist view should put some constraints on the proliferation of
the relevant parameters in cases like (17):
the worry is not so much that we’ll have too many parameters, but that there will be
no end to the addition of such parameters. The worry is that such proliferation would
make systematic semantics impossible. (MacFarlane 2009: 246)
The main open point concerns the consequences the debate between minimalism
and contextualism has for the semantics/pragmatics distinction (Bianchi 2004; Szabò
2005; Turner 1999). In a minimalist perspective, semantics and pragmatics are con-
sidered complementary research fields: semantics studies the conventional meaning
of Â�linguistic expressions, while pragmatics deals with how speakers use expressions
in context. In other words, pragmatic processes play a role at the semantic level only
in cases of indexicality – in helping to determine “what is said” by an utterance.
Â�Otherwise, they are involved at the pre-semantic level, to pick up the appropriate syn-
tactic construct in cases of ambiguity and ellipsis, and finally at the post-semantic level,
for the derivation of conversational implicatures (Perry 1998). Contextualists rethink
the distinction between semantics and pragmatics as traditionally conceived: they
claim that semantics can offer only incomplete interpretations. From this perspective,
semantics no longer has the task of giving truth-conditions: this task is now proper to
pragmatics, or to “truth-conditional pragmatics”.
We face a continuum of different proposals. At one end, from the minimal-
ist perspective, it is possible to assign truth-conditions to sentences of natural
Â�languages, relying on their conventional meaning and contextually fixing the ref-
erence of Â�indexicals and demonstratives. At the other end lies radical contextu-
alism, which claims that in order to obtain a complete proposition (i.e. complete
truth-conditions), it is always necessary to resort to pragmatic processes. Indexical-
ism and non-Â�indexical contextualism are intermediate positions: the former claims
that all truth-conditional effects can be traced to indexicals (overt or covert) at
logical form; the latter claims that the truth-conditions of an utterance are given
in a minimalist way, but the truth of the utterance is relativised to an additional
parameter, the context of assessment. According to many scholars, the non indexi-
cal Â�contextualist solution succeeds in explaining away certain apparent contradic-
tions in the linguistic practices relative to epistemic modals, knowledge ascriptions,
68 Claudia Bianchi
conditionals, ethical and aesthetic Â�discourse, without Â�jeopardising any prospect of a

systematic semantic theory.
Moreover, it is important to underline that the radical contextualist perspec-
tive indicates a kind of semantic context sensitivity that has nothing to do with the
forms of contextual dependence that minimalism can account for, i.e. ambiguity,
ellipsis, indexicality or implicit meaning. According to radical contextualism, (a) the
linguistic meaning of any expression underdetermines its truth-conditions – context
sensitivity becomes a general property of meaning; (b) the contextual factors that
could become relevant for determining the truth-conditions of a sentence cannot be
specified in advance: they are not meaning-controlled. The contribution of context
to the truth-conditions of a sentence goes beyond what is triggered by the presence
of a linguistic expression. The mode of dependence is not given – in other words it
is not determined independently of the occasion of use: it depends on context itself.
Independently of any context whatsoever, it is impossible to specify a state of affairs
that, if realised, would make a sentence like (4) or (17) true: its Â�conventional mean-
ing determines a set of truth-conditions only with regard to certain Â�assumptions,
practices, goals, ways of doing things.
References
Austin, J. (1961). Philosophical Papers. J. Urmson & G. Warnock (eds.) (1979) 3rd ed. Oxford
Â�University Press.
Bach, K. (2001). You don’t say? Synthèse 128: 15–44.
——— (2002). Seemingly semantic intuitions. In J. Keim Campbell, M.O’ Rourke & D. Shier (eds.)
Meaning and Truth: 21–33. Seven Bridges Press.
——— (2004). Minding the Gap. In C. Bianchi (ed.): 27–43.
Bezuidenhout, A. (2002). Truth-Conditional Pragmatics. Philosophical Perspectives 16: 105–34.
Bianchi, C. (2003). How to Refer: Objective Context vs. Intentional Context. In P. Blackburn et al.
(eds.) Modeling and Using Context, CONTEXT ’03, Lecture Notes in Artificial Intelligence, vol.
2680: 54–65. Springer.
——— (2006). ‘Nobody loves me’: Quantification and context. Philosophical Studies 130: 377–97.
——— (2009). Pragmatica cognitiva. I meccanismi della comunicazione. Laterza.
——— (ed.) (2004). The Semantics/Pragmatics Distinction. CSLI Publications.
Borg, E. (2004). Minimal Semantics. Clarendon Press.
——— (2007). Minimalism versus Contextualism in Semantics. In G. Preyer & G. Peter (eds.) Context-
Sensitivity and Semantic Minimalism. New Essays on Semantics and Pragmatics: 339–359. Oxford
University Press.
Cappelen, H. & E. Lepore (2005). Insensitive Semantics. A Defense of Semantic Minimalism and
Speech Act Pluralism. Blackwell.
Carston, R. (2002). Thoughts and utterances: the pragmatics of explicit communication. Blackwell.
——— (2004a). Relevance Theory and the Saying/Implicating Distinction. In L.R. Horn & G. Ward
(eds.) Handbook of Pragmatics: 633–656. Blackwell.
Contextualism 69
——— (2004b). Truth-Conditional Content and Conversational Implicature. In C. Bianchi (ed.):

65–100.
Corazza, E. (2007). Contextualism, minimalism, and situationalism. Pragmatics and Cognition (1):
115–137.
Corazza, E. & J. Dokic (2007). Sense and Insensibility Or Where Minimalism Meets Contextualism.
In G. Preyer & G. Peter (eds.) Context-Sensitivity and Semantic Minimalism. New Essays on
Semantics and Pragmatics: 169–193. Oxford University Press.
Garcia-Carpintero, M. & M. Kölbel (eds.) (2008). Relative Truth. Oxford University Press.
Gauker, C. (1997). Domain of Discourse. Mind 106: 1–32.
——— (2003). Words without meaning. MIT Press.
Gibbs, R. & J. Moise (1997). Pragmatics in Understanding What is Said. Cognition 62: 51–74.
Grice, H.P. (1989). Studies in the Way of Words. Harvard University Press.
Kaplan, D. (1989). Afterthoughts. In J. Almog, J. Perry & H. Wettstein (eds.) Themes from Kaplan:
Kölbel, M. (2008). Introduction: Motivations for Relativism. In M. Garcia-Carpintero & M. Kölbel
(eds.): 1–40.
Macfarlane, J. (2005). Making Sense of Relative Truth. Proceedings of the Aristotelian Society 105:
321–39.
——— (2007). Semantic Minimalism and Nonindexical Contextualism. In G. Preyer & G. Peter
(eds.) Context-Sensitivity and Semantic Minimalism. New Essays on Semantics and Pragmatics:
——— (2009). Nonindexical Contextualism. Synthese 166: 231–250.
Perry, J. (1986). Thought without Representation. In J. Perry (ed.) (1993) The Problem of the Essential
Indexical and Other Essays: 205–218. Oxford University Press.
——— (1998). Indexicals, Contexts and Unarticulated Constituents. In A. Aliseda, R. van Gabeek &
D. Westerstahl (eds.) Computing Natural Language: 1–11. CSLI Publications.
Predelli, S. (2005a). Contexts. Meaning, Truth, and the Use of Language. Oxford University Press.
——— (2005b). Painted Leaves, Context, and Semantic Analysis. Linguistics and Philosophy 28:
351–74.
Récanati, F. (1993). Direct Reference: From Language to Thought. Blackwell.
——— (1995). The Alleged Priority of Literal Interpretation. Cognitive Science 19: 207–232.
——— (2001). What is Said. Synthese 128: 75–91.
——— (2004a). Literal Meaning. Cambridge University Press.
——— (2004b). ‘What is said’ and the Semantics/Pragmatics Distinction. In C. Bianchi (ed.): 45–64.
——— (2007). Perspectival Thought. A Plea for (Moderate) Relativism. Oxford University Press.
——— (2008). Moderate Relativism. In M. Garcia-Carpintero & M. Kölbel (eds.): 41–62.
Sbisà, M. (2009). Contextualism without incompleteness. European Journal of Analytic Philosophy
5: 55–72.
Searle, J. (1979). Expression and Meaning. Cambridge University Press.
——— (1980). The Background of Meaning. In J. Searle, F. Kiefer & M. Bierwisch (eds.) Speech Act
Theory and Pragmatics: 221–232. D. Reidel Publishing Company.
——— (1992). The Rediscovery of the Mind. MIT Press.
Sperber, D. & D. Wilson (1986/1995). Relevance. Communication and Cognition. Blackwell.
Stanley, J. (2000). Context and logical form. Linguistics and Philosophy 23: 391–434.
Stanley, J. & Z. Szabò (2000). On Quantifier Domain Restriction. Mind and Language 15(2): 219–261.
Szabò, Z.G. (2006). Sensitivity training. Mind and Language 21(1): 31–38.
——— (ed.) (2005). Semantics Versus Pragmatics. Oxford University Press.
70 Claudia Bianchi
Taylor, K. (2001). Sex, breakfast, and descriptus interruptus. Synthese 128: 45–61.
Travis, Ch. (1975). Saying and Understanding. Blackwell.
——— (1981). The True and the False: the Domain of Pragmatics. Benjamins.
——— (1985). On What Is Strictly Speaking True. Canadian Journal of Philosophy 15: 187–229.
——— (1996). Meaning’s Role in Truth. Mind 105: 451–466.
——— (1997). Pragmatics. In B. Hale & C. Wright (eds.) A Companion to the Philosophy of Language:
87–107. Blackwell.
Turner, K. (ed.) (1999). The Semantics/Pragmatics Interface from Different Points of View. (Current
Research in the Semantics/Pragmatics Interface, vol. 1). Elsevier.
Waismann, F. (1940). Verifiability. In A. Flew (ed.) (1951) Logic and Language. Blackwell.
Wittgenstein, L. (1953). Philosophische Untersuchungen. Blackwell.
Deconstruction
Tony Schirato
Victoria University of Wellington
1.â•… Introduction
Although the term ‘deconstruction’ is closely associated with the French Â�philosopher
Jacques Derrida, deconstructive theory and practice has been appropriated and
extended by a variety of groups (feminists and marxists in particular). Most attempts
to define deconstruction inevitably return to Derrida’s writings; it must be pointed out,
however, that in addition to those versions of deconstruction that are closely related
to Derrida’s work, there are American versions of deconstruction (the so-called Yale
school of de Man, Hartman, Hillis-Miller and others, for instance), and what might be
called Neo-Marxist forms of deconstruction (involving prominent theorists such as
Ernesto Laclau and Gayatri Spivak).
Deconstruction is most easily defined in terms of what it is not: it is not a concept,
program or system of philosophy. It is perhaps best defined as a strategy of reading,
writing and analysis specifically aimed at the ‘textual unconscious’; that is to say, it
works to bring to light aspects of textuality such as idealisms, paradoxes, contradic-
tions, excesses and differences which are ‘repressed’ or passed over in silence by the
text, but which, at the same time, enable narratives, discourses and systems of thought –
that is to say, texts – to be produced.
2.â•… Historical background
There are a large number of antecedents of deconstruction, but the most prominent
are to be found in the work of Nietzsche, Freud, Saussure, Vološinov and Heidegger.
Nietzsche is its most obvious ‘founding’ influence, for two reasons: firstly, he made
the point that every text carries with it a disguised or repressed ‘will to power’, and
secondly and relatedly, this ‘will to power’, generally understood as a production of a
‘naturalized truth’, is inextricably bound up with the workings of language. Derrida’s
critique of ‘logocentricism’, that is to say his critique of the notion that texts and writ-
ing are neutral vehicles which ‘deliver up’ things and concepts in a pure and unmediaÂ�
ted form, owes a great deal to Nietzsche’s theorizing of the relationship between truth,
power and language.
Freud’s contribution to deconstruction is more difficult to plot, but one aspect
stands out as linking Freudian psychoanalysis to deconstruction: the notion of a
72 Tony Schirato
Â�
textual unconscious which always exceeds conscious intentionality. What is important
here, as far as deconstruction is concerned, is the notion that a subject can never finally
control or limit meaning.
Saussure’s main contribution to deconstruction is to be found in his insight that
meaning is not produced through identitification with a sign, nor through the phe-
nomenological concept of intentionality; rather, for Saussure, meaning is produced
within a system of signifiers, more specifically in terms of the difference between sig-
nifiers. In other words, the signifier woman does not carry any ‘natural’ meaning, but
comes to mean – to have value – through its difference from other signifiers such as
man or girl. There are two important lessons that deconstruction draws from structur-
alist linguistics: meaning is arbitrary (that is to say, not natural), and further, if every
signifier is always dependent on another signifier for its identity and value, meaning
can never be finalised, but must always ‘slip away’.
Vološinov can be read as an important precursor of deconstruction because his
work, particularly his critique of Saussure, makes the point that although every rela-
tionship between a signifier and a signified is arbitrary, at the same time it is always
motivated (that is to say, culturally produced). For Vološinov the ‘unconscious’ of any
text is ideological: meaning must appear as natural and unmediated, rather than cul-
turally produced, if it is to establish its hegemony. Vološinov and the ‘Bakhtin Circle’
are not commonly accepted as antecedents of deconstruction. But in terms of tech-
nical practice (see, e.g. Vološinov’s 1986 ‘deconstruction’ of the distinction between
synchronic and diachronic systems in Marxism and the philosophy of language, or the
similar deconstruction of certain Russian Formalist positions, attested to by Wlad
Godzich in his introduction to Medvedev’s 1985 The formal method in literary scholar-
ship) and their closeness to Nietzschean theories of language and the production of
‘truth effects’, they constitute a significant stage in the late nineteenth and early twen-
tieth century theorizing of the politics of language that in a sense anticipates many of
the central achievements of deconstruction.
Finally, the Heideggerian notion of Destruktion has close affinities with
Â�deconstruction, most particularly in its critiquing of certain key aspects ofÂ�
Western metaphysics, such as origin and presence. Heidegger’s work parts company
with deconstruction, however, in his commitment to the notion of an authentic,
pre-Â�linguistic, originary ‘being’.
Deconstruction draws from these, and numerous other sources such as late struc-
turalism, and extends them to produce what have been called ‘poststructuralist’ theo-
ries of writing, textuality and meaning. Poststructuralism moves beyond conventional
structuralist theory in two major ways: firstly, by picking up on Saussure’s notion of
meaning being produced through difference, it theorizes that meaning must be con-
stantly deferred. The reasoning behind this is straightforward enough: if, as we sug-
gested earlier, the signifier woman can only be understood in terms of its ‘other’ (man or
Deconstruction 73
girl), then meaning must always slide away, must always be denied closure because of its
dependence on what it is not (girl can be understood as ‘not boy’, and so on). This is not
to say that meanings are not always being made (they are), but rather that any attempt
to close down meaning, or to claim access to the authority of a final or originary mean-
ing, can only be achieved through recourse to idealism, or a discursive sleight of hand.
The second point that separates structuralism from deconstruction is that, although
structuralism accepts that meaning is produced through the relationship between
parts of a structure (words in a language system, for instance), it remains committed to
the notion of a closed system. Saussure’s privileging of langue over parole, for instance,
is an example of this ‘structuralist’ tendency. Deconstruction breaks down the notion
of a closed system by turning structuralist thought against itself: langue, or the system
of a language, cannot be thought without reference to its ‘other’ – in this case, parole,
or practice. A structure cannot exist outside, and uninvolved with, the play of its parts.
The notion of structure as metanarrative – as above and beyond that which it refers
to – does not make sense within structuralist logic itself. Hence the Derridean point
that structures always unravel themselves in some way: as we suggested earlier, there is
always some paradox or contradiction within a text which produces that text, but only
as long as that contradiction is able to hide or efface itself. Of course, deconstruction
itself is no exception to this rule.
3.â•… Basic tenets
Derrida has frequently made the point that deconstruction works as a kind of ‘double
writing’, both dismantling and inhabiting the structures it critiques. Texts and textual-
ity occupy a particularly central, yet ambiguous, place in deconstruction: deconstruc-
tion can almost be defined by the statement ‘there is nothing outside the (con)text’, but
text and context are highly unstable and problematical notions. Texts and contexts can
only really be understood, in deconstructive terms, as being written ‘under erasure’.
This strategy, borrowed more or less straight from Heidegger, sums up the major diffi-
culty under which deconstruction works: text and context are both ‘there’ as identities,
and ‘not there’ in that they bear the ‘trace’ (a particularly important term in decon-
struction) of their otherness, so that they must be understood as being different from
themselves. For deconstruction there is nothing outside the (con)text, but, at the same
time, there is really no pure textual or Â�contextual identity, either.
If deconstruction is forced to play what is inevitably a self-defeating game, viz. to
be used as an analytical tool, it does so through recourse to a wide variety of analytical
strategies. Probably the most important of these strategies of analysis are referred to in
the following terms: binarisms, differance, transcendental signifier, trace, supplement,
dissemination and spectrality. These are the main terms and strategies that characterize
74 Tony Schirato
deconstructive analysis; in addition, terms such as ‘writing’, ‘parergon’, ‘pharmakon’ and

‘hymen’ are also very useful. With Derrida’s work, some terms and approaches tend to
be used across a variety of texts (Â�writing, logocentric, trace), while others are specific to
one or two texts (parergon, Â�pharmakon). These terms are all related to a certain extent,
but also constitute quite distinctive analytical tools and/or approaches.
Binarisms refer to the Saussurian notion that meaning is always produced with
regard to an ‘other’ term: up/down, god/devil, good/bad, man/woman, truth/error,
rational/irrational and black/white are examples of this. These are, in deconstructive
terms, ‘binarisms’: although the two terms are mutually dependent, a hierarchy is pro-
duced so that one term is usually considered originary, at the same time being valued
over the other term. God, for instance, is considered to be independent of, privileged
above, and prior to, the devil. As deconstructionists would point out, however, those
moves could only be made, and those values posited, by ignoring certain linguistic
contradictions. Firstly, god cannot be understood as being completely independent
of the devil because that identity is ‘tied up’ with the devil; one of the ways in which
we make sense of the signifier ‘god’ is as ‘not devil’. Secondly, god cannot be prior to
the devil because of the very same reason. Therefore the value that accrues to god as
a pure, originary identity with certain distinctive attributes is undercut because that
identity is partly constituted out of, and ‘contaminated’ by, other supposedly ‘inferior’
and ‘negative’ identities.
The notion of binarisms is closely connected to the Derridean term ‘differance’, a
portmanteau neologism combining the senses of difference and deferment. Differance
refers to the way in which any term is always ‘different from itself ’; that is to say, is
partly constituted by what it is not. This means that identity is always sliding across a
chain of signifers, and is therefore constantly deferred.
Another term which reflects this notion of the constant deferral of meaning is ‘dis-
semination’. Dissemination is used by Derrida and deconstructionists to suggest an anti-
teleological and anti-linear approach to language and meaning; instead of moving in
a pre-arranged, intended or predictable fashion, meaning, for Derrida, always moves
outward and in a variety of trajectories. Here Derrida is very close to Bakhtin/Vološinov,
who emphasises that language carries a variety of histories and contexts which work
against attempts to stabilise meaning. Derrida extends this insight by pointing out that
language and meaning can only function ‘effectively’ precisely because they are ‘por-
table’; that is to say, they are not reducible to, constrained, or exhausted by, any one
context. The question of the portability of meaning is at the heart of Derrida’s critique
of Austin/Searle and speech act theory; the Derrida/Searle exchanges that relate to this
issue are reproduced in Limited Inc (Derrida 1988).
Meaning is disseminated, rather than controlled or directed, because decon-
struction demonstrates that such a controlling centre, what has been termed a ‘tran-
scendental signifier’, is always somehow implicated in the deferment of meaning. A
Deconstruction 75
Â�
transcendental signifier such as god, or say, Heidegger’s notion of a pre-linguistic
‘Being’ that Â�determines the possibility of ‘Beings’, functions as a metanarrative or struc-
tural centre, while remaining ‘detached’ from the workings of narrative or structure.
The problem with this notion is that, as we have suggested previously, the term god
is always ‘contaminated’ by its others (world, devil, evil, Â�temporality), just as being is
partly defined by, and has to be understood in terms of, the Â�functioning of ‘beings’.
Although supposedly opposite to god or Being, the world, the devil, evil, tem-
porality and beings all leave their ‘trace’, a term used in deconstruction to indicate
simultaneous presence and absence. This leads us to the consideration of another
crucial deconstructionist term, ‘supplement’. With deconstruction, identity is always
predicated on simultaneous presence and absence (the trace), and lack and excess (the
supplement). An example is to be found in Derrida’s discussion of Rousseau’s writing
on nature and culture. For Rousseau, nature is superior to culture, but at the same time
(and paradoxically) he insists that nature is only brought to its apogee, its fullest poten-
tial, through the intervention of culture (or at least certain highly desirable aspects of
culture, such as education). Derrida cites this as an example of both trace and supple-
mentarity. Firstly, nature, as presence, is marked with the trace of an absence (culture);
and secondly, nature is defined both by a lack (it needs culture to ‘fulfill’ its identity)
and an excess (nature, through being ‘inhabited’ by culture, is ‘more’ than itself).
In Specters of Marx (1994) Derrida moves deconstruction in the direction of a
more overtly political articulation and disposition via the notion of spectrality. The
ghosts referred to here are, on one level, the victims of social injustice (casualties of
wars, racism, colonialism, sexism) who haunt the institutions and discourses that stand
in for, act in the name of, or guarantee that justice (political parties, the state, the law);
they can also be understood as Marx himself, returning to his own funeral (variously
sited in Fukuyama’s The End of History and the Last Man (1992), the triumph of the
discourses of neo-liberalism and gobalization, and the fall of the Berlin Wall) which, as
Derrida explains, is also the ironic occasion of the reanimation of the promise of Marx-
ism; but they can be understood, in a more general sense, as the condition of invisibility
that make the visible, or the notion of presence, and in essence the notion of the social,
simultaneously possible (as a practical entity and effect of discourse, and a claim to
manifest the social) and utterly impossible (in the sense that any identity or supposed
manifestation of the social never corresponds to the promise from which it is derived.
Perhaps the best exemplifications of deconstruction as a practice are Derrida
(1976, 1978, 1981, 1986, 1987a, 1987b, 1989, 1994, 1997, 1998). The most accessi-
ble (although brief) introduction to deconstruction is to be found in Grosz (1989);
Gayatri Spivak’s long Preface to Of grammatology (Derrida 1976: ix–ixxxvi) remains
a standard work on the techniques of deconstruction; Leslie Hill’s The Cambridge
Introduction to Jacques Derrida (2007) is clear, up-to-date, eminently readable, and
Â�contains a useful chapter on the reception of Derrida’s work and further reading; and
76 Tony Schirato
Nial Lucy’s A Derrida Dictionary (2004) covers all the main concepts and refers, by
way of exemplification, to texts from literature and popular culture as well as philoÂ�
sophy and critical theory.
There are a number of books dedicated to Derrida’s later, more politicallyÂ�
oriented work, of which Herman Rappaport’s Later Derrida (2003) and The Late
Â�Derrida (2007), edited by Mitchell and Davidson, are particularly valuable. Finally, The
Politics of Deconstruction (2007), edited by Martin McQuillan, contains a number of
essays on Derrida’s deconstruction of and engagement with political concepts, narra-
tives, ideologies and philosophy (democracy, globalization, the idea of Europe, liberal-
ism, Marxism); in this collection, Richard Beardsworth’s commentary on and analysis
of Specters of Marx (1994), in the essay ‘The Irony of Deconstruction and the Example
of Marx’, offers a particularly acute critique regarding the insights to be derived from,
and the limitations of, Derrida’s Â�deconstruction of Marxist theory.
4.â•… Deconstruction in literature and linguistics
One of the fields most influenced by deconstruction is that of literary studies, an influ-
ence that has been vigorously contested, particularly by humanist and marxist critics.
Derrida’s work often makes use of literary texts (the writings of Artaud, Flaubert, Jabes,
Kafka, Mallarmé and Rousseau, to name just a few) as examples of the production of
a ‘double register’ of writing. It is incontestable, he writes, “that certain texts classed as
‘literary’ have seemed to me to operate breaches or infractions” in the closed circles of
metaphysics and signification. These texts, for Derrida, “operate in their very move-
ment, the demonstration and the practical deconstruction of the representation of what
was done with literature” (Derrida 1982: 69). Derrida’s exploration and (partial) privi-
leging of ‘literary’ texts is an example of the influence exerted over his work by the writ-
ings of Heidegger, and more particularly by Maurice Blanchot’s groundbreaking The
space of literature (Blanchot 1989).
Although deconstruction has been closely identified with the field of literary anaÂ�
lysis, it has been taken up, influenced, and contested, by numerous other fields and
practitioners of analysis and enquiry, including orthodox philosophy (Rorty, Levinas),
historiography (White, La Capra, de Certeau), theology (Mark C. Taylor, Altizer), marx-
ism (Laclau & Mouffe, Frow), psychoanalysis (Grosz), sociology (Jenks), Â�anthropology
(Clifford), film theory (Brunette and Wills), philosophy of science (N. Katherine
Â�Hayles, Serres), post-colonialism (Spivak, Young, Bhabha) and feminist studies (Butler,
Irigaray, Spivak).
The various areas of linguistic analysis have not been particularly quick to take
up deconstruction, despite the pioneering work of Jonathan Culler. Currently decon-
structive approaches are employed by some discourse analysts, Hallidayians and
Deconstruction 77
Â�
stylisticians such as Norman Fairclough, Bob Hodge, Gunther Kress and Terry Thread-
gold. Although most prominent discourse analysts are more likely to be influenced
by, and to work with, Foucaldian notions of power, knowledge and discourse, decon-
struction constitutes an important supplement to any theorizing of the ways in which
discourse authorizes and valorizes itself – that is to say, produces its ‘truth effects’. The
common thread here is the Nietzschean notion of a silent and carefully effaced textual
‘will to power’ which is both the condition of discourse, in the Foucaldian sense, and,
in deconstructive terms, the point of its own unravelling.
5.â•… Against deconstruction
Deconstruction has been critiqued from a number of perspectives, although by far the
most frequent criticism has been that it renders writing and texts completely meaning-
less, in the sense that any reading of a text is as ‘valid’ as any other. This is the position
assumed, to no small extent, by literary pragmatics (Carter, Simpson, Burton), which
reacts against what it sees as deconstruction’s obliviousness to grounded meanings.
This position misses the deconstructive ‘point’, which is not that meanings are not
made and shared (they are), nor that some meanings are privileged over others (they
are), but that the grounds on which meanings are privileged and legitimated are always
open to a deconstructive reading.
A more theoretically sophisticated critique of deconstruction has been mounted
by the Lacanian-Marxist Slavoj Žižek, who argues, from a Hegelian perspective, that
deconstruction does not adequately address the undecidability of identity (for Žižek,
identity is neither this nor that, but the performance of this or that), and therefore
remains caught up in a form of binarism (identity is both this and that). This position
is dealt with in more detail in Žižek (1991).
References
Blanchot, M. (1989). The space of literature. Nebraska University Press.

Derrida, J. (1976). Of grammatology. Johns Hopkins University Press.
——— (1978). Writing and difference. University of Chicago Press.
——— (1981). Dissemination. University of Chicago Press.
——— (1982). Positions. University of Chicago Press.
——— (1986). Margins of philosophy. University of Chicago Press.
——— (1987a). The postcard. University of Chicago Press.
——— (1987b). The truth in painting. University of Chicago Press.
78 Tony Schirato
——— (1988). Limited Inc. Northwestern University Press.

——— (1989). Of Spirit. University of Chicago Press.
——— (1994). Specters of Marx. Routledge.
——— (1997). Politics of Friendship. Verso.
——— (1998). Monolinguisim of the Other; or, The Prosthesis of Origin. Stanford University Press.
Fukuyama, F. (1992). The End of History and the Last Man. Free Press.
Grosz, E. (1989). Sexual subversions. Allen & Unwin.
Hill, L. (2007). The Cambridge Introduction to Jacques Derrida. Cambridge University Press.
Lucy, N. (2004). A Derrida Dictionary. Blackwell.
McQuillan, M. (ed) (2007). The Politics of Deconstruction. Pluto Press.
Medvedev, P.N. (1985). The formal method in literary scholarship. Harvard University Press.
Mitchell, W. & Davidson, A. (eds) (2007). The Late Derrida. University of Chicago Press.
Rappaport, H. (2003). Later Derrida. Routledge.
Vološinov, V.N. (1986). Marxism and the philosophy of language. Harvard University Press.
Žižek, S. (1991). For they know not what they do. Verso.
Epistemology
Filip Buekens
Tilburg University
Epistemology explores the possibilities and limits of knowledge as applied to Â�different

cognitive domains (knowledge of the external worlds, knowing one’s own mind,
understanding language, etc.). The classic definition of knowledge is that a person
knows that p iff s/he has the justified, true belief of that p (see Dancy 1985 and Dancy &
Sosa 1992 for details). A central issue is the correct characterization of the Â�justification
required for knowledge. Inaugurated by Descartes, classical foundationalism is
Â�perhaps the most influencial position in epistemology. Foundationalism divides our
beliefs into two groups: those which need support from others, and those which need
no support themselves (truths we know without relying on perceptual evidence). The
latter constitute our epistemological foundations. Foundationalism need not be asso-
ciated with cartesian rationalism, however. Empiricism identified states which do not
need justification with sensory states, our own immediate experiences of the external
world. They are infallibly true and stand on their own feet. Other beliefs are supported
by them. A central objection to both positions is that no belief is infallible – there is no
region in our thoughts which is entirely immune from the possibility of error. Recent
epistemology has therefore drawn a distinction between the possibility that all beliefs
could be false (the sceptical option) and the global claim that each particular belief
could be false without the global system of cognitive attitudes being false. The latter
claim is weaker and leads to the well-known fallibilist position in epistemology and
the philosophy of science.
Classical epistemology (the position we have just characterized) describes itself as
a philosophical discipline that explores the limits of knowledge. Epistemology is thus
seen as a part of philosophy, not science. Naturalized epistemology, a concept inaugu-
rated by W.V.O. Quine (1960), rejects the assumption that epistemology is not a branch
of science and that epistemology, as a ‘first philosophy’, is not able to Â�independently
justify claims of sciences like physics or, more mundanely, sense perception. Natural-
ized epistemology has it that epistemology is continuous with and part of scientific
inquiry, not a separate inquiry into the meanings of crucial words such as ‘to know’
or ‘to justify’, or a reach for infallible knowledge. The Quinean approach to epistemo-
logical issues is based on holism: there is a difference of degree between sentences to
whose truth we are firmly wedded on and those which we are more prone to abandon
in the light of recalcitrant evidence. All sentences are synthetic, one might say, but
80 Filip Buekens
some are more synthetic than others. This epistemological holism reflect the kind of
semantic holism that is present in contemporary philosophy of language.
The consequences of Quine’s naturalized epistemology for contemporary theo-
ries of meaning are wide-ranging, to say the least. A number of these consequences
have been drawn by Donald Davidson (1984), but one could also count Wittgenstein’s
work On certainty as having a profound influence on our current thinking about
what it means to know the meaning of a sentence, or to understand other persons.
The crucial connection between epistemology and philosophy of language is that the
evidence of one’s senses is not just what we appeal to in the justification and verifica-
tion of statements; it is also what we start from when learning a language. The notion
of evidence of one’s senses is held by empiricists to be basic in the theory of meaning.
This has led to the thesis that a statement has empirical meaning if and only if its truth
would make a difference to the evidence of our senses, a statement later relaxed by
Quine, who held that a non-observation sentence does not have its own observational
consequences, which in turn shows that there is nothing that the sentence means if
taken all by itself. To the extent that we can say something about the meaning of a
sentence, our answer is dependent upon the nature of the theory surrounding that
sentence. Sentences have no determinate meaning. It is the pattern in which they
are woven that confers Â�meaning upon them. Sentential meaning is indeterminate,
according to Quine.
Recent epistemological discussions which are relevant for semantics and
Â�pragmatics include discussions about what it means to follow a linguistic rule and
what it means to know that one follows a rule (of language) correctly. The Â�connection
between epistemological and semantic issues is obvious as soon as one thinks of
understanding as knowing the meaning of the words and sentences a speaker uses.
Interpreting Wittgenstein’s arguments on rule-following in the Philosophical investiga-
tions, a number of contemporary analytical philosophers (most notably Kripke 1982)
have written extensively on the so-called private language argument and connected it
with general considerations about rules and objectivity. The discussion centers around
two different and to a certain extent incompatible views of language. The first one
connects objectivity to a community of rule-followers. The basis for objectivity and
correctly understanding other persons lies in the present behavior of the linguistic
community. What makes deviant behavior incorrect is that it is not in step with other
persons’ behavior. In line with this communal view on (correctly) understanding
is the view that knowledge of conventions is of crucial importance in speaking and
Â�understanding. The alternative view is based on the Quinean insight that a theory
of meaning (a theory that enables us to speak and understand sentences of a Â�natural
Â�language) is a constantly evolving theory based on interpreting other persons’ Â�behavior
in the light of the overall rational character of their actions. Languages are, from this
point of view, idiolects. No person speaks the same language, and what is central in a
Epistemology 81
theory of meaning is not shared conventions or a communal set of rules one is sup-
posed to know or to follow; what confers meaning upon one’s utterances is the desire
to be understood and the evidence speakers create so as to achieve that goal. It should
not be a surprise that the idiolect view of language is defended by followers of Quine,
such as Davidson.
The relation between Quine’s holism (to know the meaning of a sentence is to
know the language that sentence belongs to) and current anti-holistic tendencies in
the philosophy of language and the philosophy of mind is discussed in J. Fodor &
E. LePore (1992). Their central anti-holistic argument is that if holism is true, one
Â�cannot understand any sentence unless one can understand all sentences. But how
could one then be in a position to learn a language? The countless reactions against
this view show that discussions on this issue are far from settled.
References
Dancy, J. (1985). Introduction to contemporary epistemology. Blackwell.

Dancy, J. & E. Sosa (1992). A companion to epistemology. Blackwell.
Davidson, D. (1984). Inquiries into truth and interpretation. Oxford University Press.
Fodor, J. & E. LePore (1992). Holism. Blackwell.
Hamlyn, D.W. (1970). The theory of knowledge. Routledge.
Kripke, S. (1982). Wittgenstein on rules and private languages. Blackwell.
Papineau, D. (1979). Theory and meaning. Oxford University Press.
Quine, W.V.O. (1960). Word and object. MIT Press.
Wittgenstein, L. (1953). Philosophical investigations. Blackwell.
Epistemology of testimony
Paul Faulkner
Sheffield University
The epistemology of testimony starts with our finding utterances, in the broad Gricean
sense, intelligible. A straight assertion is a piece of testimony, but so too is something
speaker meant, or a proposition communicated by a map or road sign. Testimony is
any intelligible utterance whose acceptance by its recipient could be the acquisition of
belief. And ‘testimony’ refers to that way of acquiring belief which is the acceptance
of intelligibly presented content. This way of forming belief is fundamental to our
cognitive lives. It is fundamental in that testimonial beliefs permeate every aspect of
our cognitive life. Is the belief that this tree is an oak perceptual or testimonial? And it
is fundamental in that if testimonial beliefs are not accorded the status of knowledge
or regarded as warranted, then the consequence is a form of scepticism. We would
not know much about history or geography. We would know little about the con-
tent of other minds. Science would not be recognised as a body of knowledge. We
would not know the identity of our parents. So the second starting point in giving an
Â�epistemological theory of testimony is presumption that the acceptance of testimony
is not merely a way of acquiring belief, it is also a way of acquiring warranted belief
and knowledge. The question is then, how is this the case?
One possibility is to regard testimonial knowledge and warrant as a species of
inductive knowledge and warrant; just as we have learnt that smoke is a reliable sign of
fire so we have learnt that bits of testimony reliably indicate their truth. What explains
our knowing that there is a fire on the basis of seeing smoke and what explains our
knowing that there is a fire on the basis of accepting testimony to this fact is that in
both cases we have grounds for thinking that one thing is evidence for the other, and it
is such evidence. (See Lackey 2008.) These grounds might be observations of the truth
of instances of testimony that support generalisations about the reliability of types.
We might think, for instance, that people tend to tell the truth on trivial matters, that
the The New York Times is reliable in its sports coverage, or that doctors are reliable on
medical matters but not matters to do with car mechanics. And these grounds might
concern the kinds of motivations that speakers have. Some speakers, we judge, might
be motivated by friendly relations, reputation, or fear of sanctions to tell the truth. We
are sophisticated in our reasoning about testimony and have good empirical grounds
for making the judgements we do. It is our having these grounds that determines
the warrant we have for our testimonial beliefs and explains how we get knowledge
through believing testimony.
Epistemology of testimony 83
We can certainly gain inductive knowledge from testimony, this cannot be

Â�disputed. What can be disputed is whether this reductive theory simply fails to recog-
nise what is distinctive about testimonial knowledge. The problem with regarding
Â�testimony as evidence for its truth is that we tell people things in order to get them to
believe things. So testimony is produced with the intention of producing belief, and
this seems to undercut its evidential value. For if the investigating officer discovers
that X’s handkerchief has been placed at the crime scene in order to get him to believe
that X did it, it ceases to be the good bit of evidence it originally appeared to him to
be. What then distinguishes an act of telling, Grice argued, is that it gives its recipient
a reason for belief, not by being a piece of evidence, but by its recipient recognising
that it is made with the intention of inducing belief. The assurance theory of testimony
claims that in recognising this the told person gains a reason for belief because he
gains the teller’s assurance that what is told is true, where the teller’s assurance is some-
thing like a promise to this effect. So what explains our knowing things on the basis of
testimony is that speakers assume the responsibility that goes with telling an audience
something. This distinguishes testimonial knowledge from knowledge inductively
acquired from testimony. (See Moran 2005.)
A problem for the assurance theory is that not all testimony is directed at an
Â�audience. Illicitly reading someone’s diary allows one to know things, but in Â�writing
the diarist does not intend that one forms any belief. Road signs do not seem to
embed intentions, but convey information nevertheless. There is something conven-
tional about understanding that seems to allow us, once the ability to understand is
in place, to acquire knowledge merely by accepting something that happens to be an
expression of knowledge. This suggests to the non-reductive theorist that anything
Â�intelligibly presented as true is testimony, not merely ‘tellings’ directed towards us.
And the knowledge that we acquire by accepting testimony cannot be reduced to
the knowledge that we possess by other means: being based on a communication of
knowledge is an epistemically distinctive route to knowledge. In this sense testimony
is a basic source of knowledge, for its recipient at a time, much like perception and
memory. And like these other faculties we have an entitlement to believe testimony
other things being equal. In doing so we get to know things, or be warranted in belief,
simply because the speaker, or someone further down the testimonial chain, knows
or is warranted in believing the things that are testimonially presented to us. Testi-
mony thereby Â�functions to transmit knowledge and warrant across persons much like
memory preserves it across time. (See Burge 2003.)
These three theories of testimony – the reductive, assurance and non-Â�reductive
Â�theories – can be compared along two axes. First, whether or not it is a Â�speaker’s know-
ing that p that explains an audience’s knowing that p on the basis of the speaker’s tes-
timony to p; that is, whether or not the theory explains our possession of Â�testimonial
knowledge in terms of knowledge transmission. Second, whether or not being
84 Paul Faulkner
Â�
warranted in accepting a speaker’s testimony to p necessitates the audience have a
particular reason to believe that p. The reductive theory answers ‘no’ then ‘yes’. What
explains the audience knowing that p is the audience’s grounds for believing that p.
These are the same grounds that supply the audience’s reason for accepting the speak-
er’s testimony to p, and the audience must have such reasons in order to be warranted.
The assurance theory answers ‘yes’ and ‘yes’. What explains the audience knowing that
p is the fact that, in telling the audience that p, the speaker assumed responsibility for
the audience believing truly, and was in a position necessary to properly assume this
responsibility, namely one of knowing that p. And in telling the audience that p, the
speaker gives the audience a particular reason to believe that p, namely the reason
provided by his telling. The non-reductive theory answers ‘yes’ and ‘no’. The speaker’s
knowing that p explains the audience’s knowing that p given that the audience believes
that p on the basis of the speaker’s testimony to p. But the audience does not need any
particular reason to believe the speaker’s testimony, only no reason not to believe it.
References
Burge, T. (1993). “Content Preservation”. Philosophical Review 102 (4):457–488.

Coady, C.A.J. (1992). Testimony: A Philosophical Study. First ed. Oxford: Clarendon Press.
Dummett, M. (1993). “Testimony and Memory”. In The Seas of Language, edited by M. Dummett.
Oxford: Clarendon Press.
Faulkner, Paul. (2000). “The Social Character of Testimonial Warrant”. Journal of Philosophy 97
(11):581–601.
———. (2007). “On Telling and Trusting”. Mind 116 (464):875–902.
Fricker, E. (1995). “Telling and Trusting: Reductionism and Anti-Reductionism in the Epistemology
of Testimony”. MIND 104 (414):393–411.
Goldberg, Sanford. (2007). Anti-Individualism: Mind and Language, Knowledge and Justification.
Â�Cambridge: CUP.
Grice, P. (1957). “Meaning”. In Studies in the Way of Words, edited by P. Grice. Cambridge, MA:
Â�Harvard University Press.
Lackey, Jennifer. (2008). Learning from Words - Testimony as a Source of Knowledge. Oxford: Oxford
University Press.
Lackey, Jennifer, and Ernest Sosa, eds. (2006). The Epistemology of Testimony. Oxford: Oxford
Matilal, B.K., and A. Chakrabarti, eds. (1994). Knowing From Words- Western and Indian Philo-
sophical Analysis of Understanding and Testimony. Edited by J. Hintikka. Vol. 230, Studies in
Â�Epistemology. Dordrecht: Kluwer Academic Publishers.
McDowell, J. (1994). “Knowledge by Hearsay”. In Meaning, Knowledge and Reality, edited by
J. Â�McDowell. Cambridge, MA: Harvard University Press.
Moran, R. (2005a). “Getting Told and Being Believed”. Philosophers’ Imprint 5 (5):1–29.
———. (2005b). “The Problems of Sincerity”. Proceedings of the Aristotelian Society 105 (3):341–61.
Shapin, S. (1994). A Social History of Truth: Civility and Science in Seventeenth Century England.
Â�Chicago: University of Chicago Press.
Williams, B. (2002). Truth and Truthfulness. Princeton: Princeton University Press.
Michel Foucault
Luisa Martín Rojo & Angel Gabilondo Pujol

Universidad Autónoma de madrid
The aim of this paper is to reflect upon Michel Foucault’s (1926–1984) contributions
to discourse studies in general, and to pragmatics in particular. In order to accomplish
this aim, we examine his central tenets in the understanding of discourse, and how his
work has been received and incorporated in those fields, both in the development of
current concepts and theoretical proposals, and in the kinds of analyses which have
been suggested.
Given that, to a certain extent, our approach is also Foucaultian, rather than
Â�presenting a mere inventory, we focus on those aspects which enabled, at a given
time and place, the development of specific objects of study, theoretical models,
and Â�techniques of analysis, and we also raise several theoretical problems and
questions in this field. In the first place, we deal with those Foucaultian con-
cepts we consider to have shaped a new understanding of discourse, in a pro-
cess through which discourse has emerged as the object of a field of knowledge
(Sections 2 & 3). In addition to this, we discuss how this new understanding has
introduced changes in the analytical practice, allowing the adoption of particular
perspectives, such as a ‘critical’ perspective (understood in the widest sense) and
the emergence of particular aims, such as the ‘interventional aim’ (taking part
in the present discursive and social order). Both can be seen as consequences of
analysts’ increasing awareness of and concern about the social and political con-
sequences of discourse, and of discourse analysis itself (Section 4). (See Mchoul &
Grace 1993, for a similar attempt; for a reviews of Â�Foucault’s work as a whole see,
among others, Dreyfus & Rabinow [1982]).
One of the difficulties we had to face in writing this review is the plurality of
readings of Foucault’s work. This generally acknowledged multiplicity has been associ-
ated with different traditions, and cultural and geographical areas. Rorty, for example,
assigns to the European continent what he considers a Nietzschean reading, while he
attributes a liberal (reformist) one to the Americas (in fact North America). But these
multiple readings can also be derived from Foucault’s texts themselves, the choice
depending on which topics attract readers, in a process of reading which constitutes a
philosophical journey (Enrici 1999). As we will see below, these multiple options seem
to have their roots in the Protean nature of some of the main Foucaultian concepts
86 Luisa Martín Rojo & Angel Gabilondo Pujol
(i.e. discourse itself; see, 2.1), which is also linked to Foucault’s procedure of
Â�‘problematization’ (that is, questioning what is generally assumed). And, finally, this
plurality also explains how the assessment of his legacy causes frequent controver-
sies, and why different authors, from different and even opposed approaches, refer
to his work, and claim to have been influenced by him. This is attested not only in
the more radical discursive approaches, in which the relativist-constructionist per-
spective is stronger, but also in ‘constructivist structuralist’ approaches, concerned
with the Â�constraining role of social structures, as well as with the active process of the
Â�production of social practices which can transform social structures.
2.â•… Foucault and the discursive turn
As Maingueneau (1998) points out, not all of the numerous trends within dis-
course studies have been directly influenced by Foucault’s thinking, although his
proposals have not gone unnoticed. In fact, as this author remarks, the Archéologie
du savoir and the 13th issue of the French Journal Langages, an inaugural issue in
which what has come to be known as L’Ecole française d’analyse du discours was pre-
sented, appeared simultaneously. This school looked for an ‘epistemological break-
down’, which Â�contributed to the emergence of a new ‘science’ of ideology, founded
simultaneously on Â�structural linguistics, Marxism, and psychoanalysis. Nevertheless,
this coincidence in time does not entail a comparable and immediate coincidence in
the way this school and Â�Foucault approach discourse. Hence, for Maingueneau, the
interaction between these Â�developments, and Foucault’s thinking is progressive and
oblique (Maingueneau 1998).
We are dealing, then, with indirect influences and with ‘conditions of possibility’,
which enable the emergence of a new understanding and new objects of study at a
particular time and place. Within this context, we regard as one of the most significant
contributions of Foucault to linguistic thinking and to discourse studies, his clarifying
exploration of the consequences for western epistemics of the break with two classi-
cal views of language and thought (See Les Mots et les Choses). These classical views
were the understanding of language as a ‘mirror of mind’, and the correlative view of
mind as a ‘mirror of nature (‘reality’)’, which accurately reflects relations in the objec-
tive world. Together these views meant the correlative consideration of words as mere
labels or names of concepts, and the understanding of those concepts as mere internal
reflections of external realities.
The view of language as a ‘representation of a representation (thought)’ (Foucault
1966) implies an emphasis on the representative function of language, and the idea
that the internal organization of language has to reproduce the internal organization
of thoughts (speaking was understood as ‘prononcer les idées’ (‘speaking is Â�uttering
Michel Foucault 87
ideas’/‘to express ideas’); see also Ducrot 1968: 18–34). This view of language as a
limited correlate of thought stifled the development of linguistic thinking (Ducrot
1968: 18–19; Foucault 1968: 101; see also Harris 1988).
In Les Mots et les Choses, Foucault presents the break with this way of view-
ing Â�language as a ‘doorway to modernity’ (‘le seuil du classicisme à la modernité’;
Â�Foucault 1966: 315). And his archaeological procedure allows him to perceive it as an
Â�epistemological step which places language at the center of mental, representative, and
cognitive fields, bringing about the so-called ‘linguistic turn’. This step forward means,
as Foucault points out, the correlative questioning of three beliefs: the understanding
of language and thought as two separate phenomena; the view of knowledge as objec-
tive and unembodied; and the nomenclaturist theory of meaning vs. those based on
arbitrariness. Foucault understood this epistemological step as a required ‘condition of
possibility’, which has made possible not only the development of linguistics, but also
of discourse analysis as a field of study.
Thus, the development of structural linguistics is inextricably linked to the end of
these views. Language is now understood as a means for communication. The aim in
the study of language is to find its own units, and its own internal order. But Foucault
also establishes a productive link between this break and the emergence of discourse
analysis as a field of study:
a task that consists of not – or no longer treating discourses as groups of signs

(signifying elements referring to contents or representations) but as practices that
systematically form the objects of which they speak. Of course, discourses are
composed of signs; but what they do is more than using these signs to designate
things. It is this more that renders them irreducible to the language (langue) and to
speech. It is this ‘more’ that we must reveal and describe (Foucault 1969/1972: 49)
However, as we will see in the following sections, Foucault goes even further, exploring
the consequences of the power of language in the formation of the objects to which
it refers, that is exploring the relationship between discourse and the production of
knowledge, and exploring the relation between discourse-knowledge, and processes of
domination, and the constitution of subjectivity (the constitution of the subject as its
own object, as a domain of possible knowledge). Foucault is interested in the history
of subjectivity as a mode in which, in the game of truth, the subject experiences itself.
He calls the process through which the subject constitutes itself subjectivization. “The
question is one of determining what the subject must be, what conditions are imposed,
what status it is to have, and what position it is to occupy in reality or the imaginary, in
order to become the legitimate subject of one type of knowledge or another. In short, it
is a matter of determining its mode of ‘subjectivization’” (Florence 1994: 315).
These are the three axes on which Foucault’s writings have been created: from
the archeology of knowledge (Les mots et les choses, a most decisive work), through
the genealogy of power (L’ordre du discours and Surveiller et punir), to the ethics and
aesthetics of existence (in which the modifications of the initial plan of the Histoire
de la sexualité are decisive). They are all part of a more general project. In order to
see the unity of this project, the key works are the texts on language and literature
(see Gabilondo 1996, for a unique edition of these significant texts), published
between Histoire de la folie à l’âge classique, in 1961, and Naissance de la clinique
(1963) and Les mots et les choses. The relations between these initial texts and those
corresponding to the period before Foucault’s death (June 25th 1984), reflect and
specify what he devotes his whole life to: the subject and its constitution, and the
repercussions of this constitution on the processes and domains that constitute what
is to count as ‘truth’. Knowledge, Power and the Self are a three-fold base of Foucault’s
problematization of thought. By means of this problematization what was familiar
and assumed is questioned, and all the answers given are not treated as solutions, but
reopened as further questions.1 It is, then, a matter of transforming states of affairs
into problems to which a number of solutions have attempted to respond. In this
sense, Foucault seeks not only to analyze the ideas, behaviors and ideologies of an
epoch, but the problematizations (observing the questions, and the answers given,
and reformulating the questions). Even Foucault’s style in his writings, which breaks
the usual order of constituents and uses syntagmatic patterns which produce dra-
matic pragmatic effect, is a means in this problematization, provoking a new way of
thinking (as it is shown in Casanovas 1986: 93). In this way he forces language, trying
to force thought to the limits. In fact, it is almost impossible to write about Foucault’s
thought without experiencing the limits of language.
Finally, Knowledge, Power and the Self are also seen as pivots of the constitution
of experience. Thus, a link is established between the history of concepts and different
ways of being a subject. Instead of the notion of experience he proposes the notion of
experimentation, as an action in which the subject and the object of the observation
are involved. In this way, the analytical task is not understood as an external con-
templation of an object, but requiring the involvement of the analyst (and it is closer
to an essay than to an experiment). In this paper, we consider experimentation and
problematization as major terms in Foucault’s works, and as we will see, both have
a key role not only in the emergence of particular concepts, but also of a particular
analytical task.
ïœ±.â•… In fact, shortly before his death, Foucault, in an interview with François Ewald, characterizes
his current work as ‘problematization’ (vs. deconstruction) (Foucault 1984: 18). The term denotes
“the ensemble of discursive and nondiscursive practices that make something enter into the play
of the true and the false and constitutes it as an object of thought (whether in the form of moral
reflection, scientific knowledge, political analysis or the like)” (Foucault 1994b: 667).
Michel Foucault 89
2.1â•… Discourse as a practice

Foucault considers that a history focused on problematizations is necessary, because it
would take charge of the production of signification, as well as of the modes of signi-
fication. In this way, we should problematize what is understood as discourse. Such a
problematization shows in Foucault’s work that discourse is not limited to its content,
independent of its functioning and its effects; and, even less, limited to its devices,
and internal connections and to the persuasive or demonstrative strategies used in it.
What is at stake then is not a matter of representation of a preexisting object not even
of creating through discourse an object that does not exist. Rather, what is at stake is a
set of practices, discursive or not, that makes something enter the ‘game of truth and
falsehood’ (regardless of its form, i.e. moral reflection, scientific knowledge, political
analysis, etc.) (Foucault 1994b: 670).
Thus, in the problematization of discourse, the task that confronts us is to take
charge of the modality of existence of language according to which it is not to be
reduced to that which it signifies. What one is concerned with is the fact of language.
We have to take up the notion of enounced (énoncé) – often translated as ‘statement’ –
at its very root. Foucault considers that the statement is a ‘function of existence’, rather
than a unity of a discourse or a structure. (It is a function because it brings about
effects, more than merely to represent things; see McHoul & Grace 1995: 36–39, for
distinctions between statements and sentences, propositions, speech acts, etc.). The
‘enounced’ implies a modality of existence of a group of signs, i.e. “a modality that
allows it to be in relation with a domain of objects, to prescribe a definite position to
any possible subject, to be situated among other performances, and to be endowed
with a repeatable materiality” (Foucault 1969/1972: 107). Description of the enuncia-
tive function requires considering its conditions of existence, and, in fact, this is the
specific object of the archeological analysis. The analysis is, then, a historical analÂ�
ysis of the conditions under which the enounced has come into existence, and of its
implications.
Thus, we shall call discourse “a group of the statements (énoncés) in so far as
they belong to the same discursive formation”, which is in its turn considered as the
general enunciative system to which a group of verbal realizations belongs (Foucault
1969/1972: 107–108; and 117). The introduction of the concept of ‘discursive forma-
tion’ allows us to find some patterns and regularities (established relations), and then
to speak of clinical, political discourse, etc. Thus, the enunciative system to which such
a group of discourses belong includes: a subject (a position that can be filled by vari-
ous individuals), a referential (which is not exactly a fact, state of affairs, or object, but
a principle of differentiation or formation), as well as an associated field (a domain of
coexistence for other statements), a materiality (which is not only the substance, but
the status of enunciations and the possibilities of their use and appropriation).
Hence, discourse is a wider concept than a mere group of statements: the relations
established within the enunciative system give place to a particular way of organizing
knowledge through discourse. That view confers a dynamic and processual character
to discourse, captured by the concept of ‘discursive practice’. This concept emphasizes
the materiality of discourse and its capacity for action (discourses do things rather
than merely to represent things and events) – in contrast to an immanent and idealized
approach. Discourse does not reflect extrinsic conditions, but rather produces them:
discourse relates elements, concepts, and makes it possible for certain non-Â�discursive
elements to constitute themselves as objects. And it is precisely for this reason, that its
production is neither free nor spontaneous (“one cannot speak of anything at any time”,
Foucault 1972: 44). At the same time, it is recognized that many extrinsic elements
and agents play their part in the production of discourses. However, this process, by
means of which social structures constrain discourses, is not conceived in simplistic
or deterministic terms. In fact, no one single element could determine the emergence
of a discourse (neither economic, nor normative, nor juridical practices, etc.), but the
set of their relations, which are, in their turn, established within discourse.2 This non-
deterministic and non-mechanical interplay between non-discursive and discursive
practices means in Foucault’s view that the logic that produces discourses is not what
directs the operations that build institutions, dominations and relations. And, hence,
it is not legitimate to reduce constitutive practices of the social sphere to the logic that
rules the production of discourses. The question is, according to Roger Chartier, how
can the relations between discourse production and social practices be grasped (see
Chartier 1995 for a confrontation with Foucault’s concept of practice)? The theory that
is then relevant is what comes out of practices. We have to take this into account in
order to study discourse in Foucault.
The term ‘practice’ does not only mean ‘realization’ (an act in contrast to a
Â�system), but the fact that the production of discourse implies rules (formation rules).
Foucault defines discursive practice, as a “body of anonymous, historical rules, always
Â�determined in the time and space that have defined for a given period, and for a given
ïœ².â•… Discursive relations are not, as we can see, internal to discourse: they do not connect Â�concepts
or words with one another; they do not establish a deductive or rhetorical structure between
Â�propositions or sentences. Yet they are not relations exterior to discourse, relations that might
limit it, or impose certain forms upon it, or force it, in certain circumstances to state certain things.
They are, in a sense, as the limit of discourse: they offer it objects of which it can speak, or rather
(for this image of offering presupposes that objects are formed independently of discourse), they
determine the group of relations that discourse must establish in order to speak of this or that
object, in order to deal with them, name them, analyze them, classify them, explain them, etc. These
relations Â�characterize not the language (langue) used by discourse, nor the circumstances in which
it is deployed, but discourse itself as a practice. (Foucault 1969/72: 46).
Michel Foucault 91
social, economic, geographical, or linguistic area, the conditions of operation of the

enunciative function” (Foucault, 1969/1972: 117). This notion demands a dynamic
approach to the materiality of discourse, even if some interpretations of Foucault’s
work have not seen this. An approach like this encompasses both discursive actions
and the production of rules. As a result, a mechanical view of the materiality of
discourse should be avoided.
The set of rules which governs discourse functioning (‘formation rules’) has four
levels at which discursive formations are shaped (objects, enunciative modalities, con-
cepts, strategies), and they grasp the generative power of discourse, as they are the
rules for what it is possible to know. Thus, these rules are, to a certain extent, intrinsic
to discourse, though they are not purely linguistic (grammatical), nor purely extrin-
sic or material. In fact, they establish the connection between both domains. Besides,
socio-historical conditions have to be taken into account to describe these rules.
However, this is not sufficient. The horizon of the search is the discursive event
(Foucault 1969: 38–39), and its description consists of determining its conditions of
appearance and its correlation with other discursive events. Discourses are character-
ized by their exteriority, plurality and character of being practical. Discourses act. They
do things.
In L’Archéologie du Savoir, Foucault focuses on the functioning of discourses. He
does not offer a method, but a way of dealing with discourse, which is linked to certain
analytical practices. As a consequence, this work presents the network of concepts we
have been considering, and which is the core of this proposal. At this point, it should
be underlined that, as Hodge (1984) says, it is possible to identify at least two embrac-
ing meanings of discourse in Foucault: ‘discourse as process’, which emerges in relation
to its generative power, and ‘discourse as a system of rules’ of production and control
(and restriction) of knowledge and speech. This more static view of discourse seems to
be related to the cumulative effect of knowledge, which is captured by the concept of
‘archive’ (which is not an amorphous collection of texts collected in a given period, but
texts grouped together according to their multiple relations, and with specific ‘regu-
larities’; that is regularities in the group of relations established between a discursive
practice, like psychiatry and the status and juridical conditions of jurisprudence, or
the types of public assistance, etc. (Foucault 1969: Ch. 5). However, again different
interpretations are left open, and as Maingueneau (1998: 4) claims, the archive is also
linked to the idea of production and transformation of knowledge.
In this case, the plurality of readings is particularly clear, from more static to more
dynamic interpretations of Foucault’s concept of ‘discourse’, from some which are
agent-less, and more depersonalized, to others socially involved, and which are mainly
concerned with the interplay between the exteriority and the interiority of discourse.
It seems to us that this multiplicity of readings has its roots in the Protean nature of
Foucaultian concepts, which is fully coherent with his undogmatic understanding of
concepts as tools which operate in different ways at different moments, depending on

the nature of the phenomena to be explored. It is precisely this shifting nature of some
Foucaultian concepts, such as ‘discourse’ that is essential in ‘problematization’, given
that it allows us to explore and reach new interpretations, and to shape new objects of
knowledge.
2.2â•… Discourse, knowledge and power

The view of discourse as social practice, shared by almost all the current trends in
discourse analysis, is inextricably linked to Foucault’s understanding of the relation
among discourse, knowledge and power. Nothing can be understood separately,
because everything is related. Thus, it is usually emphasized that Foucault opens
up an entire field of experiences, and points out the necessity to experience the
limits of language. However, the matter should not be reduced to the existence
of ‘Â�experiences’ (for example, the experience of madness, disease, sexuality), nor
to ‘knowledges’ (i.e. psychiatry, medicine, sexology), nor to the ‘power’ that is
exerted in institutions that endeavor to control the individual. Rather, it is a matter
of knowing exactly how this form of knowledge and this sort of power have come
into existence; i.e. how these experiences are related to knowledge and to power
(Foucault 1994a: 147–148).
In relation to power, it is not enough to ask what power or freedom are; we have
to consider power as actualized in power relationships, and liberation as embod-
ied in practices of freedom; as strategies for governing both one’s own and others’
behaviour. This means that it is not so much a matter of the analysis of a specific
power, but of the analysis of strategic relations, techniques of governing and states
of domination.
The same is true of the concept of reason. Instead of the idea of the existence of a
founding act of reason, Foucault puts forward the idea of the self-creation of reason
and discusses the need to analyze the forms of rationality (and, irrationality is, after
all, a form of rationality). Also in regard to the subject, the question is to study the
ways in which the subject could be inserted as an object of knowledge in a truth game
in which it has a relation to itself: through which forms of rationality, in which historic
conditions and with what cost.
Power and knowledge need and imply one another, and their productive relations
are incarnated in discourse. According to Foucault, discourse produces power, rein-
forces it, but also exposes it, makes it fragile and permits its localization and deten-
tion (Foucault 1976: 123). Indeed, “the relations of power are, above all, productive”
(Foucault 1975b: 2–5). Instead of locating power in the state apparatus, Foucault sees
a multiplicity of ‘relations of force’; instead of its subordination to the economic order,
he proposes its integration into the mode of production; and instead of a power that
Michel Foucault 93
gives place only to ideological knowledge, he charts a power that produces the real.
That favors, or encourages, certain practices.
Finally, the power/knowledge dyad needs the participation of a third axis,
Â�discourse, which, in Foucault’s view, is inextricably involved in the exercise of power
relations, and in the production of forms of rationality. The emphasis on the generative
power of systems of knowledge and belief attributed to discourse entails the conceptu-
alization of discourse as a type of action. Thus, Foucault proposes studying it in terms
of the techniques through which it is exercised, and, in particular, those mechanisms
which produce and put into circulation discourses which appear as true, and which
convey particular forms of power.
In this framework we have to understand Foucault’s ‘process of objectivization’, in
which the three axes examined play a significant role in the emergence of an object
of field of knowledge. An example can be found in Foucault (1975a), in which the
establishment of criminology, and other human sciences, like clinical medicine, is
understood in terms of a shift in the understanding of ‘illegal practice’ related to the
socio-political and economic changes which took place in the second half of the eigh-
teenth century. These transformations had different consequences – among them, the
higher juridical and moral value placed on property relations, and a correlative exten-
sion and refinement of punitive practices (stricter methods of surveillance, a tighter
classification of the population, more efficient techniques of locating and obtaining
information). These changes in ways of punishment and a new organization of the
power to punish were the bases of the appearance of new fields of knowledge, Â�especially
criminology. As a consequence, two processes of objectivization emerged at the end of
the eighteenth century. First, the definition of offenses, the fixing of a scale of penalties,
rules of procedure, the definition of the role of magistrates (Foucault 1975a). To carry
out all of this, new discourses were generated (exhaustive laws, sufficiently Â�precise for
each type of offense, which entail the consolidation of a legal jargon). Second, there
was a scientific objectivization, by which the criminal was defined as an object of study.
This second process also required the production of new discourses, through which
normality and abnormality are clearly dissociated, and in which the rules of normality
to be followed were made explicit. These rules sought to homogenize, and to classify
all kinds of behavior, both linguistic and non-linguistic.
Thus, the formation and accumulation of new types of knowledge produce a
Â�multiplication of the effects of power. From this point of view every discursive practice
can contribute to creating, reinforcing or questioning the present status quo and the
values which support it.
In current discourse analysis, the question of power is presented as inseparable
from the question of knowledge (social representations), and even as the main focus
of interest in the analysis. And the way it is dealt with echoes Foucault’s views. Thus,
the role of discourse in the shape of social cognition is generally upheld in Â�discourse
Â�
analysis. Through discourse, different views of the social world are built up, they
become consolidated, and they are persuasively transmitted, although not every
Â�discourse is equally socially influential. However, as ‘critical’ analyses show, the ques-
tion of power not only entails the regulation and control of discourses, but also how
control over individuals, social groups, and classes is exercised through them (e.g. by
a precise division between us and them – such as immigrants, delinquents, women,
etc. –, by the projection of a negative and biased representation of them, or by their
exclusion or rejection). In fact, the goal of the analysis in many critical analyses has
been to explore how the ‘subjection of discourses’, and the ‘subjection of individuals and
social groups’ interact.3 Indeed, Foucault’s work, claiming that the subject is constituted
by relations of power, provides a key explanation of how power relationships permeate
subjectivity (Foucault 1986, 1981). By the internalization of dominant and legitimized
discourses – and the knowledge they convey – this knowledge becomes essential in
building up a particular view of the social world, and in defining identities; that is,
in determining in what sort of world individuals are living, as well as who are they,
and what they are like. Unfortunately, this dimension has not always been studied in
depth, not even in critical approaches. (For the understanding of certain strategies for
Â�domination – especially discursive strategies – and their consequences on Â�individuals,
see Martín Rojo 1998: 95–97, 2010).
Today’s authors often approach this generative power through its relation to
Â�ideology, and ideological manipulation. This is, for instance, the case in van Dijk’s dis-
cursive approach to social cognition (1998). In turn, discursive psychology incorporates
a subjective dimension, particularly through the concept of ‘interpretative repertories’,
proposed by Wetherell and Potter. These repertories are understood as methods for
making sense and managing the speaker’s position in a particular interaction. And
we see how this concept echoes Foucault’s concepts of discursive formation, and con-
ditions of possibility, from the moment repertories are defined as culturally familiar
and habitual lines of recognizable themes, common places and tropes (Â�Wetherell &
Potter 1988; Wetherell et al. 1987). Even if this approach is closer to Â�Foucault’s Â�process
of subjectivization, the analysis is not focused on how dominant and disciplining
discourses are internalized by individuals. The process called by Foucault ‘normaliza-
tion’, a narrowing and impoverishment of human possibilities, linked to the exercise of
power and its internalization, are not fully considered. Instead, to consider its role in
ïœ³.â•… Power is exercised by exercising destructive forces (executions, occupations) or by unin-

terrupted constraints imposed in practices of discipline and training. The exercise of power on
Â�discourse, individuals and groups, produces gestures, habits, and attitudes, often by structuring the
space and time in which people live (producing subjected and practised, docile bodies (Foucault
1975)).
Michel Foucault 95
the process of self-formation, the analyses proposed by this approach mainly look for
variability in the way we fashion our subjectivity: that is (in a discursive approach) in
accounts and formulations: “tracking the emergence of different and often contradic-
tory or inconsistent versions” (Wetherell 1998: 395). The emphasis on this variability
has several effects: while it captures the dynamic, kaleidoscopic and argumentative
nature of identity and ideological stances, at the same time it blurs the interaction
between identity phenomena and social positions and processes.
‘Constructivist structuralist’ stances (such as in CDA, for instance), are mainly
concerned with the constraining role of social structures in the production of
Â�discourses and with the active process of the production of social practices which can
transform social structures. As a result, the proposed analyses not only explore the
role of discursive practices within social practices (discourse as a way of acting in a
particular context), but they also (and even mainly) focus on how discourses represent
social practices, and what the social consequences of such representations are. Hence,
the relationship between discourse and knowledge is mainly seen from the perspec-
tive of its social implications: that is, how discourse produces knowledge, and to what
extent the circulation of discourse contributes to the reinforcement, justification, and
legitimation of particular social representations, beliefs, values and ideologies. These
analyses reveal the power of the construction and reproduction of social structure and
social order attributed to discourse. Thus, for example, changes in the social position
of women, such as their presence in public life, and their increasing independence,
have allowed the emergence of particular discourses and repertories, denouncing,
for example, the prevalent androcentrism in the workplace. The Â�production and
Â�legitimation of these discourses could play a part in social change.
The general acknowledgment of the role discourse plays in the production of
knowledge and in the exercise of power is certainly also reinforced by the view of
discourse as a social practice. And from that assumption it follows, as Foucault has
emphasized, that discourse is socially valued, and, as a consequence, its production, its
reception and circulation, are governed by social rules and constraints. A Â�discursive
order emerges, resulting from ‘interventional’ and appropriation moves, and the
Â�limitations of access to discourse, which are frequent and socially established.
3.â•… The order of discourse
Foucault’s proposal of an order of discourse is inextricably linked to the view of

Â�discourse as an asset with social worth, and is mainly focused on the procedures of
regulation of the production, reception, and circulation of discourse. As a result of
this regulation, it is socially established which discourses can be produced and dis-
tributed, within which context, which features of the discourse can authorize them,
and which features de-authorise them and prevent their circulation. For Foucault,
discourse Â�analysis is not a matter of searching – in a specific domain – for who con-
trols discourse, and who suffers from it, who knows, and who is ignorant. It is, rather,
a matter of deciphering it, thinking of it in terms of spatial metaphors (discursive
regions, domains, order of discourse, displacement, transformation, the polyhedron
of intelligibility). These metaphors are deciphering tools, through which it is possible
to grasp the key places in which discourse is transformed through and by power rela-
tionships.4 What Foucault tries to approach are the conditions in which ‘enunciation’
takes place, given the fact that it is possible to intervene in the economy of discourse,
by means of procedures which control, select, organize, and distribute discourses.
These procedures aim to exercise the power of discourse, to control the possibility
of its emergence, and to avoid its materiality (1971: 10–11). Faced with this, Foucault
asks: “Is there anything dangerous in the fact that people speak, and their discourses
proliferate indefinitely? (Foucault 1971: 10). Foucault establishes three main areas of
control in the discursive universe, which are accomplished through different discur-
sive procedures. They explain why discourses do not emerge and do not flow freely,
why their production and reception is hindered or helped, and why the value assigned
to them is weighted (Whittaker & Martín Rojo 1998). They are the following:
A. Control of discourse power. This takes place through prohibition/censorship, as

well as through the neutralization of the representations (knowledge) discourses
convey. Martín Rojo in her development of Foucault’s proposal shows that this
delegitimation may be detected at three points of discourse conformation: at the
sources of discourse (questioning his/her authority and legitimacy); at other’s rep-
resentations and ideologies (questioning his/he truthfulness and objectivity); at
the discourse itself (presenting it as vulgar or unsuitable). As a result, polarization
(‘division and rejection’ in Foucault’s terms; like: we vs. they; true vs. false; adequate
vs. inadequate) seems to be the main discursive strategy (see for examples, Martín
Rojo 1995; Martín Rojo & van Dijk 1997).
B. Control of the conditions of circulation of discourses. This control takes place
when the presence or the circulation of specific discourses is blocked or limited
within a particular social domain or institution. Thus, if we examine what happens
in some discursive domains, like the media, parliamentary discourses, and even
in everyday conversation, we observe that there is a pervasive unbalanced repro-
duction of discourses: while dominant/legitimated/authorized discourses are
constantly evoked, the discourses of minorities are ignored. Discourses which are
ïœ´.â•… These metaphors are part of the spatial thinking of Foucault, which plays a key role in the
changes he introduced in the way of thinking.
Michel Foucault 97
reproduced become the source of new discursive events (Â�intertextuality), through

which they continue to exist, but are also transformed, while those discourses
which are nor hegemonic often become silenced. If we examine, for instance,
which discourses have been reproduced in relation to the anti-capitalist and
anti-globalization movement after police repression following the G8 Meeting in
Genova, we will find very few examples in which the discourse of these movements
and their representatives is evoked. As a consequence, their political positions and
ideology remain confused, and it is easy to bring about the de-ideologization and
the criminalization of the movement.
Such control over discourse flow reinforces social inequality: the different
Â�distribution of discourses entails a different distribution of knowledge. Some rep-
resentations, values, and ideologies are constantly reproduced, while others are
silenced.
C. Control of the conditions of production of discourses, by establishing restric-
tions and rules – that is, sociolinguistic, linguistic, and rhetorical constraints- in
different social contexts. As a result of such regulation, access to discourses that do
not meet these conditions can be blocked or, at least, hindered. These proceedings
respond to the strategies of “appropriation of discourses” (Foucault 1971: 39–42),
by means of which those who have access to the means of production of discourse
establish the norms (rules, patterns) of production and circulation of discourses.
From such forms of intervention, by means of these procedures, it follows that the pro-
duction, circulation and reception of discourses are not random but ordered. The influ-
ence of this Foucaultian ‘order of discourse’ is evident in current discourse analysis, in
sociolinguistics and pragmatics. On some occasions, this influence arrived through
Bourdieu’s proposal of the ‘linguistic market’. However, in other cases, it is direct and
explicitly claimed, like for example, in Fairclough’s proposal of institutional and societal
‘orders of discourse’ (Fairclough 1992), referring to the totality of discursive practices
within a particular domain (that is a discursive region or institution), and the interdis-
cursive relations and articulation of the discursive formation. And the same is true of
Martín Rojo’s proposal of a ‘social order of discourse’ (1997a), which is mainly focused
on the social implications of the discursive order which emerges from the applica-
tion of Foucaultian procedures, preventing some social groups (i.e. Â�minorities) from
producing, circulating, and legitimating their own discourses and linguistic practices.
4.â•… Rethinking the analytical practice
As a result of the increasing awareness of the relations between discourse, power and
knowledge, following Foucault’s work, the analytical practice is also reconsidered.
In this sense, the analysts’ concern with the impact of research, and, in particular, their
awareness of social reflexivity seem to lead to a new understanding of the analytical
task, which is related to Foucault’s concept of experimentation. That is, analysis is no
longer understood as a mere description or contemplation of an object, but requires
the involvement of the analyst. However, it is difficult to give more details of Fou-
cault’s influence on the critical turn, today attested in discourse studies, as well as in
Â�pragmatics and in sociolinguistic approaches.
It is certainly true that the ideological meaning of Foucault’s work (that is, for
instance, his position on freedom, social change and social movements, and domi-
nation) has been evaluated in different ways. However, Foucault’s approach to dis-
course is a ‘critical’ one, and his influence can be seen in current approaches. To think
critically is to problematize concepts, to call into question evidence and postulates, to
break habits and ways of acting and thinking, to dissipate the familiar and accepted,
to retrieve the measure of rules and institutions, to show the techniques of production
of knowledge, and the techniques of domination, and also the techniques of control of
discourse. Then starting from this (re)problematization it is possible for citizens to take
part in the formation of a political will (see 1994b: 676–677). In a word, it is a matter
of the genesis, formation and history of concepts.
In fact, Foucault’s procedure of problematization underlies critical Â�perspectives,
but also some of the critical reactions that this kind of analysis has provoked.
Those critiques seem also to respond to a broad tendency to problematize, but also
Â�re-problematize disciplines, institutions, and discourses. As in Foucault’s thinking, it
is not a question of deconstruction, aimed at the creation of new and subsequent con-
secrations. Problematization should be a never-ending process of monitoring social
practices, and, among them, especially research.
A critical perspective is also related to experimentation, and, in addition to making
us aware of the generative power of discourse, increases analysts’ involvement in some
kind of ‘discursive intervention’, by developing an ‘interventional’ social and political
analysis. These attempts have often been found in the last few decades, in sociolinguis-
tics, pragmatics, and discourse analysis. They appear in regulations of linguistic prac-
tices, in guidelines, in linguistic policies for affirmative action and non-discriminatory
language, but also in the analysis of discourse showing linguistic exclusion in different
domains (not only in politics and in education). Among the different aims of these
studies, two stand out: to explore the social implications of discourses; and to take part
in the present discursive order. Both aims seem to be controversial, and both are inex-
tricably related. The analysis of the construction of a representation through discourse
does not end with the study of the discursive procedures and the linguistic resources
involved, but with the observation of their social implications. It is taken for granted
that knowledge of these results can increase speakers’ reflexivity and awareness, and
that could have effects on their communicative behavior, enriching the climate of
Michel Foucault 99
Â�
public debate about the acceptability and social consequences of discourse. This is
what explains analysts’ involvement in education (see, for a wider account, related to
social reflexivity, Martín Rojo 2001: 43–48).
Once research has contributed to showing how discourse is regulated in a
Â�particular society or in specific domains (discursive regions), it is possible to intro-
duce some changes in the social order of discourses. The study of discourse and
the linguistic market can become engaged in specific, discourse-focused struggles,
in particular in relation to social exclusion and the control of discursive produc-
tion. This is, in fact, the main reason why the study of how discourse is produced
and monitored in a society should be considered a social practice: the analysis itself
Â�creates the conditions of possibilities for some new discourses and representations.
For instance, by showing the implications of the use of terms, like ‘illegal Â�immigrants’,
it is possible to create the conditions for a different understanding of migrations and
cultural diversity.
At the moment, the misinterpretation suggesting that changes in discourse bring
about changes in social conditions is obviously a simplification, and a narrow and
deterministic view of the discursive and socio-political order (For a wide theoreti-
cal grounding, see Chouliaraki & Fairclough 1999; Martín Rojo 2001). But changes
in linguistic uses could, in conjunction with other social transformation, lead us to
question some beliefs, usually presented as ‘natural’, and could create the conditions
of possibility for new social representations (e.g. women’s empowerment). Now we
can see to what extent the ‘problematization procedure’ leads us to rethink the role of
natural evolution and of cultural intervention in linguistic and discursive changes, and
re-opens the pervasive debate between different kinds of universalism and relativism,
and between more relativist-constructionist and more social determinist perspectives.
Besides this guiding influence of Foucault in relation to critical perspectives and
the interventionalist aim in current analyses, there are other, more specific, attempts
to incorporate a Foucaultian approach and concepts in the analysis. Within discourse
analysis, we find outstanding examples in the ‘French discourse analysis tradition’,
in the consideration of the effects of discourses in ideological and social position-
ing, particularly in Pêcheux’ work (1982, 1988), and in further developments, like in
Maingueneau’s study of discursive formation (1998), and in Coutine’s notion of text
heterogeneity and ambivalence (1981). Outside the French tradition, the incorpo-
ration of Foucault’s perspective is clear in critical discourse analysis, and in specific
domains (the study of gender relationships, particularly in Lazar (1993), and Â�Martín
Rojo (1998); political discourse, organizational discourse, etc.). This is often signaled
in Fairclough’s early analysis (1992), but it is also present in Kress (1993), in Maas
(1989) and Jäger (Jäger & Jäger 1993), who incorporate a historical perspective; see also
Â�Fairclough & Wodak 1997). However it is also evident in qualitative sociolinguistics.
The widening of the understanding of discourse as a social practice which produces
and organizes knowledge, and through which power is exercised, has strengthened
all the approaches interested in the relationships between language and context. As
Duranti notes (1997: 11–12), that is particularly clear in linguistic anthropology: in
the study of speaker’s moves within conversations – the significance attributed by
Foucault to spatial metaphors (‘region’, ‘domain’, etc.) helps to understand how speak-
ers’ spatial configurations are limited by power relationships; but also at a macro-
level, Foucault has contributed to a deeper understanding of the regulations of the
Â�production and circulation of discourses in institutions (see, for instance, Sarangi &
Roberts 1999).
Pragmatics is possibly the field which seems less directly influenced by Foucault.
Nevertheless, his influence can be perceived, especially in relation to the continuing
debates between the interactionalist and cognitive approaches. In particular, Fou-
cault’s ideas can be detected in the criticism that the approaches more detached from
social conditions and social relations (and which often claim to be more ‘linguistic’
and ideologically neutral), are in fact operating on the basis of a theoretical but also
ideological choice by means of which the role of social structures and conditions is
dismissed, while pragmatic phenomena are considered linked to cognitive and indi-
vidual phenomena (like speaker’s intentions). However, a deeper incorporation of the
Foucaultian triad discourse-knowledge-power in interactionalist approaches could
be fruitful, in particular in the study of conversational inferences in cross-cultural
communication (following, for instance, Gumperz’s proposals). In fact, in asymmetric
interactions in which some members of the majority (doctors, social workers, teach-
ers, employers) act as gate-keepers, misunderstandings and conflicts can be seen as
consequences of the moves of appropriation of discourse and knowledge, and as a
means of domination. Finally, reading Casanova’s study (1986) of stylistic ‘fractures’
in Foucault’s writings, used by this author as a way of producing new perspectives and
new interpretations, we see how useful his views could be in the study of the organiza-
tion of conversation, of the cooperative building up of meaning (conversational infer-
ences) and social relations (politeness). One of the reasons for this limited interest in
Foucault’s work could be that pragmatics has not yet started a deep internal process of
problematization.
5.â•… Conclusions
In this paper, we have tried to show that, although difficult to assess precisely, the influ-
ence of Foucault’s thinking in the study of discourse is both deep and wide. Foucault
charted a completely new understanding of discourse in its relation to knowledge and
power, and proposed some influential procedures and guiding notions, such as prob-
lematization or experimentation. His understandings of and approaches to discourse
Michel Foucault 101
helped to create a new and interdisciplinary field of study, discourse analysis, and led
to a critical perspective and an interventional aim in the analysis of discourses.
His polyhedral and Protean contribution has led to different developments, and
to multiple readings. This multiplicity is in line with the instrumental and Â�strategic
nature attributed to concepts, and the pervasive aim of problematization and
re-problematization.
Finally, we have seen how his influence could be even more productive, in
Â�particular in relation to pragmatics, not only because the study of Foucault’s
Â�discursive practices could be revealing for the relation between modalities of think-
ing and the use of pragmatic procedures, but also because of the need for a process
of Â�problematization of the discipline, which can contribute more to rethinking the
Â�relationship between language and context, and to a wider process of production of
a new global theory of language. In this process of change, the critical turn seems to
be a necessary step.
References
Bourdieu, P. & L. Wacquant (1992). An Invitation to Reflexive Sociology. Polity Press.

Casanovas, P. (1986). Notas sobre pragmática y metapragmática en el lenguaje literario de Michel
Foucault. In R. Maíz (ed.) Discurso, poder, sujeto. Lecturas sobre Michel Foucault: 69–96.
Â�Universidad de Santiago de Compostela. (1998). Foucault’s pragmatics. Paper read at Sixth
International Pragmatics Conference.
Chartier, R. (1995). Généalogie et architecture de l’oeuvre: Foucault lecteur de Foucault. In Les
Cahiers de la Ville Gillet 3: 188–203.
Chouliaraki, L. & N. Fairclough (1999). Discourse in Late Modernity. Rethinking Critical Discourse
Analysis. Edinburgh University Press.
Courtine, J.-J. (1981). Analyse du discours politique. Langages 62: 69–96.
Dreyfus, H.L. & P. Rabinow (1982). Michel Foucault: Beyond structuralism and hermeneutics.
Â�University of Chicago Press.
Ducrot, O. (1968). Le structuralisme en linguistique. In O. Ducrot et al. Qu’est ce que c’est
l’estructuralisme?: 15–96. Le Seuil.
Duranti, A. (1997). Linguistic anthropology. Cambridge University Press.
Enrici, A. (1999). Lecturas Continentales de Foucault. La interpretación de Michel Foucault desde
una división continental de su lectura. Revista Espacios 13, año 4: 26–35.
Fairclough, N. (1992). Discourse and Social Change. Polity Press.
Fairclough, N. & R. Wodak (1977). Critical Discourse Analysis. In van Dijk (ed.) Discourse as social
interaction: 258–284. Sage.
Florence, M. (1994). Foucault, Michel, 1926. In G. Gutting (ed.) The Cambridge Companion to
Â�Foucault: 314–319. Cambridge University Press.
Foucault, M. (1966). Les Mots et les Choses. Une Archéology des Sciences Humaines. Gallimard; [The
Order of Things. Routledge, 1970.].
——— (1969). L’Archéologie du Savoir. Gallimard; [The Archeology of Knowledge. Routledge, 1972.].
——— (1971). L’ordre du Discourse. Gallimard.

——— (1975a). Surveiller et punir. Naissance de la prison. Gallimard; [Discipline and Punish.
The Birth of the Prison. Penguin Books, 1991.].
——— (1975b). Pouvoir-corps, Quel Corps 2: 2–5.
——— (1976). La volonté de savoir. Histoire de la sexualité, vol. I. Gallimard; [The History of Sexuality,
vol. I: An Introduction. Penguin, 1981.].
——— (1986). Pourquoi étudier le pouvoir: la question du sujet. In H. Dreyfus & P. Rabinow (eds.)
Michel Foucault. Un Parcours Philosophique. Gallimard.
——— (1994a). Omnes et singulatim: vers une critique de la raison politique. In Dits et écrits, Paris:
Gallimard, vol. IV: 134–161.
——— (1994b). Le souci de la verité. In Dits et écrits, Vol. IV: 668–678. Gallimard.
Gabilondo, A. (1990). El discurso en acción. Foucault y una ontología del presente. Anthropos.
——— (1996). Introducción y compilación, Michel Foucault, De lenguaje y literatura: 9–59. Paidós
I. C. E./U. A. B.
——— (1999). La creación de modos de vida, Introducción, traducción y edición, Michel Foucault,
Estética, ética y hermenéutica. Obras esenciales III: 9–35. Paidós.
Harris, R. (1988). Language, Saussure and Wittgenstein. How to play games with words. Routledge.
Hodge, R. (1984). Historical semantics and the meaning of discourse. Australian Cultural Studies
2(2): 124–130.
Kress, G. (1993). Against arbitrariness: the social production of the sign as a foundational issue in
critical discourse analysis. Discourse & Society 4(2): 169–191.
Jäger, S. & M. Jäger (1993). Aus der Mitte der Gesellschaft. Duisburg: Diss.
Lazar, M. (1993). Gender, discourse, and semiotics: the politics of parenthood representations.
Â�Discourse & Society11(3): 373–401.
Maas, U. (1989). Sprachpolitik und politische Sprachwissenschaft. Suhrkamp.
Maingueneau, D. (1984). Genèses du discours. Pierre Mardaga.
——— (1998). Archéologie et analyse du discours. Readings of Foucault. Paper read at Sixth
Â�International Pragmatics Conference.
Martín Rojo, L. (1995). Division and rejection: from the personification of the Gulf conflict to the
demonisation of Saddam Hussein. Discourse & Society 6(1): 49–79.
——— (1997a). El orden social de los discursos. Discurso 21–22: 1–37.
——— (1998). Intertextuality and the construction of a new female identity. In M. Bengoechea &
R. Solá Buil (eds.) Intertextuality/Intertextualidad: 81–98. University de Alcalá de Henares.
——— (2001). New developments in Discourse Analysis: discourse as social practice. Folia Lingüística
XXXV/1–2: 41–96.
——— (2010). Constructing Inequality in Multilingual Classrooms. De Gruyter.
Martín Rojo, L. & T. Van Dijk (1997). “There was a problem and it was solved!” legitimating the
expulsion of ‘illegal’ migrants in Spanish Parliamentary discourse; Amsterdam: Discourse &
Society 8(4): 563–606.
Mchoul, A. & W. Grace (1993). A Foucault Primer. Discourse, power and the subject. UCL Press.
Pêcheux, M. (1982). Language, Semantics and Ideology. Macmillan.
——— (1988). Discourse — structure and event? In C. Nelson & L. Grossberg (ed.) Marxism and the
interpretation of culture: 633–650. Macmillan.
Sarangi, S. & C. Roberts (1999). Talk, work, and institutional order. Discourse in medical, mediation
and management settings. Mouton.
Michel Foucault 103
Van Dijk, T. (1996). Discourse, power, and access. In C.R. Caldas-Coulthard & M. Coulthard (eds.)
Texts and practice. Readings in critical discourse analysis: 84–104. Routledge.
——— (1998). Ideology. Sage.
Wetherell, M. (1988). Positioning and interpretative repertories: Conversation analysis and
Â�post-structuralism in dialogue. Discourse & Society 9 (3): 387–412.
Wetherell, M. & J. Potter (1988). Discourse analysis and the identification of interpretative Â�repertories.
In C. Antaki (ed.) Analysing Everyday Explanation. Sage.
Wetherell, M. et al. (1987). Unequal egalitarianism: A preliminary study of discourse concerning
gender and employment opportunities. British Journal of Social Psychology 26: 59–71.
Whittaker, R. & L. Martín Rojo (1998). A dialogue with bureaucracy: register, genre and information
management as constraints on interchangeability. Journal of Pragmatics 31: 149–189.
H.P. Grice
Frank Brisard
Paul Grice is one of those remarkable twentieth-century philosophers’ philosophers

who have greatly influenced styles of philosophical thinking (and writing1) without
ever, or so it seems, gaining even the slightest form of notoriety outside the academic
fraternity. In this capacity, he is in the company of equally ‘unmemorable’, more or
less contemporaneous, thinkers like J.L. Austin, A.J. Ayer, G.E. Moore, or Gilbert
Ryle, most of them based, at one time or another, in Oxford, and most of whom were
actively – though in some cases only indirectly or even adversely – involved in the
development of something like a world center of ‘ordinary language’ philosophy.
One of the more serious drawbacks of this disposition, however, is the relative unfa-
miliarity, in the orthodox reception of Grice, with the global (integrated) Â�picture that
is being presented, notably in much of his contribution to the philosophy of Â�language.
This has largely resulted in the proposal of a whole series of counterexamples to the
original ‘Meaning’ (Grice 1957) hypotheses and in some fiddling around with Grice’s
conversational maxims (and sometimes with the Cooperative Principle itself), as well
as in very technical discussions concerning the status of different types of meaning
(semantic, pragmatic). More recently, and in particular following the first (1986) edi-
tion of Sperber & Wilson’s Relevance, linguistic debates have focused upon the precise
nature of the interface that is postulated to exist between the domains (or rather, the
methodologies) of semantics and pragmatics. This interface, then, defines a number
of conditions that are hoped to replace various extra (perhaps superfluous) theoretical
concepts that were, in the past, prompted by the ‘explosion’ of meaning (Turner 1999),
i.e. its fragmentation into numerous more or less improvised distinctions of meaning
types that proved more often than not to be dependent on fairly Â�contingent contextual
parameters. In general, therefore, linguists working in the line of Grice have tended
to focus almost exclusively on the theme of ‘pragmatic intrusion’, or the idea that the
old Gricean dichotomy between what is ‘said’ and what is ‘implicated’ can be refined
by distinguishing between minimal propositions, their ‘expanded’ forms, and genuine
ïœ±.â•… Here’s a quick sample of what the received view of Gricean writing may look like: “grice, n.
Conceptual intricacy. ‘His examination of Hume is distinguished by erudition and grice.’ Hence,
griceful, adj. and griceless, adj. ‘An obvious and griceless polemic.’ pl. grouse: A multiplicity of grice,
fragmenting into great details, often in reply to an original grice note” (in Dennett 1987).
H.P. Grice 105
inferential (argumentative) work on the part of the hearer, as guided by the conver-
sational hypothesis. As such, Grice is one of the prime instigators of a ‘Â�radically prag-
matic’ take on meaning that suggests the viability of maintaining a rigorous Â�theoretical
distinction between semantics and pragmatics, while ensuring a paradigmatic conti-
nuity in the formal study of syntax, semantics, and pragmatics.
Let us not forget, though, that radical pragmatics is largely based on the Â�ultimately
unjustified (cf. Grice 1978: 119) assumption “that it is more generally feasible to
strengthen one’s meaning by achieving a superimposed implicature, than to make
a relaxed use of an expression”. Not only does this pre-theoretical orientation hinge
Â�crucially on the acceptance of (propositional) logic as a kind of universal (and suf-
ficient) semantics, which is somehow acknowledged in much of pragmatic work. But
many of these and similar assumptions made by Grice himself (and actually presented
as suppositions at work in language users’ own understandings of utterances) are also
directly related to the generally presumptive or ‘projective’ nature of communication
and interpretation, which cannot be explained without reference to the rational prop-
erties that should be ascribed to speech participants. Most existing overviews of Grice’s
work, especially those targeted at a linguistic and/or cognitive audience, choose to
ignore issues of this rational grounding of Grice’s philosophical project. In what fol-
lows, I will try to fill out this gap (but not fill it in), indicating links to Grice’s views on
(philosophical) psychology, ethics, and metaphysics. It will be suggested, though hardly
argued, that it is these nonlinguistic considerations solely that can provide the neces-
sary and ultimate rationale for Grice’s rationalist account of meaning. The structure
of this exposition will explicitly follow some of the lines set out in Grandy & Warner’s
(1986a) excellent introduction to Grice’s thinking. Insofar as possible, I will refer to
Grice’s original publications, in order to convey a feeling of the historical progression of
his thinking. It should be pointed out, though, that most of his papers directly Â�relevant
to the development of linguistic pragmatics can be found in Grice (1989), which also
contains an important ‘Retrospective epilogue’.
1.â•… Life
Grice (1913–1988) was academically formed in Corpus Christi College, Oxford, where
he began his philosophical curriculum, including the study of Plato and Aristotle (and
more specifically of the latter’s Nicomachean Ethics), with W.F.R. Hardie. Grice Â�himself
acknowledges the strong influence that Hardie seems to have had on the develop-
ment of a sense of rationality informing much of his own philosophical work in later
times, including the belief “that philosophical questions are to be settled by reason,
that is to say by argument” (Grice 1986: 46; emphasis in original). This early period is
already of some interest to the student of Grice’s ideas, because in it we see the germs
106 Frank Brisard
of Grice’s preoccupations with moral theory, especially the kind that is of Aristotelian
descent, as well as his relative dissatisfaction with the way in which logic and logical
Â�connections are traditionally conceived of as exhausting the range of tools needed
for solid Â�argumentation. His demonstration of the largely argumentative structure of
discourse (or conversation), moreover, will figure among Grice’s main contributions to
the study of (nonlogical) meaning in natural language.
In Oxford, Grice spent the bulk of almost thirty years teaching. In this capacity, he
managed to have a considerable impact on (at least) one of his students, later colleague
and friend, Peter Strawson,2 as well as on the occasional sabbatical visitor, such as John
Searle. In 1939, he became a Fellow of St John’s College, where he would remain until
the presentation of his William James lectures in 1967. During this time, Oxford tried
to recover from a rude awakening caused by Ayer’s introduction of logical Â�positivism,
as a new style of linguistic philosophy. Probably the most renowned answer to this
challenge came from what would later be known as ‘ordinary language’ philosophy,
in actuality made up of several heterogeneous groups of philosophers, many of whom
were at one point or another concerned with the particularities of linguistic usage and
their possible relevance to the study of philosophical problems. Grice himself mainly
took part in the discussions that took place on Saturday mornings under the lead-
ership of Austin, more affectionately remembered as ‘The Play Group’. Behind these
language games, there was first and foremost a very strong legalist interest, oriented
towards the types of meaning distinction that can be found in sentence pairs like I shot
your donkey accidentally and I shot your donkey inadvertently. Presumably, none of
these discussions included the words semantic or pragmatic.
Whether the inspiration came from Austin, Ryle, or Wittgenstein, ‘ordinary
Â�language’ philosophers in Oxford stressed the idea that discourse must be grounded in
some collection of metaphysical judgments, or world picture. Austin, who felt sympa-
thetic to Moore’s ‘Defence of common sense’ (Moore 1959), recast this idea in terms
of a natural metaphysic that could be discovered by the philosophical Â�investigation
of detailed features of ordinary, i.e. nontechnical, discourse. Only then, or so the
Â�argument would go, could a sound foundation of philosophical thinking be achieved.
Austin’s own main concerns were truth and truth telling, and in his philosophical
sketches he certainly displayed less respect for logic than, e.g., Grice. Still, this over-
all picture left room for much diversity. The exact relationship between linguistic
Â�phenomena and specific philosophical theses would remain a heated topic of debate,
ïœ².â•… With Strawson, Grice would publish ‘In defense of a dogma’ (Grice & Strawson 1956). Later,
Grice also indicates his collaboration with Strawson on predication and Aristotelian categories,
some of which appeared in Individuals (Strawson 1959). Grice repeatedly expressed his respect
for the work of Strawson, e.g. in Grice (1981), where he proposes a ‘conversationalist’ alternative
to one of Strawson’s influential contributions to formal semantics, the analysis of presuppositions.
H.P. Grice 107
even among the members of the Play Group. Moreover, they had to defend Â�themselves
against the perennial accusation of decadent linguistic snobbery haunting (this newest
brand of) ‘English Futilitarianism’. From outside, it may indeed have looked at times
as if a decaying academic establishment, immersed in a classical education, was try-
ing to keep control of a discipline that threatened to escape its ancient clutches. It
would do so by cultivating an idiom that could only be acquired from inside (the
establishment) and that was supposed to suggest the refined linguistic sensitivities of
its Â�speakers. Ultimately these scholars, it was thought by some, are not concerned with
the nature of reality, the proper subject of philosophy, but with its mere representation
or appearance.
In 1967, Grice presented the William James lectures in Harvard (revised in Grice
1989). In that year, he was also appointed Professor of Philosophy in the University of
California at Berkeley. Globally, this period marks the beginning of Grice’s increased
interest in more technical treatments of linguistic phenomena (as well as the Â�beginning
of a more productive period in his life, in terms of producing tractable records). Grice
had published an important essay on ‘Meaning’ (Grice 1957), which set the stage for
subsequent discussions of (and distinctions between) ‘utterer’s meaning’ and ‘utter-
ance-type meaning’. The intentional structure of meaning assignments was herewith
revealed (cf. Schiffer 1972 on such meaning or ‘M-intentions’). But it is only in the
William James lectures, which received a great deal of attention in America, that Grice
would develop, and partly revise, his analysis of the relevant conditions under which
meaning can be held to rely on (the structure and content of) a speaker’s M-intention.
This elaboration is already succinctly outlined in Grice (1968). At the same time, it
was this very project which had somehow forced Grice to find his philosophical for-
tune in the United States from then on, where he would benefit from closer and more
intensive contacts with experts in logic and linguistics. For Grice, Noam Chomsky and
W.V.O. Quine are two of his more notable models in this respect. Both scholars have
concerned themselves with finding out what a ‘suitable theory’ could be (in formal
syntax and the philosophy of science, respectively), and they are both advocates of a
strong methodological apparatus for tackling the more intractable regions of philo-
sophical investigation. The methodology in question, when it comes to studying the
grammar of ordinary discourse, should ideally lead the researcher to see that grammar
as reflecting properties of an underlying logical form. Grice deplored the fact that,
though Quine and Chomsky are so strongly united in their quest for methodological
rigor, they never seemed to agree on the theoretical fundamentals for the analysis of
natural language. Grice himself would try to integrate their respective positions in an
unfinished undertaking that aimed at presenting a syntax-cum-semantics with mini-
mal use of transformations. These investigations in formalistic philosophizing were
never published. Ironically, Grice, who appreciated the modest level of sophistication
marking his own technical machinery, eventually decided to retreat from an overly
108 Frank Brisard
formalistic treatment of natural language, partly prompted by Hilary Putnam’s remark

that his philosophy had become too formal.
Towards the end of his life, Grice became more and more engaged in areas outside
the philosophy of language, including philosophical psychology/biology (dealing with
the Aristotelian theme of ‘life’), metaphysics, and Kantian ethics (with Judith Baker).
These domains, as Grice remarked, are much less amenable to a formal treatment.
And so we return to Grice’s initial qualms with the hegemony of logical thinking in
Â�argumentation, when technology starts to squeeze a developing philosophical ‘style’
out of operation. Today, what seemed relevant to Grice’s own preoccupations has
become even more so in the light of subsequent evolutions in neo-Gricean pragmat-
ics (and related areas of interdisciplinary research). If a formal (i.c., logical) apparatus
“began life as a system of devices to combat woolliness [it] has now become an instru-
ment of scholasticism” (Grice 1986: 61; emphases in original). If Grice was dissatisfied
with the unnecessary proliferation of meanings attributed to certain natural language
expressions, he would certainly have been with the terminological inflation associated
with the increasingly technocratic order of current linguistic pragmatics.
A short autobiographical note was included in Grice (1986). In that same volume
(Grandy & Warner 1986b), a list of Grice’s main (book- and article-length) publica-
tions and ‘unpublications’ can be found as well. Many of his manuscripts are deposited
in the Paul Grice Archives, UC Berkeley.
2.â•… Language
2.1â•… Meaning
‘Meaning’ (Grice 1957) constitutes Grice’s first serious attempt to distinguish between
what is involved when we say (roughly) that ‘something means something’, and the
concerns expressed by the collocation that ‘someone means something’. The first
type of meaning,3 as in Smoke means fire or That guy means trouble, is called ‘natural
Â�meaning’, in that it points at a ‘causal’ relation between a sign and what it means. This
type of meaning is not dependent on anything some instance may have meant by the
sign in question. The second, ‘nonnatural meaning’ or ‘meaningNN’ defines instances
ïœ³.â•… In a truly pragmatic spirit, Grice distinguishes between two types of use of the English expres-
sion mean(ing). In this sense, the present emphasis on language use should not be confused with
how Grice will, later on, draw a distinction between logical and derivative aspects of ‘meaning’ (the
philosophical object, not the linguistic expression; cf. the Prolegomena to ‘Logic and conversation’
in Grice 1989).
H.P. Grice 109
of intentional communication, as in When a man says yes, he means yes, as opposed to

the incidental transfer of information. MeaningNN will ultimately have to be explained
in terms of one or another occasional sense of mean, or rather, language, as a human
institution, functions as an artificial substitute for natural signs. The best way to go
about this explanation, according to Grice, is to reduce the so-called ‘timeless’ mean-
ing of a linguistic expression to instances of intentional communication in which an
agent – an utterer – is involved on a particular occasion of the use of that expression.
For Grice, the content of a meaning (intention) is identified with the intended effect
that an utterance (qua object or event) should trigger.
In his paper, Grice cuts up the concept of utterer’s (or speaker’s) meaning into
three parts: A speaker ‘meansNN’ something by x, which can be any meaningful sign
(from a mere hand wave to full-blown linguistic phrases or sentences), if and only if
she intended x to produce an effect in the hearer by means of the recognition of this
intention. The effect at issue may be anything from a belief to an action prompted by
the utterance of x, and the recognition involved in interpreting meaningful Â�behavior
reveals the reflexive (yet nonparadoxical; cf. Grice 1957: 384) nature of communi-
cative intentions.4 Later, utterance-type, or sentence, meaning is seen as equated,
though in a largely underdetermined way, with underlying M-intentions. The result
of this move is that utterance-type meaning is effectively handled as standardized or
Â�conventionalized5 utterer’s meaning, which fits in with the Gricean program of Â�tracing
semantics back to (propositional attitude) psychology. This account is meant, by Grice,
to supplant so-called ‘causal’ theories of meaning, which can only deal with standard,
semiotic meaning and which in any case do not seem to be able to go beyond the
identification of tendencies, attributed to utterances, to produce certain Â�(cognitive
or behavioral) effects in an audience. In contrast, Grice considers that what users
ïœ´.â•… For a critique of the nesting or iteration of (representational within) self-referential commu-
nicative intentions, see, among others, Strawson (1964), Searle (1965), and Schiffer (1972), who
generally object to the ‘infinite regress’ affecting this type of reasoning. For a defense of reflex-
ivity, see, e.g., Levinson (1983: 16–18) and Bach (1987) – a reply to Recanati (1986). Grice himself
(1969: 158) remarks that “one cannot have intentions to achieve results which one sees no chance
of achieving”. In practice, Grice therefore argues, rationality stops the vicious cycle suggested by the
reflexivity of communicative intentions via a principle of calculability, which will show up again in
the discussion of implicatures and which states that a speaker cannot, and will not, expect a hearer
to calculate complex meaning intentions and sub-intentions that are contextually implausible.
ïœµ.â•… The conventionalization of utterer’s meaning, yielding utterance-type meaning, may be a
theoretical turn that is due more to Schiffer’s (1972) interpretation than to Grice himself. It is
Â�‘de-essentialized’ in Davidson (1986). Grice (1969: 160ff.) already suggested, a.o. contra Searle, that
the conventional meaning of a sentence may not always be relevant in determining meaningNN and
that, in general, the meanings of sentences can only be seen as special cases of meaning something
by an ‘utterance’ (in Grice’s extended sense).
110 Frank Brisard
do (or should) meanNN by a particular utterance is brought about by the audience’s

Â�recognition of an M-intention, and that this recognition is presented as a reason, not
a cause, for believing or doing something. Such reasons, as opposed to causes, are
intrinsically argumentative and thus constitute the core of Grice’s rational account of
communication.
In Grice (1968), a subtle modification is introduced with respect to the original
proposal. Instead of the straightforward ‘interventionist’ semantics devised Â�previously,
in which a speaker exerts a direct influence upon thoughts and/or actions of her
audience, the hearer, in this second version, should only intend to do something, or
think that the speaker believes something, as an effect of understanding an utterance.
Accordingly, Grice is able to distinguish the direct intended effect from an indirect
one, e.g. (for indicatives) that the hearer thinks that p herself (an effect which, even if it
is indirect, may actually still constitute the prime interest of an informing utterance).
Grice also pays considerable attention to the problem of (grammatical) mood in the
determination of utterance-type meaning. The general idea is that, if the interpreta-
tion of what the speaker meant by x hinges on the identification of an M-intended
response, generic differences in types of response will correlate with generic
Â�differences in what is meant (Grice 1969: 165). Mood operators, like the indicative or
the Â�imperative, mark (quite literally as distinct auxiliary representations in syntactic
form) how a Â�psychological state, taken as a ‘radical’ propositional6 content, is to be
construed in relation to the hearer: in terms of a belief (one type of generic response)
or an action. Thus, the contextual ‘occasion-meaning’ of an imperative like Pass the
salt could be noted as follows: An utterer U meansNN !(the salt is passed) if and only
if U produces some utterance x such that she M-intends the hearer (i) to think that
U intends (to bring it about) that ‘the salt is passed’, and (ii) to intend herself that ‘the
salt is (Â�somehow) passed’ partly because of, or through, her recognition of U’s ‘first’
M-intention. Insofar as the meanings of different sentence types tend to correlate with
different types of speech act, moreover, the latter can henceforth be distinguished on
the basis of the various propositional attitudes that can be expressed.7 The other side
ïœ¶.â•… I use the term ‘proposition’ in an unexceptional, perhaps Fregean, sense, much like Grice. For a
warning against the problematic character of this notion, see Grandy & Warner (1986a: 9, 28–30).
ïœ·.â•… Habermas (1998: Ch. 5) questions the feasibility of characterizing illocutionary types only
Â�according to the kind of representation of states of affairs and the speaker’s corresponding proposi-
tional attitude. More precisely, he doubts that an analysis of the satisfaction conditions for Â�different
states of affairs represented in the propositional component could do the job of classifying speech
acts in an unequivocal way. He points out in this respect that some speech acts may not belong to
the same type even though they meet the same satisfaction conditions and express identical propo-
sitional attitudes. What would be needed in addition, then, is a specification of the mode through
which a speech act achieves its illocutionary purpose, something like an authorization condition.
H.P. Grice 111
of the speech act equation, that of propositional content, is in turn simply reduced to
mental content (see especially Searle 1986), so that we get a general picture of language
as a vehicle for the communication of thoughts, which are analytically more basic, and
not as the medium of thought itself.
Further refinements are proposed in Grice (1969) and Grice (1982), which
Â�witness an increasing complexity in the technical apparatus that is being deployed.
In Grice (1969), a clear typology of meaningNN specifications is provided. First, the
timeless meaning of a ‘complete’ utterance-type is what would usually be considered
the conventional meaning of a sentence. On any concrete use of a sentence, parts
or the whole of that (token) sentence may turn out ambiguous, though. To resolve
the ambiguity, one needs to refer to the ‘applied’ timeless meaning, which is what a
Â�sentence means here, i.e. on a particular occasion of use. The ‘occasion-meaning’ of an
utterance-type, furthermore, refers to how speakers may mean something by uttering
a sentence beyond the composite conventional meaning of that sentence. Finally, all
of these meaning types, as Grice purports to show, will depend upon, but not neces-
sarily coincide with, a fourth conception of meaning, that of the ‘utterer’s occasion-
meaning’, or what the speaker intended on a particular occasion by uttering x. This
fourth Â�distinction is not only theoretically motivated, in that the whole Gricean proj-
ect depends upon the connection between linguistic meaning and intentionality, it is
also needed if one wishes to deal with instances of indirect meaning (cf. Section 2.2).
For ‘word-meaning’, the same distinctions apply, given that a word or nonsentential
phrase is taken as an ‘incomplete’ utterance-type.
The main challenge in the remainder of this enterprise proved to be utterance-
type meaning, or rather sentence meaning as structured utterance-type meaning. The
notion of a ‘procedure’ that a person may have in her ‘repertoire’ (Grice 1968) turns
out of central importance here.8 For any (communicative) procedure to be Â�successful,
an agent (speaker or otherwise) must assume that her audience will recognize the
Â�intention by means of taking a particular token as instantiating the procedure in
Â�question. Just executing the procedure then counts as giving the audience a reason to
think or intend to do something (routinely associated with that procedure). For this to
work, the agent must also assume that the knowledge of such procedures is shared or
mutual (Lewis 1969; Smith 1982). Of course, when dealing with sentential utterances,
Crucially, such modes do alter illocutionary meaning and are thus not merely extra-linguistic, or
institutional.
ïœ¸.â•… Strictly speaking, a procedure is nothing specifically linguistic or even communicative. If I
see food and use my hands to bring it to my mouth, that’s a procedure. Thus, Grice introduced yet
another concept that allows the reduction of linguistic meaning to general patterns of (purposeful)
behavior.
112 Frank Brisard
which are compositional, there can be no individual procedure attached to every

sentence of a language, because of the generative nature of natural Â�grammars. The
interpretation of sentences, as utterance-types, is therefore dependent upon ‘Â�resultant
procedures’, which are generated by a finite stock of ‘basic procedures’. The latter apply
to particular utterance-types that are components of sentences (lexicon) and to par-
ticular ways of combining utterance-types (syntax). In Grice’s project, there is a real
need to pay attention to syntax, as a set of basic procedures, and incorporate it into his
own meddling with meaning, which explains his interest in Chomsky’s theory.
2.2â•… The conversationalist hypothesis

In Grice (1968: 225), reference is made to the distinction between what a speaker may
‘say’ and what she may ‘implicate’ (imply, indicate, suggest, etc.). The program defined
by this distinction is directed at, among other things, a clarification of the notion of
conventional meaning, the basis of semantics. This type of meaning, and specifically
lexical meaning, ultimately reduces to utterer’s meaning, too. But what a speaker ‘says’
in an utterance, by using a certain combination of words, does not necessarily exhaust
the actual meanings conveyed by it (and intended by the speaker). There may be more,
or even completely different, shades of meaning that she wishes to implicate. What is
important, then, as a matter of methodology, is to make sure that these extra meanings,
as functions of the use of words, are not confused with the words’ conventional mean-
ings. This is where the Gricean antecedent of the semantics/pragmatics dichotomy is
born. Levinson (2000: 13) refers to this as “cutting up the semiotic pie”. (For a critique
of this hypothesis, see especially Cohen 1977.) Grice was particularly apprehensive
about the proliferation of senses affecting certain extravaganzas of ordinary language
philosophy. His Modified Occam’s Razor thus reads: “Senses are not to be multiplied
beyond necessity” (Grice 1978: 118–119). It specifically concerns lexical senses, like
the ones that can be distinguished for the verb believe,9 but also, and perhaps more
importantly, real and derivative meanings of some grammatical counterparts of Â�logical
connectives, like and (pure conjunction, temporal sequencing,…), or (exclusive vs.
inclusive), and if (material vs. strict).
There are quite a few reasons to believe that Grice’s longstanding interest in
the philosophy of language is very much tailored to the proper analysis of condi-
tional sentences, and in particular of the relation between the material conditional
ïœ¹.â•… If one (merely) ‘believes’ something (and says so much), it might be implied that one does not
really ‘know’ whether that something is true or not. This represents one of the prime cases to be
handled by an analysis of presuppositions. According to Grice, the ‘presupposition’ in question is
a matter of the use of believe (i.e. that one is using this instead of a stronger verb like know), not
of its meaning.
H.P. Grice 113
of Â�propositional logic and the indicative conditional that is found in many natural
Â�languages. Any debate on the distinction between meaning and use might thus be seen
as parasitic upon this more basic concern with meaning as truth-conditional (what is
‘said’), which might point to a potential incompatibility with earlier views on Â�meaning
as intentional structure.10 Grice’s point for all of the logical particles listed above is
that strictly nonlogical inferences do not belong to their semantics but develop out
of Â�typical, sometimes even necessarily associated, features of their use. This does not
mean, however, that the way in which such meanings are to be derived is identical
for all of these (and other) expressions. In fact, we may distinguish between Â�different
‘genera’ and ‘species’ of meaning, depending on whether they are seen as part of the
linguistic code (conventional) or calculated on the basis of rational principles of
conversation.
In Grice (1975) and (1978), this fragmentation of meaning is shown to be a
Â�general phenomenon, not just applicable to ‘extraneous’ meanings of logical words
but to any kind of non-truth-conditional inference based on what is ‘said’. The anaÂ�
lysis refers, first and foremost, to the assumption of conversational cooperation,
voiced by the Cooperative Principle: “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 principle is constituted
by four more specific maxims, echoing the Kantian categories of Quantity (informa-
tiveness), Â�Quality (truth), Relation (or Relevance), and Manner. Each of these, or any
of their combinations, may guide the discovery of nonlogical inferences in an utter-
ance, on the assumption that the speaker is in fact being cooperative. Under this con-
dition, then, any kind of breaching or flouting of a maxim, or of maxims, will prompt
the hearer to set up an argumentation (i.e. a series of linked propositions) so as to
Â�safeguard the original assumption of cooperation. Thus, if a professor is asked to write
a letter of reference about one of her students and limits her remarks to the observa-
tion that the student has regularly attended tutorials, the conversational implicature
for this particular context would be that the student in question is not a very good
one. This implicature can come about because it is obvious, to any audience familiar
with the genre at hand, that not enough information has been provided (Quantity).
ïœ±ïœ°.â•… The central part played by truth conditions in the characterization of what is ‘said’ betrays
a preoccupation with truth that is much closer to Austin’s than might be gathered at first blush.
Consider the following revealing quote in this respect: “In my eyes the most promising line of
answer [to the question of the relation between formal logic and natural language] lies in building
up a theory which will enable one to distinguish between the case in which an utterance is
Â�inappropriate because it is false or fails to be true, or more generally fails to correspond with the
world in some favored way, and the case in which it is inappropriate for reasons of a different kind”
(Grice 1989: 4).
114 Frank Brisard
Â�
Conversational implicatures, in general, are calculable because the argumentation
that leads one from observing a flouted maxim to construing a (nonlogically) derived
Â�supposition is Â�transparent and proceeds along rational lines. In addition, they are
Â�cancelable/defeasible (Â�nonmonotonic), nondetachable (i.e. based on content rather
than form, except for Manner implicatures), and nonconventional or pragmatic.
The resulting typology for those aspects of meaning that do not fall under what is
‘said’ looks as follows. Conversational implicatures, i.e. those calculated on the assump-
tion of cooperation, come in two varieties, particularized and generalized. The first type
depends essentially on contextual information that is needed to arrive at a plausible
interpretation of what the speaker may have meant over and above what she ‘said’. The
second, generalized conversational implicatures (GCIs), share all the Â�properties of their
particularized counterparts but also have some sort of default status, in that they apply
regardless of contextual specifications (unless they are explicitly overridden). A GCI
for expressions containing a quantifier like some, for instance, specifies that ‘not all’ of
the members of a designated set are meant, even though the interpretation of some in
terms of the stronger quantifier all is not logically incompatible. This type of implica-
ture is in the process of acquiring a whole new theory of its own (Levinson 2000), which
stresses the idea that not all of the Gricean inferential work belongs to a rhetorical level
of pragmatics, and that in fact some implicatures labeled as GCIs, including cases of
disambiguation, fixing reference, and generality-narrowing, go into the determination
of what is ‘said’ and thus happen before semantic interpretation. Finally, implicatures
can also be conventional (and still not belong to the realm of what is ‘said’), if they are
strictly speaking determined by the conventional meanings of the words to which they
attach without being ‘part of ’ those conventional meanings (insofar as these are truth-
conditional). Conventional implicatures may look like GCIs, because they, too, are sup-
posed to hold over various contexts, but they are not in principle calculated on the
assumption of rational conversational behavior, and they are probably much harder (if
not impossible) to defeat. An example of a conventional implicature is the contrastive
sense of but (Grice 1961), which attaches to the use of this expression in (virtually) all
contexts but cannot properly be seen as a component of its truth-conditional Â�meaning
(which is conjunctive). All in all, though, it does seem that many of these distinctions
are obscured or at least complicated by the fact that it is ultimately not too clear what is
meant by ‘what is said’, suggesting a less than “systematic philosophical theory of lan-
guage” (Grice 1989: 4) on the part of Grice.
2.3â•… Rationality
Implicatures demonstrate the rational character of conversational behavior in two ways.
First, their calculation rests on transparent chains of propositions, which Â�constitute an
argument, the stuff that rationality is made of. Secondly, their very identification relies
H.P. Grice 115
on an original assumption of rationality in the behavior of speech participants. This is

the Cooperative Principle. It is portrayed as rational, which means that we should not
merely think of it “as something that all or most do in fact follow but as something that
it is reasonable for us to follow, that we should not abandon” (Grice 1975: 48; empha-
ses in original). It is not an empirical fact but a rationalist, idealized, and perhaps
transcendental assumption, a foundation for Grice’s theory of communication that
betrays its moral embedding. But above all, something like the Cooperative Principle
Â�demonstrates that people carry around a host of assumptions with them, expecta-
tions or projections that should help them deal with the world and that actually turn
them into ‘persons’ and allow them to find ‘happiness’. The conception of a Â�person,
thus defined, is not a matter of finding some spatio-temporal continuity related to
an Â�‘organism’s’ bodily existence, but rather of construing a chain of mnemonic states
(memory) as constitutive of what will then be interpreted as a Â�personal identity (Grice
1941). It is a notion explicitly borrowed from Locke and meant to approach the
Â�problem of consciousness from a nonbehaviorist position, where propositions about
the self are indeed possible. One such proposition would be that persons do project.
The concept of rationality that is entertained by Grice, and by others in and out
of his wake (like Davidson), is a classically Platonic-Aristotelian one. It may be con-
ceived of as a regulator, directing and controlling pre-rational (biological) impulses. It
is first of all a category of everyday psychology, “which can be regarded as underlying
our ordinary speech and thought about psychological matters, and as such will have
to be a part of folk-science” (Grice 1991: 127).11 In defining rationality, Grice refers
to the condition that, for a human being to count as a person, she should have a set
of ‘Â�evaluative principles’ that allow her to make rational choices, i.e. choices that may
count as rational. The idea behind this is that the behavior of ‘agents’, originally con-
ceived of as pure automatons or robots, can be predicted on the basis of a preconceived
design of configurations (physical reactions or psychological states) in response to
incoming input, but only partly so. While Grice may appreciate certain behaviorist
explanations for their attempt to relate psychological concepts to “appropriate forms
of behavior” (Grice 1991: 124), he does not condone the urge of such dispositional
accounts to exclude these very psychological concepts from their theories proper.
For this would also exclude the reflexivity that is necessary to describe meaningful
Â�behavior. The way the robot is designed to work – its “genetic” makeup, so to speak –
does not necessarily reflect how it actually works, precisely because the robot has also
been programmed to be self-regulating. This is where its evaluative principles come
in, which determine, for each individual occasion, whether or not to follow one or
the other of the robot’s “biological” inclinations. Some of these evaluative principles
ïœ±ïœ±.â•… Here, Grice offers a variation on Moore’s defense of ‘common sense’.

116 Frank Brisard
may be replaced or modified, and the robot even has the capacity to invoke completely
new, subsidiary ones, based on previous experiences with its environment. However,
a small number of such principles is immune to revisions, and these must constitute
the core of what it means to be a person, functioning within a society of other persons.
Evaluative principles that are essential to the constitution of a rational agent are also
necessarily generalized, i.e. they should count for all agents (in similar situations).
If ‘happiness’, then, is one such principle, the notion of ‘personal happiness’, if it is
to serve a regulating function, can only be a derivative of the system constitutive of
happiness in general. We might say that this aspect of Grice’s substantiation of Â�rationality
represents his Kantian inclinations. In passing, we may also note that Â�rationality,
thus conceptualized, is more like a cluster of principles than one Â�monolithic notion.
This cluster may include the ‘end’ of being happy, next to a host of other evaluative
principles (or values).
What are the implications of this ‘psychological’ approach to meaning? For one,
it may demonstrate that the way in which a communicative system like language is
used by humans can never be entirely predictable. This is not so much because there
are irrational components to language use, which there certainly are as well, but rather
because linguistic habits can always be negotiated and exploited by speakers in Â�specific
circumstances. In fact, many of the so-called pragmatic meaning phenomena depend
on this notion of exploitation, and what they show is not a defect of the linguistic
system/model (say, semantics) but rationality at work. Thus, semantics should not be
chucked out just because there are pragmatic meanings. On the contrary, a Gricean
explanation of ‘pragmatics’ should start from the validity of a semantic theory. More
generally, the psychological theory envisaged by Grice contains provisions for a Â�feeling
of interest involved in our ascriptions of mental states to other minds, and such inter-
ests betray a concern for those others. Accordingly, a truly rational person is not some
neutral observer of external goings-on, but a ‘passionate’ participant with a number
of strong motivations (also affecting communicative interaction) which that per-
son regards as self-justified. One of these, the assumption that people generally use
Â�language to refer truthfully to states of affairs, may actually serve as the basis for truth-
conditional semantics. Semantics can therefore be seen as the professionalized account
of an ordinary principle which we might call ‘charity’ and which can be read as a lin-
guistic manifestation of the general Humean12 affect of curiosity, which is ‘the love of
truth (values)’. Humean Projection, or the propensity of the mind “to spread itself on
ïœ±ïœ².â•… Hume’s Treatise of human nature discusses the origins of theories of truth and language
in terms of the passions underlying them. In this light, Hume compares philosophy to hunting
or ‘playing’ in general, actions which provide the kind of pleasure that is indispensable to the
Â�development of intellectual curiosity and which are, as such, the prime manifestations of the drive
and ‘desire’ that mark human activity.
H.P. Grice 117
Â�
external objects” (Baker 1991: 4), thus figures prominently in Grice’s Â�epistemological
discussions. In the process of interpretation, properties of the mind are projected unto
the world, and this economy of passions is regulated by principles like charity (which
is a kind of trust in the utterer and, as such, a moral category; see Section 3).13 It is
exactly this ‘metaphysical routine’ which also shows that semantics is not autonomous,
as it is itself constituted by and through Projection. Here, Grice diverges from more
classical conceptions of formal semantics, à la Frege, Russell, or even Husserl, in that
the latter would disavow any attempt to reduce the problem of truth (semantics) to
matters of truthfulness. Truth values in standard formal Â�semantics are not Â�necessarily
‘values’ in Grice’s sense, the latter marking the act of ‘telling the truth’ rather than ‘truth’
itself. Clearly, Grice sees Humean Projection as a principle of semantic Â�organization,
too, tracing meaning back to psychology once more.
3.â•… Value and the new metaphysics
3.1â•… Creature construction

Grice’s conception of everyday psychological explanation offers terra firma for his
Â�analysis of utterer’s meaning (Grice 1982). If we wish to realize certain ends (e.g. sur-
vive, or be happy), we ‘know’ that an ability to produce beliefs in others will help
us meet those ends. As M-intentions are extraordinarily effective ways of achiev-
ing just this, I will do something, like produce an utterance, such that my audience
will Â�recognize the underlying M-intention, adopt a corresponding belief, and Â�perhaps
treat that belief as a reason for doing something else. The point, now, is that we do
have the ability to M-intend, i.e. this is an everyday psychological fact. And we
expect this Â�ability to be present in others as part of their ‘pre-rational’ makeup. We
are all ‘designed’ to M-intend, not because that ability is programmed into us like
an Â�automatic reflex, but because “it is rational for us to be so ‘designed’” (Grandy &
Warner 1986a: 23). The structure that allows us to M-intend is pre-rational because it
is always there and not subject to revision, and it is rational because it has genitorial
justification.
ïœ±ïœ³.â•… Charity serves as an anti-deception clause in the game of communication but is itself not
specifically linguistic, as always with Grice. This is Grice’s answer to the concerns of what we
may call the Obstinate Egoist, who will maintain that (the structural pursuit of) self-interest, or
treating others exclusively as means to one’s own ends, is all that is needed to escape the demands
of morality. Rational persons, Grice would counter, generally expect that others will keep their part
of an agreement beyond what could be calculated out of self-interest. They simply cannot escape
that expectation and still count as rational. Similarly, interlocutors cannot escape the Cooperative
Principle and still be seen as ‘appropriate’ partners in speech.
118 Frank Brisard
A pre-rational structure is, in a way, a transcendental construct, an object

of Â�metaphysics. Grice’s Carus lectures on the conception of value, written and
Â�delivered in 1983 (see Grice 1991), address such structures in view of his construc-
tivist defense of value. Humean Projection explains “the growth of conceptualization
and Â�representation in progressively more complex creatures” (Baker 1991: 2). This
type of metaphysical argumentation is called ‘creature construction’, a topic which
Grice also tackles in his unpublished lectures on ‘Language and reality’ (Urbana,
1970–1971) as well as in ‘Method in philosophical psychology (From the banal to the
bizarre)’, included in Grice (1991) and based on a 1975 APA address. Grice wants
to demonstrate the theoretical validity of (his) psychological concepts by develop-
ing sequentially distinct psychological theories for various types of ‘pirot’. Pirots are
Â�creatures constructed by genitors, or semi-gods (e.g. a philosophical psychologist)
who design14 living things. Such design is implemented in animal stuff (‘flesh’) and
provides a number of procedures, basic and evaluative, that are deployed in the inter-
action of our creatures with their environment. Thus, Grice acknowledges the basic
behaviorist position of understanding what creatures are in terms of what they do. In
turn, ‘very intelligent rational pirots’, with the capacities for thought and action, can
put themselves in the genitorial position and, correspondingly, observe that they are
themselves designed to realize certain built-in, and more or less constant, ends, insofar
as they are allowed by concrete circumstances.15 As already noted, it is also assumed
here that, given the dynamic nature of context, pirots are assigned the capacity to rede-
sign themselves, in order to cope most effectively with the ever-changing character
of the environment. That is why the rational creatures that we will call ‘persons’ are
also conceived of as ‘end-setters’, i.e. creatures with the ability to adopt new ends and
eliminate old ones on the basis of a limited set of evaluative principles. The latter are,
of course, only properties ascribed to creatures from a strictly Â�genitorial Â�perspective,
which betrays the genitors’ own natural (or rational) disposition to take the world to
be a certain way. But the rationality of this projection itself lies precisely in the fact
that we, as genitors, have good reasons to suppose that these evaluative Â�principles are
present in (Gricean) end-setters, like ourselves. By projecting such Â�rational structure
upon a creature’s ‘diagram’ for dealing with the world, the accidental property of being
rational, observable in humans as an empirical fact, is transformed into an essential
ïœ±ïœ´.â•… Genitors are not full-fledged gods because they cannot and do not actually create anything.
The ‘design’ or ‘construction’ work they engage in should be taken as a redistribution of (perfectly
observable) properties (Baker 1991: 7).
ïœ±ïœµ.â•… The ultimate reference point for these features of their design is ‘survival’. The term does not,
however, refer to any biological conception of ‘staying alive’, but rather to a vitalistic one (Grandy
& Warner 1986a: 31).
H.P. Grice 119
one. By this transformation (a process which Grice called ‘Metaphysical Transubstan-

tiation’; cf. Baker 1991: 5–14 and Sbisà 2001: 200–202), humans – as a biological type –Â�
reconstitute themselves as (cultural) persons.
A ‘good’ person is someone who is good ‘as a person’, or good at doing what
a Â�person should, i.e. finding reasons, constructing arguments, and many more
operations derivative of these. Constructing creatures also means building a world
that depends critically on the attribution of ends (finality), in the sense of what such
creatures ought to do (this ought may be Humean as well). And while Grice’s genitorial
account may present itself as entirely mechanistic, he insists that questions of finality
and value (evaluation), which lie beyond the biologically useful, will ultimately prove
irreducible. A strange union of cybernetics and vitalism is the outcome of this Â�routine –
see Grice’s John Locke lectures (1979), delivered in Oxford, on reasons, ends, and
happiness (also given as the Immanuel Kant lectures in 1977 and posthumously
published in Grice 2001), as well as some as yet unpublished work, like ‘Probability,
desirability, and mood-operators’ (1972) and his collaborative project with Baker on
Kant’s ethics.
3.2â•… Absolute value (Kantotle)

In Grice’s account, rationality attaches to humans only accidentally (Grice
1986: 102). It is what we may deduce from our observations of humans in interac-
tion with the world, including acts of communication with other humans. At this
particular level, the maxims that elaborate Grice’s Cooperative Principle describe
empirical facts of conversational behavior. But of course their status is not limited
to this type of description. The maxims, and the Cooperative Principle itself, appeal
to moral categories as well and can be seen as communicative instances of the evalu-
ative principles needed for humans to count as persons (specifically, in linguistic
interaction). No new properties are invented here, and a purely mechanistic account
of conversational behavior might arrive at the same set of principles. But Grice’s
account is thicker, as a result of his metaphysical routine, and it effectively ‘grounds’
the rational project in the essentially reflexive or self-justifying character, not just
of the M-intentions involved in language, but of every manifestation of purposeful
conduct, i.e. of behavior that needs to be recognized as intended rather than acciden-
tal. In his discussion of value, Grice eventually holds that all evaluative principles,
whatever they may be and including their potential derivations, are necessarily true
(Kant). This allows him to maintain what is probably one of the bolder positions in
contemporary metaphysics, viz., the idea that relative personal values, if they spring
from persons who are ‘good at’ being persons (in conditions of freedom), borrow
their constructed objectivity from the argument for absolute (irreducible) value:
‘What seems good to the good man is good’ (Aristotle). Such absolute value is criti-
cal in the projection involved in creature construction, because absolute value, and
120 Frank Brisard
nothing less, is what we as rational beings legitimately demand. But even outside the
domain of ethics, these observations still stand. Specifically, M-intentions, ground-
ing Grice’s theory of meaning, are pre-rational structures, and they can only be fully
motivated with reference to the demands of morality, i.e. “the necessity of rational
agents accepting and acting on certain imperatives (in so far as they act rationally)”
(Grandy & Warner 1986a: 38).
Grice’s recourse to psychology in explaining meaning involves a number of
Â�argumentative steps that need to be individually validated. In the end, however,
there is nothing left to validate but the validation procedure itself. At this point, we
may ask why certain ‘ordinary’ ways of thinking, using concepts from both ordinary
Â�psychology and ordinary moral reasoning, serve the purpose of justifying such highly
abstract theoretical constructs so well. The answer to this is metaphysical, not scien-
tific. Grice’s investigation has shown that there are necessary concepts and categories
which rational beings cannot avoid applying to reality (including their perception of
other rational beings). He develops a kind of “ontological Marxism” (Grice 1991: 131),
based on the simple tenet ‘They work therefore they exist’, for theoretical entities
which we can quite liberally go about finding: “The entities in these categories are
entia realissima. We discover these categories by discovering what parts of everyday
psychology are entrenched” (Grandy & Warner 1986a: 30–31). Any identification of
entia realissima, or real objects, happens on the basis of the evaluative principles we
have discovered in the course of creature construction. As a direct result of this, the
structure of M-intentions can only become as complex as the demands of rationality
(implied in creature construction) would allow it to if these entities are to stay ‘real’,
and infinite regress is not really an issue (cf. Section 2.1). Moreover, the metaphysical
entities that come out of the construction routines – and propositions may form one
class of them – derive their truth from the fact that they can be ‘justifiably accepted’
within a metaphysical argument. Truth itself, as the key notion in semantic theory, is
still allowed a meaningful part, but only to the extent that it is seen as emerging from
a common human activity. In distinguishing between science (presumably including
semantics) and metaphysics, Grice stresses the role of the people involved in theory
and creature construction, respectively: a small expert elite vs. humanity in general.
For Grice, only the latter group may bestow absolute legitimacy upon the outcome of
that construction, which is in line with the overall pragmatic orientation of his intel-
lectual program.
4.â•… Concluding remarks
Grice’s recourse to psychology in explaining meaning may at first seem perfectly

Â�compatible with the heavy emphasis on mind and mental constructs in the Â�cognitive
H.P. Grice 121
Â�
sciences, where Grice is still one of the prime sources of inspiration for tackling
Â�phenomena conveniently called pragmatic (see Bach 1999). It is therefore tempt-
ing to see in Grice a forerunner of the radical turn from the philosophy of language
to the philosophy of mind that has marked the second half of the twentieth cen-
tury. Yet we have noticed that quite a few premises generally adopted in the cogni-
tive sciences are not shared by Grice. For one, the appeal to truth values grounding a
semantic theory of meaning should be seriously qualified, and a belief in the modular
Â�architecture of the brain does not by itself warrant a transposition of that principle onto
Â�semantic Â�organization, which is manifestly not autonomous for Grice. Also, the kind of
Â�materialistic reductionism/eliminativism typifying much of present-day research into
the relation between mind and brain is too mechanistic and in any case unwarranted
by the Gricean perspective, because it does not provide for the interests that are crucial
to explain any act of interpretation. Brain scholars do not investigate persons, they take
organisms or systems as their objects of study. And the processing concerns that go
with this biological perspective are not identical to the inferential paths that Grice relies
on in constructing his own argumentation. Many ‘neo-Gricean’ models have been set
up in various areas of investigation, e.g. that of figurative language (cf. Searle 1993), to
produce empirically verifiable predictions on the ‘behavior’ of the brain (or of parts of
it, the alleged modules). Yet processing models are not what is needed to pursue Grice’s
linguistic goal of eliminating ambiguity in meaning theory by maintaining a sharp dis-
tinction between semantics and pragmatics. Finding out which kinds of information
are prerequisite to which kinds of meaning assignment (semantic or Â�pragmatic), as
illustrated, e.g., in Grice’s account of implicatures, is not the same as locating, tracking,
and comparing physiological reactions to various types of linguistic input.
On a more serious note, Grice wants to make sure that the style of psychology
adopted in his inquiries is not to be confused with that of scientific (or cognitive)
psychology, with its experimental bias. What is more, it is not to be expected that
pirotology, as a brand of everyday psychology, may some day provide a template for
the elaboration of its scientific counterpart. On the contrary, Grice explicitly warns
against the possible excesses of cognitivism (as the belief that ‘knowledge’ and ‘infor-
mation’ are the only viable objects in the study of behavior): “We must be ever watchful
against the devil of scientism, who would lead us into myopic over-concentration on
the nature and importance of knowledge, and of scientific knowledge in particular”
(Grice 1991: 161). A more fabulous illustration of this principle is provided at the very
end of that same article, ‘Method in philosophical psychology’. Of course, it is taken
entirely out of context here:
The very eminent and very dedicated neurophysiologist speaks to his wife. ‘My (for at
least a little while longer) dear,’ he says, ‘I have long thought of myself as an acute and
well-informed interpreter of your actions and behaviour. I think I have been able to
identify nearly every thought that has made you smile and nearly every desire that has
122 Frank Brisard
moved you to act. My researches, however, have made such progress that I shall no
longer need to understand you in this way. Instead I shall be in a position, with the aid
of instruments which I shall attach to you, to assign to each bodily Â�movement which
you make in acting a specific antecedent condition in your cortex. No longer shall I
need to concern myself with your so-called thoughts and feelings. In the Â�meantime,
perhaps you would have dinner with me tonight. I trust that you will not resist if
I bring along some apparatus to help me to determine, as quickly as possible, the
Â�physiological idiosyncracies which obtain in your system.’ I have a feeling that the lady
might refuse the proffered invitation. (ibid.)
5.â•… Further reading
Apart from the references already given, a number of additional (standard and new)
treatments of, or comments on, Gricean thinking deserve to be cited here. To start
with, many important papers relating to Grice’s philosophical project have been
Â�collected in Kasher (1998). For a recent monograph on Grice, see Chapman (2005).
On the linguistic side, Bach & Harnish (1979) present a comprehensive and sys-
tematic theory of communication within the broader framework of social interaction,
adapting Grice’s notion of (reflexive) M-intentions and thereby challenging concep-
tions of speech act theory as entertained by the likes of Austin, Searle, and Sadock.
Grandy (1989) and Neale (1992) go some way to coordinating a number of linguisti-
cally important issues in the work of Grice, in a period in which a lot of that material
was still inaccessible. Schiffer, an early proponent of Grice (cf. above), has over the
years been led to abandon the intentionalist program once and for all (Schiffer 1987).
Sperber & Wilson (1995) have used the Gricean theory of implicature to develop
their own, heavily cognitively oriented, research project, dubbed Relevance Theory.
The turn to a processing interpretation of Gricean procedures is symptomatic for
much of the empirical investigation that has been conducted in Â�neo-Griceanism over
the past couple of decades, including in the relatively young discipline of ‘experi-
mental pragmatics’. Obviously, this reorientation has also had a decisive impact on
psychological models of communication and interpretation, of which Clark (1996)
is a good example. Clark’s study basically reinterprets the Gricean and other prag-
matic theories of meaning in the light of his perspective on language use as a joint or
Â�coordinated activity type. Finally, Horn’s (1989) critical study of negative expressions,
which include such diverse categories as refusals, contradictions, lies, and irony, is a
classic example of the kind of (philosophically inspired) linguistic pragmatics that
has developed out of Grice’s preoccupation with the details of language use. Its reis-
sue adds a comprehensive state-of-the-art preface surveying past work on negation.
From a philosophical point of view, a larger picture of the analytic program
Â�entertained by Grice is presented in Black (1973), Davidson & Harman (1975), and
H.P. Grice 123
MacKay (1972). The volume edited by Bar-Hillel (1971) proposes an early appre-
ciation of general philosophical problems conjured up by Grice’s (and others’) logi-
cal approach to the acquisition and use of natural languages. Rationality and the
Cooperative Principle are the subjects of studies by Attardo (1997a, 1997b) and
Sarangi & Slembrouck (1992). Neale (1990) offers a (neo-)Gricean treatment of
descriptions, a classic Russellian problem, while Avramides (1989) attempts to show
how Grice’s meaning and communication theory fits in with the concern, typical
of Â�twentieth-century philosophy, with knowing ‘other minds’. Davis (1998), lastly,
offers a fierce yet bizarre criticism of Grice’s implicature model.
References
Attardo, S. (1997a). Competition and cooperation: Beyond Gricean pragmatics. Â�Pragmatics and
Â�Cognition 5: 21–50.
——— (1997b). Locutionary and perlocutionary cooperation: The perlocutionary cooperative
Â�principle. Journal of Pragmatics 27: 753–779.
Avramides, A. (1989). Meaning and mind: An examination of a Gricean account of language. MIT
Press.
Bach, K. (1987). On communicative intentions: A reply to Recanati. Mind & Language 2: 141–154.
——— (1999). Grice, H. Paul. In R.A. Wilson & F.C. Keil (eds.): 359–360.
Bach, K. & R.M. Harnish (1979). Linguistic communication and speech acts. MIT Press.
Baker, J. (1991). Introduction. In P. Grice, 1–21.
Bar-Hillel, Y. (ed.) (1971). Pragmatics of natural languages. Humanities Press.
Black, M. (1973). Meaning and intention: An examination of Grice’s views. New Literary History 4:
257–279.
——— (ed.) (1965). Philosophy in America: Essays. Allen & Unwin.
Chapman, S. (2005). Paul Grice: Philosopher and linguist. Palgrave Macmillan.
Clark, H.H. (1996). Using language. Cambridge University Press.
Cohen, L.J. (1977). Can the conversationalist hypothesis be defended? Philosophical Studies 31:
81–90.
Cole, P. (ed.) (1978). Syntax and semantics, vol. 9: Pragmatics. Academic.
——— (ed.) (1981). Radical pragmatics. Academic.
Cole, P. & J.L. Morgan (eds.) (1975). Syntax and semantics, vol. 3: Speech acts. Academic.
Davidson, D. (1986). A nice derangement of epitaphs. In R.E. Grandy & R. Warner (eds.), 157–174.
Davidson, D. & G. Harman (eds.) (1975). The logic of grammar. Dickenson.
Davis, W.A. (1998). Implicature: Intention, convention, and principle in the failure of Gricean theory.
Cambridge University Press.
Dennett, D.C. (ed.) (1987). The philosophical lexicon. Blackwell.
Grandy, R.E. (1989). On Grice on language. Journal of Philosophy 86: 514–525.
Grandy, R.E. & R. Warner (1986a). Paul Grice: A view of his work. In R.E. Grandy & R. Warner
(eds.): 1–44.
Grandy, R.E. & R. Warner (eds.) (1986b). Philosophical grounds of rationality: Intentions, categories,
ends. Clarendon.
124 Frank Brisard
Grice, H.P. (1941). Personal identity. Mind 50: 330–350.

——— (1957). Meaning. Philosophical Review 66: 377–388.
——— (1961). The causal theory of perception. Proceedings of the Aristotelian Society, Supplementary
Volume 35: 121–152.
——— (1968). Utterer’s meaning, sentence meaning, and word-meaning. Foundations of Language
4: 225–242.
——— (1969). Utterer’s meaning and intentions. Philosophical Review 78: 147–177.
——— (1975). Logic and conversation. In P. Cole & J.L. Morgan (eds.): 41–58.
——— (1978). Further notes on logic and conversation. In P. Cole (ed.): 113–127.
——— (1981). Presupposition and conversational implicature. In P. Cole (ed.): 183–197.
——— (1982). Meaning revisited. In N.V. Smith (ed.): 223–243.
——— (1986). Reply to Richards. In R.E. Grandy & R. Warner (eds.): 45–106.
——— (1989). Studies in the way of words. Harvard University Press.
——— (1991). The conception of value. Clarendon.
——— (2001). Aspects of reason. [Ed. by R. Warner.] Clarendon.
Grice, H.P. & P.F. Strawson (1956). In defense of a dogma. Philosophical Review 65: 141–158.
Habermas, J. (1998). On the pragmatics of communication. Polity Press.
Horn, L.R. (1989). A natural history of negation. University of Chicago Press. [Reissued in 2001, CSLI
Publications.]
Kasher, A. (ed.) (1998). Pragmatics: Critical concepts. Routledge.
Levinson, S.C. (1983). Pragmatics. Cambridge University Press.
——— (2000). Presumptive meanings: The theory of generalized conversational implicature. MIT Press.
Lewis, D.K. (1969). Convention: A philosophical study. Harvard University Press.
MacKay, A.F. (1972). Professor Grice’s theory of meaning. Mind 81: 57–66.
Moore, G.E. (1959). A defence of common sense. In Philosophical papers: 32–59. Allen & Unwin.
Neale, S. (1990). Descriptions. MIT Press.
——— (1992). Paul Grice and the philosophy of language. Linguistics and Philosophy 15: 509–559.
Ortony, A. (ed.) (1993). Metaphor and thought. (2nd ed.) Cambridge University Press.
Recanati, F. (1986). On defining communicative intentions. Mind & Language 1: 213–242.
Sarangi, S.K. & S. Slembrouck (1992). Non-cooperation in communication: A reassessment of
Gricean pragmatics. Journal of pragmatics 17: 117–154.
Sbisà, M. (2001). Intentions from the other side. In G. Cosenza (ed.) Paul Grice’s heritage, 185–206.
Brepols.
Schiffer, S.R. (1972). Meaning. Clarendon.
——— (1987). Remnants of meaning. MIT Press.
Searle, J.R. (1965). What is a speech act? In M. Black (ed.): 221–239.
——— (1986). Meaning, communication, and representation. In R.E. Grandy & R. Warner (eds.):
209–226.
——— (1993). Metaphor. In A. Ortony (ed.): 83–111.
Smith, N.V. (ed.) (1982). Mutual knowledge. Academic.
Sperber, D. & D. Wilson (1995). Relevance: Communication and cognition. (2nd ed.) Blackwell.
Strawson, P. F. (1959). Individuals: An essay in descriptive metaphysics. Doubleday.
——— (1964). Intention and convention in speech acts. Philosophical Review 73: 439–460.
Turner, K. (1999). Introduction: From a certain point of view (seven inch version). In K. Turner (ed.)
The semantics/pragmatics interface from different points of view, 1–18. Elsevier.
Wilson, R.A. & F.C. Keil (eds.) (1999). The MIT encyclopedia of the cognitive sciences. MIT Press.
Hermeneutics
Piet Van de Craen

Free University of Brussels
One of the tasks of Hermes, the messenger god of the ancient Greeks, was to bring
the gods’ messages to the mortals. In order to do this he was confronted with a double
task: “[h]e had to understand and interpret for himself what the gods wanted to con-
vey before he could proceed to translate, articulate, and explicate their intention to
mortals” (Mueller-Vollmer 1985: 1). The Greek verb hermeneuein may be translated
as ‘to interpret’ but, in fact, in ancient Greek the verb carried three meanings: “(1) to
express aloud in words, that is, ‘to say’; (2) to explain, as in explaining a situation; and
(3) to translate, as in the translation of a foreign tongue” (Palmer 1969: 13). Today, the
hermeneutic enterprise in modern human and social sciences refers to the interpreta-
tion and understanding of written and spoken discourse as well as human behavior. It
cuts across various disciplines, such as linguistics, literary analysis, philosophy, (social)
psychology, psychiatry, sociology, and anthropology.
2.â•… The origins of hermeneutic thinking
Although the School of Alexandria in late antiquity performed interpretation, it

was not until the Renaissance and Reformation that hermeneutics as a discipline
came to the foreground. Four branches of human science lie at the basis of her-
meneutic thinking: theology, philology, jurisprudence and philosophy. In the 16th
century a number of protestant theologians, such as Flavius, Luther and Melanch-
ton, considered the Scriptures as insufficiently understood. A linguistic and her-
meneutic training was proposed as a remedy. At about the same time, the interest
in classical texts from antiquity gave rise to a number of philological methods
dealing with the reconstruction of the original – or more truthful – version of a
text. Third, the interest in Roman Law, especially the study of the Code of Justin-
ian, ultimately lies at the basis of legal hermeneutics, which developed together
with philology. Finally, following Aristotle’s analysis of logic in his essay On Inter-
pretation, the Enlightenment philosophers of the 18th century tried to develop
hermeneutics as a general method for all fields of knowledge based on interpreta-
tion. This philosophical approach deals with the foundations of philosophy itself
126 Piet Van de Craen
and, consequently, of all science. Their approach can be seen as a forerunner of

the universality claim of 20th century philosophical hermeneutics (cf. Mueller-
Vollmer 1985; Habermas 1971). It is clear that the origins of hermeneutic thinking
are firmly rooted in the study of texts, be they of theological, philological, legal or
philosophical origin. The basic problem in all of these cases is the same, namely
what makes an interpretation correct and/or objective and by what standards can
this be measured?
3.â•… Some aspects of the evolution of hermeneutic thinking
In the 18th and 19th centuries, hermeneutics as biblical exegesis had a consid-
erable influence on philology, to such an extent that those systems of interpreta-
tion became the methods of, first, classical philology, and later, philology at large.
Schleiermacher, who considered hermeneutics as ‘the art of understanding’ favored
a ‘general hermeneutics’ based on the examination of sentence structure and con-
text of meaning. According to him an utterance could only be understood by
reconstructing it by taking the place of the speaker (Shleiermacher 1959). By the
end of the 19th century Dilthey, further elaborating Schleiermacher’s ideas, had
come to see hermeneutics as the core discipline for all human and social sciences
(Geisteswissenschaften). Dilthey maintained that while nature can be explained,
man must be understood with respect to his lived experience (cf. Dilthey 1976).
For Heidegger, hermeneutics was neither a philological method nor a methodology
for the Geisteswissenschaften. His ‘hermeneutics of Dasein [Being]’ is an attempt
to explain human existence itself. Here the role of language is quite essential since
‘being is speaking’ (cf. Heidegger 1962). Gadamer looked at language in a different
way. He does not deny the encompassing power of language (cf. Gadamer 1966),
but he strongly emphasizes the relationship between, on the one hand, language
and experience, and on the other hand, language and thought. This led him to reject
language form as the most essential part in favor of its expressive power (Gadamer
1975). With Apel and Habermas hermeneutic thinking acquired a social dimension
and their approaches are often referred to as ‘critical hermeneutics’. Apel’s ultimate
goal in performing a ‘normatively methodologically relevant philosophical herme-
neutics’ was emancipation. This means, first, to make transparent the underlying
individual and social processes of social actions and, second, to develop one’s con-
sciousness accordingly (cf. Apel 1976). Habermas’s interpretation of hermeneu-
tics refers to the competence to understand communication or, in other words, to
understand symbolic utterances grounded in and by society. Understanding seen
from the speaker’s viewpoint becomes the capacity to recognize the acceptability of
speech acts (Habermas 1976, 1981).
Hermeneutics 127
4.â•… The nature of the hermeneutic enterprise and the hermeneutical circle
Two basic aspects are related to any hermeneutical act, i.e. interpretation and
understanding. But how can we know when something is correctly interpreted and
fundamentally understood? Since understanding is basically a referential process –
something is understood by comparing unknown to known ‘things’ – there is a dia-
lectical interaction between the act of understanding and the ‘thing’ to be understood.
The notion of ‘objectivity’, as used in the natural sciences, gets a different meaning
here. This problem is known as the hermeneutical circle.
Suppose the meaning of a sentence is to be investigated. Unless the meaning of
its parts is understood, no interpretation is possible. Conversely, the parts can only
be understood through understanding the whole. Consequently, meaning arises from
the relation between part to whole, grounded in lived experiences. As a result, objec-
tive meaning does not exist but is always the outcome of some kind of interaction
process. This non-objectivity does not imply that meaning is impossible to grasp: the
lived experiences – of whatever kind – are in a sense objectified. The famous words by
the sociologists Thomas & Znaniecki (1958) “if men define situations as real, they are
real in their consequences”, stressing the importance of the lived experiences, are to be
remembered here.
The logical contradiction which seems to be entailed by the hermeneutical circle
is overcome, first, by emphasizing that logic in itself cannot account for the process of
understanding and, second, by presupposing that elements of intuition and some kind
of foreknowledge is part of all acts of understanding.
5.â•… Linguistics and hermeneutics
19th century philologists were deeply affected by hermeneutic thought and method.
Boeckh maintains that philologists must gain understanding of (1) the writing, the
symbol of the thing signifying, (2) language, the thing signifying, and (3) the thing
signified (cf. Boeckh 1968). However, despite obvious links between philology and
hermeneutics, the rise of linguistics as an autonomous science, towards the end of
the 19th and the beginning of the 20th century, did not create room for hermeneuti-
cal reflection. Most 20th century structuralists, such as Bloomfield, Harris, Hjelmslev
and Šaumjan considered language as an autonomous system independent from social
spatio-temporality. These scholars found support in positivist or neo-positivist meth-
odological approaches. In the 1950s and 1960s, Chomsky’s psychologically inspired
mental claims did not change the intellectual climate in favor of hermeneutics either. It
can be said that 20th century mainstream linguistics had little or no interest in herme-
neutics. However, though the link is rarely made explicitly, traces of Â�hermeneutic
thinking are to be found in traditions as divergent as Saussure’s structuralism, Sapir’s

linguistic anthropology, some trends in cognitive linguistics, and most clearly in
Â�conversation analysis.
5.1â•… Structuralism
In studying Schleiermacher’s and Boeckh’s ideas it is difficult not to be reminded
of Saussure’s positions expressed in the Cours de linguistique générale. However,
Â�Saussure’s methodological position is not quite clear, especially with respect to the
role of the social aspects of language. He acknowledges the social aspect of language,
while he fails to realize the consequences, as he argues that la langue rather than la
parole encompasses the social aspects of language. In discussing the various aspects of
signs, Saussure is indirectly influenced by hermeneutic thought, but he did not man-
age to incorporate it into his linguistics. Nor did his student and disciple Bally, who
was intrigued by expressive and affective language use. But how can expressive and
affective language use, clearly belonging to the parole, be incorporated in the study
of the langue? Bally concludes that this requires a new field which he calls stylistics
(cf. Bally 1965), a field which adopted the elements of hermeneutic thought which
Saussure’s theoretical position could not handle.
5.2â•… Linguistic anthropology

Towards the end of his life, Sapir wrote a number of stimulating articles, which are
obviously influenced by an interpretive way of looking at language and communica-
tion. Language, Sapir says “can discover meanings for its speakers which are not sim-
ply traceable to the given quality of experience itself but must be explained to a large
extent as the projection of potential meanings into the raw material of experience”
(Sapir, in Mandelbaum 1949: 10). What linguists should do is to restate their findings
in terms of, for instance, social psychology so that we can understand human behavior.
Here, Sapir anticipated the interpretation of anthropology and linguistics as cogni-
tive sciences. That is why the linguist needs the psychiatrist and why Sapir advocated
what he called ‘psychiatric’ research. By this he meant research about language and
communication intended to understand humans and human behavior. It could easily
be argued that Edward Sapir was the first linguist who incorporated and promoted
hermeneutic ideas in his approach.
5.3â•… Cognitive linguistics

Although Lakoff does not mention hermeneutics in introducing his experiential
Â�realism – as opposed to objectivism or rationalistic thought – his ideas are obviously
hermeneutic in nature. The key term he uses with respect to thought is embodiment,
which refers to conceptual systems resulting from bodily experience “grounded in
Hermeneutics 129
Â�
perception, body movement, and experience of a physical and social character” (Lakoff
1987: xiv). Apart from this, thought is also imaginative. But this implies that, when
humans categorize the world, be it verbally or otherwise, an interpretative process
based on embodiment and imagination is at work. Consequently, cognitive processes
are hermeneutic in nature in the sense that experiential realism cannot be reached
without the interpretation of the surrounding world.
5.4â•… Conversation analysis

By far the most obvious hermeneutic approach in current linguistic practice is to
be found in research on conversation analysis. Conversation analytic methodology,
rooted in the ethnomethodologists’ return to an interpretive approach in sociology,
is empirical and inductive, with an emphasis on “the interactional and inferential
consequences of the choice between alternative utterances” in order to “discover the
systematic properties of the sequential organization of talk, and the ways in which
utterances are designed to manage such sequences” (Levinson 1983: 287). Some
approaches concentrate on conversational style and its influence on human relations,
be they of ethnic gender or relational origin (cf. Tannen 1982, 1986, 1991). This echoes
research on the ethnography of communication, and carries interesting implications
of great importance for those who believe that the way people talk influences human
development and behavior. Conversational style can be seen as a product of human
development. This means that socialization and/or developmental processes deeply
influence conversational style and may even influence brain organization. The latter
can be explained by referring to the way connections in the brain are formed (Luria
1966; Edelman 1987, 1989). If further research would corroborate these hypotheses,
it would mean that a hermeneutic or interpretative methodology is basic to human
development and to an understanding of mankind.
References
Apel, K.-O. (1976). Transformation der Philosophie. Suhrkamp.

Apel, K.-O., C. von Bormann, R. Bubner, H.-G. Gadamer, H. J. Giegel & J. Habermas (1971).
Â�Hermeneutik und Ideologiekritik. Suhrkamp.
Bally, C. (1965). [1925]. Le langage et la vie. Droz.
Boeckh, P. (1968). [1886] On interpretation and criticism. University of Oklahoma Press.
de Saussure, F. (1972 [1916]) Cours de linguistique générale (1972 ed.). Payot.
Dilthey, W. (1976). Selected writings. Cambridge University Press.
Edelman, G. (1987). Neural Darwinism. Basic Books.
——— (1989). The remembered present. Basic Books.
Gadamer, H.-G. (1966). Man and language. In: H.-G. Gadamer (1976): 59–68.
——— (1975). [1960/1965] Truth and method. Sheed & Ward.
——— (1976). Philosophical hermeneutics. University of California Press.

Gumperz, J.J. (1982). (ed.) Language and social identity. Cambridge University Press.
Habermas, J. (1971). Der Universalitätsanspruch der Hermeneutik. In Karl-Otto Apel et al.: 120–159.
——— (1976). Communication and the evolution of society. Beacon Press.
——— (1981). Theorie des kommunikativen Handelns. Suhrkamp.
Heidegger, M. (1962). [1949]. Being and time. Harper & Row.
Lakoff, G. (1987). Women, fire and dangerous things. University of Chicago Press.
Levinson, S. C. (1983). Pragmatics. Cambridge University Press.
Luria, A. (1966). Higher cortical functions in man. Basic Books.
Mandelbaum, D. (1949). Selected writings of Edward Sapir. University of California Press.
Mueller-Vollmer, K. (1985). Introduction. Language, mind and artifact: an outline of hermeneutic
theory since the Enlightenment. In K. Mueller-Vollmer (ed.) The hermeneutics reader: 1–53.
Blackwell.
Palmer, R. (1969). Hermeneutics. Northwestern University Press.
Schleiermacher, F. (1959). Hermeneutik. Carl Winter Universitätsverlag.
Tannen, D. (1982). Ethnic style in male-female conversation. In J. Gumperz (ed.): 217–231.
——— (1986). That’s not what I meant! Morrow.
——— (1991). You just don’t understand. Morrow.
Thomas, W. & F. Znaniecki (1958). [1918/1920] The Polish peasant in Europe and America. Dover.
Indexicals and demonstratives*
Eros Corazza
Carleton University
In thinking and talking about items in our surroundings, we often rely on the
Â�context of use and thought. We thus succeed in talking and thinking about some-
thing because we are in a given context. We are, we could say, Â�context-bound crea-
tures. We share that with other organisms. Yet language is species-Â�specific. And
natural languages such as French, English or Japanese, could not be Â�conceived
independently of the context in which utterances are produced. First of all, in
natural languages we have tools whose specific function is to exploit the context
of use in order to select the things we aim to talk about. We can use the very same
words and yet refer to very different things, just as we can use the same hammer
to nail different nails. When you use ‘I’ you refer to yourself, whereas when I use
it, I refer to myself. I cannot refer to you using ‘I’. We use the very same linguistic
expression with the same conventional meaning to refer to different individuals.
It depends on who uses it that determines who the referent is.
If, in addressing Anya and Sue, John says “You come and you stay”, to understand
John’s request Anya and Sue must grasp who the addressees of the two uses of ‘you’
are. If the first occurrence of ‘you’ stands for Anya, John is asking her to follow him,
while if it stands for Sue, John is asking Sue to follow him. The same happens with
other expressions. If tomorrow John wants to say what he said today using ‘today’, he
is likely to use ‘yesterday’. If in different contexts we want to express the same thought
and/or say the same thing we often have to change the expression we use.
Indexicality, broadly understood, is the study of the expressions relying on the
context of use to select items of discourse. Hence, one of the questions we face is:
Q1
What counts as an indexical expression?
*For comments and discussions I’d like to thank Jérôme Dokic, Joana Garmendia, Kepa Korta, John
Perry, María de Ponte, Marina Sbisá, and Stefano Predelli. The comments of two reviewers have also
been useful. Research for this chapter has partly been sponsored by grants of the Spanish Ministry
of Science and Innovation (FFI2009-08574), the University of the Basque Country (GIU 08/25) and
the Social Sciences and Humanities Research Council of Canada (SSHRC 410-2010-1334).
132 Eros Corazza
This amounts to determining which, among the expressions of natural language, count
as indexicals and which don’t.
Another important question is:
Q2
What counts as the context of use?
For if an indexical expression determines its referent only relative to a Â�context of

use, we must define what counts as context of use.
In this chapter I’ll mainly try to deal with these two questions.
2.â•… Indexical expressions
Among the obvious expressions which may switch reference with a change in
Â�context, we have personal pronouns: ‘my’, ‘you’, ‘she’, ‘his’, ‘we’, …; Â�demonstrative
pronouns: ‘this’, ‘that’, ‘those’, …; complex demonstratives: ‘this book’, ‘that lady near
the window’, …; adverbs: ‘today’, ‘yesterday’, ‘now’, ‘here’, …; adjectives: ‘actual’ and
‘present’; and possessive adjectives: ‘my daughter’, ‘their car’, … (cf. Kaplan 1977). In
his seminal work “Thought”, Frege claims that: “If someone wants to say today what
he expressed yesterday using the word ‘today’, he will replace this word with ‘yes-
terday’” (Frege 1918: 332). He further argues that sentences containing expressions
like ‘here’ and ‘there’ are not complete expressions of thought insofar as the same
sentence can be true in one situation and false in another: “In all such cases the mere
wording, as it can be preserved in writing, is not the complete expression of the
thought; the knowledge of certain conditions accompanying the utterance, which
are used as means of expressing the thought, is needed for us to grasp the thought
correctly” (Frege 1918: 332).
These expressions have been termed, following Peirce, indexicals. They are the
paradigm of context-sensitive expressions, i.e. those expressions which rely on the
context of use to select an item of discourse. Reichenbach (1947) suggests that index-
icals are token reflexive for they can be defined in terms of the locution “this token”,
where the latter (reflexively) self-refers to the very token used. So, ‘I’ can be defined
in terms of “the person who utters this token”, ‘now’ in terms of “the time at which
this token is uttered”, ‘this pen’ in terms of “the pen pointed to by a gesture accom-
panying this token”, etc. The reference of an indexical expression depends on its
particular linguistic meaning: “the utterer of this token” is the linguistic meaning –
the character (Kaplan 1977) or role (Perry 1977) – of ‘I’, while “the day in which this
token is uttered” is the linguistic meaning of ‘today’. The meaning of an indexical
can be understood as a rule which one needs to master in order to use an indexical
correctly. Â�Furthermore, the linguistic meaning of an indexical can be represented

as a function taking as its Â�argument the context and giving as its value the refer-
ent/content (this is Kaplan’s famous content/character distinction). The meaning of
an indexical, thus, directs us to some aspects of Â�context in order to fix the object
of discourse.
Cappelen & Lepore (2005) propose a test that can be used to distinguish between
indexical expressions and non-indexical ones.1 They propose the Inter-Contextual
Â�Disquotational Indirect Report test (ICD for short).
ICD
Take an utterance u of a sentence S by a speaker A in context C. An Inter-
Contextual Disquotational Indirect Report of u is an utterance u* in a context
C* (where C* ≠C) of “A said that S”.2
According to the ICD test it is quite easy to establish whether or not an expression
can be classified as an indexical. If the occurrence of an expression e in an utterance
blocks the disquotational indirect report (i.e. it makes the report of the relevant utter-
ance false) then we have evidence that e is an indexical expression. On the other hand,
if e does not block the disquotational indirect report it is not an indexical expression.
Take for instance ‘I’, ‘this’ and ‘today’ in (1), (2), and (3):
(1) Jane: “I am rich”
(2) Jane: “This is cute”
(3) Jane: “Today is sunny”
1.â•… To be sure, Cappelen & Lepore use the ICD test to determine whether an expression is context
sensitive and, as such, contributes a value into the proposition expressed only relative to a context
of use. They claim that if an expression doesn’t pass the ICD test the value it contributes is the same
regardless of the context of use in which it occurs. I’m using this test in a more narrow and modest
way. That is, I use it merely to determine whether an expression is an indexical or not. Hence, I
leave open the question whether some expressions can be context sensitive without being indexical
expressions and, as such, their contribution into the proposition expressed also depends on the
context of use, i.e. that a change of context is likely to trigger a change of the expression’s semantic
contribution. In other words, I do not take, pace Cappelen & Lepore, the ICD test to settle issues
between semantic minimalism and contextualism.
2.â•… “Put intuitively, we suggest using such reports to test for context sensitivity as Â�follows: If the
occurrence of an expression e in a sentence tends to block the disquotational indirect reports
(i.e. render such reports false), then you have evidence that e is context sensitive.” (Cappelen &
Lepore 2005: 88)
134 Eros Corazza
The ICD test gives us:

(4) Jane said that I am rich
(5) Jane said that this is cute3
(6) Jane said that today is sunny4
Since (4), (5) and (6) do not capture what Jane said (i.e. as reports of what Jane said
they are false), ‘I’, ‘this’, and ‘today’ are indexical expressions. Consider now:
(7) Jane: “Igor is ready”
(8) Jane: “Igor is a foreigner”
(9) Jane: “Aristotle is a philosopher”
The ICD test delivers:
(10) Jane said that Igor is ready
(11) Jane said that Igor is a foreigner
(12) Jane said that Aristotle is a philosopher
Thus, ‘ready’, ‘foreigner’, ‘Igor’, ‘Aristotle’, and ‘philosopher’, unlike ‘I’, ‘this’, and ‘today’,
are not indexicals. These reports may be incomplete or underdetermined insofar as
they don’t state what Igor is ready for, vis-à-vis which country or place Igor is a for-
eigner, etc. Yet, unlike reports (4), (5), and (6), they are not blatantly false. The ICD
test can thus be adopted as a reliable guide which helps in distinguishing indexical
expressions from non-indexical ones.
3.â•… Demonstratives, pure indexicals, and essential indexicals
It is often the case, however, that the linguistic meaning of indexical expressions such
as ‘this’, ‘that’, ‘she’, etc., together with context, is not enough to select a referent. These
expressions are often accompanied by a pointing gesture or demonstration, and the
referent will be what the demonstration demonstrates. Kaplan (1977) distinguishes
between pure indexicals (‘I’, ‘now’, ‘today’, …) and demonstratives (‘this’, ‘she’, …). The
former, unlike the latter, do not need a demonstration-or directing intention, Kaplan
(1989)-to secure the reference.
3.â•… Since, as we’ll see in the next section, the reference of a demonstrative is fixed by the accompa-
nying demonstration (or directing intention) we are here assuming that a change in context entails
a change both in the demonstration and in the demonstratum, for two people can refer to the same
object using a demonstrative.
4.â•… It goes without saying that a change in context entails a change in the day of the utterance.
In their paradigmatic use pure indexicals also differ from demonstratives insofar
as the latter, unlike the former, are perception based. When one says ‘I’, ‘today’, etc., one
doesn’t have to perceive oneself or the relevant day in order to competently use and
understand these expressions.5 On the other hand, to competently use and understand
‘this’, ‘she’, etc., one ought to perceive the referent or demonstratum.
Furthermore, a demonstrative, unlike a pure indexical, can be a vacuous term.
‘Today’, ‘I’, etc. never miss the referent. Even if I do not know whether today is Â�Monday
or Tuesday and I am an amnesiac, when I say “Today I am tired” I refer to the relevant
day and to myself. By contrast, if one says “She is funny” whilst hallucinating, or “This
car is green” whilst pointing to a man, ‘she’ and ‘this car’ are vacuous.
Besides, pure indexicals cannot be coupled with sortal predicates, while ‘this’ and
‘that’ are often used to form complex demonstratives like “this book”, “that lady with
a brown hat”, …. Sortal predicates can be considered to be universe narrowers which,
coupled with other contextual clues, help us to fix a reference. If one says “This liquid
is green” whilst pointing to a bottle, the sortal ‘liquid’ helps us to fix the liquid and not
the bottle as the referent. Moreover, personal pronouns which work like demonstra-
tives (e.g. ‘she’, ‘he’, ‘we’, …) have a built-in or hidden sortal. ‘She’, unlike ‘he’, refers to
a female whilst ‘we’ usually refers to a plurality of people, of whom one will be the
speaker (see Corazza 2002a).
A further distinction comes from Perry (1979). Among the category of Â�indexicals
he also distinguishes essential indexicals (‘I’, ‘here’, and ‘now’). The latter cannot be
replaced without destroying the force of explanation. For, uses of essential indexicals
seem to have a special, privileged and primitive function. Their use is intrinsically tied
to the time/location the agent uses them.6 They are tied, one could say, to the agent of
the utterance’s egocentric coordinates. An individual, Jane, may believe that someone
is making a mess without realizing that she herself is making a mess and thus without
adjusting her behavior and acting accordingly. One can look into a mirror and say “His
zipper is open” without realizing that his own zipper is open and thus without bother-
ing to close it. One can know that the meeting starts at 13:30 without realizing that
it is 13:30 now and thus missing it. Only when one comes to entertain the thought
expressed by “My zipper is open” is one likely to feel embarrassed and close it. Only
when one comes to entertain the thought one would express by “The meeting starts
now” would one hurry to attend the meeting. Perry’s central thesis can be encapsulated
5.â•… The proto-perceptions – i.e. the various sensations one feels when engaged in an activity – that
may be involved when one is acquainted with oneself (or in self-Â�consciousness) don’t count as per-
ception insofar as one doesn’t perceive oneself the way one perceives an object in front of oneself.
6.â•… On the way indexicals expressions play a privileged epistemic and cognitive role see also
Castañeda (1966, 1967, 1969, 1989).
136 Eros Corazza
into the following: essential indexicals have a cognitive impact, for they trigger self-
centered behavior.
4.â•… Indexicals as singular terms
Indexicals are generally conceived as singular terms which contribute a referent to what
is said. According to the direct reference view (see Kaplan 1977 and Perry 1977), utter-
ances containing indexicals express singular propositions, i.e. propositions whose con-
stituents are the referents of the indexicals. As such, indexicals are usually characterized
as expressions whose interpretation requires the identification of some element of the
utterance context, as stipulated by their linguistic meaning. Thus an utterance of “I am
tired” expresses a proposition containing the referent of the first person pronoun, and
one understands it insofar as one knows to whom the term ‘I’ refers in the context in
which it is uttered. The linguistic meaning governing the use of the indexical – such as
“the agent of the utterance” qua meaning of ‘I’, ‘the day of the utterance’ qua meaning of
‘today’ – does not feature as a constituent of the proposition expressed.
If indexical expressions are characterized as singular terms contributing their
referents into what is said (i.e. the proposition expressed), adjectives such as ‘local’,
‘distant’, ‘actual’ – not to mention count nouns like ‘(a) foreigner’, ‘(an) enemy’, ‘(an)
outsider’, ‘(a) colleague’ – would not fall into the same category, for they do not contriÂ�
bute a referent to the proposition expressed. Yet they are, undeniably, context-Â�sensitive
expressions. ‘Local’, ‘foreign’ and ‘native’ in:
(13) A local pub is promoting foreign beer
(14) A native speaker should correct your essay
do not contribute a specific individual or individuals to the proposition expressed.

Hence they are not singular terms. It should be evident that context-Â�sensitivity
does not merely concern singular terms. It is worth distinguishing between indexi-
cals qua singular terms, contributing their referents to the proposition expressed,
and contextuals qua expressions which, though context-sensitive, are not singular
terms (see Vallée 2003). Adjectives such as ‘tall’, ‘big’, ‘small’, ‘old’, etc. are also con-
text-sensitive, insofar as one is only tall/small/big/old… relative to a comparison
class. John may be too small to play tennis and yet too tall to be a jockey, whilst
Jane may be too old to join the army and too young to take early retirement.
Proper names, like indexicals, also contribute individuals into the proposition
expressed. As such they are singular terms as well; yet they are not indexicals (but see,
among others, Recanati 2003 for a different view). Nouns like ‘Monday’, Â�‘February’
and the like also seem to contribute specific individuals in the proposition expressed.
They are best viewed in the same light as count nouns, i.e. as nouns such as ‘lemon’,
‘frog’ and ‘table’ (see Corazza 2002). As such, they can be used to build singular terms.
This happens when they are coupled with an indexical expression such as ‘this’, ‘next’,
‘last’ and contribute in forming complex demonstratives of the form ‘next week’,
‘last Â�Saturday’, ‘next Christmas’. This peculiarity parallels the way count nouns can
Â�participate in building complex demonstratives such as ‘these lemons’, ‘that table’,
‘this pen’. King (2001), however, defends the view that complex demonstratives are
quantified terms.7
As we already saw, one of the features of indexicals differentiating them from
other referential expressions (e.g.: proper names: ‘Plato’, ‘Paris’, … mass terms: ‘silver’,
‘water’, … terms for species: ‘frog’, ‘raspberry’, … and so on) is that they are usually
used to make reference in praesentia. That is, a use of an indexical exploits the pres-
ence of the referent. Usually in a communicative episode involving an indexical the
referent is in the perceptual field of the speaker and contextual clues are used to raise
the referent to salience (see e.g.: Q. Smith 1989, Sidelle 1991 and Predelli 1998 for a
discussion of indexicals used to refer to objects not present, e.g. answering machines,
post-it notes, etc.).
5.â•… Indexicals and anaphors
When indexicals are not used to make reference in praesentia they exploit a previ-
ously fixed reference. ‘That boy’ in “That boy we encountered yesterday was in trou-
ble with the police” does not refer to someone present. The same with ‘here’ in: “John
visited Rome. Here he discovered many interesting things”. In cases like this, the
indexical makes reference in absentia. One can thus distinguish between the context
of utterance and the context of reference fixing. In our example, the speaker and the
hearer appeal to a past context to fix the reference. The gap between the two contexts
would be bridged by memory. Another way to handle examples like this would be to
argue that, in such cases, the indexical expression works like an anaphoric pronoun
linked to a tacit initiator. Actually, indexicals can fix the reference in an anaphoric
way. Consider, for instance, cases of historical time:
(15) On New Year’s Eve 1834 John visited his mother, now an old sick woman
in which ‘now’ does not pick out the time of the utterance. Instead, it picks out the
time (New Year’s Eve 1834) when John visited his mother. In this example, the time
picked out by ‘now’ is explicitly stated by the utterance, but in some cases it can simply
be presupposed.
7.â•… See Corazza (2004) for a criticism of King’s view.

138 Eros Corazza
Most interesting cases can be found with the historical present. If, during a history
class discussing the recent invasion of Iraq, the lecturer says:
(16) Now Bush and Blair begin their illegal invasion of Iraq
‘now’ does not refer to the time of the utterance – it refers to the time of the USA/GB
invasion. The time picked out, however, is not explicitly stated in the utterance (and it
need not even have been stated in previous utterances). If the lecturer assumes that all her
students know the period under discussion, she does not need to mention the relevant
date, viz. March 20, 2003. If, in the process of proving a theorem, one states:
(17) Now I prove lemma S
‘now’ does not pick out the time of utterance.8 At this very moment one might stop
one’s lecture and continue proving the theorem during next week’s lecture. In these
cases ‘now’ seems to go proxy for something like ‘at this point of the proof ’.9
The same (or a similar) phenomenon occurs with spatial indexicals such as
‘here’. If we consider the following passage from California: The Ultimate Guidebook
(Riegert, 1990):10
(18) If an entire neighborhood could qualify as an outdoor museum, the Mount
Washington district would probably charge admission. Here, just northwest of
downtown, are several picture-book Â�expressions of desert culture within a
few blocks
‘here’ does not refer to the location of the author, Ray Riegert, when he wrote (18), or
to the location where the guidebook is read. It refers to the Mount Washington district.
Similarly, pointing to Venice on a map one can say:
(19) We spent last week-end here
and refer to Venice and not the location at which (19) is uttered. ‘Here’ can also be used
to refer to an imaginary location: watching a movie that takes place on an imaginary
distant planet, one can say:
(20) The Martians’ ancestors first landed here.
In other uses, ‘here’ does not even refer or purport to refer to a spatial location,
as in:
(21) Here Prokofiev always stops playing
8.â•… Whether one takes the time of the utterance to correspond with the time of writing or the
time of producing and/or decoding the message does not matter. In this example ‘now’ picks out
none of these times.
9.â•… Similar examples can be found in Q. Smith (1989).
10.â•… This example comes from Predelli (1998).
for one is likely to mean that Prokofiev stops playing at that particular point in
the composition. Notice that one could reach the same goal in using ‘now’ instead
of ‘here’.
In considering these cases one need not, pace Quentin Smith (1989), give up
the Â�traditional Kaplanian framework insofar as we can consider ‘here’ and ‘now’ as
demonstratives referring to the place and time demonstrated. The referent would
be an item raised to salience. In that way ‘now’ and ‘here’ would refer on the Kap-
lanian model. Yet they wouldn’t refer as pure indexicals do. They would rather be
considered as demonstrative. Hence, this analysis jeopardizes the traditional dis-
tinction between demonstratives and pure indexicals. According to this interpreta-
tion, ‘here’ and ‘now’ are ambiguous between being used as essential indexicals or
demonstratives.11
Furthermore, one could claim, pace Quentin Smith, that the relevant contex-
tual parameter to which the character of ‘now’ and ‘here’ is sensitive is not the loca-
tion and time of the utterance, but an intentional time and/or location (see Predelli
1998). One could thus maintain that the referent of ‘now’ and ‘here’ corresponds to
the intended contextual parameter. In that case, the character takes as argument a
relevant contextual aspect and gives as value the relevant (intended) referent. Thus,
the contextual parameters to which ‘now’ and ‘here’ are sensitive may differ from
the time and location of the utterance and so their referent may also differ from the
location and/or time where the utterance occurs.12 Although this position does not
distinguish between the indexical and the anaphoric use of ‘now’ and ‘here’, like
the position emerging from Kripke, Lasnik and Lewis, it commits itself to the view
that a single token of ‘now’/‘here’ may be sensitive, in principle, to infinitely many
contexts (times and locations). Since the relevant time and location qua param-
eters of ‘now’ and ‘here’ may differ from the time and location of the utterance,
they must be contextually furnished. Following this suggestion, ‘now’ and ‘here’
can be viewed as contextually ambiguous. The ambiguity of the indexicals is not at
the level of character but at the level of context.
Another interpretation which doesn’t jeopardize the pure indexicals/demon-
stratives distinction and doesn’t commit itself with a context ambiguity, is to
11.â•… This would possibly be the interpretation endorsed by the generative grammar school (see for
instance Kripke 1977, Lasnik 1976 and Lewis 1979).
12.â•… Predelli (1998) defends a position similar to this. On his account, ‘now’ and ‘here’ may not
refer to the time and location of utterance, for they can be sensitive to an intentional context,
which may differ from the context of utterance. Following Predelli’s position, the characters of
‘now’ and ‘here’, for instance, are sensitive to intentional parameters. He thus rejects the thesis
that contextual parameters such as time and location are always identical with the time and
Â�location of the utterance.
140 Eros Corazza
assume that in examples like these, ‘now’ and ‘here’ can be interpreted as ana-
phoric pronouns. We can characterize this as the anaphoric interpretation thesis
(see Corazza 2004):
AI
When a use of ‘now’/‘here’ does not select the time/location of the utterance
as referent, ‘now’/’here’ works like an anaphoric term inheriting its reference
from another noun phrase.
In other words, when ‘now’ and ‘here’ are not used as indexicals picking out the
time and location of the utterance, they do not have referential independence; they
work as anaphors inheriting their value from an antecedent to which they are linked
and coindexed. (15) can be interpreted as
(15) a. On New Year’s Eve 18341i John visited his mother, now1d an old, sick woman
where it is stressed how the reference of ‘now’ depends on the reference of its anteced-
ent, ‘New Year’s Eve 1834’.13 A NP is always marked as (semantically) independent
or dependent, i.e. NPi/d, with the superscripts ‘i’ standing for independent and ‘d’ for
dependent. A proper name is always marked as independent, while an indexical is
ambiguous between being independent or dependent.14 If it is independent, then it
13.â•… It is worth mentioning that, while coindexation entails coreference, non-coindexation does not
entail non coreference. If, in answering the question “Who left?” one says:
(i) He1 put John2’s coat on,
what is said may well be consistent with ‘he’ and ‘John’ being coreferential (see Fiengo & May 1984: 3).
14.â•… One could claim that ‘I’ is always referentially independent. Castañeda, though, claims that ‘I’
can also work in a dependent (anaphoric) way: “To complicate things further there is also the depen-
dent quasi-indexical use of the first-person pronoun. Consider:
(i) I believe that I am heading for trouble
On one interpretation, (i) is simply the first-person instance or filler of the quasi-Â�indexical schema:
(ii) X believes that he himself is heading for trouble
Clearly, the variable ‘X’ is instantiated into ‘I’ and this, by the agreement of grammatical person
… mandates that the quasi-indicator ‘he himself ’ be replaced with ‘I’. Thus, the second, the quasi-
indexical ‘I’ in (i) has a meaning that includes the general meaning of ‘he himself ’. The role of the
quasi-indicator ‘he himself ’ is to depict first-person reference by the speaker. It is, therefore, not
entirely vicarious as is the first-person pronoun in (iii)[Satan believes of me that I am grouchy]. Yet
it is an important difference. The quasi-indicator is a mechanism for the attribution, by depiction, of
indexical reference, not for making it. Thus, the quasi-Â�indicator ‘I’ in (i) represents a self-attribution
of the making of first-person reference.” (Castañeda 1983: 322)
is used deictically. If it is dependent, then it is an anaphora (or a bound variable). In

the latter case, one must look for an antecedent. Within this framework, anaphoric-
ity is not represented by indices (subscripts), but by dependence and independence
markers (superscripts). In other words, the superscripts, not the subscripts, rep-
resent anaphoricity or non-anaphoricity, i.e. semantic dependence/independence,
not coreference. To illustrate how coindexation (and thus coreference) differs from
anaphoricity, we can focus on identity statements. In “Tully = Tully”, the two occur-
rences of the proper name must be coindexed. This is constrained by the grammar
of ‘=’. Think of mathematical statements such as “2 + 3 = (2 – 1) + 4”, where the two
occurrences of ‘2’, although they are automatically coindexed insofar as we have two
occurrences of the same numeral, are not anaphoric upon one another. The Â�correct
representation should be “21i + 32i = (21i – 13i) + 44i”. Since proper names are never
Â�anaphoric, they are always referentially independent; yet they can be coindexed.
Let us now consider (16) [Now Bush and Blair begin their illegal invasion of Iraq].
Since ‘now’ does not pick out the time of utterance it must, according to the thesis pro-
posed (AI), function as an anaphoric term; but then, where is the antecedent? Accord-
ing to history, Bush and Blair began their war against Iraq on March 20, 2003, so this
fact should set the value of ‘now’; yet since there is no term in our utterance explicitly
referring to March 20, 2003, where could the value of ‘now’ come from? One solution
that comes to mind is that ‘now’ inherits its reference from a tacit Â�initiator, for it is
presupposed by the discourse situation that the facts described happened on March 20,
2003. In such cases, the speaker may not be fully aware of the relevant presupposed
information. One may have forgotten that the relevant time is March 20, 2003. Yet one
may rely on historical facts to fix the tacit, Â�presupposed, information. The latter need
not be fully stored in one’s memory.
The notion of tacit initiator can be spelled out as follows. An anaphora inherits
its reference from the NP, the antecedent, to which it is linked. The antecedent fixing
the reference initiates the anaphoric chain. For this reason the antecedent can also be
characterized as the initiator of the reference. An anaphoric chain can be initiated by a
tacit reference, in which case the antecedent of the anaphora is not expressed. As such,
it is a tacit initiator or a tacit antecedent. We can adopt the following notation: if the
antecedent is explicit, the indexical has a non-zero number as a subscript, while if the
antecedent is implicit the subscript is a zero (NP0d).15
15.â•… So in our syntax we can represent the whole range of indexical NPs as follows:
Dependence + (non-zero) number index ⇒ anaphora with explicit initiator/antecedent

(NPnd)
Dependence + zero index ⇒ anaphora with tacit initiator/antecedent (NP0d)
Independence + (non-zero) number index ⇒ deictic (independent) term (NPni)
142 Eros Corazza
We can now return to our example (16). If we assume that ‘March 20, 2003’ is the
tacit initiator, (16) can be regimented as:
(16) a. It is March 20, 20030i. Now0d Bush and Blair begin their Â�illegal invasion
of Iraq
‘March 20, 2003’ can be viewed as the tacit initiator (for this reason it gets ‘0’ as a
subscript) from which ‘now’ qua anaphoric term inherits its value. It is a platitude
that during a linguistic interchange a great deal of information is unexpressed and is
conveyed tacitly. Without this phenomenon, communication would be extremely dif-
ficult and slow, if not impossible, thus the existence of tacit initiators should not come
as a total surprise. These initiators may be understood as what is taken for granted
during a speech act. If a cooperative speaker does not believe that her audience is
aware of this presupposed information, she would simply express it in her utterance.
Presuppositions are usually viewed as the part of discourse or speech denoting prop-
ositions whose truth is taken for granted. In uttering, for instance, “Jane’s hat is red”,
one conveys two propositions: the proposition that Jane has a hat and the proposi-
tion that it is red. While the former proposition is presupposed, the latter is asserted.
In many cases, however, the presupposed propositions are merely taken for granted
and nothing in the utterance itself may trigger or bring them to salience.16 As an
approximation of the phenomenon under discussion, we can follow Stalnaker when
he argues that:
The notion of common background belief is the first approximation to the notion
of pragmatic presupposition … A proposition P is a pragmatic presupposition of a
speaker in a given context just in case the speaker assumes or believes that P, assumes
or believes that his addressee assumes or believes that P, and assumes or believes that
his addressee recognizes that he is making these assumptions, or has these beliefs.
(Stalnaker 1974: 49)
If the speaker does not assume that her audience is aware of the presuppositions in
place, she can easily express them; thus, the tacit initiator could be expressed by the
speaker – this accounts for the difference between (16) and (16a). (16) contains a tacit
initiator, whilst (16a) contains an explicit initiator, i.e. that which is presupposed in
16.â•… For a convincing and detailed account of presuppositions, see Stalnaker: “The distinction
between presupposition and assertion should be drawn, not in term of the content of the proposi-
tion expressed, but in terms of the situation in which the statement is made – the attitudes and
intentions of the speaker and his audience. Presuppositions, on this account, are something like
the background beliefs of the speaker – propositions whose truth he takes for granted, or seems to
take for granted, in making his statement.” (Stalnaker 1974: 48).
(16) is made explicit in (16a). Both (16) and (16a) express the proposition that on
March 20, 2003 Bush and Blair begin their illegal invasion of Iraq.17
We should stress, though, that unlike the third person pronoun ‘s/he’, pure indexi-
cals like ‘now’ and ‘here’ cannot act as bound anaphors. They cannot behave as bound
variables either. Consider:
(22) a. Jane won the lottery and she is now rich
b. Jane thinks that she is rich
In (22a) ‘she’ is not bound by ‘Jane’, whereas it is in (22b). To stress this point we can
replace ‘Jane’ by a quantifier. We would thus have:
(23) a. Everyone won the lottery and she is now rich
b. Everyone thinks that she (herself) is rich
In (23a) ‘she’ is not bound by the quantifier ‘everyone’: it works like a free variable,
while in (23b) ‘she (herself)’ is bound and works as a bound variable.18 If we consider
‘here’ and ‘now’, they are never bound the way ‘she’ can be in (23b). Thus, if ‘here’ and
‘now’ work as anaphoric terms, they are unbound anaphors.19 Some pronouns cannot
be bound either. Yet, ‘them’ and ‘we’ in:
(24) a. John said that Sue expected them to win the competition
b. I told Jane that we will go to the movie tonight
17.â•… One could object that the notion of a tacit initiator qua anaphora antecedent cannot be
Â�explained in terms of pragmatic presuppositions, for the notion of the antecedent of an anaphor
belongs to semantics, not pragmatics. It has been argued (Huang 2000), though, that an exhaustive
study of anaphora cannot be done in purely semantic or syntactic terms, insofar as an anaphora’s
antecedent cannot be selected uniquely on the basis of syntactic and semantic considerations.
Pragmatic factors often Â�operate in Â�determining the antecedent of an anaphor. This is particularly
evident when one Â�concentrates on Asiatic languages, which rely more heavily on pragmatics than
English, Italian, French, etc.
18.â•… Actually, (23a) and (23b) would be regimented as follows:
(23) c. ∀x (x won the lottery) & y is now rich

d. ∀x (x thinks that x is rich)
19.â•… In particular, they do not fall into the category of anaphors according to the principles A,
B and C of Binding Theory:
Principle A: anaphors must be bound in their governing category.

Principle B: pronouns must be free in their governing category.
Principle C: other NPs must be free in all categories.
144 Eros Corazza
can be interpreted as anaphoric pronouns.20 If so, the interpretation of ‘them’ in (26a)

is that it is anaphoric on ‘Sue’ and ‘John’, while the interpretation of ‘we’ in (26b) is that
it is anaphoric on ‘I’ and ‘Jane’. Yet, they are unbound – pronouns can never be bound
by split antecedents.21 They are unbound anaphors. (24a–b) could be represented as:
(24) c. John1i said that Sue2i expected them1+2d to win the competition
d. I1i told Jane2i that we1+2d won the lottery
To be sure, one could still claim that even in this case ‘them’ does not work in an ana-
phoric way, but instead works as a demonstrative selecting John and Sue as referent
because they have previously been raised to salience. The same consideration would
apply for ‘we’ in (26b). In that case, a representation like (16c) and (16d) would not be
appropriate, insofar as the superscript ‘d’, as we saw, signals that the pronoun is referen-
tially dependent, i.e. that it inherits its reference from another/other NP(s).
Needless to say, in the case of unbound anaphora the anaphoric chain is not
merely syntactically constrained: semantic and pragmatic considerations also enter
into the picture.22 On this issue one can quote recent studies, notably Huang (2000),
which convincingly show that:
(i) syntax and pragmatics are interconnected to determine many of the processes
of anaphora that are thought to fall within the province of grammar, and (ii) the
extent to which syntax, semantics, and pragmatics interact varies typologically.
(Huang 2000: 205)
There seems to exist a class of languages (such as Chinese, Japanese, and Korean)
where pragmatics appears to play a central role which in familiar European languages
(such as English, French, and German) has hitherto been alleged to be played by
grammar. In these ‘pragmatic’ languages, many of the constraints on the alleged
20.â•… In this sentence ‘we’ is ambiguous insofar as it could mean “the speaker and the addressee”.
21.â•… “Although pronouns with split antecedents are coindexed with those antecedents, they are
never bound by them. This predicts that the distribution of split antecedents will not be limited by
Binding Theory.” (Fiengo & May 1994: 39)
22.â•… As an example of a pragmatic constraint on anaphora we can quote the general pattern of
anaphora, i.e. the fact that “reduced, semantically general anaphoric expressions tend to favor locally
coreferential interpretations; full, semantically specific anaphoric expressions tend to favor locally
non-coreferential interpretations”. (Huang 2000: 214) As examples, consider the following contrasts:
(i) a. Mozart1i adored his1/2d music b. He1i adored Mozart’s2/*1d music
(ii) a. The bus1i came trundling round the bend. The vehicle1d almost flattened
a pedestrian
b. The vehicle1i came trundling round the bend. The bus2/*1d almost flattened
a pedestrian
grammatical processes are in fact primarily due to principles of language use rather
than rules of grammatical structure. (Huang 2000: 213)
Be that as it may, if one accepts the existence of unbound anaphors inheriting their
reference from another NP, one can maintain that ‘now’ and ‘here’ are systematically
ambiguous: they are either pure indexicals or unbound anaphors having either a tacit
or explicit initiator. On the other hand, one could claim that coreferentiality (and thus
coindexation) in the examples proposed is accidental, i.e. the relevant NP is coreferen-
tial with another NP not in virtue of inheriting its reference but simply because the lat-
ter has been previously raised to salience (they are thus referentially independent). One
is then forced to embrace the thesis that ‘now’ and ‘here’ in the example discussed, like
all alleged unbound anaphors, actually work like Â�demonstrative expressions selecting
as referent a location and time differing from the time and location of the utterance. To
summarize, the anaphoric interpretation proposed, unlike the demonstrative interpre-
tation: (i) maintains that there is a structural difference between unbound anaphora
and demonstrative reference (the former is referentially dependent while the latter is
referentially independent) and (ii) fits within the distinction between pure indexicals
and demonstratives. Besides, the position advocated here does not commit itself to the
view that ‘now’ and ‘here’ are contextually ambiguous. The anaphoric interpretation is
thus cheaper than the demonstrative interpretation.
The position proposed here presents the following advantage over the position
advocated by Predelli and the position attributed to Kripke, Lasnik and Lewis. On the
view defended in this section ‘now’, ‘here’ (and similar indexicals) can be used either as
indexicals or as anaphoric terms. As such, they are systematically ambiguous, but their
ambiguity rests on their being used either deictically or anaphorically. It does not rest
on a multiplicity of contexts.23 Hence, the position advocated doesn’t multiply contexts.
It simply rests upon an existing common phenomenon, the phenomenon characterized
as unbound anaphora (sometimes labeled cross-sentential or discourse anaphora). For
instance, “Mary1i believes that she1d is lucky” contains an intra-sentential anaphora,
whilst in “John1i won a million dollars last night in Monte Carlo. Even though he1d
won all that money, he1d was still not very pleased”, we have a case of cross-sentential
or discourse anaphora, i.e. the anaphor ‘he’ inherits its value from an antecedent, ‘John’,
appearing in another sentence.
23.â•… Insofar as the present tense is interpreted using ‘now’, we can claim that it can work in an
anaphoric way as well. If, watching a video, one claims that “Arsenal are dominating the game”,
the present tense can be viewed as anaphoric on a tacit initiator referring to the time the game
was played. No doubt, more should be said on this issue, but time and space prevent me from
discussing it any further.
146 Eros Corazza
Since the picture defended appeals to existing phenomena, it comes cheaper than
the position claiming that indexicals are contextually ambiguous. It is thus more eco-
nomical than the Kripke-Lasnik-Lewis-Predelli position which posits ambiguities by
multiplying contexts. According to the latter view, the relevant context may thus differ
from the context of utterance. The moral is that the Â�Kripke-Lasnik-Lewis-Predelli view
violates a principle of parsimony and Ockham’s Razor, by multiplying contexts and
ambiguities beyond necessity.
If our aim is to understand and explain communicative interaction, we must
take into consideration discourse anaphora, which may extend beyond utterances of
a single speaker. Hence we should also consider interpersonal anaphora, for it is not
uncommon that, in everyday communication, anaphors that inherit their Â�reference
from an antecedent uttered by someone else are often used. Consider:
(25) Mary: “I’ll be in Paris1i on New Year Eve”
John: “I’ll be there1d anxiously waiting for you”
In this case, ‘there’ inherits its reference from a token made by someone else, Mary.
If we consider a telephone conversation such as:
(26) Mary (in Paris): “It is still snowing”
John (in Lyon): “In that case, I won’t be able to be there0d before midnight”
we have ‘there’ inheriting its reference, Paris, from another sentence. In that other
phrase, though, the reference to Paris is not explicitly made – it is for this reason
that the pronoun ‘there’ has ‘0’ as a subscript. Mary tacitly refers to Paris. To borrow
Perry’s (1986) terminology, Paris is an unarticulated constituent of what Mary says.
If Mary uttered, “It is still snowing here”, Paris would be an articulated constituent
of what she says and ‘there’ would be anaphoric and coreferential with ‘here’. Cases
like this suggest that we do have anaphoric terms inheriting their value from a tacit
initiator. It is difficult to see how (26) could be interpreted without taking ‘there’ to
be anaphoric on (and coreferential with) a presupposed tacit reference to Paris.
Cases where anaphors inherit their value from tacit initiators are not confined to
spatial and temporal indexicals. In:
(27) Jane is hoping for a baby, whereas he0d is far from excited by the idea
‘he’ seems to inherit its value from a tacit initiator (NP like ‘Jane’s partner/husband/…’
would suffice), for the speaker of (27) tacitly refers to Jane’s partner. ‘Jane’ is relevant in
helping the audience to select Jane’s partner/husband/… as the object of discourse and
thus in bringing to salience the referent the tacit initiator stands for.
In the case of pointing to a map (say, pointing to the spot representing Venice)
and saying ‘here’, one can think of ‘here’ as going proxy for ‘this location’ and thus
that ‘here’ is used as a demonstrative picking out the demonstrated location, instead
of picking out the location where the utterance occurs. The anaphoric interpretation,
however, allows us to reach the very same conclusion without supposing that ‘here’
works as a demonstrative. For it can be argued that ‘here’ works as an anaphora, and
that an utterance like (19)[We spent last week-end here] should be analyzed as:
(19) a. This/that location is/represents/… Venice0i. We spent last weekend here0d
Similarly, we can stress the anaphoric reading, taking (19a) as equivalent to (19b):
(19) b. This/that locationis/represents/… Venice0i. We spent last weekend there0d
where ‘there’ stresses the anaphoric interpretation. (19a) and (19b) both express the
proposition that we spent last weekend in Venice.
When ‘now’ and ‘here’ are not used to pick out a time and location, they belong
to the idiomatic use of language, i.e. that peculiar use which cannot be captured by a
general rule. As such they do not work in their usual, indexical, way to single out the
time/location of the utterance. It is for this reason that ‘now’ and ‘here’ are often inter-
changeable without any loss or gain in a communication. My New Shorter Oxford Dic-
tionary reads “at this point in an argument, a situation, etc.; at this juncture” under the
entry for ‘here’, while the entry for ‘now’ reads, “at an important or noteworthy place
in an argument or proof or in a series of statements”.24 Again, even in these cases, ‘now’
and ‘here’ can be interpreted as anaphors; (17) and (21) can be analyzed as:
(17) a. We are at this point of the argument/proof/…0i. Now/here0d
I am going to prove lemma S
(21) b. We are at this point of the sonata/concert/…0i. Here/now0d
Prokofiev always stops playing
The following chart should summarize the various distinctions we encountered so far:
Indexicals
Anaphors (dependent) Deictic (independent)
Bound Unbound Pure Indexicals Demonstratives
Essential non-Essential Simple Complex
24.â•… If I am right in considering these examples as idiomatic uses of language, it may be better
not to treat them in the same way as the other examples. It is by mere accident that we use spatio-
temporal terms to describe non-temporal items. Is it simply because it takes time to run through a
proof or a sonata that we adopt the temporal framework? When the proof and sonata are written,
we adopt the spatial framework for similar Â�reasons. These phenomena are linked to our cognitive
architecture; they are not Â�intrinsic to the semantics of our language. Thanks to Predelli for this
suggestion.
148 Eros Corazza
6.â•… Indexicals and contexts
To fully appreciate the specific role indexicals have within natural languages, it is
worth taking into consideration the various ways context can be exploited both in
a linguistic interchange and when cashing out an utterance’s truth-conditions. To
begin with we can distinguish between presemantic, semantic, and postsemantic (or
“content supplementing”, to borrow Perry’s recent characterization) use of context.
At the presemantic level context is used to figure out the meaning of words,
to disambiguate, to determine which homonym is used, to resolve polysemy, to
decide which language is used, etc. If, for instance, one hears or reads “David
went to the bank” one needs to know who among the many Davids (David
Kaplan, David Braun, David Israel, David Matheson, etc.) went to the bank
and whether he went to the financial institution or near the river. If one comes across
“John saw Jane with the binocular”, before processing its truth Â�conditions one ought
to determine whether it’s of the form “S[John VP[saw NP[Jane with a binocular]]]” or
of the form “S[John VP[saw NP[Jane] PP[with a binocular]]]]”. If one hears ‘Ich!’, one
ought to know whether it is an utterance made by an English speaking person and
thus expressing some kind of disgust or the Â�utterance Â�produced by a German speaking
person meaning ‘I’. These choices happen at the presemantic level.25
At the semantic level, context is exploited in order to determine the referent of
indexical expressions. Once all expressions are disambiguated, structural ambiguity
and polysemy resolved, the language fixed, etc. context remains relevant, for the con-
tent or referent of indexical expressions can be determined only relative to the context
of the utterance itself. As we saw, following Kaplan (1977), indexicals have a linguistic
meaning (character) which can be represented as a function taking as argument context
(whose parameters are: agent, time, location, demonstratum, and possible world) and
giving as value the referent (content). The linguistic meaning of ‘today’, we saw, can be
characterized as “the day in which this token is uttered” which takes as argument the
relevant day and gives as value that very day. We cannot fix the referent of an indexical
expression at the presemantic level: we don’t have infinitely many ‘I’s or ‘now’s standing
for infinitely many individuals or times. A language like that would be, if not impos-
sible, extremely different from natural languages such as English, Russian, Navajo, etc.
Furthermore, ‘I’ and ‘now’ are not ambiguous terms. They are not proper names either.
If David Kaplan and David Beckham say “I am happy”, they are not using the same
25.â•… The presemantic, semantic, and postsemantic distinction doesn’t reflect the way in which an
utterance is processed. From the audience perspective, disambiguation, for instance, can occur
after the sentence has been uttered and the literal (or possible) meaning(s) processed, and the
hearer relies on the whole discourse situation to disambiguate a word and/or a structure.
name ‘I’ to refer to themselves. They use the same tool that, given the context it is used
in (in our case the speakers), refers to different individuals. If today we say “Today is
sunny” and tomorrow we repeat “Today is sunny” we don’t use two homonyms stand-
ing for different days. We use the very same word with the very same linguistic meaning
or character to designate different days. Indexicals are particular words working in a
specific way. In English we have just one ‘I’ and one ‘now’ that we all use to refer to dif-
ferent people and moments depending on the context in which we happen to be. This
seems to be a universal phenomenon across natural languages. If we translate ‘I’ into
French we have ‘je’ which presents the very same semantic properties.26
At the postsemantic level, context is relevant when we try to cash out an utter-
ance’s truth-value. If one were to hear “It’s raining” one ought to determine where it
is raining. For, rain occurs at times and locations. Yet, in “It’s raining” no location is
picked out by an element of the utterance (unlike in “It’s raining here” or “It’s raining
in London” where the relevant place is referred to by ‘here’ and ‘London’). There are at
least three ways one can represent the truth-conditions of an utterance of:
(28) It’s raining
One can either represent its truth-conditions as (22a), (22b), or (22c):

(28) a. An utterance u of “It’s raining” is true iff it’s raining
b. An utterance u of “It’s raining” is true iff it’s raining in l
c. An utterance u of “It’s raining” is true iff it’s raining in the Â�situation
(location) of u
(28a) captures the minimalist approach proposed by Cappelen & Lepore (2005).
(28b) captures Perry’s (1986) view according to which the relevant location is an unar-
ticulated constituent of the proposition expressed, viz. the location enters the proposi-
tion expressed without being specified by any linguistic element either at the surface or
at the deep (logical form) level.27 (22c), on the other hand, captures the view according
to which the proposition expressed can be minimal, i.e. without the location enter-
ing it, viz. without the relevant location being a (unarticulated) constituent of the
proposition expressed. Yet the truth value of the proposition depends on the situation
(location) vis-à-vis which the proposition is evaluated. The truth value of the minimal
26.â•… For argument’s sake I ignore the case where we have to distinguish between “you” as the
second person singular pronoun and “you” as the second person plural pronoun. By the same
token I’m ignoring cases where, like in French or Spanish, we distinguish between the informal “tu/
tú” and the formal “vous/ustedes”.].
27.â•… This also captures the position defended by so-called indexicalists (e.g. Stanley 2001) that
posit a hidden argument (or indexical) in the logical form of “It’s raining”. In that case the location
is articulated. These subtleties, as interesting as they can be, transcend the scope of this chapter.
150 Eros Corazza
proposition that it is raining is thus relative to a situation: the very same (minimal)
proposition can be true vis-à-vis one situation and false vis-à-vis another situation.
This latter position can be characterized as a form of relativized minimalism, inso-
far as a proposition is not, pace the Fregean tradition, true/false absolutely, but only
true/false vis-à-vis the situation in which it is evaluated.28 Another position would be
to embrace a form of radical contextualism in which we would get pragmatic intru-
sion into semantic. In that case the relevant location would enter at the semantic level
via some pragmatic processes of free enrichment. This interpretation would, though,
jeopardize the traditional, Grice-inspired, pragmatics/semantics distinction.29
In dealing with the problem of context sensitivity, in general, and indexicality, in par-
ticular, we saw that a few distinctions need to be made. We first saw how the tradi-
tional Kaplanian framework can be adopted to deal with indexical reference on one
side, and in distinguishing between pure indexicals and demonstratives, on the other
side. To deal with various deviant cases, like answering machines and post-its, where
an indexical is not used to refer to an item pertaining to the context of use (e.g. when
‘now’ or ‘here’ are not used to refer to the time and place of the utterance), it suffices
to appeal to the distinction between deictic (independent) reference and anaphoric
(dependent) reference. This reflects the fact that indexicals can be used either in an
independent (deictic) way or in a dependent (anaphoric) way. We further saw that in
order to fully appreciate the way indexicals work within natural languages we have to
distinguish between the various ways utterances can rely on context. With the distinc-
tion between presemantic, semantic, and Â�postsemantic context exploitation we can
further highlight how indexicals constitute a particular semantic phenomenon and
how the traditional, Kaplan-inspired, treatment is still the best game in town when
discussing how references get fixed with the use of an indexical expression.
References
Barwise, J. & J. Etchemendy (1987). The liar: An essay on truth and circularity. Oxford
Carston, R. (2002). Thoughts and utterances. Blackwell.
28.â•… For a defense of this position see Corazza (2007) and Corazza & Dokic (2007). A similar posi-
tion is proposed by Recanati (2007, 2008). See also Gauker (2003), MacFarlane (2007 and 2009),
Predelli (2005a and 2005b).
29.â•… For a defense of such a position see in particular Sperber & Wilson (1986) and Carston (2002).
Castañeda, H.-N. (1966). ‘He’: A study in the logic of self-consciousness. Ratio 8: 130–157.
Castañeda, H.-N. (1967). Indicators and quasi-indicators. American Philosophical Quarterly 4:
85–100. Reprinted in H.-N. Castañeda (1989): 206–31.
Castañeda, H.-N. (1968). On the logic of attributions of self-knowledge to others. The Journal of
Philosophy 65: 439–56.
Castañeda, H.-N. (1983). Knowledge and self: A correspondence between R.M. Adams and
H.-N. Castañeda. In J. Tomberlin (ed.). Agent, language, and the structure of the world, 293–309.
Hackett Publishing Company.
Castañeda, H.-N. (1989). Thinking, language, and experience. The University of Minnesota Press.
Corazza, E. (2002). Temporal indexicals and temporal terms. Synthese 130 (3): 441–460.
Corazza, E. (2002). ‘She’ and ‘He’: Politically correct pronouns. Philosophical Studies 111 (2): 173–96.
Corazza, E. (2003). Complex demonstratives qua singular terms. Erkenntnis 59 (2): 263–283.
Corazza, E. (2004). On the alleged ambiguity of ‘now’ and ‘here’. Synthese 138: 289–313.
Corazza, E. (2007). Contextualism, minimalism, and situationalism”. Pragmatics & Cognition 15 (1):
115–37.
Corazza, E. & J. Dokic (2007). Sense and insensitivity: or where minimalism meets contextualism”.
In G. Preyer & G. Peter (eds.). Context-sensitivity and semantic minimalism: Essays in semantics
and pragmatics, 169–193. Oxford University Press.
Devlin, K. (1991). Logic and information. Cambridge University Press.
Frege, G. (1919). Thought. In M. Beaney (ed.) (1997). The Frege reader, 325–345. Blackwell.
Gauker, C. (2003). Words without meaning. MIT Press.
Huang, Y. (2000). Anaphora: A cross-linguistic study. Oxford University Press.
Kaplan, D. (1977). Demonstratives. In J. Almog et al. (eds.) (1989). Themes from Kaplan, 481–563.
Kaplan, D. (1989). Afterthoughts. In J. Almog et al. (eds.) (1989). Themes from Kaplan, 565–614.
King, J. (2001). Complex demonstratives. MIT Press.
Kripke, S. (1977). Speaker’s reference and semantic reference. In P.A. French et al. (eds.) (1979).
Contemporary perspectives in the philosophy of language, 6–27. University of Minnesota Press.
Lasnik, H. (1976). Remarks on Coreference. Linguistic Analysis 2: 1–22. Reprint in H. Lasnik (1989).
Essays on anaphora, 90–109. Kluwer.
Lewis, D. (1979). Scorekeeping in a language game. In D. Lewis (1983). Philosophical papers 1,
MacFarlane, J. (2005). Making sense of relative truth. Proceedings of the Aristotelian Society 105:
321–39.
MacFarlane, J. (2007). Semantic minimalism and nonindexical contextualism. In G. Preyer &
G. Peter (eds.). Context-sensitivity and semantic minimalism: Essays in semantics and Â�pragmatics,
MacFarlane, J. (2009). Nonindexical contextualism. Synthese 166: 231–250.
Perry, J. (1977). Frege on demonstratives. The Philosophical Review 86 (4):474–97. Reprinted in
J. Perry (2000). The problem of the essential indexical and other essays, 1–26. CSLI Publications.
Perry, J. (1986). Thought without representation. Proceeding of the Aristotelian Society 60: 137–52.
Reprinted in J. Perry (1993). The problem of the essential indexical and other essays, 205–225.
Perry, J. (2001). Reference and reflexivity. CSLI Publications.
Predelli, S. (1998). Utterance, interpretation, and the logic of indexicals. Mind and Language
13 (3): 400–414.
152 Eros Corazza
Predelli, S. (2005a). Contexts, meaning, truth, and the use of language. Oxford University Press.
Predelli, S. (2005b). Painted leaves, context, and semantic analysis. Linguistics and Philosophy 28:
351–74.
Récanati, F. (2003). Direct reference. Blackwell.
Récanati, F. (2007). Perspectival thought. A plea for (moderate) relativism. Oxford University Press.
Récanati, F. (2008). Moderate relativism. In M. Garcia-Carpintero & M. Kolbel (eds.) Relative truth,
Reichenbach, H. (1947). Elements of symbolic logics. Free Press.
Sidelle, A. (1991). The answering machine paradox. Canadian Journal of Philosophy 81 (4): 525–539.
Smith, Q. (1989). The multiple use of indexicals. Synthese 78: 167–91.
Sperber D. & D. Wilson (1986). Relevance. Blackwell.
Stanley, J. (2000). Context and logical form. Linguistics & Philosophy 23: 391–434.
Stalnaker, R. (1974). Pragmatic presuppositions. In M. Munitz & P. Unger, P. (eds.), Semantics and
Philosophy. New York University Press. Reprinted in R. Stalnaker (1998). Context and Â�content,
Vallée, R. (2003). Context-sensitivity beyond indexicals. Dialogue 42: 79–106.
Intensional logic
Paul Gochet
University of Liège
1.â•… The distinction between intension and extension
The distinction between intension and extension is rooted in the Aristotelian tra-
dition. Aristotle built up a logic whose smallest units are general terms like ‘Man’,
‘Animal’, and ‘Mortal’. With each general term is associated a concept or characteristic
property in Â�virtue of which the term is ascribed to an individual. This is called the
intension of the term. The set of individuals which happen to fall under the above-
mentioned concept or to possess the corresponding property is called the extension
of the term. The copula linking two general terms (‘Men are mortal’) can also be read
intensionally or extensionally. Intensionally, the sentence could be paraphrased as
‘mortality belongs to humanity’. Extensionally, it means ‘the class of men is included
in the class of Â�mortals’. Porphyry (232–300) designed a tree which showed the pos-
sibility of inverting the intensional Â�connection into an extensional one and anticipated
the controversial law according to which extension and intension are in inverse ratio
to one another.
The logic of terms can be interpreted intensionally or extensionally. Aristotle
chose the former. Boole opted for the latter; and his choice turned out to be a very
fruitful one. As Schröder noticed, it is hard to think of a property shared by triangle,
melancholia and sulfuric acid, yet there is a class to which all those entities belong,
namely the complement of the class which is the extension of ‘men’. The extensional
reading of the copula of particular statements (‘Some men are white’) is also easier to
capture in extensional terms (it can be paraphrased into ‘The intersection of the class
of men with that of white things is not empty’).
2.â•… The principle of extensionality and its failures
In its standard version, the system which constitutes the backbone of modern logic,
i.e. pure predicate calculus with identity relation contains three laws which make up the
principle of extensionality as applied to singular terms; to propositions understood as
linguistic entities i.e. as sentences; and to propositional functions (also called ‘open
sentences’), respectively.
A clear formulation of the principle of extensionality can be found in Russell’s
Inquiry into Meaning and Truth, although the terminology is outdated: Russell talks
154 Paul Gochet
about ‘propositional functions’ for what is now called ‘sentential functions’ or ‘open
sentences’.
“The principle of extensionality has two parts”, Russell writes,
The truth-value of any function [any truth-function] of a proposition depends only
upon the truth-value of the argument, i.e. if p and q are both true or both false,
then any sentence containing p remains both true or false, as the case may be, if q is
substituted for p.
The truth-value of any function of a [propositional function] depends only on the
extension of the function, i.e. if whenever ϕ x is true, ψ x is true, and vice versa, then
any sentence about the function ϕ remains true or false as the case may be, if ψ is
substituted for ϕ. (Russell 1940: 168)
To be complete, we should add the law of identity which governs singular terms:
For all x and all y, if x = y then ϕ x implies ϕ y.
The extensionality principle holds only when the syntactic constructions involved
are the three standard constructions admitted in the predicate calculus: predication
(substituting singular terms to the individual variables of the propositional func-
tions); introduction of truth-functional connectives such as ‘not’, ‘and’, ‘or’, ‘if … then
…’; and quantification. It fails for constructions involving modalities such as ‘It is nec-
essary that’ or ‘It is known (believed) that’. Here is a famous example due to Russell:
From ‘George IV wished to know whether Scott was the author of Waverley’ and from
‘Scott = the author of Waverley’, it does not follow that George IV wished to know
whether Scott was Scott.
3.â•… The Frege-Carnap treatment of intensional contexts
In Über Sinn und Bedeutung, Frege (1892) claimed that the cognitive difference
between sentences such as ‘The morning star is identical to the morning star’ and ‘The
morning star is identical to the evening star’ hinges on a difference between sense
and reference: the singular term – later called definite description – ‘The morning
star’ has the same reference as ‘The evening star’, but not the same sense. The contrast
between sense and reference applies also to sentences. The two identity statements
are said to refer to the same truth-value but to express different propositions (not
understood here as linguistic entities but as senses or information contents).
Frege was well aware of the problems we illustrated. He thus took on to develop
his theory of sense and reference further to account for the failures of the principle of
extensionality. According to Frege (1949 [1892]: 87), sentences and singular terms do
not have their usual references and senses when they occur within oblique construc-
tions such as indirect discourse. A verb of propositional attitude (‘A believes that …’)
or, to use Frege’s words, a verb introducing indirect discourse (‘A says that …’) shifts
its reference and sense: what was the sense in direct discourse becomes the reference
in indirect discourse and a new sense has to be conjured up to play the role left by the
sense: “In indirect (oblique) discourse we speak of the sense, e.g. of the words of some-
one else. From this it becomes clear that also in indirect discourse words do not have
their customary nominata; they here name what customarily would be their sense”.
The semantic ambiguity between the senses and references of expressions conjec-
tured by Frege to account for the failure of extensionality in oblique contexts succeeds
in blocking the unwanted inferences and in rescuing substitutivity in those contexts
on the proviso that the substitute and the expression substituted for share more than
reference, extension, or truth-value, i.e. that they share also sense. This is a first step
toward introducing a principle of intensionality to deal with what became known later
as intensional contexts.
In Meaning and necessity, Carnap (1956: 10) formulates a fully-fledged seman-
tic theory which systematically deals with the intension and the extension of singular
terms, general terms and sentences. The basic opposition lies in the contrast between
F-truth and L-truth: “A sentence S1 is L-true in the semantical system S if and only if
[…] its truth can be established on the basis of the semantical rules of the system S alone,
without any reference to (extra-linguistic) facts”. This informal definition leads to the
following one inspired by Leibniz: “A sentence is L-true in the system S if and only if S1
holds in every state description (i.e. in every class of sentences in S which contains for
every atomic sentence either this sentence or its negation but not both). A sentence is
F-true, i.e. factually true, if and only if there is at least one state description in which it
holds and at least one in which it does not hold.” (1956: 12).
Equipped with the notions of L-truth and F-truth, we define the notions of L-equiv-
alence (L-truth of a biconditional) and F-equivalence (F-truth of a biconditional) from
which identification criteria for intension and extension are easily derived. If two des-
ignators (singular terms, general terms, or sentences, respectively) are L-equivalents,
their intensions (individual concepts, properties or relations-in-intension, proposi-
tions, respectively) are identical. If two designators are F-equivalents, their exten-
sions (individuals, classes of n-tuples, truth-values, respectively) are identical.
4.â•… The problem of hybrid contexts
The rules of exchangeability within intensional contexts are more stringent than those
which license substitution in extensional contexts. This guarantees that the intension
of the whole remains unchanged if a subexpression is replaced by one with the same
intension. A problem arises, however, when the same subexpression occurs both in
an extensional and in an intensional context. Frege argues that a term or a sentence
156 Paul Gochet
undergoes a reference shift when it occurs in the scope of verbs such as ‘He says that’,
‘He believes that’, ‘He hopes that’ or in modal contexts (‘It is necessarily the case that’).
The semantic ambiguity allegedly triggered by verbs of propositional attitudes cre-
ates problems. Consider the sentence: ‘Jimmy Carter walks and Tom believes it’. As
P. Tichy (1978: 155) observes, “if Frege were right in maintaining that outside epistemic
contexts sentences stand for truth-values, no proposition would be mentioned in the
first half [of the sentence]; consequently there would be nothing for ‘it’ to refer back
to”. A similar problem arises in connection with the logical rules of existential
Â�generalization (from Fa you may derive ∃xFx).
The force of Tichy’s point becomes obvious if we replace ‘it’ by its antecedent. We
obtain: ‘Jimmy Carter walks and Tom believes that Jimmy Carter walks.’ The two occur-
rences of ‘Jimmy Carter walks’ do not share the same reference, due to the reference shift
triggered by the verb ‘believes’. Hence the two occurrences do not co-refer. The use of
the anaphoric pronoun ‘it’ is illegitimate. There is no co-occurrence to capture.
Frege’s account of sense and reference in indirect discourse aroused strong oppo-
sition from Davidson: “If we could recover our pre-Fregean semantic innocence”,
Davidson writes, “it would seem to us plainly incredible that the words ‘The earth
moves’ uttered after the words ‘Galileo said that’, mean anything different, or refer to
anything else, than is their wont when they come in different environments” Â�(Davidson
1968–9: 141). Davidson’s objection is convincing. In indirect discourse, the words of a
reported speech do not differ in meaning from the same words in direct speech. The
speaker who reports somebody else’s speech may not endorse the truth of the propo-
sition he reports, nor its presuppositions. But the meaning and the reference of the
terms remain the same.
5.â•… I ntensional constructions in natural language:

The montagovian paradigm
In Tarskian semantics (cf. Tarski 1956), the meanings of predicates which share the
same extension but whose intensions differ, such as ‘cordates’ (animals with a heart)
and ‘rinates’ (animals with kidneys), are not distinguished. To give a Â�satisfactory
semantics of natural language we need a more fine-grained semantics in which differ-
ences in intension are reflected.
The difference between extension and intension is important at the level of
the lexicon, as shown by the example above. But it is also important at the level of
the grammatical constructions. We have to distinguish between extensional and
intensional grammatical constructions. The combination of ‘vegetarian’ and ‘pilot’
in Â�‘vegetarian pilot’ is extensional. It can be accounted for in set-theoretical terms:
the set of ‘vegetarian pilots’ is the intersection of the set which is the extension of
the predicate ‘vegetarian’ and that which is the extension of the predicate ‘pilot’.
This is not so for ‘good pilot’ which means ‘good’ as a pilot. Here ‘good’ is not
predicated of the individuals which happen to be pilots but of the property of
being a pilot. On both scores the predicate calculus and the extensional seman-
tics which go with it prove to be too coarse-grained to represent the semantics of
Â�ordinary language.
This should not be blamed on Tarskian semantics since Tarski intended to deal
with formal languages only. He thought ordinary language to be too messy to be given
a rigorous semantics. Around 1970, Richard Montague took up the challenge of giv-
ing a rigorous syntax and semantics of natural language. His first paper on the subject
bears the provocative title ‘English as a formal language’. The final form of Montague’s
treatment can be found in his paper ‘The proper Treatment of Quantification in ordi-
nary English’ (see Montague 1974).
Montague’s intensional model is a quintuple 〈A,I,J, ≤ ,F〉 such that A,I,J are non
empty sets (respectively a set of domains, a set of possible worlds and a set of moments
of time), the relational predicate ‘≤’ is defined over the Cartesian product I × J and F is
a function having as its domain the set of all constants.
In their Introduction to Montague Semantics, Dowty, Wall & Peters bring out the
main achievement of Montague’s semantics quite clearly: “With the advent of Kripke’s
semantics for modal logic (taking possible worlds as indices), it became possible for the
first time to give an unproblematic formal definition of intension for formalized lan-
guages. When we have done this for a formal language resembling English, we will have
produced a theoretical construct (intension of an expression) that does what meaning
does”, in so far as a meaning of an expression is something that determines, for any type
(of expression), place and possible situation, the denotation of the expression in that type,
place and situation (Dowty et al. 1981, quoted by Thayse et al. 1989: 96–97).
It is clear that we go from intension to extension. We have to understand the mean-
ing of a sentence before assessing its truth-value. Montague’s semantics formally cap-
tures this dependence. Intensions are seen as functions from indices (possible worlds
and moments of time) to extensions. The intension of an individual constant (such as
the definite description ‘the President of the U.S.’) is a function which, given a world
and a time, picks up an individual. The intension of a one-place predicate constant or
intransitive verb is a function which, given a world and a time, picks up a set (or its char-
acteristic function). The intension of a sentence is a function which, given a world and
a time, picks up a truth-value. We now have a clear picture of how intension determines
extension. If I know for each possible world and each moment of time which individual
is picked up by ‘the president of the U.S.’ and if I take the actual world and 1944 as argu-
ments, I will obtain Harry Truman as a value.
The view that intension determines extension is, however, an oversimplification.
As Putnam stresses, extension is determined socially and not individually, owing to
158 Paul Gochet
the division of linguistic labor (Putnam 1975: 246). I shall however refrain for belabor-
ing this point here.
Montague’s account of intensions as functions from possible worlds to exten-
sions suffers from another defect. It does not meet the findings of psycholinguists who
study production and comprehension of language. To remedy this shortcoming, van
Lambalgen and Hamm, following in the footsteps of Moschovakis (1993), identify the
sense of an expression with the algorithm that computes the expression’s denotation
(van Lambalgen & Hamm 2005: 49).
6.â•… Inadequacies of the standard semantics of intensional logic
Set-theory is the paradigm of extensionality. One of its axioms, the axiom of exten-
sionality, stipulates that two sets are identical if and only if they share the same mem-
bers. To that extent, sets are individuated in terms of their members, not in terms of
some resemblance or even family resemblance which their members might share.
This situation does not change fundamentally when we bring possible worlds and
moments of time to bear on the matter. Properties are now being individuated in
terms of the instances they have in different possible worlds. This forces us to say that
‘x is sold’ and ‘x is bought’ have the same intension to the extent that they apply to
the same objects in all possible worlds. Hence a theory of properties and propositions
that is obtained by combining possible worlds with set theory is bound to remain too
coarse-grained.
The standard conception of a set, i.e. the iterative conception, makes it meaningless
to speak of sets which are members of themselves. As Boolos (1971: 220) puts it: “Here
are some things. Now we bind them up into a whole. Now we have a set. We do not
suppose that what we come up with after combining some elements into a whole could
have been one of the very things we combined”. This is not the case with properties. We
allow the self-attribution of properties. We feel entitled to say that ‘to be crazy is crazy’
and we count as valid the inference: ‘Everything has the property of being self-identical,
therefore the property of being self-identical has the property of being self-identical’.
To circumvent this problem, several authors have looked for a model of intensional
logic elsewhere than in set theory, for instance we could try to adapt the models built up
for the interpretation of the lambda calculus. As opposed to set theory, lambda calcu-
lus is an intensional theory (as long as no axiom of extensionality has been imported
in it). Church (1941: 3) observes in his monograph that
It is possible […] to allow two functions to be different on the ground that the rule of
correspondence is different in meaning in the two cases although always yielding the
same result when applied to any particular argument. When this is done we shall say
that we are dealing with functions-in-intension.
For instance the expression x(x+1) describes another function than x2 + x even though
they correspond to the same graph.
7.â•… Hyperintensionality
We have seen that Montague’s account of intensions is not fine-grained enough to

account for intentional constructions such as ‘believes that…’ , ‘is aware that’. From
‘A believes that if a woman walks no man talks’, we cannot derive that ‘A believes that
if a man talks no woman walks’. Yet the two conditionals share the same intension in
Montague’s account of intension since they are true in the same possible worlds, actu-
ally in all possible worlds. (the example is borrowed from R. Muskens).
To cope with this problem, R. Thomason enriched Montague’s type theory and
introduced a new type for proposition (Thomason 1980) which denotes objects which
are not constructed from other objects such as possible worlds. Thomason introduced
a function designed to send propositions to their extensions. That function is con-
strained by meaning postulates.
In 2005, R. Muskens spelled out a fully-fledged theory which provides a account
of intension that is fined-grained enough to avoid the identification of ‘p’ with ‘¬¬p’ or
‘p ^ q’ with ‘q ^ p’. Moreover he explains how sense (intension) can be ‘a recipe for find-
ing referents’. Developing views foreshadowed by Moschovakis (Moschovakis 1994)
and van Lambalgen and Hamm (van Lambalgen & Hamm 2005), Muskens managed
to recast Thomason’s meaning postulates in the form of a logic program.
The higher-order logic which is needed to capture the semantics of natural
Â�language might make the reader skeptical about the possibility of systematically
accounting for senses in terms of program. Higher logic is known to be undecid-
able and the unification algorithm wich is the core of logic programming has been
proved to be undecidable beyond first-order. This objection against Muskens’ enter-
prise Â�however can be rebutted. A new kind of unification called patterns unification
has been invented which is decidable in polynomial time. (Miller 1991, referred to by
Â�Muskens 2005: 490–491). This important result does not affect the undecidability of
higher order logic.
Will not the rehabilitation of senses as objects of intentional verbs fall prey to
the puzzles raised by hybrid contexts? If we take propositions as objects of belief,
awareness etc. can we say that ‘Galileo believed that the earth is round and the earth
is round’ without committing ourselves to the view that the first occurrence of the
italicised sentence denotes a proposition while the second denotes a truth-value?
The answer is ‘No’ to the first question and ‘Yes’ to the second. Both sentences can
be said to denote propositions. The price to pay is this: we have to take propositions
rather than sentences as bearers of truth and falsity. Once we have a fine-grained
160 Paul Gochet
criterion of identification for propositions, as it is the case in Muskens’s account, this

move seems quite unproblematic.
As opposed to property theory, Muskens’ approach remains within the main-
stream model theory based on set theory but it brings in several innovations designed
to make it possible to use type theory without commiting oneself to the axiom of
extensionality (Muskens 2007).
8.â•… Propositional attitudes and pragmatics
The problem of defining substitutivity principles within the scope of verbs of proposi-
tional attitudes has not been solved, however fine-grained the concept of propositions
employed. Niiniluoto & Saarinen (1982) argue that all attempts to solve the problem
in semantic terms are doomed to failure; in their opinion, the problem is pragmatic.
The question of ‘how does the meaning (truth-conditions) of Tully is short con-
tribute to the meaning of Herb believes that Tully is short?’ calls for a pragmatic, not for
a semantic answer. We get along, they say, with propositional attitude attributions as
we do with indexicals such as ‘here’, by relying on the context.
Identifying the relevant pragmatic factors for the interpretation of clauses occur-
ring within the scope of a verb of propositional attitude involves a lot of guesswork.
The mechanisms undergoing such guesswork “draw from the speaker’s intentions,
the auditor’s ability to guess those intentions, and other parameters of equally com-
plicated and uncontroversially pragmatic nature” (Niiniluoto & Saarinen 1982: 159).
In his early work, Quine discussed extensively the problems raised by the failure
of the principle of substitutivity of identity in belief contexts examplified in this well-
known example: from ‘Philip believes that Cicero denounced Catilina’ and ‘Cicero =
Tully’, ‘Philip believes that Tully denounced Catilina’ does not follow. Quine initially sug-
gested a semantic treatment of the problem which bears some resemblance with Frege’s
solution. Revisiting the problem in a late essay, he suggested a pragmatic approach. In his
essay ‘Promoting extensionality’ he writes: “When someone ascribes a propositional atti-
tude to someone, he impersonates that person to some degree. The subordinate clause
of the construction is uttered from the subject’s point of view, somewhat as if from the
subject’s mouth. No wonder substitutivity of identity fails: the subject poor fellow, didn’t
know the things were identical” (Quine 1994: 145).
A fully-fledged pragmatic theory has been developed by Recanati in ‘Opacity
and the attitudes’. He shows that most of the facts which Frege tries to account for in
semantic terms, by positing intensional entities, can be dealt with in pragmatic terms
by carefully distinguishing the perspective of the ascriber of propositional entities
from that of the ascribee. Making appropriate use of the ascriber-ascribee contrast
would require us to shift from what might be described as the ‘ascriber’s world’ to the
‘ascribee’s world’, but the ontology would remain that of the ascriber all along, that is,
the singular terms would refer to the same objects, whether we were talking about
the actual world or about the ascribee’s belief world (Recanati 2000). This tallies with
Davidson’s requirement.
9.â•… Intension, compositionality and context-dependence
The principle of compositionality [the meaning of the whole is a function of the

Â�meaning of the parts and the grammatical structure] explains the property of language
which Wittgenstein describes in these terms in the Tractatus: “I understand the propo-
sition without having its sense explained to me” (Wittgenstein 1960 [1921]: 41). Lan-
guage production and language comprehension are clear examples of rule-governed
behviour. Yet in some areas of language, production and comprehension should be
conceived of as a goal-directed behavior. Accomplishment verbs are a case in point. To
understand the meaning of an accomplishment verb phrases such as ‘cross the street’
or ‘build a house’, we need to bring in a goal-plan structure (van Lambalgen & Hamm
2005: 44). The NP ‘the house’ need not refer to an existing house (2005: 47). Here the
meaning of the part is determined by the meaning of the whole. Hence the principle
of compositionality which works inside out does not suffice. It has to be supplemented
by contextual factors which work outside in. Van Lambalgen and Hamm claim that
what gets computed by the sense of an expression (or set of expressions) is not sim-
ply a reference but “a model in which these expressions can be given a denotation”
(Van Lambalgen & Hamm 2005: 50).
References
van Benthem, J. (1988). A manual of intensional logic. Center for the Study of Language and
Â�Information, Stanford.
Boolos, G. (1971). The iterative conception of a set. The Journal of Philosophy 68: 215–231.
Bealer, G. & U. Mönnich (1989). Property theories. In D. Gabbay & F. Guenthner (eds.) Handbook
of philosophical logic 4: 133–251.
Carnap, R. (1956). Meaning and necessity. University of Chicago Press.
Chierchia, G. (1985). Formal semantics and the grammar of predication. Linguistic Inquiry 16:
417–443.
Chierchia, G., B. Partee & R. Turner (1989). Properties, types and meaning, 2 vols. Kluwer.
Church, A. (1941). The calculi of lambda conversion. Princeton University Press.
Cocchiarella, N. (1988). Predication versus membership in the distinction between logic as language
and logic as calculus. Synthese 77: 37–72.
——— (1989). Conceptualism, realism, and intensional logic. Topoi 8: 15–34.
Davidson, D. (1968–9). On saying that. Synthese 19: 130–146.
162 Paul Gochet
Dowty, D., R. Wall & S. Peters (1981). Introduction to Montague Semantics. Springer.
Frege, G. (1949) [1892]. On sense and nominatum. In H. Feigl & W. Sellars (eds.) Readings in philo-
sophical analysis: 85–102. Appelton Century Crofts.
Gamut, L.T.F. (1991). Logic, language and meaning, vol. 2. Intensional Logic and Logical Grammar.
University of Chicago Press.
Heny, F. (ed.) (1981). Ambiguities in intensional contexts. Reidel.
Hintikka, J. (1969). Models for modalities. Reidel.
Hintikka, J. & M. Hintikka (1989). The logic of epistemology and the epistemology of logic. Reidel.
Lambalgen, M. van & F. Hamm (2005). The Proper Treatment of Events. Blackwell.
Miller, D. (1991). A Logic Programming Language with Lambda-abstraction, Function Variables
and Simple Unification. Journal of Logic and Computation 1: 497–536.
Moschovakis, Y. (1993). Sense and denotation as algorithm and value. In J. Oikkonen & J. Väänänen
(eds.) Lecture Notes in Logic: 210–249. Association for Symbolic Logic, A.K. Peters Ltd.
____ (1994). Sense and Denotation as Algorithm and Value. Logic Colloquium 90 (Helsinki 1990),
vol.2 of Lectures Notes in Logic: 210–249. Springer.
Muskens, R. (2005). Sense and the computation of reference. Linguistics and Philosophy 28:
473–504.
____ (2007). Intensional Models for the Theory of Types. Journal of Symbolic Logic 72: 98–118.
Niiniluoto, I. & E. Saarinen (eds.) (1982). Intensional logic. Acta Philosophica Fennica 35.
Orilia, F. (1999). Predication, analysis and reference. Clueb.
Putnam, H. (1975). The meaning of ‘meaning’. In H. Putnam Mind, Language and Reality, Philosophi-
cal papers, vol.2: 215–271. Cambridge University Press.
Quine, W.V. (1994). Promoting Extensionality. Synthese 98: 143–151.
Recanati, F. (2000). Opacity and the Attitudes. In P. Kotatko & A. Orenstein (eds.) Knowledge,
Â�Language and Logic. Questions for Quine: 367–407. Kluwer.
Rijke, M. de (ed.) (1997). Advances in Intensional Logic. Kluwer.
Russell, B. (1940). An inquiry into meaning and truth. Allen & Unwin.
Salmon, N. (2002). Puzzles about Intensionality. In D. Jacquette (ed.) A Companion to, Philosophical
Logic: 73–85. Blackwell.
Tarski, A. (1956). Logic, semantics, metamathematics. Oxford University Press.
Thayse, A. (ed.) (1989). From modal logic to deductive databases, vol. 2. Wiley.
Thomason, R. (1980). A Model Theory for Propositional Attitudes. Linguistics and Philosophy 4:
47–70.
Tichy, P. (1978). Two kinds of intensional logic. Epistemologia 1: 143–162.
Vergauwen, R. (1993). A metalogical theory of reference. University of America Press.
Wittgenstein, L. (1960) [1921]. Tractatus-logico-philosophicus, with a new Translation by
D.F. Pears & B.F. McGuinness and with the Introduction by B. Russell. Routledge & Kegan Paul.
Zalta, E. (1988). Intensional logic and the metaphysics of intensionality. MIT Press.
Modal logic
Paul Gochet
1.â•… The development of modal logic
Standard elementary logic studies the deductive inferences which rest on two sets of
logical constants: the connectives ‘not’, ‘and’, ‘either … or …’, ‘if … then …’ and the
quantifiers ‘all’ and ‘some’. Modal logic extends the logical lexicon in order to account
for the validity or non-validity of inferences which depends on the presence of modal
terms such as ‘necessarily’ or ‘possibly’. With respect to arguments involving these
terms, intuition cannot be trusted. Even the best logicians and philosophers fall prey
to fallacies in this area. For instance, Etchemendy (1990: 87) discovered a hidden
Â�fallacious move in Tarski’s account of logical consequence, where he went from a state-
ment of the form ‘Necessarily if p and q, then not r’ to a statement of the form ‘if p then
necessarily if q then not r’.
Modal logic goes back to Aristotle, the Megarics and the Stoics. On Aristotle’s
account, categorical propositions assert or deny that a predicate belongs to a Â�subject.
In Analytica Priora 1 (29b 29–35), he states that there is a difference between belong-
ing, necessarily belonging and being able to belong. Modalities qualify the onto-logical
relation of belonging that obtains between attribute and substance. Aristotle’s modali-
ties are ontic modalities: they are rooted in metaphysical features of reality. To explain
not necessarily belonging, one has to bring in the metaphysical concept of accident. To
explain being able one has to bring in the notion of potentiality.
Aristotle’s theory provides clues to differentiate between valid and invalid modal
syllogisms. For example, the syllogism ‘Some S is necessarily not P. All S are necessar-
ily R. Therefore some R is necessarily not P’ is a valid syllogism. Another discussion of
modal statements can be found in Aristotle’s De interpretatione (Chapters 12 and 13)
where a square of opposition sums up the logical relations obtaining between ‘possible
to be’, ‘necessary to be’, and ‘impossible to be’.
A distinction must be made between the ontic modalities mentioned above and
the notion of logical necessity which occurs in Aristotle’s (An. Pr. 24a 18) definition of
a syllogism (“A syllogism is discourse in which certain things being stated, something
other than what is stated follows of necessity from their being so”). Logical modali-
ties are sometimes called alethic modalities. The link tying together the premises of
a valid syllogism to its conclusion can be described in terms of truth preservation (a
164 Paul Gochet
syllogism being valid if all substitutions for its three terms which preserve truth in the
premises, preserve truth in the conclusion too).
In Aristotle’s time, the Megaric logician Diodorus of Chronos, took part in a
famous discussion on the Master argument, which claims that the following three
propositions cannot be true together:
a.â•… Every proposition concerning the past is necessary
b.â•… The impossible does not follow from the possible
c.â•… Something that neither is nor will be is possible.
Disagreement arose as to which of the three had to be abandoned to restore
Â�consistency. The very formulation of the Master argument bears witness to a shift from
ontic to temporal modalities. Diodorus came to define ‘It is possible that A’ as ‘A or it
will be the case that A’ and ‘It is necessary that A’ as ‘A and it will be the case that A
forever’ (see Section 3).
In the Middle Ages philosophers came to grips with inferences whose logical struc-
ture contains both modal terms and quantifiers. Can we from ‘It is possible that every-
body comes’ validly infer that ‘Everybody is such that possibly he comes’? Â�Buridan’s
answer is negative. It could be that everyone is God, (since God could Â�annihilate
all creatures) but it does not follow that ‘everything could be God’ (for no creature
could be).
Logicians bring inferences based on ontic, temporal, deontic and epistemic
modalities together under the heading modal logic. This does not, however, do
Â�violence to the use of the term ‘modality’ by grammarians. In English grammar, for
instance, some unity is found in the field in the category of modal auxiliaries. In ‘If
A is a Â�triangle, it must have three sides’, must expresses logical necessity. In ‘John will
come’, will expresses futurity. In ‘Thou shall not kill’, shall expresses deontic interdic-
tion. In ‘He may be at home’, may expresses an epistemic modality: the sentence can be
paraphrased into ‘It is compatible with what I know that he is at home’ (Palmer 1979).
2.â•… Irving Lewis’ contribution
In his Begriffschrift (1879), Frege left the study of modal terms aside. In Principia math-
ematica (1910–1913), Whitehead and Russell pursued Frege’s program of founding
mathematics on logic, and ignored modality. Their system of propositional logic rests
upon two primitive ideas: the one-place connective ¬ (‘not’) and the two-place con-
nective ∨ (‘Either … or …’). The connective → (‘If … then …’) is defined as ‘Either it
is not the case that … or …’. This definition goes back to Philo, a pupil of Diodorus.
It makes the conditional true if it has a false antecedent or a true consequent. Philo’s
reading of the conditional significantly differs from that of Diodorus. Diodorus under-
stood ‘if…then…’ in the modal sense as ‘It is impossible that… and not…’.
Modal logic 165
Whitehead and Russell interchangeably use the connective ‘if then’, which belongs
to the object-language, and the predicate ‘implies’, which belongs to the metalanguage.
It has been argued that this practice confused use and mention. In the first occurrence,
p and q stand for sentences, whereas in the second they stand for the names of
these sentences. Thus, if in the following theorem of the propositional calculus, ¬ p →
(p → q), the two readings of the arrow are blended, you can paraphrase the formula as
follows: ‘If it is not the case that p, then p implies any proposition q whatsoever’. So if
we take the Philonian conditional as Â�expressing implication, we get a very weak sense
of implication. In 1918, Irving Lewis spelled out a propositional logic which contained
a sign for a connective that was designed to capture a sense of implication closer to
Diodorus’ view, as primitive. He called this implication ‘strict implication’ and distin-
guished it from Russell’s material implication by using the symbol .

Lewis adopted the logistic method, which requires the use of formal language in
which proofs take place through operations on symbols according to precise rules.
Rather than depending on a pre-theoretic meaning, the axioms of a formalized calcu-
lus bestow meaning upon the primitive symbols. Any interpretation is admissible if it
makes the set of axioms true. Thus, different sets of axioms can confer different mean-
ings on the strict conditional ‘If … then …’ just as different sets of axioms for geometry
can confer different meanings upon the primitive terms ‘point’, ‘line’, and ‘plan’.
Lewis spelled out a whole set of axiomatic systems which he called S1–S8. Each
of them imposes a different meaning on the strict conditional and indirectly on the
modal operators ‘necessarily’ (, the ‘box operator’) and ‘possibly’ (, the ‘diamond
operator’). In the current practice today, the box operator is taken as primitive and
the diamond operator is defined as ¬  ¬. The strict conditional is ‘defined in terms

of the Philonian and the box operators in the following manner:  (p → q). Scott
(1971) showed how to recast Lewis’ account in order to safeguard it against the
Â�use-mention confusion.
3.â•… What is modal logic all about?
To interpret formulas of a modal language, we use a model made up of a set of pos-

sible worlds W, an accessibility relation R defined on W, and a valuation function V
which gives the truth-value of each propositional atom in each ‘possible world’. To
grasp what is proper to modal logic, however, we do not have to be so specific. We can
content ourselves with a more neutral definition and say that modal language is a way
of talking about relational structures. A relational structure is simply a set together
with a collection of relations on that set. One might wonder whether modal logic is
really needed to talk about relational structures. Why not use the classical predicate
calculus for that purpose (taking sets as values of variables and interpreting the dyadic
predicate ‘Rxy’ as expressing the accessibility relation)? The answer is this: When we
166 Paul Gochet
interpret a modal formula at a certain point (world) we dismantle the formula stepwise
by applying the clauses of the recursive definition of truth.
We scan only the points that are accessible from the current point. In other terms,
we adopt an internal perspective on the relational structure. In contrast, when we
interpret a quantified formula of the predicate calculus, we stand outside the rela-
tional structure and scan the information it contains from an external vantage point
(Blackburn, de Rijke & Venema 2001: XII–XIII).
Even if a modal propositional axiomatic system can be translated into a fragment
of non modal first-order logic, important differences remain between the two formal-
isms. Usually modal logic has good computational properties (decidability, expressiv-
ity) which first-order theories lack (see Wolper 1983).
4.â•… Quantified modal logic
The invention of formal systems of quantified modal logic goes back to 1946. Two
foundational papers appeared in succession in the Journal of Symbolic Logic: ‘Func-
tional Calculus of First-Order Based on Strict Implication’ by Ruth C. Barcan (Barcan
1946) and ‘Modalities and Quantification’ by Rudolf Carnap (Carnap 1946). Quantified
modal logic provides the formal tools to capture an important distinction foreshad-
owed by Aristotle, the distinction between de dicto and de re constructions illustrated in
the contrast between (a) and (b) below (adapted from Analytica Priora (II, 21, 67a16ff):
(a) It is necessary that every triangle has angles equal to two right angles.
(b) Of every triangle it is necessary that it has angles equal to two right angles.
The first sentence entails the second if we subscribe to the principle (c), proposed by
Ruth Barcan – and referred to as the Barcan formula.
(c) If necessarily everything is P then of everything it is necessary that it be P.
In quantified modal logic the following principle holds:

(d) ∀x∀y(x = y⊃  x = y)
When the modal operator ‘’ is read as ‘it is necessary’, the principle holds asÂ�
illustrated by Kripke’s example (d′):
(d′) If Hesperus = Phosphorus then necessarily Hesperus = Phosphorus
When the modal operator ‘’ is read as ‘A knows that’ it does not, as shown by Lauri
Carlson’s example (d″) :
(d″) ∃ x ∃ y (x = y ∧ ¬ the police knows that x = y)
Modal logic 167
Trivial transformations show that (d″) is the negation of (d) in which the belief-Â�
operator has been substituted for the necessity-operator.
For the interpretation of quantified modal logic we have to enrich the model with
a domain of individuals D for the interpretation of the singular terms and with an
assignment function g for the interpretation of the individual variables. ‘We may also
introduce different domains, one for each possible world. This helps us accommodate
singular terms such as ‘Pegasus’ which refers to an individual in some worlds but not in
the real world (Hughes & Cresswell 1996; Fitting & Mendelsohn 1998; Garson 2006).
5.â•… Tense logic
Standard logic deals with propositions whose copula is the tenseless ‘is’. It is not
equipped to reflect the logical structure of inferences which involve tenses and preÂ�
dicates referring to time. It fails to account for the difference between ‘John married
Mary, Mary is a widow, ergo John married a widow’, which is unsound and ‘John
Â�married Mary, Mary is a half-breed, ergo John married a half-breed’, which is sound.
There are two ways of representing the logical form of tensed sentences. One
could use a propositional tense logic in which modal operators are meant to capture
the future and the past. One could also use an applied predicate calculus containing
the predicate ‘x is earlier than y’ in which the variables ‘x’ and ‘y’ range over instants.
A crucial question arises here: can one of these two accounts be reduced to the
other? The philosophical significance of this question is best understood against
the background provided by McTaggart’s famous distinction between A-series and
B-series. According to McTaggart, events or temporal positions may be ordered either
as past-present-future or by the binary relation earlier than. In the first case they form
what McTaggart calls the A-series. In the second, they form what he calls the B-series.
McTaggart assumed that the reality of time rests upon change and he argued that
unless the characteristics of the A-series (past-present-future) can change (as is the
case if the future becomes present and the present becomes past), nothing can change.
Should the A-series be reducible to the B-series, we would be led, he thinks, to see time
as a timeless tapestry in which events are stuck once for ever, and we would have to
admit, as he boldly did, that time is unreal (Mc Taggart 1908).
Prior (2003 [1968]) made a major step forward toward the solution of this
problem. First he constructed a calculus in which ‘it is true at instant a that p will
be true’ is defined by ‘there exists some instant b which is later than a and p is true
at b’. Next he constructed another calculus in which ‘there exists some instant b
which is later than a and p is true at b’ is defined by ‘it is true at instant a that p will
be true’. (The two definitions differ only in so far as their definiens and definiendum
are swapped). Can we conclude that the conception of time based on the triplet
168 Paul Gochet
past-present-future and the conception of time based on the binary relation earlier
than are interdefinable and therefore reducible to one another? Prior adduces an
argument showing that this hasty conclusion is unwarranted. He observes that the
first reduction rests on manipulations of definitions only. The second, in contrast,
rests on axioms (which can bear a Â�truth-value). Hence the two reductions are not
on an equal footing.
6.â•… From tense logic to pragmatics
Lyons (1977) pointed out that the participants in a language event must be able to
control two different frames of temporal reference:
1. the deictic frame which relates the time of the event that is being described to
the context of the utterance,
2. the frame of reference which locates the time of the event in relation to
calendar time.
The difference is quite straightforward. The sentence ‘End of the lease: 1–1–1993’ does
not tell you if the event is past, present or future relatively to the utterance. On the
contrary, ‘the end of lease is recent’ or ‘the lease has come to an end’ does. Deictic
Â�temporal reference is of crucial importance if we adopt Bar-Hillel’s (1970) concep-
tion of pragmatics: the study of the way in which changes of context (situational or
Â�linguistic) influence meaning.
The semantics of the tenses of natural language relies heavily upon indices. In his
Elements of Symbolic Logic, Reichenbach brought to the foreground the fact that the
tenses of natural language are very often referential and anchored in an index which is
the point of speech. The first layer of the grammatical tenses of many natural languages
can be obtained by distinguishing three different positions in relation to the point of
speech: respectively ‘before the point of speech’, ‘simultaneous with the point of speech’
and ‘after the point of speech’ ( Reichenbach 1974 [1947]). Gabbay’s (1974) formal
semantics of the tenses of English makes systematic use of a pair of indices (one index
for the utterance time and the other for the time of evaluation).
Among the various indices, there is one which raises important questions, namely
‘now’. “The problem of Now”, Carnap says, “worried Einstein seriously”. “He explained
that the experience of the Now means something special for men, something different
from the past and the future, but that this important difference does not and cannot
occur within physics” (Carnap 1963: 37).
Neither the term ‘now’ nor the term ‘then’ existed in the language of Prior’s tense
logics considered above. To remedy this lack of expressive power, Goranko (1996)
extended the language of temporal logic with two new symbols (down arrow ↓ and
Modal logic 169
up arrow Â�↑) which capture the deictic indexical ‘now’ and the anaphoric indexical
‘then’, respectively. The first symbol is used to fix a point of reference. The second,
which is used when we want to refer back to the point of reference, is the reference
pointer. With that apparatus we can formally express ‘now will always fail to be the
case then’ which is the logical rendering of the colloquial sentence ‘now will never
occur again’ (Goranko 1996). (For a formal semantics for ‘now’, see Kamp 1971 and
Gabbay 1974, 1976).
The asymmetry between the two definitions of time, the absence of ‘now’ in the
language of physics, and the meaningfulness of the sentence ‘now will never occur
again’ all point towards the same fact. They highlight the irreducibility of pragmatics
to semantics.
References
Ackrill, J.L. (ed.) (1963). Aristotle’s Categories and De interpretatione. Oxford University Press.
Aristotle: see Ackrill (ed.) (1963). and Ross (ed.) (1928).
Barcan Marcus, R. (1946). A functional calculus of first order based on strict implication. The Journal
of Symbolic Logic 11: 1–17.
Bar-Hillel, Y. (1970). Aspects of Language. North Holland.
——— (1971). Pragmatics of natural languages. Reidel.
Blackburn, P., M. de Rijke & Y. Venema (2001). Modal Logic. Cambridge University Press.
van Benthem, J. (1977). Tense logic and standard logic. Logique et Analyse 80: 395–437.
——— (1983). The Logic of Time. A Model-Theoretic Investigation into the Varieties of Temporal
Â�Ontology and Temporal Discourse. Reidel.
Carnap, R. (1946). Modalities and Quantification. The Journal of Symbolic Logic 11: 33–64.
——— (1963). Carnap’s Intellectual autobiography. In P.A. Schilpp (ed.) The Philosophy of Rudolf
Carnap. Open Court.
Fitting, M. & R. Mendelsohn (1998). First-Order Modal Logic. Kluwer.
Gabbay, D. (1974). Tense Logic and the Tenses of English. In J. Moravcsik (ed.) Logic and Philosophy
for Linguists: 177–186. Mouton.
——— (1976). Investigations in modal and tense logics. With applications to Problems in Philosophy
and Linguistics. Reidel.
Gabbay, D. & J. Woods (eds.) (2006). The Handbook of History of Logic, vol.7 Logic and the Modalities
in the Twentieth Century. Elsevier.
Garson, J. (2006). Modal Logic for Philosophers. Cambridge University Press.
Goranko, V. (1996). Hierarchies of Modal and Temporal Logic with Reference. The Journal of Logic,
Language and Information 5: 1–24.
Hughes, G.E. & M.J. Cresswell (1996). A new introduction to modal logic. Methuen.
Kamp, H. (1971). Formal Properties of ‘Now’. Theoria 37: 227–273.
Lewis, I. & C.H. Langford (1959) [1932]. Symbolic logic. Dover.
Lyons, J. (1977). Semantics, vol. 2. Cambridge University Press.
McTaggart, J.M.E. (1908). The Unreality of Time. Mind: 457–474.
Moisil, G. (1972). Essai sur les logiques non chrysipiennes. Académie Socialiste de Roumanie.
170 Paul Gochet
Palmer, F.R. (1979). Modality and the English modals. Longman.

Reichenbach, H. (1974) [1947]. Analysis of Conversational Language. In J. Moravcsik (ed.) Logic and
Philosophy for Linguists: 122–141. Mouton.
Ross, W.D. (ed.) (1928). The works of Aristotle: Analytica Priora. Clarendon.
Scott, D. (1971). On engendering an illusion of understanding. Journal of Philosophy 68 (21): 787–807.
Wolper, P. (1983). Temporal Logic Can Be More Expressive. Information and Control 56: 72–99.
Model-theoretic semantics
Paul Gochet
1.â•… The meeting of two different approaches to semantics
Quine once argued that semantics had split into two provinces “so fundamentally dis-
tinct as not to deserve a joint appellation at all” (1953: 131): the theory of meaning
whose main concepts, apart from meaning itself, are synonymy, significance and ana-
lyticity, and the theory of reference whose main concepts are those of naming, truth,
denotation and extension. Katz (1972: 182) expresses full agreement with this view
and describes the term ‘semantics’ as ambiguous.
Semantics in the second sense originated in Tarski’s work (see Tarski 1956). The
revival of semantics in the first sense started with Katz & Fodor (1971), who present a
theory developed in the wake of Chomsky’s generative grammar.
Just as Chomsky had come to grips with the problem of accounting for the human
syntactic competence which manifests itself in the ability to identify the members of
the potentially infinite set of syntactically well-formed expressions of a given natural
language, Katz & Fodor set out to reconstruct “the speaker’s semantic ability to inter-
pret any of the infinitely many sentences of his language” (Katz & Fodor 1971: 487).
Wittgenstein had already been struck by our ability “to understand the proposition
without having had its sense explained” (1921: 4.021), and he tried to account for this
peculiarity of language by appealing to the picture theory of propositions (“A proposi-
tion shows its sense”, 4.022). Katz & Fodor, who observe that “the most characteristic
feature of language […] is its ability […] to produce and understand sentences never
before encountered” (1971: 475), introduced a new component in semantic theory
which was meant to explain this feature, i.e. a system of projection rules which guide
(a) selections among the senses which words have in isolation (e.g. in The man hits the
colorful ball, the occurrence of ‘hits’ blocks the reading of ‘ball’ as a formal gathering
for social dancing), and (b) the amalgamation of elements from different sets of paths
(in the sense given to the latter in transformational grammar but enriched with lexical
pieces of information) when the selection restrictions are satisfied.
Katz & Fodor are concerned with senses, ambiguities, anomalies, and paraphrases
in synonymous sentences. They deal with natural language. On the other hand, Tarski
deals with artificial languages free of ambiguities and anomalies, and he focuses on
reference and extension. Yet, when dealing with the semantics of formal languages,
Tarski addresses the very problem which Katz & Fodor solve with their projection
172 Paul Gochet
rules, but he uses an altogether different tool to generate the compound meanings of
the potentially infinite set of sentences by finite means: the recursive definition of the
semantic notions of satisfaction and truth.
For these two approaches to meet, a final step had to be taken: bridging the gap
between natural and formal languages. This was done by one of Tarski’s students, Â�Richard
Montague, whose basic claim is: “I reject the contention that an important theoretical
difference exists between formal and natural languages” (1974: 19). Â�Montague consid-
erably enriched both the syntax and the semantics of formal language in order to turn
it into a fragment of natural language. He did much to reunify the two disconnected
provinces of semantics described by Quine. Independently of this later rapprochement,
there is a great deal to be learned from Tarski’s foundational work in semantics. As
it stands, Tarski’s semantics of formal languages provides us with a straightforward
account of the meaning of logical constants (connectives and quantifiers) of natural
languages in so far as the meaning of these constants coincides with their extension.
2.â•… The basic notions of Tarski’s semantics
Tarski understands ‘semantics’ as “the totality of considerations concerning those

Â�concepts […] which express certain connexions between the expressions of a language
and the objects and states of affairs referred to” (1956: 401). As typical examples of
semantic concepts he mentions denotation, satisfaction and definition:
The expression ‘the victor of Jena’ denotes Napoleon; snow Â�satisfies the condition ‘x
is white’; the equation ‘x3 = 2’ defines (determines uniquely) the cube root of the
number 2 (Tarski 1956: 401)
It is clear from these examples that the denotatum hooked up to the world of history
by ‘the victor of Jena’ is the referent of the definite description used to denote him, as
opposed to its sense. By analogous considerations, it can be shown that what is the
target of the condition ‘x is white’ in the world is not only its expected satisfiers but
also the satisfiers of every odd condition which happens to share its satisfiers with ‘x
is white’. This clearly indicates that what is at stake here is the extension as opposed to
the intension of ‘white.’
Semantics does not link linguistic expressions to extralinguistic entities as directly
as a label in front of a statue in a museum does. It expresses these links linguistically
in a metalanguage which contains both expressions referring to entities in the world
and expressions referring to linguistic symbols, the latter often being the very symbols
inserted between quotation marks.
Let us consider a formal language L with a lexicon consisting of (1) logical sym-
bols (a countable set of individual variables x, x’, x’, …, the connectives ¬, ∨, →, ↔⇔
and &, the quantifiers ∃ and ∀, the identity symbol = and parentheses) and (2) non-
logical symbols (a finite set of n-place predicates), and with a syntax allowing three
grammatical constructions, viz. (1) a predication generating open sentences such as

Fx1…Fxn by combining n-place predicate constants with n variables, (2) quantification
binding free individual variables, and (3) the introduction of unary or binary connec-
tives. To cover cases of arbitrary complexity, we have to give a recursive definition of
a (well-formed) formula. The basis of the recursion stipulates that a n-ary predicate
constant followed by n occurrences of individual variables is a formula. To this we add
a recursive clause for each construction (e.g. If A and B are formulas, then A∧B is a
formula. If A is a formula, then ∀xA is a formula).
We turn language L into a first-order logical calculus C by adding the usual Â�axioms
and inference rules of first-order logic including the axioms governing the identity
symbols. We turn this first-order calculus into the first-order theory T of partial order
by adding predicate constants (and eventually functions symbols) together with
Â�axioms proper to the theory such as, e.g. the following three:
∀x(x ≤ x)
∀x∀y((x ≤ y ∧ y ≤ x) → x = y)
∀x∀y∀z((x ≤ y ∧ y ≤ z) → x ≤ z)
We are now ready to introduce all the basic concepts of Tarskian semantics for the
theory T.
We call interpretation I of the theory T the ordered pair 〈D,F〉 formed of a non-
empty domain D over which the individual variables of T range together with a
function F which associates a set of n-tuples from Dn with each n-place predicate
constant.
An ordered pair of individuals – let us say 〈0, 1〉 – satisfies or fails to satisfy a two-
place predicate such as ‘x is smaller or equal to y’. We ought to be able to deal with open
formulas of any finite length, such as a conjunction of a thousand open sentences. To be
prepared for cases of arbitrary complexity, we arrange the individuals of our non-empty
domain in arbitrary infinite sequences and we use these sequences as satisfiers for our
open sentences. We also order the infinite set of individual variables of language L.
Let us use the symbol S(xi) for the individual item which is correlated with di in
the sequence of individuals S, i.e. 〈d1, d2….〉. We can now give a recursive definition of
what it means to say that a sequence S satisfies a formula A with respect to an interpre-
tation I. There is a clause for each atomic predicate constant, a clause for each connec-
tive and a clause for each quantifier. These clauses read as follows.
1. If A is an atomic formula of the form Rx1…xn, S satisfies A with respect to I if and
only if the n-tuple of individuals 〈d1…dn〉 denoted by 〈S(x1)…S(xn)〉 belongs to
the set of n-tuples which the interpretation function F associates with R.
2. If A is ¬ B, for some B, then S satisfies A if and only if S does not satisfy B.
3. If A is B ∨ C, for some formula B and C, then S satisfies A if and only if either
S satisfies B or C or both. The definition of &, → and ↔ are Â�analogous, mutatis
mutandis.
174 Paul Gochet
4. If A is of the form ∃xiB, for some variable xi and some formula B, then S satisfies
A if and only if there is at least a sequence S’ which differs from S as far as the i-th
member of S is concerned and S’ satisfies B. There is an analogous clause for the
universal quantifier.
We can now define truth in terms of satisfaction. A formula A is true under interpreta-
tion I if it is satisfied by all sequences. It is false if it is satisfied by no sequence.
A model M of an axiomatized theory T is an interpretation I (〈D,F〉) in which all
the axioms are true under I. The interpretation of the theory T with the three axioms
mentioned above in which the domain D is the set of natural numbers and in which R
is interpreted as the relation smaller or equal to is a model of that theory.
A sentence B is a logical consequence of a set of formulas S (S |= B) if and only if
every model M of the set S is also a model of the sentence B.
An inference ‘S therefore B’ is valid if the conclusion B is a logical consequence of
S. The Tarskian notion of validity is important for linguists. As Kameyama observes,
“Grammatical reasoning is governed by the Tarskian notion of valid inference in stan-
dard logic” (Kameyama 1996: 110). We shall see later that another notion of validity
is needed to account for the pragmatic reasoning involved in the interpretation of
natural language.
This definition of logical inference given by Tarski is a major step forward. It
shows that we can describe the logical knot which ties the premises of a deduc-
tive inference to its conclusion without bringing in the modal notion of necessity.
Â�Aristotle said : ‘A syllogism is a discourse in which, certain things being stated,
something other than what is stated follows of necessity from their being so’.
Â�Taking advantage of Tarski’s semantics we can reduce the modal notion ‘follows
of necessity’ to a simpler notion, namely the notion of generality: ‘the conclusion
is true whenever the premises are true’. This result of Tarski was anticipated by
Bolzano in 1833.
3.â•… The scope and limits of Tarski’s semantics
To appreciate the novelty of Tarski’s definition of truth, one should set it off against
the classical correspondence theory which states that truth lies in the agreement
of sentences with facts. In the traditional view, different truths are expressed by
‘snow is white’ and by ‘grass is green’. Undoubtedly they differ as far as meaning
is concerned. But they do not differ as far as truth is concerned. Hence we should
not be reluctant to say that they are all satisfied in the same way. They are satis-
fied by all sequences (and false sentences are satisfied by none). Appealing to facts
and explaining truth as Â�correspondence with facts has no explanatory value. What
are facts? If we reply that facts are what makes sentences true, our explanation is
Â�circular (see Gochet 1980).
Tarski’s definition does not erase the difference between the way distinct sentences
such as ‘Snow is white’ and ‘Grass is green’ are hooked onto the world. This point was
highlighted by Davidson in these terms: “[s]ince different assignments of entities to
variables satisfy different open sentences and since closed sentences are constructed
from open, truth is reached, in the semantic approach, by different routes for different
sentences. All true sentences end up in the same place, but there are different stories
about how they got there” (Davidson 1984: 48–49).
Tarski’s semantics is compositional. The recursive definition of satisfaction shows
how the meaning of the whole depends on the meanings of the parts together with the
syntactic structure. Admittedly it deals with extension only. The differences in inten-
sion are ignored. For instance, no distinction is made by Tarskian semantics between
cordates (animals with a heart) and rinates (animals with kidneys). These predicates
have the same extension (they are satisfied by the same individuals) but they do not
have the same intension. In Muskens 2005 it is shown how this limitation can be over-
come within the Tarskian paradigm (if we adopt Muskens’s approach described in
‘Intensional Logic’). For a thorough examination of the differences between semantics
inspired by Chomsky and his school and Montague’s semantics see Sandu and Aho
(Sandu & Aho 2009).
A semantics that takes only extension into account is not eo ipso deficient. There
are terms whose meaning is exhausted by an account of their extension.The point is
made forcefully by Garcia-Carpintero (1993). This is the case for truth-functional con-
nectives such as ‘not’, ‘or’, ‘if then’ (at least when ‘if p then q’ is understood as ‘not p or
q’) or quantifiers (‘something’, everything’, ‘nothing’).
Are there other quantifiers whose meaning can be given by spelling out their
extension? The reply is affirmative as shown in the next section.
4.â•… Generalized quantifiers
Recent accounts define quantifiers in set-theoretic terms. Barwise & Etchemendy

(1989) break with the common treatment of quantifiers (or more generally Â�determiners).
Instead of treating them syncategorematically, i.e. as symbols which do not have a
meaning in isolation, they assign a binary relation between subsets A and B of the
domain of discourse D of the model M under consideration. We find a definition of
the usual quantifiers:
Every = { 〈A,B〉 | A ⊆ B }.
Some = { 〈A,B〉 | A ∩ B ≠ ∅ }
No = { 〈A,B〉 | A ∩ B = ∅ }
176 Paul Gochet
In addition, we find an analogous set-theoretic account of quantifiers which had

always been left out by proponents of the standard treatment:
Most = { 〈A,B〉 | card(A ∩ B) > card(A–B) }
Consider an inference whose premise is of the form ‘all P are Q and S’. We can infer ‘all
P are S’. The inference still holds if we substitute the following determiners for ‘all’: ‘at
least two’, ‘at least three’, ‘most’, ‘many’, ‘both’. But it ceases to hold for ‘few’, ‘no’, ‘neither’,
‘at most two’, ‘exactly two’.
An account of this difference in the logical inferences licensed by the two classes
of modifiers can be given within the framework of model-theoretic semantics. The
difference hinges on a set-theoretic property which Barwise & Etchemendy define in
these terms:
We say that a relation R on subsets of D is monotone (increasing) in its second
coordinate provided that for all A,B⊆D, if 〈A,B〉 ∈R and B⊆B’ then 〈A,B’〉 ∈R’. The
inferences built on the schema mentioned above are valid – i.e. the conclusion is
tied to the premise by a relation of logical consequence – in exactly those cases in
which the determiner is monotone increasing in its second argument.
(Barwise & Etchemendy 1989: 226)
5.â•… The layers of contexts
It is useful to distinguish the situational context which is needed to interpret deictic

terms such as ‘Now’, ‘Here’, ‘I’ etc. from the linguistic context (co-text) of a sentence
which is needed to interpret the anaphoric pronouns. (The standard example is given
by the so-called ‘donkey sentences’ such as ‘Every farmer who owns a donkey beats
it’ which resist regimentation into standard predicate calculus.) Research carried out
within the fields of Artificial Intelligence and computational linguistics have shown
that still another kind of context plays a crucial role in the interpretation of anaphoric
pronouns, namely the commonsense knowledge shared by speakers and hearers.
Consider the following example (borrowed from Kameyama): ‘John hit Bill. He
was severely injured’. Grammatically the subject should be preferred to the object as
an antecedent candidate for ‘he’. Yet 46 speakers of 47 chose ‘Bill’ as antecedent. The
grammatical preference for the grammatical subject was superseded by a piece of
causal knowledge about the world, the knowledge that, as Kameyama puts it, ‘hitting
may cause injury on the hittee but less likely on the hitter’.
This piece of common sense knowledge can in turn be superseded by a new piece
of information about the world. If we learn that Bill is not a normal warm-blooded real
person but is instead the indestructible cyborg called ‘Terminator’ we shall revise our
first choice and interpret ‘he’ as referring to the hitter (John).
It is worth observing that the kind of reasoning at work here rests on a concept
of valid consequence which differs from Tarski’s concept. According to the standard
definition of validity, ‘each model of the premises is a model of the conclusion’; accord-
ing to the concept of validity we need here, “Each most preferred model of the premises
is a model for the conclusion” (Kameyama 1996: 110). Common sense reasoning is
defeasible, reasoning in standard logic is not defeasible.
Salience is still another factor which plays a role in the choice of an antecedent for
an anaphoric pronoun like ‘he’. Consider the following discourse: ‘Babar went to a bak-
ery. The baker greeted him. He pointed to a blueberry pie’. In Kameyana’s experiment,
three speakers said that the pronoun ‘He’ referred to Babar but ten of them said that it
referred to the baker. Hence the more salient singular term was preferred. Sometimes
there is a tie between the pros and the cons. When this happens the discourse is infelici-
tous, to use a word borrowed from Speech Act Theory.
We have described the interplay of various pragmatic constraints which deter-
mine the choice of an antecedent for an anaphoric pronoun. It may also happen
that the choice is made on the basis of conventional presuppositions attached to
lexical items. In this case the choice is indefeasible. The following example brings
that out quite clearly. In ‘John hit Bill, Mary hit him too’, the anaphoric pronoun
must refer to ‘Bill’, the preference for the grammatical subject over the object
notwithstanding.
Kameyama draws a sharp distinction between combinatorics and preferences.
Combinatorics determine all and only possible interpretations. Preferences prioritize
the possibilities. This led him to posit two subsystems: a grammar subsystem which
handles indefeasible possibilities and a pragmatics subsystem which is made up of
(mostly) defeasible preferences.
Kameyama adopts a dynamic approach to semantics. He set up a discourse pro-
cessing architecture which accounts for the incremental disambiguation of underspec-
ified expressions as long as the discourse is not over. In the last section we shall briefly
describe the very elaborate model of context which he proposes.
6.â•… A model-theory for contexts
A model for Context is a 6-tuple 〈φi, Di, Aι, Ii, L, K〉 in which φi stands for the preferred
interpretation of the last utterance. It is a sort of record of the surface structures of the
previous sentences; Di is a set of discourse states that the discourse has been about. It
is made up of successive updates which bring this or that entity into focus; Ii is a set
of indexical anchors which help us interpret deictic terms; Ai is a partial order of enti-
ties and propositions (ordered by salience); L represents linguistic knowledge; and K
represents knowledge about the world.
178 Paul Gochet
While the discourse is in progress these six items change but not at the same
pace. Both linguistic knowledge and knowledge about the world evolve slowly. Mutual
understanding requires that the linguistic competence and the background knowl-
edge of the discourse participants overlap. Psycholinguistic experiments show that the
interpretation of utterances goes faster when it is based on linguistic knowledge than
when it is based on knowledge about the world.
References
Barwise, J. & J. Etchemendy (1989). Model-theoretic semantics. In M.I. Posner (ed.) Foundations of
cognitive science: 207–243. MIT Press.
Bonzon, P., M. Cavalcanti & R. Nossum (2000). Formal Aspects of Context. Kluwer.
Bolzano, B. (1972). [1833] Theory of science. Blackwell.
Buvac, S. & I. Mason (1993). Propositional Logic of Context. Proceedings of the IIth National Confer-
ence on Artificial Intelligence 65: 197–247. AAAI Press.
Cooper, R. (1996). The Role of Situations in Generalized Quantifiers. In S. Lappin (ed.) The
Handbook of Contemporary Semantic Theory: 65–86. Blackwell.
Davidson, D. (1984). Truth and interpretation. Oxford University Press.
Etchemendy, J. (1990). The concept of logical consequence. Harvard University Press.
Hobbs, J. (1978). Resolving Pronoun Reference. Lingua 44: 311–338. [Also in B. Grosz,
K. Spark-Jones & B. Webber (eds.) (1986). Readings in natural language Processing: 339–352.
Morgan Kaufman.]
Garcia-Carpintero, S-M. (1993). The grounds for the model-theoretic account of the logical proper-
ties. Notre Dame Journal of Formal Logic 34: 107–131.
Gardenfors, P. (1988). Generalized Quantifiers, Linguistic and Logical Approaches. Reidel.
Gochet, P. (1980). Outline of a Nominalist Theory of Propositions. Reidel.
Muskens, R. (2005). Sense and the computation of reference. Linguistics and Philosophy 28:
473–504.
Kameyama, M. (1996). Indefeasible Semantics and Defeasible Pragmatics. In M. Kanazawa,
Ch. Pinón & H. de Swart (eds.) Quantifiers, Deduction and Context: 111–138. CSLI.
Kanazawa, M. (1996). Indefeasible Semantics and Defeasible Pragmatics. In M. Kanazawa,
Ch. Pinón & H. de Swart (eds.) Quantifiers, Deduction and Context: 111–138. CSLI.
Kanazawa, M. & Ch. Pinon (eds.) (1994). Dynamics, Polarity and Quantification. CSLI.
Katz, J. (1972). Semantic theory. Harper & Row.
Katz, J. & J. Fodor (1971). [1963] The structure of a semantic theory. In J. Rosenberg & C. Travis
(eds.) Readings in the philosophy of language: 472–514. Prentice Hall.
Keenan, E. (1996). The Semantics of Determiners. In S. Lappin (ed.) The Handbook of Contemporary
Semantic Theory: 41–63. Blackwell.
Keenan, E. & D. Westerstahl (1997). Generalized Quantifiers in Linguistics and Logic.
In J. van Benthem & A. ter Meulen (eds.) Handbook of Logic and Language: 835–893. Elsevier.
Lascarides, A. & N. Ascher (1993). Temporal Interpretation, Discourse relations, and Commonsense
Entailment. Linguistics and Philosophy 16: 437–493.
McCarthy, J. (1993). Notes on Formalizing Context. In R. Bajcsy (ed.) Proceedings of the Â�Thirteenth
International Joint Conference on Artificial Intelligence: 555–562. Morgan Kaufmann.

Quine, W.V.O. (1953). From a logical point of view. Harvard University Press.
Sandu, G. & T. Aho (2009). Logic and Semantics in the Twentieth Century. In L. Haaparanta (ed.)
The Development of Modern Logic. Oxford University Press.
Tarski, A. (1956). Logic, semantics, metamathematics. Oxford University Press.
Wittgenstein, L. (1921). Tractatus logico-philosophicus. Routledge & Kegan Paul.
Charles Morris
Susan Petrilli
University of Bari
1.â•… Morris’s behavioristics and pragmatics
Pragmatics is defined by the American philosopher Charles Morris (Denver, Colorado,

1901–Gainesville, Florida, 1979) as the study of the pragmatical dimension of semiosis,
i.e. of the relations of signs and interpreters. As we shall see below, this dimension is one
of three, the other two being the syntactical and the semantical.
Morris began his studies in engineering, biology, psychology and philosophy.
After having finished his science degree in 1922, he completed a Ph.D. in philosophy
at the University of Chicago in 1925, where he taught from 1931 to 1958. Morris’s pro-
fessional activities included teaching at Harvard University, the New School of Social
Research, Rice University, the University of Chicago, and the University of Florida,
Gainesville where he finished his academic career as Professor Emeritus. Morris had
a strong interest in biology which coincided with the beginning of his studies on signs
or ‘symbolism’, as he said in the 1920s. His Ph.D. dissertation Symbolism and Real-
ity, of 1925 (but only published in 1993 preceded by a German translation in 1981),
includes a chapter entitled ‘Some Psychological and Biological Considerations’. THere-
fore the terms ‘symbolism’ and ‘biology’ appeared very early in his work. Also, in the
preface to his book of 1932, Six Theories of Mind, Morris clearly stated his intention
to develop a general theory of symbolism on the convinction that the mind identifies
with the symbolic process.
In accordance with the project for unified science as ideated by logical Â�positivism,
Morris combined pragmaticism and logical empiricism in a doctrine he named
Â�scientific empiricism (cf. Morris 1937). In this perspective and under the influence of
George H. Mead (1863–1931) who in 1922 had already published the article ‘A BehaÂ�
viorst’s Account of the Significant Symbol’ (Journal of Philosophy), Morris gradually
delineated what he called behavioristics.
Particularly significant for a study of Morris’s relation to American pragmatism
and behaviorism in the years of his intellectual formation is his book of 1932, Six
TheoÂ�ries of Mind. This offers a survey of different theories in American philosophy on
the connection between mind and world, framing Morris’s own relation to American
pragmatism with specific reference to Charles S. Peirce (1839–1914), William James
(1842–1910), John Dewey (1859–1952), and to behaviorism. From this point of view,
another important booklet is Morris’s Logical Positivism, Pragmatism and Â�Scientific
Charles Morris 181
Empiricism (1937). In 1938 he published his groundbreaking contribution to the

Â�science of signs, Foundations of the Theory of Signs, the article ‘Scientific Â�Empiricism’
(both in the International Encyclopedia of Unified Science), and ‘Peirce, Mead and
Pragmatism’ (Philosophical Review) in which he insists on the affinity between Peirce
and Mead, or between the original pragmatism of the former and the more recent
version of the latter. As these publications make evident pragmatism or pragmaticism
also strongly influenced the development of Morris’s semiotics. (‘Semiotic’ in Mor-
ris’s terminology. However, we shall use the term ‘semiotics’, except when citing him
directly). Morris lists the following points in common between Peirce and Mead: the
importance attributed to the theory of signs; the thesis of the inseparability of thought
and semiosis; and of the connection between thought and action; the importance
attributed to finalism, chance and creativity in the mind-world relationship. On the
other hand, the distinguishing element between Peirce and Mead according to Morris
is what he described as Peirce’s metaphysical approach by contrast with Mead’s contex-
tual and situational approach.
Morris continued working on such issues in his volume of 1970, The Pragmatic
Movement in American Philosophy which proposes a global reconsideration of the
development of the basic ideas characterizing pragmatism in the U. S. A., contextual-
ized in historical terms as well. Morris’s own relation to pragmatism also emerges in
retrospect offering us a fuller understanding of his own collocation with respect to
this movement. In a note to the last section in Chapter II on pragmatic semiotics and
its behavioral orientation, he makes the following statement which he had already
repeated on various occasions:
My own work in Signs, Language and Behavior, supplemented by the later study
Signification and Significance, might be regarded as a way to organize and to extend
the contributions to semiotic by the various pragmatist philosophers. It was not done,
however, with any such goal explicitly in mind. My own work started from Mead and
not from Peirce; the influence of Dewey, Lewis, Peirce (and Rudolf Carnap), came
later, and in that order (Morris 1970: 47)
While our special interest in the present article is not Morris’s interpretation of either
Peirce or Mead, it is important to underline that Morris accepted Mead’s behaviorism
(or pragmatism) and recognized many points in common between Mead’s approach to
pragmatism and Peirce’s. Morris distinguished his own specific formulation of behav-
iorism, or behavioristics, from the physicalist thesis of the Unity of Science Movement
preferring a biological framework, and beyond Mead his approach may be associated
with Peircean pragmatism more than with James’s (though he recognized his debt to
the latter in his article ‘William James Today’, 1942a).
The biological basis of behavioristics was fundamental in determining Morris’s
critical stance with respect to the tendency toward dogmatism and reductionism char-
acteristic of the physicalist project. His biological perspective was also important in
182 Susan Petrilli
determining the boundaries of semiosis which he extended and at once circumscribed

to the world of organisms. In fact, Morris’s semiotics offers a general description of
sign as embracing all that belongs to the world of life. His aim was to develop an
approach to semiotics able to deal with all types of signs and to this end, as is partiÂ�
cularly evident in his book of 1946, Signs, Language and Behavior, he chose to con-
struct his terminology within a distinctly biological framework.
Morris referred to semiotics as a ‘science of behavior’, intending a ‘science’, ‘disciÂ�
pline’ or ‘field’ yet to be developed and not the philosophical-psychological trend
known as behaviorism. He pointed out that his version of behaviorism derived mainly
from George H. Mead, Edward Tolman and Clark L. Hull, while the term ‘behaviorÂ�
istics’ for the science he was delineating was adopted from Otto Neurath. Differently
to other behaviorists who applied the tenets of psychology as developed in the study of
rats to the study of men (as one of Morris’s reviewers protested), these scholars worked
at a general theory of behavior, or ‘behavioristics’, says Morris, intended to describe the
behavior of both men and rats, while accounting for their differences.
In his paper of 1948, ‘Signs about Signs about Signs’, Morris explained his relation-
ship with Peirce regarding the behavioral sciences. He declared that his position in
Signs, Language and Behavior, i.e. the idea of studying signs in the context of ‘behav-
ioristics’, did not originate from Peirce but from Mead. Indeed, from this viewpoint
Signs, Language and Behavior may be considered as a development on Mead’s book,
Mind, Self, and Society: “I never heard Mead in lecture or conversation refer to Peirce.
Only later did I work earnestly at Peirce, Ogden and Richards, Russell, and Â�Carnap,
and still later at Tolman and Hull. All of these persons influenced in various ways the
formulation of SLB” (Morris 1971 [1948]: 445). All the same, Morris explained that in
historical perspective Peirce was a modeling influence on Signs, Language and Behav-
ior, for though its orientation did not derive from Peirce, it was in effect “an attempt to
carry out resolutely” his approach to semiotics (ibid.).
The term ‘behavioristics’ had already been introduced in Foundations of the The-
ory of Signs where in fact Morris stated that the question of the sign is framed “from
the point of view of behavioristics”. However, he also added that “this interpretation of
the definition of sign is not… necessary”, explaining that
It is adopted here because such a point of view has in some form or other (though not
in the form of Watsonian behaviorism) become widespread among psychologists, and
because many of the difficulties which the history of semiotic reveals seem to be due
to the fact that through most of its history semiotic linked itself with the faculty and
introspective psychologies (Morris 1971 [1938]: 21)
By comparison with Foundations of the Theory of Signs, Morris further developed and
consolidated the relation between biology, behaviorism and semiotics in Signs, Lan-
guage and Behavior. However his recourse to biology for semiotic terminology did not
imply ‘biologism’ in a negative sense. In other words, there is no tendency in Morris’s
Charles Morris 183
work to reductionism (the temptation of reducing a plurality of universes of discourse

to one only, for the case in point that of biology), the expression of some metaphysical
fancy or of a naturalistic vision of the existent that loses sight of the historico-social
dimension of semiosis. From this point of view, Morris’s attitude was completely dif-
ferent from that of the logical empiricists or neopositivists who developed an explicitly
physicalist orientation in reductionist terms.
Morris developed a pragmatic conception of meaning which led to his focus not
only on signs but also on values. Paths of Life appeared in 1942, The Open Self in 1948.
In these works he focused on the preferential behavior of humankind describing the
‘fundamental choices’ operated in different cultures. Varieties of Human Value was
published in 1956 and collects the results of Morris’s experimental research on values.
In his book of 1964, Signification and Significance. A Study of the Relations of Signs and
Values, he continued his work on values in relation to his studies on signs and consoli-
dated the connection between semiotics and axiology. The English term ‘meaning’ has
a dual accceptation referring not only to the semantic dimension of signs (significa-
tion) but also to the valuative (significance): “that which something signifies and the
value or significance of what is signified” (Morris 1964: vii). Moreover, this book con-
firms Morris’s approach to semiotics as an “interdisciplinary enterprise” (p. 1) focused
on signs in all their forms and manifestations, relatively to human and non-human
animals, normal and pathological signs, linguistic and nonlinguistic signs, personal
and social signs.
In Foundations of the Theory of Signs Morris divides semiotics into the three
branches of syntactics, semantics, and pragmatics, which correspond respectively to
the three dimensions of semiosis, the syntactical, the semantical and the pragmatical.
This trichotomy is fundamentally the result of two main influences: logico-empiricism
and behaviorism on the one hand, and the pragmatic philosophy of Mead and Peirce,
on the other (cf. Morris 1970), which explains why Morris had already understood the
importance of not separating pragmatics from semiotics, nor therefore the pragmati-
cal dimension of semiosis from the syntactical and semantical dimensions, as early as
1938. That meaning is generally associated with the semantical dimension of semio-
sis when, on the contrary, it is present in all three dimensions is the result of a misÂ�
understanding. When Morris claims that syntactics deals with relations among signs
this does not exclude that it involves meaning, which too is part of the relation among
signs. Similarly, as much as pragmatics focuses on the relation of signs to interpreters, as
says Morris, it too deals with signs and therefore with meanings (cf. Rossi-Landi 1994
which includes his paper of 1967, ‘Sul modo in cui è stato fraintesa la semiotica estetica
di Charles Morris’). Morris worked from a semiotical perspective so that his focus was
not on psychological or sociological relations but on semiosical relations considered
from the viewpoint of the interpreter, the interpretant, and of the relations between
interpreters and interpretants. That Morris also focused on the semantical dimension
of semiosis distinguished his own approach to behaviorism from that adopted by the
184 Susan Petrilli
structuralist Leonard Bloomfield (1933, 1974) who, on the contrary, in his effort to
avoid ‘mentalism’ and to keep faith to the behavioristic approach to language was rather
skeptical of semantics. This led to the consequence that semantical issues were long
neglected by American structuralists.
SEMIOSIS
Semantical Dimension
of semiosis
Syntactical Dimension
DESIGNATION
of semiosis
Pragmatical Dimension
DENOTATUM
of semiosis
OTHER SIGN
SIGN VEHICLE
VEHICLES
INTERPRETANT
INTERPRETER
SEMIOTIC
SYNTACTICS SEMANTICS PRAGMATICS
The top part of the diagram represents semiosis, the process by which something func-
tions as a sign. At the bottom ‘semiotic’, the general science of signs, is represented
with its tripartition into the subordinate sciences of syntactics, semantics and prag-
matics. Distinguishing between semiotic and semiosis in Foundations Morris states
that, “semiotic as a science makes use of special signs to state facts about signs; it is a
language to talk about signs” (Morris 1971 [1938]: 23), and indeed one of the primary
tasks he set himself was to establish a sign system to talk about signs.
This diagram shows how the sign-vehicle, the object that functions as a sign,
relates to a designatum and eventually a denotatum. This relation concerns the seman-
tical dimension of semiosis. However, the sign is also the relation to an interpreter
which in response to the sign produces an interpretant. This is the pragmatical dimen-
sion of semiosis. Moreover, the sign must necessarily relate to other sign-vehicles, this
being the syntactical dimesion of semiosis. The sign involves all three dimensions of
semiosis, always. And, indeed, only for the sake of analysis is it possible to distin-
guish between the relation of the sign-vehicle to the designatum (and eventually the
Charles Morris 185
Â�
denotatum), the relation of the sign-vehicle to other sign-vehicles, and the relation of
the sign-vehicle to the interpreter which is such only if endowed with an interpretant.
Consequently, as stated above, the science of signs with its three branches, semantics,
syntactics and pragmatics, focuses on semiosis and its three dimensions, the semanÂ�
tical, syntactical and pragmatical.
To reduce meaning solely to the semantical dimension instead of tracing it through-
out all three dimensions of semiosis is to reduce the sign totality to one of its parts only,
in the case of semantics to the relation of designation and denotation. Similarly, the
relation of the sign to other signs does not only concern the syntactical dimension in a
strict sense to the exclusion of the pragmatical and the semantical, just as the relation of
the interpreter to other interpreters does not uniquely concern the pragmatical dimen-
sion to the exclusion of the syntactical and the semantical. Each time there is semiosis
and, therefore, a sign, all three dimensions are involved and are the object of semiotics.
In Foundations, Morris establishes a correspondence between the three branches
of semiotics and three philosophical orientations: these are ‘formalism’ or ‘symbolic
logic’ which is made to relate to syntactics, empirism which is related to semantics,
and pragmatism to pragmatics. Indeed, Morris, citing Peirce, James, Dewey and Mead,
has ‘pragmatics’ derive specifically from ‘pragmatism’. And, in fact, Chapter V, entitled
‘Pragmatics’, opens with the following statement:
The term ‘pragmatics’ has obviously been coined with reference to the term ‘pragmatism’.
It is a plausible view that the permanent significance of pragmatism lies in the fact that
it has directed attention more closely to the relation of signs to their users than had
previously been done and has assessed more profoundly than ever before the relevance
of this relation in understanding intellectual activities. The term ‘pragmatics’ helps
to signalize the significance of the achievements of Peirce, James, Dewey, and Mead
within the field of semiotic. At the same time, ‘pragmatics’ as a specifically semiotic
term must receive its own formulation. By ‘pragmatics’ is designated the science of the
relation of signs to their interpreters. ‘Pragmatics’ must then be distinguished from
‘pragmatism’, and ‘pragmatical’ from ‘pragmatic’ (1971 [1938]: 43)
2.â•… M
orris’s pragmatics and Peirce’s pragmaticism:
Towards a ‘behavioral semiotic’
Morris may be considered as one of the main continuators of pragmatism (after

Dewey, Mead and Clarence I. Lewis [1883–1964]), understood as a set of doctrines
and methods originally elaborated by Peirce and James. A second generation prag-
matist, Morris recovered and relaunched this particular philosophical trend at a time
when neoÂ�positivism was dominating in America, a movement to which he himself
contributed with his participation in the Unity of Science Movement.
186 Susan Petrilli
Pragmatic philosophy is native to the United States and was officially introduced
in a public lecture entitled ‘Philosophical Conceptions and Practical Results’ delivered
by James in 1898. However, the originator of American pragmatism was Peirce. In his
search for the origins he considered Nicholas St. John Green the ‘grandfather’ of this
movement, while the latter evoked the Scotchman, Alexander Bain, author of Emotions
and Will (1859), urging the importance of applying his definition of belief as ‘that upon
which a man is prepared to act’ (CP 5.12 – CP refers to Peirce’s Collected Papers, cf.
Peirce 1931–1958). Peirce published an original nucleus of three writings on the topic
of pragmatism in 1868 in Journal of Speculative Philosophy (cf. CP 5.213–263, 264–317,
318–357), subsequently developed in a series of six articles published in Popular Science
Monthly between 1877–78 under the general title ‘Illustrations of the Logic of Science’
(cf. CP 5.358–387, 5.388–410, 2.645–660, 2. 669–693, 6. 395–427, 2.619–644). Peirce
himself dated the birth of pragmatism back to the meetings held by the ‘Metaphysical
Club’ in his study, or James’s, between the end of 1871 and the beginning of 1872 in
Cambridge (Mass.).
Pragmatism re-evaluated the importance of action in cognitive processes keeping
account of discoveries in various fields, above all psychology, sociology, and biology
with particular reference to the work of Charles Darwin (1809–1882). The influence
of Darwinian biology is obvious in Peirce’s essay of 1877 ‘Fixation of Belief ’, where he
states that logicality in regard to practical matters may result from the action of natu-
ral selection (cf. CP 5.366). Peirce was trained and worked as a physical scientist and
privileged the scientific experimental method which he employed and developed in
his research. Pragmatism for Peirce, says Morris (cf. 1970: 20), is essentially the pro-
posal to adopt the pragmatic maxim in philosophy, variously formulated throughout
his writings (cf. CP 5.402, 5.9, 5.412), so that philosophy may gain the progressive and
cumulative character of empirical science.
In the pragmatic perspective, mind (or spirit or thought) is not a substance as in
Cartesian dualism, nor a process or an act as understood by idealism, nor a set of rela-
tions as in classical empiricism, but a function or activity exercised by verbal and non-
verbal signs. Therefore, says Morris in Six Theories of Mind, given that mind and sign
or symbolic processes are interrelated and even identify with each other, a theory of
signs and verbal language is necessary for an adequate understanding of the Â�workings
of the mind.
More than as a comprehensive philosophy, at its origin pragmatism emerges as a
method to ascertain ‘How to Make our Ideas Clear’ (Peirce in a paper of 1878). Here
he describes meaning in terms of the practical verifiability of the truth of an asser-
tion, arising from our conception of the sensible effects of things. For a clear under-
standing of the object of our conception we must have an understanding of its effects
and eventual practical bearings (cf. CP 5. 401–402). James developed this aspect of
the theory of meaning into a theory of truth. He interprets pragmatism in terms of
Charles Morris 187
instrumentality and has knowledge depend on the needs of action and emotions.
For James, that which has satisfying practical consequences is true, which leads him
to emphasize the practical value of religious faith, the will to believe, the reasons of
the heart (cf. The Will to Believe, 1897, and Pragmatism, 1907, in James 1975–1988).
Ferdinand C. S. Schiller (1864–1937) (cf. 1907) oriented his own approach in James’s
direction asserting the relativity of knowledge to personal or social utility. Dewey
also worked the problem of meaning and knowledge into his own version of pragma-
tism, which he denominated ‘experimentalism’ or ‘instrumentalism’. In Italy, pragma-
tism was developed along Peircean lines by Giovanni Vailati (1863–1909) and Mario
Calderoni (1879–1914), and along Jamesian lines by Giovanni Papini (1881–1956)
and Giuseppe Prezzolini (1882–1982).
Peirce returned to the question of pragmatism in a set of seven lectures Â�delivered
at Harvard, by initiative of James (cf. CP 5.14–40, 5.41–65, 5.66–92, 5.93–119,
5.120–150, 5.151–179, 5.180–212). In these lectures he identified pragmatism with the
logic of abduction and the theory of inquiry and implicitly, therefore, with logic and
semiotics.
In his Monist articles of 1905 (CP 5.411–437, 5.438–463, 4.530–572), Peirce dis-
tanced himself from pragmatism as conceived by James and Schiller, and coined the
substitute term ‘pragmaticism’ for his own position (CP 5.414–415). He rejected the
idea of ‘Doing’ as ‘the Be-all and the End-all of life’ (CP 5. 429). Unlike vulgar pragma-
tism, meaning is conceived in terms of a general law of conduct independently from
the particular circumstances of action. Peirce worked at a general theory of meaning.
In the Appendix to Signs, Language and Behavior a paragraph is dedicated to
Peirce’s contribution to semiotics (1971 [1946]: 337–340). Morris observes that Peirce
connected sign-processes with processes involving mediation or ‘thirdness’, indeed
equated them, and established a relation of identification between sign processes and
mental processes. From this point of view there is an obvious discrepancy between
Peirce’s triadic approach and the approach adopted by behaviorism when based on
two-term relations between stimuli and responses, unless the stimulus-response rela-
tion is recognized as being mediated by a third factor with a ‘reinforcing’ function.
On the one hand, Morris searched for sign-processes in the wide class of media-
tion processes; on the other, he was convinced that mediation processes did not
always involve signs. He formulated this restriction on the basis of the claim that sign-Â�
processes are mediation processes in which the mediation factor is an interpretant,
and the interpretant is not always present. It is our belief that this so-called ‘restriction’
or, better, ‘specification’, is in line with the Peircean perspective (cf. Ibid.: 338).
Another aspect of Peirce’s semiotics evidenced by Morris concerns the fact that,
as anticipated, sign-processes and mental processes identify with each other, therefore,
sign-processes are not restricted to behavioral or actional situations alone. Â�Furthermore,
it must also be remembered that for Peirce, as for example in ‘Prolegomena to an
188 Susan Petrilli
Apology for Pragmaticism’ (1906), while mental processes and semiosis are described
as coinciding, “thought is not necessarily connected with a brain. It appears in the work
of bees, of crystals, and throughout the purely physical world” (CP 4.551).
Peirce’s definition of sign in terms of mind was considered as a limit by Morris,
an example of ‘idealistic metaphysics’. In a manuscript of 1903, Peirce says “A sign is a
representamen of which some interpretant is a cognition of a mind” (CP 2.242). But
Peirce did not provide a criterion for determining when the mind or thought inter-
venes, being the condition for determining when something is or is not a sign. And
while in ‘Questions Concerning Certain Faculties Claimed for Man’ of 1868, Peirce
stated that “every thought is a sign” (CP 5.253), so that thoughts and signs identify to
the extent that thought is not possible if not in signs, in ‘Some Consequences of Four
Incapacities’ (published in the same year), he stated that thinking never occurs with-
out the presence of something which acts as a sign, where signs instead emerge as an
instrument of thought (cf. CP 5.283).
Lastly, Morris significantly rejected Peirce’s definition of the sign, of all signs, as
giving rise to other signs. That signs are continuously translated or interpreted in a
subsequent sign cannot be used as a defining criterion, for this involves a form of
circularity wherewith a sign is defined as something which generates a sign which
generates another sign, etc. Morris’s comment is the following: “Signs, at least at the
human level, do frequently generate a series of sign-processes, but I see no reason why
this fact about signs should be incorporated into the definition of ‘sign’ itself ” (Morris
1971 [1946]: 339).
The aspect Morris found most interesting about Peirce’s work (in spite of what
he believed were his mentalistic limitations) was the latter’s emphasis on behavior.
Peirce maintained that to determine the meaning of a sign we must identify the habits
of behavior it produces, which in fact resounds in Morris’s own orientation. In Morris’s
view, Peirce had the merit of rejecting old Cartesian mentalism and replacing it with
the concept of habits of behavior and, therefore, of directing semiotics toward a more
adequate account of sign-processes.
Another merit according to Morris is Peirce’s refusal to establish a distinct sepaÂ�
ration between human sign processes and nonhuman animal sign processes.
Morris drew two fundamental lessons from Peirce’s studies on signs: firstly, it is
not necessary for a theory to belong to some psychological or philosophical behaÂ�
vioral trend to achieve important results in semiotics; secondly, semiotics must not
be grounded in a mentalistic order to achieve the status of science, but rather in behaÂ�
vioristics understood as a general theory of behavior, including sign behavior.
Morris’s orientation as regards behavioristics and biology emerges clearly through
confrontation with Peirce. Peirce dealt with problems which could not be ignored in
semiotics and underlined the importance of terminological issues for Â�methodoÂ�logy.
Morris too was critical of what he called uncertain and ambiguous terms (such as
Charles Morris 189
the mentalistic) and constructed his own terminology from a science that stud-
ies the behavior of human and nonhuman animal life objectively, that is, biology.
Â�Consequently, he returned to Peirce with the intention of moving beyond him.
Morris rejected critics like Bentley who did not appreciate his efforts to continue
and develop Peirce’s ideas. Bentley analysed Morris’s relationship to Peirce through
Dewey’s critical reading of Foundations of the Theory of Signs. As reported by Â�Morris
in his 1948 paper ‘Signs about Signs about Signs’, Dewey accused him of having
“inverted Peirce’s position” as regards the terms ‘interpreter’ and ‘interpretant’ (1971
[1948]: 445), to which Morris replied, as reported above, that though the orientation of
Signs, Language and Behavior did not derive from Peirce, the position expressed in it is
in effect an attempt to carry out resolutely the latter’s approach to semiotics.
On replying to Bentley, Morris was also replying to Dewey who accused him of
falsifying Peirce. According to Morris, Dewey did not understand the connection
established by Peirce between the concepts of ‘interpreter’ and ‘interpretant’. Dewey
considered the sign-interpretant relation as a relation in a sign system prescinding
from the sign-interpretation relation and, therefore, from the role of the interpreter
(an organism for Morris) when something functions as a sign. There is no sign without
an interpretant or an interpreter, for the interpretant is the effect of a sign on an inter-
preter. Indeed, given that the interpreter does not subsist as such if not as a modifica-
tion ensuing from the effect of a sign in an open chain of interpretants, the interpreter
is also an interpretant and, therefore, a sign. The correspondence between man and
sign, interpreter and interpretant is explained by Peirce in ‘Some Consequences of
Four Incapacities’, and does not imply that one of the two concepts forming these pairs
can be eliminated for each term evidences different aspects of semiosis.
Another important observation made by Morris is that Peirce used the term
‘interpretant’ with different meanings: in fact, he distinguished between ‘immediate
interpretant’, ‘dynamical interpretant’ and ‘logical final interpretant’. Dewey’s critique
of Morris was determined by his misunderstanding of the different ways Peirce’s inter-
pretant may be understood, despite Morris’s attempt to reduce ambiguity by introduc-
ing the term significatum (designatum in Foundations) alongside it. Dewey employed
the term ‘interpretant’ with the same meaning Morris attributed to significatum, with-
out realizing that for the latter ‘interpretant’ indicated the effect of a sign on an inter-
preter. Morris also pointed out that Dewey himself in other contexts in fact underlined
this aspect of the concept of sign, and cited as proof a passage from Logic where Dewey
speaks of a preparatory disposition to act in a certain way in relation to the sign.
Morris’s discussion of the theses advanced by Dewey and Bentley concerning his
relation to Peirce indicates just how important this relation was to him. The relation
between Peircean and Morrisian semiotics is of continuity such that we may hypoÂ�
thesize that Morris chose a biological basis as a means of reinforcing such Â�continuity.
Apart from Morris’s critique of the mentalistic and metaphysical aspects of Peirce’s
190 Susan Petrilli
work, there are no substantial differences in the general orientation of Morris’s

Â�behavioristics and Peirce’s pragmaticism.
Although Morris was in accord with Peirce’s description of semiosis as the “action
of a sign” (CP 5.473), he limited his own concept of action to behavior directed toward
a goal. This approach implies the presence of a subject, an organism, its goal-oriented
behavior and a goal-object, that is, a ‘final object’ outside semiosis.
In Morris’s example (cf. 1971 [1946]: 83), the sound of a buzzer is a sign for a dog
trained to seek food when the buzzer has sounded or some time after. In this situation
semiosis is the action of the buzzer which provokes the action of seeking food, or, as
in the case of Pavlov’s dog, which acts on its glands causing it to salivate. But for this to
occur there must be a preceding semiosis, that through which the dog interprets a cer-
tain noise as the sound of a buzzer.This ‘simple’ perception is already an interpretation
in itself and, therefore, semiosis. A sound, a noise hits the ear in a certain way and the
first action it provokes is the interpretation which identifies that noise as a particular
sound connected with the need for food. The first inference: the noise interpreted as
the buzzer sound, or as ‘that certain sound’; the second: ‘therefore there is food’; only
the third, the reaction that consists in directing attention to seeking food is finalized
or goal-seeking behavior.
Certain ants which nourish themselves with droplets of honeydew secreted by
aphids (myrmecophilous) on vibrating their antennas against the rear end of the latter,
can actually begin ‘milking’ the aphids (Linneus called them the ant’s cows) thanks to
a (mistaken) identification of a sign (cf. Sebeok 1979: 13; 1986: 144). The ant exchanges
the hind end of an aphid’s abdomen which acts as an icon (or better effigy), for the
head of another ant, which sparks off behavior involving the intraspecific sharing of
food between giver and accepter: this leads to the secretion of food on the basis of a
misunderstanding.
In our terminology, the action of a sign must first provoke a response or interÂ�
pretation that identifies a given sign and only subsequently respond to it through goal-
seeking behavior. For a person another person’s wet raincoat may be a sign. Semiosis
occurs on the basis of an inferential interpretation: ‘therefore, it’s raining’; which pre-
cedes the response of picking up an umbrella to attain the goal of leaving the house
without getting wet.
Morris’s second example: a driver must accomplish a series of previous semioses
before being able to complete his goal-seeking behavior directed toward arriving at a
given town; before responding to the information that the road is blocked and, there-
fore, turning off on a side-road. In other words, before reaching the decision to deviate
his course, thereby enacting a semiosic chain with which he reacts to the informer’s
words as though he had actually encountered the obstacle blocking his route, the driver
must first accomplish the following semioses: interpret the vocal sounds emitted by the
informer as words of a language he recognizes given that he identifies its phonemes and
Charles Morris 191
monemes; formulate the conclusion that the informer is informing him to the effect
that “the road is blocked”; that this is exactly what the informer is saying with reference
to the route the driver is moving along, and not something else; that the informer is
trustworthy so that what he says may be interpreted as being true (which involves other
semioses and other inferences); and even before stopping to listen to the informer,
the driver must accomplish such interpretations as the following: ‘this is a person’, ‘his
Â�gestures mean “stop”’, ‘his stop signs are turned to the driver in question’, etc.
We may conclude that as ‘sign activity’, semiosis already occurs when we perceive a
stimulus and identify it (which involves the risk of error): perception and identification
are reactions to the sign, therefore, they are interpretants (identification interpretants)
which precede interpretation of that sign at the level of ‘answering comprehension’ –
according to our examples, the action of food-seeking or turning off into another road.
Morris forestalls the objection that his definition of action as goal-oriented behav-
ior presupposes the subject, its behavior, the goal-object, etc. outside semiosis, by
maintaining that ‘perception’ is a very vague and ambiguous term: it precedes sign
processes just as it may be considered a sign process itself. Consequently, a definition
of sign (similarly to all semiotic terminology) would not be well founded on a theory
of perception. Furthermore, not all response-sequences provoked by a stimulus-object
constitute sign behavior: “A person reaching for a glass of water is not prepared to act
in a certain way because of a sign, but simply is acting in a certain way to an object
[…]. The glass of water may of course itself become a sign”, for example, “a sign of a
certain person’s kindness” (Morris 1971 [1946]: 135).
No doubt we may agree with Morris if his observations are understood in the
sense that any behavior, including perceptive behavior, may be sign behavior, as he
admitted in Foundations of the Theory of Signs where he clearly stated that nothing is
a sign in itself but can become a sign in given processes and on given conditions. But
in Signs, Language, and Behavior Morris maintained the opposite: “the semiotician,
however, must avoid making all response-sequences cases of sign-behavior if he seeks
to formulate a behavioral criterion for signs in terms of response-sequences” (Ibid.:
135). And we must also observe that Morris is referring both to sign behavior as well
as to non-sign behavior, to ‘responses’, forgetting the (questionable) distinction which
he had established himself earlier between ‘reactions’ and ‘responses’. On our part, we
believe that the action of reaching for a glass of water is the consequence of respond-
ing to a sign, or, better, to an array of signs of a cultural order, as well as the perceptive
interpretation of the object in question.
According to Morris the object acting as a sign is already given as a sign to a
subject who must simply respond to it. According to Peirce, the object is a ‘dynam-
ical object’, something that provokes a process through which the object gradually
manifests itself as an object, beginning from its initial manifestation as a ‘ground’. The
Â�latter is completely undetermined and as such susceptible to multiple determinations;
192 Susan Petrilli
it is ambiguous to the point of requiring an initial more or less elaborate semiosis of

disambiguation.
We have not yet said anything about goal-oriented behavior, about having a need
and identifying an object able to satisfy it, which also involves preliminary semioses:
the driver interprets his desire, obligation or convenience to direct himself toward a
given destination, decides that it is best to go by car, that a given route and not another
is the best one to follow, etc. Also, for the dog to seek food it must feel ‘the signs of
hunger’. How such signs are produced is a question that leads back to the ‘biologi-
cal basis’ of semiosis where certain parts of the organism behave as the interpreters,
through certain interpretants, of certain sign interpreteds at a level which is no longer
the macrobiological.
Hunger is a subjective sensation that signals the request for food by the organism
when its tissues are impoverished of nutritional substances. More precisely: adeno-
sinemonophosphate, or cyclic AMP omnipresent in biological regulation in most
organisms, signals a nutritional crisis, in other words, that the carbon-source has been
exhausted. Thomas A. Sebeok, who under certain aspects may be considered as one of
Morris’s most original successors, underlines the double function of AMP, that is, the
endocrine and endosemiotic:
[…] and cyclic AMP [may be] taken as a sign significantly substituting for something
else – aliquid stat pro aliquo. There is a genuine triadic relation here involving
mediation, not reducible to a combination of dyadic relations between pairs of objects,
in a word, semiosis (Sebeok 1979: 23)
3.â•… Pragmatic philosophy in the United States
Morris’s book of 1970, The Pragmatic Movement in American Philosophy, is interest-

ing for at least two reasons: firstly, it presents a survey of the basic ideas informing the
development of pragmatism in the U. S. A. and does so in historical perspective, focus-
ing primarily on the thoughts of four American philosophers who form the original
nucleus and propelling force of this movement – Peirce, James, Dewey, and Mead;
secondly, by implication this study presents an overview, contextualization and inter-
pretive key for a retrospective reading of the development of Morris’s own work as a
semiotician and philosopher relating to this particular movement.
In line with his predecessors and coherently with his own method of inquiry, in
his analysis of the factors occasioning the development of American pragmatic philoÂ�
sophy Morris distinguished between the problematic and the unproblematic. All prob-
lems occur in a context where certain features – objects, meanings and beliefs – are
unproblematic, therefore taken for granted in the attempt to solve the problem. What
is problematic in one context may be unproblematic in another context and viceversa.
Charles Morris 193
Morris had already used this method to develop a general theory of signs, signs to
talk about signs, when he distinguished between analyzed signs, the object of study,
and unanalyzed signs, the unproblematized terminological apparatus drawn from the
language of biology, used to study that object (cf. 1971 [1948]: 435). In his book of
1970 he identified four main factors in American culture considered as unproblematic
by the pragmatists, which provided the occasion for the development of pragmatic
philosophy: (1) scientific method; (2) philosophical empiricism; (3) evolutionary bioÂ�
logy; (4) the humanist democratic ideal (for the pragmatists freedom and democracy
were subject to scientific inquiry). These four themes form the context and framework
for the solution to philosophical problems. They combine differently to influence all
the major pragmatists in varying degrees, and account for both the unity and diver-
sity characterizing the American pragmatic movement: e pluribus unum is the motto
for the American nation which Morris evokes and transfers to its most characteristic
philosophy. In fact, the American pragmatists are distinguished by important differ-
ences in personality, in characteristic problems treated, and even in proposed solu-
tions, while they unite to form a single and unique philosophical movement. Common
orientations which characterize pragmatic philosophy as summarized by Morris, and
which he developed throughout his own research include: behavioral semiotics; semi-
otically interpreted logic; epistemology understood as the theory of inquiry; axiology
conceived as the study of preferential behavior (that is, of prizing and appraising); a
view of experience as an integral part of the cosmos; and a semiosical theory of mind
(cf. Morris 1970: 142–143).
Morris made a point of underlining the distinction between pragmatism, the
practical and action. Peirce developed the term ‘pragmatism’ from Kant’s use of
the term ‘pragmatisch’ in Critique of Pure Reason to express a ‘relation to some
definite purpose’. For Peirce pragmatism is concerned with the way ‘rational cog-
nition’, or, knowledge, relates to ‘rational purpose’, or, human action or conduct
(cf. CP 5.412). It is neither concerned with ‘practicalism’, that is, with the prac-
tical, nor with different types of practices. In Morris’s description pragmatism is
focused primaÂ�rily upon one aspect of human behavior: “intelligent action, that is,
purposive or goal-seeking behavior as influenced by reflection” (1970: 10). This is
the aspect he most developed in Signs, Language and Behavior which is specifically
dedicated to the inextricable interrelation between signs and human conduct. For
Peirce rational purpose is “self-controlled conduct”, that is, “conduct controlled by
adequate deliberation”. Pragmatism, or as Peirce preferred, pragmaticism, was con-
cerned then with the relation between “deliberate conduct” and “the intellectual
purport of symbols” (cf. CP 8.322, 5.442). Therefore, intelligent, purposive behav-
ior, a distinctive and important part of human behavior, is the aspect privileged by
American pragmatism, which makes it unique among other modern philosophies.
In Morris’s view the focus on intelligent purposive behavior explained various other
194 Susan Petrilli
Â�
characteristics of the pragmatic movement including the stress on the type of signs
that occur in reflective inquiry.
Following a tradition as old as the Stoics, Morris embraced a broad conception of
philosophy as including consideration of the methods of inquiry (methodology, com-
prising theory of signs), a doctrine of the nature of value, such as ethical and aesthetical
values (axiology), and a picture of man and the world (cosmology). He structured this
book of 1970 so as to demonstrate the philosophical status of the pragmatic movement
in America.
The distinctive trait of the pragmatists’ inquiry into the nature of meaning is the
belief that there is an intrinsic connection between meaning and action. If it is granted
that meaning comes in signs, then the connection between meaning and action (or
behavior) implies that semiotics, the general study of signs, should be developed as
an actional or behavioral theory. However, Morris believed that the American pragÂ�
matists did not work out a comprehensive theory of meaning, and that not even Peirce
proposed a comprehensive formulation of the nature of meaning and of the intrinsic
relation of meaning to action. Consequently, on Morris’s account, no fully worked
out behaviorally-oriented semiotics was ever developed by the pragmatists. We might
comment that Morris’s own research, particularly as formulated in Signs, Language
and Behavior and in Signification and Significance, is an attempt at fulfilling this pre-
cise goal (even if he does not say so explicitly), consequently bringing American
pragÂ�matism to its full development. His description of the pragmatic movement in
Â�America, as in the following statement which expresses the situation in a nutshell, may
in fact be applied to the general orientation of his own research:
A behavioral semiotic might then be considered as the foundation of pragmatism,

and pragmatism regarded as a philosophy that attempts to deal with the traditional
problems of philosophy on this foundation. In this way some precision could be given
to the term ‘pragmatism’, namely, pragmatism would be philosphy erected upon the
basis of a behavioral semiotic (Morris 1970: 16–17)
Morris dedicated a consistent part of his research and writings to behavioral semi-
otics, therefore to the study of man’s intelligent, purposive or goal-oriented behavior,
committed to controlling his future in the direction of his values. However, Morris
wished to reach a comprehensive understanding of man and of his relations to the
world, of man in the totality of his interests and actions, as in other occasions we have
also stated of Peirce, so that while he analysed the signs of knowledge and of reflective
inquiry, the signs of rational thought and behavior, he also emphasized the importance
of nonrational, even antirational factors of behavior. In the context of his value theory
(axiology), largely developed as a theory of preferential behavior, Morris evidenced
other types and value-orientations, for example, the Buddhist personality ideal, beyond
Charles Morris 195
the ‘Promethean’ or ‘pioneer’ type of personality favored by the pragmatists – under this
aspect remaining closer to Peirce and James than to Dewey and Mead.
Morris aimed at developing a global approach to man and his signs which led
to his interest for all types of signs – beyond the rational and goal-seeking, signs of
the irrational, of mysticism, man-cosmos symbols, malfunctioning signs, pathologi-
cal signs, the signs of social alienation, of mental illness. And, indeed, in addition to
the books and articles discussed in this paper, still others – including, Paths of Life
(1942), The Open Self (1948), as well as his poetry, or what Sebeok calls his ‘wisdom
literature’, collected in the volumes Festival (1966) and Image (1976) – evidence the
comprehensiveness of Morris’s behavioral semiotics as developed in the light of his
pragmatic philosophy.
4.â•… Pragmatics, signs and values
Morris defined pragmatics as the study of the relations of sign vehicles to interpreters
or more simply as “the relations of signs to their users” (1938c). Unlike Rudolf Â�Carnap
(1939) who restricted the field of pragmatics to verbal signs only to include nonlin-
guistic signs much later (1955), Morris’s conception of pragmatics concerns both ver-
bal and nonverbal signs. John L. Austin (1962) and John Searle (1969) also limited
their interest in the pragmatical dimension to verbal signs. On the contrary, Morris
extended it to include the ethic and aesthetic dimensions as well. As we have men-
tioned above, Morris’s interest in the relation of signs to values is closely connected
with Â�pragmatics as the science of the study of the relation of signs to their interpreters.
In Signification and Significance (1964), Morris related sign theory systematically
to value theory wherewith he consolidated the close connection between semiotics
and axiology. He underlined the vague and ambiguous nature of the concept of mean-
ing commonly used to cover (at least) intention, signification, and value, and main-
tained that the nature of meaning could only be adequately understood in the light of
a theory of semiotics.
With reference to the relation between sign theory and value theory we may dis-
tinguish between so-called ‘interpretation semiotics’ (a trend counting such exponents
as Peirce, Welby, Morris, Bakhtin, Rossi-Landi) and ‘identification semiotics’ or ‘equal
exchange semiotics’.
With reference to decodification semiotics, Rossi-Landi evidenced the relation
between sign model (in this case, of Saussurean derivation) and economic value the-
ory. The connection between linguistics and economics, or semiotics and economics,
promised a fuller understanding of the disciplines in question as well as of their inter-
relations (cf. Rossi-Landi 1975b). Saussure referred specifically to economic value as
196 Susan Petrilli
conceived by Marginalistic Political Economy of the School of Lausanne and, therefore,

to equal exchange value. On the contrary, in the light of historico-dialectical material-
ism and the Marxian critique of exchange value, associated with semiotical categories
developed from Peirce, Vološinov(-Bakhtin) (1929) and Morris, Rossi-Landi (1961,
1968) criticized the Saussurean approach and, therefore, the limits of linguistic theory
based on equal exchange value theory.
In the introduction to his 1954 Italian translation of Morris’s book of 1938, Rossi-
Landi stated that after Foundations, Morris’s research developed in two different direc-
tions. One consisted in elaborating the notion of sign and a general sign theory; the
other dealt with the problem of value (cf. Morris 1938c, It. trans. 1999: 62). In Signi-
fication and Significance, Morris at last united these two areas of his research: he had
worked on values almost as much as he had worked on signs, and rejected the idea that
the mere fact of working on signs gave one the right to judge about values. And, in fact,
he had dedicated a large part of his work specifically to value theory, in particular to
the problem of ethical and aesthetic value judgments, as in his book of 1956, Varieties
of Human Value.
The opening paragraph in Signification and Significance analyses the various
senses in which the expression ‘to have meaning’ may be understood: that is, as having
a given linguistic meaning, a given signification, semantic meaning, on the one hand,
and as having value, of being significant, evaluative meaning, on the other. The term
‘meaning’ is doubled into ‘signification’, the object of semiotics, and ‘significance’, the
object of axiology. To consider signs and values together meant to deal with the prob-
lem of the relation of semiotics to axiology insofar as they concern different aspects
of the same process, human behavior. And by evidencing the ambiguous nature of the
term ‘meaning’, Morris’s work testifies to the semiotic consistency of human signifying
and behavioral processes:
That there are close relations between the terms ‘signification’ and ‘significance’ is
evident. In many languages there is a term like the English term ‘meaning’ which
has two poles: that which something signifies and the values or significance of what
is signified. Thus if we ask what the meaning of life is, we may be asking a question
about the signification of the term ‘life’, or about the value or significance of living or
both. The fact that such terms as ‘meaning’ are so widespread in many languages (with
the polarity mentioned) suggests that there is a basic relation between what we shall
distinguish as signification and significance (Morris 1964: vii)
Morris believed that the problem of the relation between signs and values, semiotics
and axiology had not yet received adequate attention, and observed that Peirce too was
concerned with semiotics more in its logico-cognitive aspects than in relation to axi-
ological issues. In truth, Peirce’s writings (cf. 1923) did not fail to deal with the prob-
lem of values. In line with pragmaticism, his cognitive semiotics cannot be separated
from man’s social behavior and from the totality of his interests. For Peirce the prob-
lem of knowledge necessarily involved orientations and issues of an evaluative order,
Charles Morris 197
and consequently his so-called ‘cognitive semiotics’ implies the ethical-pragmatic or

valuative-operative dimension of signs.
As a global science of signs, semiotics must focus upon all aspects of semiosis
without ignoring what is understood by the terms ‘signification’ and ‘significance’.
We suggest the term ‘semioethics’ (which replaces our previous proposal of ‘teleose-
miotics’, cf. Petrilli 1998b) for that tendency to push beyond the neutral descriptive
and cognitive limits of the science of signs, though the term ‘semiotics’ is adequate if
understood as the general study of signs. In relation to human signs, semioethics con-
tributes to a global understanding of human beings in the entirety of their relations to
themselves, to the world, and to others. Therefore, with respect to ‘semiotics’, the term
‘semioÂ�ethics’ is intended to highlight that particular orientation in sign studies that is
not purely descriptive, that does not claim to be neutral. From the viewpoint of the-
ory of knowledge, semioethics evidences the axiological dimension of sign processes
beyond, or, better, in conjunction with the strictly logical-cognitive aspects.
From the viewpoint of the relation between signs and values, other authors in addi-
tion to Peirce (Welby, Ogden, Richards, Bakhtin, Vailati, Rossi-Landi) may be read in a
similar key. Victoria Welby (1837–1912) oriented her research on signs and meaning so
strongly in the direction of value theory and the problem of significance that she coined
the neologism ‘significs’, which she preferred to both ‘semantics’ and ‘semiotics’.
But to return to Morris: how did he connect the threads of his two lines of research
on signs and values, more commonly divided between the competencies of semioti-
cians and of philosophers? Signification and Significance introduces a few terminoÂ�
logical innovations regarding the components of semiosis:
–â•fi Sign or sign vehicle, the object acting as a stimulus for sign behavior;
–â•fi Interpreter, any organism acted upon by the sign vehicle. This extension of the
concept of interpreter to include any organism whatever, and, therefore, any type
of sign behavior beyond the human, implies extending semiotics beyond the
social behavior of man and, therefore, beyond the limits established by Saussurean
sémiologie. This orientation in semiotic studies is developed specially by Sebeok
the ideator of ‘zoosemiotics’, ‘biosemiotics’ and ‘global semiotics’.
–â•fi Interpretant, the disposition to respond to a certain type of object as the result of
a sign stimulus.
–â•fi Signification, the object to which the interpreter responds through an interpretant,
that is, the signified object which as such cannot function simultaneously as stimu-
lus. Here, signification replaces what Morris variously called designatum (1938c)
and significatum (1946), while the concepts of interpreter and interpretant remain
constant. That the object of signification cannot function as a stimulus does not
mean that what gives itself to direct experience cannot be signified. The point is
that only a part of an object can be perceived directly; and this is the part that func-
tions as the stimulus or sign vehicle. Instead, the part not fully perceived functions
198 Susan Petrilli
as the signified object, the object of signification. We say that ‘this is a desk’ on the
basis of our limited experience of the object in question, that part which is per-
ceived directly and interpreted as a sign of the fact that we are dealing with a desk
on the basis of the hypothesis (implying the risk of error) that there exist parts we
do not actually see – the back of the desk, its underside, the drawers, etc.
–â•fi Context, the set of circumstances in which semiosis takes place.
Another important specification in this section dedicated to identifying the funda-

mental components of semiosis concerns (though not directly) the role of definition in
the cognitive process. Morris did not intend to define the sign, but to establish the con-
ditions according to which something may be recognized as a sign. This operational or
pragmatic attitude toward the cognitive object demystifies the role generally assigned
to definition. It is not a question of defining the object as a condition of its knowabil-
ity, but of identifying situations in which we deal with signs. Authors like Welby and
Vailati, who criticized excessive trust in the cognitive importance of definition, had
already worked in a similar sense. Vailati observed that definition did not necessarily
testify to our knowledge about something as evidenced by our difficulty in defining
precisely that which we know best: think of the difficulties involved in defining such
words as ‘hot’, ‘cold’, ‘black’, etc. (Petrilli 1988: 47–56; 1990b: 339–340; 1998a: 173–219;
Ponzio & Petrilli 1998).
In line with his plan to theorize the connection between values and the sign
dimension of behavior, Morris qualified signification as designative, prescriptive, and
appraisive signification, respectively exemplified with the words ‘black’, ‘ought’, and
‘good’; and action, following Mead, as perceptual, manipulatory, and consummatory
action, where these three types of action and signification correspond to each other in
the order indicated.
If values are considered in relation to action and preferential behavior and not as
absolute, independent entities, the study of action is inevitable when passing from sign
theory to value theory. Value is studied by Morris in terms of a ‘value situation’ which
involves both social and individual values, and is regarded as any situation in which
preferential behavior occurs. As such a value situation is inherently relational.
Values are described as objectively relative properties, that is, they are properties
of objects (in a wide sense of this term) relative to preferential behavior. Morris classi-
fies them as object, operative, and conceived values corresponding to the tripartition of
signification and of action.
The term ‘value’ is used in different contexts to signify different aspects of value
situations, that is, situations involving preferential behavior. Morris discusses three
usages which he considered as basic. ‘Object value’ is applied to objects and as such
is objectively relative, it comes into play in perceptual action. ‘Operative value’ sig-
nifies the direction of preferential behavior, it holds wherever there is a direction
of Â�preference at choice points as regards objects and behavior and is correlated to
Charles Morris 199
Â�
manipulatory action. ‘Conceived value’ concerns preferential behavior as accorded to
a signified object or situation. Ideally it guides our effective choices and corresponds to
consummatory action. Morris specified that object and operative values do not neces-
sarily involve signs, that is, an object of signification, while, on the contrary, conceived
Â�values can only exist as signified values and, therefore, necessarily involve signs.
Moreover, Morris identified three dimensions of value: detachment, dominance,
dependence which correspond respectively to the classification of action as percep-
tual, manipulatory, or consummatory, and of signification as designative, prescriptive,
or appraisive.
Morris’s theoretical horizon is far more complex and articulate than we have
reported in this paper in which we have simply traced its outlines. However, what we
do wish to underline is the fact that his studies on the relation of signs and values iden-
tify correspondences among notions established in the context of sign theory, action
analysis (Mead) and value theory.
It is a question of reading the correspondences that relate the two faces of the
same process, as though we were looking at the correspondences in writing on the
two different sides of the same sheet of paper. Morris’s research concerns a fact of
communication: communication among the order of signs and the order of values,
and, therefore, among the practitioners of the disciplinary fields concerned with such
aspects of behavior.
References
Austin, J.L. (1962). How to do things with words. Clarendon Press.

Bakhtin, M.M. (1986). Speech genres and other late essays. University of Texas Press.
——— (1993). Toward a philosophy of the act. University of Texas Press.
Bernard, J., M. Bonfantini, J. Kelemen & A. Ponzio (eds.) (1994). Reading su Ferruccio Rossi-Landi.
Semiosi come pratica sociale. Edizioni Scientifiche Italiane.
Bloomfield, L. (1933). Language. Allen & Unwin.
——— (1974). Linguistic aspects of science. University of Chicago Press.
Carnap, R. (1939). Foundations of logic and mathematics. University of Chicago Press.
——— (1955). On some concepts of pragmatics. Philosophical Studies 6: 89–91.
Dewey, J. (1938). Logic: the theory of inquiry. Henry Holt & Co.
——— (1939). Theory of valuation. In Encyclopedia and unified science. University of Chicago Press.
Eschbach, A. (ed.) (1981). Zeichen über Zeichen über Zeichen. Gunter Narr.
James, W. (1975–1988). The works of William James. Harvard University Press.
Hardwick, C.S. (ed.) (1977). Semiotic and Significs. The correspondence between Charles S. Peirce and
Victoria Lady Welby. Indiana University Press.
Mead, G. H. (1922). A behavioristic account of the significant symbol. Journal of philosophy
19: 157–163.
——— (1934). Mind, self, and society. University of Chicago Press.
——— (1938). The philosophy of the act. University of Chicago Press.
Morris, C. (1932). Six theories of mind. University of Chicago Press.
200 Susan Petrilli
——— (1937). Logical positivism, pragmatism and scientific empiricism. Hermann.

——— (1938a). Scientific empiricism. In Encyclopedia and unified science: 63–75. University of
Â�Chicago Press.
——— (1938b). Peirce, Mead, and pragmatism. Philosophical review XLVII: 109–27.
——— (1938c). Foundations of the theory of signs. In C. Morris (1971): 15–71. [It. trans. Lineamenti
di una teoria dei segni, intro., comment & trans. by F. Rossi-Landi, 1954; new ed. by S. Petrilli.
Lecce: Manni, 1999.].
——— (1942a). William James today. In H.M. Kallen (ed.) In commemoration of William James: 178–
187. Columbia University Press.
——— (1942b). Paths of life: Preface to a world religion. Harper [new printing: University of Chicago
Press, 1973.].
——— (1946). Signs, language and behavior. In C. Morris 1971: 73–397.
——— (1948a). The open self. Prentice-Hall.
——— (1948b). Signs about signs about signs. In C. Morris 1971: 434–455.
——— (1956). Varieties of human value. University of Chicago Press.
——— (1964). Signification and significance. A study of the relations of signs and values. MIT Press.
——— (1966). Festival. George Braziller.
——— (1970). The pragmatic movement in American philosophy. George Braziller.
——— (1971). Writings on the general theory of signs. Mouton.
——— (1976). Image. Vantage Press.
——— (1988). Segni e valori. Significazione e significatività e altri scritti di semiotica, etica ed estetica.
[It. trans and intro. by S. Petrilli.] Adriatica.
——— (1993 [1925]). Symbolism and reality. A study in the nature of mind. John Benjamins.
Peirce, C.S. (1923). Chance, love and logic. Harcourt.
——— (1931–1958). Collected papers of Charles Sanders Peirce, 8 vols. The Belknap Press of Harvard
University Press.
Petrilli, S. (ed. & intro.) (1987a). Per Ferruccio Rossi-Landi, Il Protagora 11/12: XXVII.
——— (1987b). Il contributo di Rossi-Landi allo studio di Charles Morris. In S. Petrilli 1987a:
105–114.
——— (1988). Significs, semiotica, significazione. Adriatica.
——— (1990a). On the materiality of signs. In A. Ponzio (1990): 365–401.
——— (1990b). The problem of signifying in Welby, Peirce, Vailati, Bakhtin. In A. Ponzio (1990):
313–363.
——— (1991). From Peirce (via Morris and Jakobson) to Sebeok: interview with Thomas A. Sebeok.
In T.A. Sebeok (1991): 95–105.
——— (ed. and intro.) (1992). The correspondence between Charles Morris and Ferruccio Rossi-Landi,
Semiotica 8(1/2).
——— (1993). Signs and values: For a critique of cognitive semiotics. Journal of pragmatics 20: 239–251.
——— (1994). Signs and values in linguistic work. In J. Bernard et al. (eds.): 135–159.
——— (1998a). Su Victoria Welby. Significs e filosofia del linguaggio. Edizioni Scientifiche Italiane.
——— (1998b). Teoria dei segni e del linguaggio. Graphis.
——— (1999). Charles Morris’s biosemiotics. Semiotica 127(1/4): 67–102.
——— (2001). In the Sign of Charles Morris. Recherches Sémiotiques/Semiotic Inquiry, vol. 21, n. 1–2:
3–26.
——— (2005). Percorsi della semiotica. Graphis.
——— (ed. and intro.) (2008). Approaches to Communication. Trends in Global Communication Studies.
Atwood.
Charles Morris 201
——— (2009). Signifying and Understanding. Mouton De Gruyter.

——— (2010). Sign Crosswords in Global Perspective. Semioethics and Responsibility. Transaction.
Ponzio, A. (1990). Man as a sign. Essays on the philosophy of language. [Intro., trans. & ed. by S. Petrilli.]
Mouton de Gruyter.
——— (1993). Signs, dialogue and ideology. [Ed. and trans. by S. Petrilli.] John Benjamins.
——— (1995). Segni per parlare dei segni/Signs to talk about signs. [trans. by S. Petrilli.] Adriatica.
Ponzio, A., O. Calabrese & S. Petrilli (1993). La ricerca semiotica. Esculapio.
Ponzio, A., P. Calefato & S. Petrilli (1994). Fondamenti di filosofia del linguaggio. Laterza.
Ponzio, A. & S. Petrilli (1998). Signs of research on signs. Semiotische Berichte 22(3,4).
——— (1999). Fuori campo. II segni del corpo tra rappresentazione ed eccedenza. Mimesis.
——— (2000). Il sentire della comunicazione globale. Meltelmi.
——— (2001). Thomas Sebeok and the Signs of Life. Icon Books.
——— (2003). Semioetica. Meltemi.
——— (2005). Semiotics Unbounded. Interpretive Routes through the Open Network of Signs. Toronto
University Press.
——— (2007). Semiotics Today. From Global Semiotics to Semioethics, a Dialogic Response. Legas.
——— (2008). Lineamenti di semiotica e di filosofia del linguaggio. Graphis.
Rossi-Landi, F. (1953). Charles Morris. Bocca.
——— (1961). Significato, comunicazione e parlare comune. Marisilio.
——— (1968). Il linguaggio come lavoro e come mercato. Bompiani. [Language as Work and Trade.
Bergin and Garvey, 1983.].
——— (1975a). Charles Morris e la semiotica novecentesca. Feltrinelli.
——— (1975b). Linguistics and economics. Mouton.
——— (1978). On some post-morrisian problems. Ars Semeiotica 3: 3–32.
——— (1985). Metodica filosofica e scienza dei segni. Bompiani.
——— (1992). Between signs and non-signs. John Benjamins.
——— (1994 [1972]). Semiotica e ideologia. Bompiani.
Searle, J. (1969). Speech acts. Cambridge University Press.
Sebeok, T.A. (1979). The sign and its masters. University of Texas.
——— (1981). The Image of Charles Morris. In A. Eschbach (ed.): 267–284.
——— (1990). Semiotics in the United States. Indiana University Press.
——— (1991). American signatures. Semiotic inquiry and method. University of Oklahoma Press.
Vailati, G. (1967). II metodo della filosofia. Saggi di critica del linguaggio. Laterza.
——— (1987). Scritti. 3 vols. Arnaldo Forni.
Vološinov, V.N. (1929). Marksizm i filosofija jazyca. Leningrad. [Marxism and the philosophy of
Â�language, Harvard University Press, 1973.].
Welby, V. (1983 [1903]). What is meaning? (Studies in the development of significance). John Benjamins.
——— (1985 [1911]). Significs and language (The articulate form of our expressive and interpretative
resources). John Benjamins.
Notation in formal semantics
Walter De Mulder
1.â•… Objectives
The aim of this text is not to present all the different systems of notation used in
Â�theories of so-called ‘formal semantics’. One could ask whether such a project would
be feasible or even sensible. Not only do the notations used in ‘formal semantics’
vary from framework to framework, they also change with the development of the
Â�corresponding theories and from article to article – as can be expected of any living
field of inquiry.
The first objective of this text is to introduce some of the most currently used
notational devices in Montague semantics, which can still be considered to be
the Â�reference theory in formal semantics. In this way, this text hopes to give some
Â�necessary background for the other entries on formal semantics. A second aim is to
answer a question reminiscent of one that is asked in conversation analysis. There the
question is: how can we find a transcription that reflects all the relevant factors of an
utterance, meta- and paralinguistic clues included. Here a parallel question could be:
given a natural language sentence, how do we transcribe or translate it into a formal
language? Is there some kind of ‘automatic’ translation procedure that provides the
interpretation of any natural language utterance? In order to answer this question, we
will closely follow Cann (1993), an excellent description of what such a translation
looks like. In the end, however, we will briefly analyze the role of context, to show that
such a translation cannot be as ‘mechanical’ as is sometimes suggested and that it is
impossible to neglect the role of context in deciding which proposition is expressed by
a natural language sentence.
2.â•… Principles
In Montague’s opinion there is no essential divide between natural language and the
artificial languages used in logics: “I consider it possible to comprehend the syntax
and semantics of both kinds of languages within a single and mathematical precise
theory” (Montague 1974: 222). To present such a theory is Montague’s aim.
A first step is to translate natural language sentences into a formal language. This
translation into a logical language cannot be the end point of the theory, however:
the ultimate aim of the translation of the natural language sentences into the logical
Â�language is to assign to the natural language sentences the same interpretation as the
one that is given to their logical translations, essentially by making use of the resources
of set theory (for set theory, see Partee, Ter Meulen & Wall 1987). The benefit provided
by translating natural language sentences into logical formulas is the elimination of
the vagueness and ambiguity inherent in natural language sentences. When natural
language sentences are translated into logical formulas, they inherit their clear and
univocal interpretation.
The basic notion in the enterprise of assigning an interpretation to natural
Â�language sentences is truth: to say what a sentence means is to say what situations
the sentence can denote, and thus to say what the world is like if the sentence is true.
Consequently, the basic meaning of a sentence is reduced to the truth conditions of the
proposition it expresses.
Montague semantics respects two further constraints. First of all, it respects the
principle of compositionality, which says that the meaning of a sentence is a Â�function
of the meanings of its parts. Indeed, in view of the finite resources of the brain,
the fact that we can understand an infinity of sentences, can only be explained
if new sentences are nothing but a combination of known elements from a finite
list. As a consequence, syntax actively contributes to the meaning of the sentence.
Â�Considerations such as these lead Montague to demand that for each syntactic rule
there be a semantic rule corresponding to it, an idea that is known as the rule-by-rule
hypothesis.
3.â•… The basis: Predicate logic and model theory
Taking into account the rule-by-rule hypothesis, the starting point of the translation
procedure has to be a syntactic analysis of the natural language sentence. Such an anal-
ysis can be delivered by different syntactic theories, such as a version of government
and binding theory (Chierchia & McConnell-Ginet 1990), or of categorial grammar.
But since we follow Cann’s (1993) exposition of Montague semantics, we will adopt a
context-free grammar, like he does. For a sentence such as (1).
(1) Paul eats an apple,
such a grammar gives the following rule:

(2) GR1: S → NP1 Vt [+ fin] NP2
204 Walter De Mulder
where the lexicon stipulates:

(3) NP1 = {Paul, … }
Vt = {eat, … }
NP2 = {apple, … }.
Then the first thing an explicit semantic theory has to do, is to provide a series of
Â�translation rules, one for each syntactic construction. The appropriate rule for (1) is
the one associated to rule GR1 in (2):
(4) TGR1: S → Pred’(N1’,N2’)
The logical symbols in this formula, which should be familiar from predicate logic
(see Partee, Ter Meulen & Wall 1987), are replaced by forms that belong to the logical
language into which the natural language is translated:
(5) Paul → Paul’
eat → eat’
apple → apple’
As already said above, the final result of this translation procedure, (eat’(Paul’,apple’)),
does not by itself provide an interpretation of the natural language sentence – as could
be suggested by some theories in generative semantics and some theories of logical form
in more recent versions of generative grammar which stop their analysis at this stage.
To provide an interpretation, Montague semantics takes up the idea that the basic
meaning of a sentence is to be identified with its truth conditions. Since the Â�logical
translation is unambiguous, it is possible to say exactly for each component of a Â�sentence
what must correspond to it if the sentence is to have the truth conditions it has.
This is the function of the third part of the theory, model theory. As the name
Â�suggests, the logical translations of sentences are not directly linked to real situations,
but to mathematical models, composed of sets of entities and functions defined on
those entities. More specifically, a model is composed of an ontology and an assign-
ment function. The ontology stipulates what objects exist and, thus, what can be talked
about in the language that is interpreted with respect to the model. The denotation
function essentially has the role of assigning names to the entities whose existence was
stipulated by the ontology. A model for the sentence in (1) would thus.
–â•fi first, stipulate the existence of two entities, a person and an apple, noted PAUL
and APPLE respectively; it follows from the theory that PAUL and APPLE are not
symbols, but entities in the model;.
–â•fi second, assign a name to each of them, roughly:
Paul’ denotes PAUL
an apple’ denotes APPLE.
It remains now to assign an interpretation to the symbol eat’. The logical translation in
(4) suggests that this verb marks a relation between two entities such that the second
gets eaten by the first. The meaning of the verb is thus conceived on the model of the
predicate that translates it, as a function, a relation that is either one-one or many-one,
but that in any case for each argument gives only one value and in this way provides an
unambiguous interpretation. With respect to the model that is being set up, this means
that an expression like eat’ denotes a set of ordered pairs composed of the argument
and its value, in our case symbolised as {<PAUL, APPLE>} – where {a,b} symbolises a
set composed of the elements a and b and <a,b> symbolises the idea of an ordered pair
of which a is the first and b the second member. This gives us the proper definition of
truth in the model set up thus far: the sentence in (1) will be true if the ordered pair
<PAUL, APPLE> is in the set that constitutes the interpretation of eat’, where PAUL
and APPLE are the denotations of Paul’ and an apple’ respectively.
4.â•… Semantic types
The analysis as presented thus far is not satisfactory, since in most syntactic theories,
sentences such as (1) are not directly analysed as in (2), but are analysed in two steps,
as in (6):
(6) S → NP VP
VP → V NP
However, no element in the logical language used until now gives a direct Â�interpretation
to the VP constituent, whereas Montague’s theory demands a one-to-one correspon-
dence between syntactic and semantic rules in order to maintain the compositionality
principle.
The solution to this problem cannot be just to provide an interpretation that
Â�corresponds to the VP, since it can be expected that similar problems will be posed by
other expressions if the compositionality principle and the rule-by-rule hypothesis are
to be maintained. What is needed is a more principled solution. This is provided by
coupling each grammatical category to a corresponding semantic type in what is then
called a typed logical language.
This happens by means of a recursive procedure, where a restricted set of basic
Â�elements allows the construction of non-basic categories via an operation that
Â�combines them. The two primitive types are noted e and t and the complex type is
noted <a,b>; it is a type if each of a and b are a type. In this notation, a represents the
input type and b the output type of the operation that combines the basic types into
the complex type <e,t>.
It remains to convert the syntactic categories of the grammar into the Â�appropriate
logical types (an operation that is much simplified if one already uses a categorial
Â�grammar). For this, the basic equivalences are given by the following type assignment
function:
(7) type (S) = t
type (NP) = e
This means, roughly, that t is the type for expressions that denote truth values
whereas e is the type for expressions that denote entities. The other categories are then
Â�constructed in accordance with the rule of functional application:
(8) Rule of functional application:
If f is an expression of type <a,b>, and a is an expression of type a, then f(a) is an
expression of type b, or, slightly more informally, an <a,b> combines with an a to
form a b.
This rule allows us to find the type corresponding to an intransitive verb, i.e. an
Â�expression of the category Vi in the grammar. An intransitive verb produces a sentence
when it gets combined with a noun phrase, which means, in the language of the types,
that it produces an expression of type t when combined with an expression of type e.
Consequently, it must be itself of type <e,t>, since <e,t> x e = t.
One could think that a transitive verb should be translated in the typed lan-
guage as <[e,e],t>, since a transitive verb needs two NPs, two expressions of type e
to give an S, an expression of type t. However, this notation would imply a return to
the flat structure in (2). The adequate notation that reflects the hierarchical struc-
ture in (6) is <e,<e,t>>, where the transitive verb is interpreted as an expression
that takes an expression of type e (an NP) to produce an expression that itself needs
another expression of type e to produce an expression of type t. In other terms, the
transitive verb combines with its object to produce an expression that has to be
combined with a subject to Â�produce a sentence. In this way, all categories can be
assigned a semantic type:
(9) type (S) = t
type (VP) = <e,t>
type (NP) = e
type (Vi) = <e,t>
type (Vt) = <e,<e,t>>
type (Vdt) = <e,<e,<e,t>>> (where Vdt is a ditransitive verb)
etc.
It remains to be seen how these typed expressions can be interpreted using model
Â�theory. We already said that a transitive verb denotes a set of ordered pairs. An intransi-
tive verb, on the other hand, denotes a set of entities: the denotation of walk, e.g. is the
set of all things that walk. In fact, this is the set of all elements of which it is truly said
that they walk. For this reason, an intransitive verb, or a one-place predicate, such as
walk is said to denote a characteristic function, a function that maps all the Â�elements
of some domain either into the true or into the false. The set of entities A given in the
model is mapped onto the set of truth values {0,1}, which is formally noted as DtDe,
where Da symbolises the set of possible denotations for expressions of type a and DaDb
is a notation for the function from the set of possible denotations for expressions of type
b to the set of possible denotations for expressions of type a. This finally suggests the
followig rule for the assignment to types of appropriate denotations from the models:
(10) De is the domain A of entities existing in the model
Dt = {0,1}
If a and b are types then D<a,b> is DbDa
For a transitive verb, a two-place predicate, which is an expression of type <e,<e,t>>,
this means that its denotation is D<e,<e,t>> = D<e,t>De = DtDeDe = ({0,1}A)A, and thus a
function that, when applied to an entity, yields a function that picks out a set of enti-
ties related to the entity already mentioned. This seems intuitively right: eat denotes a
function that applies to a specific apple to yield the characteristic function that denotes
the set of entities in A that eats this apple. This can alternatively be noted as: eatM = a
function from De to D<e,t>, where the expression AM = B symbolizes that the seman-
tic value of a well-formed expression A in model M is B.
5.â•… Lambda abstraction and lambda conversion
The discussion above implies that the VP eat an apple is an expression of type
<e,t> and, consequently, denotes a set of entities. Such sets are normally noted as,
e.g. {xx eats an apple}. They can at first sight be translated into our logical language
as (eat’(an apple’))(x). This is an open formula, however, with a free variable. Such an
expression does not denote a characteristic function. To make it do so, the lambda
operator is used, which is an abstraction of the variable used in the expression and turns
the expression into one that denotes a characteristic function: λx [(eat’ (an apple’))] (x).
In short:
If Φ is an expression of type t containing an unbound instance of a variable x of type
e, then λ x[Φ] is a well-formed expression of type <e,t>. (Cann 1993: 116)
Briefly, λx [(eat’(an apple’))](x) denotes the set of all entities x that eat an apple.
Since lambda expressions are expressions of type <e,t>, it is normal that when they
are combined with expressions of type e, they yield expressions of type t, this being an
instance of the rule of functional application. This is what is called lambda conversion:
(11) Lambda conversion:
λx [(eat’ (an apple’))(x)] (Paul’) = (eat’ (an apple’))(Paul’)
It still must be pointed out how lambda expressions get interpreted by model Â�theory.
For a start, they contain variables, whose interpretation cannot be given by the
Â�denotation assignment functions introduced above, since these give constant names to
entities in the domain of the model. The interpretation of variables is thus to be given
by a new function, the variable assignment function, usually noted g, which assigns to
each variable an element from the model as its value. This gives the following semantic
formulation for lambda conversion:
(12) λx[Φ](a)M,g = [Φ]M,ga/x (Cann 1993: 128)
where the assignment function ga/x is the same as g except that a, the semantic value of
a, an expression of type e, in M with the variable assignment function g, is substituted
for x.
6.â•… Quantifiers
Now that we have given the instruments needed to analyze the VP, it is time to turn
to the NPs. Until now, we have considered Paul and an apple equally as expressions
of type e. This is not appropriate, however. Consider, first of all, the NP an apple. It is
simply wrong to consider it as an expression of type e, if only because it is a complex
expression, composed of a determiner a(n) and a common noun. Since both can be
combined to other elements, the compositional dictum and the rule-to-rule Â�hypothesis
require that we assign a separate interpretation to both elements.
Let’s start with the common nouns: these denote sets of entities. The noun apple,
for instance, denotes the set of all apples. This means, of course, that a common noun
is of type <e,t>. It remains, then, to assign a proper type to the determiner. To find out
which one, consider first (13), with a quantified pronoun expression:
(13) Someone eats an apple.
As such, the pronoun does not deliver an entity and is thus not of type e. It denotes,
rather, a function from VP denotations to sentence denotations, or from sets to truth
values, and this means that it is of type <<e,t>,t>. This, then, gives the cue for the type
that is to be assigned to determiners. Since they are expressions that take a common
noun, of type <e,t>, to furnish a quantified NP, of type <<e,t>,t>, they must be of type
<<e,t>,<<e,t>,t>>.
The characterisation of quantified NPs as functions from sets to truth values,
implies that they denote sets of sets of entities. This is confirmed by the translation of a
sentence containing a into our logical language. A sentence such as (14).
(14) A soldier dies

is translated as (15), following the translation rules provided by the theory:

(15) ∃x [soldier’(x) & die’(x)].
This formula, however, contains no translation of the noun phrase as such; it is a

Â�contextual definition of the determiner a and the noun phrase that contains it. Since
it contains no translation of the noun phrase as a proper part, it does not respect the
compositionality dictum. To find the contribution of the noun phrase as such, lambda
abstraction is applied, in two steps:
(15bis) 1. λP [∃x [soldier’ (x) & P(x)]] (die’)
2. λQ [λP [∃x [Q(x) & P(x)]]] (soldier’) (die’)
This gives the final translation of a in our logical language:
(15ter) a: λQ [λP [∃x [Q(x) & P(x)]]]
This translation can be seen to confirm the denotation for the quantifier and the quanti-
fied NP proposed above. The quantifier clearly denotes a relation between the sets denoted
by the two expressions of type <e,t>, the common noun and the intransitive verb. To be
specific, the quantified NP picks out the intersection between the set denoted by soldier’
and the set denoted by die’ according to the model. This in turn provides the possibil-
ity of defining quantifiers by specifying the properties of the relations between the sets
involved. This is the starting point of the theory of generalised quantifiers (see Â�Barwise &
Cooper 1981; Barwise & Etchemendy 1989, and the essays in Gärdenfors 1987).
We illustrated the case of the quantified NP with a sentence in which this NP
Â�functions as a subject. The reason is that in cases where the NP functions as an object,
there is an additional complication. Indeed, the type of the transitive verb is <e,<e,t>>.
Consequently, it cannot be combined with the quantified NP, if this is of type <<e,t>,t>.
To make the combination possible, the type of the transitive verb has to be changed
into <<<e,t>,t>,<e,t>>. This operation is called type raising and can be shown to be
Â�justified for independent reasons (Cann 1993: 177–180; Partee 1986).
7.â•… Intensionality
Until now, the semantic values of all expressions have been defined as entities or sets
of entities that may be said to exist in the model. Since Frege, however, it is well-
known that the meaning of an expression cannot always be equated with the entities it
denotes. Frege noted that if such were the case, then the substitution of an expression
by another expression denoting the same object should not affect the truth value of the
sentence as a whole, as can be seen in (16):
(16) The morning star rises in the east.
The evening star rises in the east.
Since both subject NPs denote the same object, the substitution of the one for the
other does not affect the truth value of the whole. Nevertheless, one may already have
the feeling that the second sentence is less “natural” than the first, and this intuition is
confirmed by Frege’s observation that in an assertion of identity such as (17),
(17) The morning star is the evening star,
the NPs cannot exclusively denote the entities they refer to. For if that were the
case, (17) should be a necessary truth, whereas in reality it can be used to report a
scientific discovery. So Frege concluded that the sense of an expression is to be dis-
tinguished from its reference. The term reference, however, is confusing, since it is
also used to refer to the act of picking out one particular object of reference, whereas
the semantic value of an expression is mostly equated with the set of its possible
referents. Therefore, model-theoretic semantics has adopted the terms extension
and intension from Carnap (1947). The extension of an expression is then the set of
entities in the model which the expression can denote, whereas the intension is the
property that identifies the entities in the extension. If, for instance, the extension of
soldier’ in the model is a set of entities as defined by the model, then its intension is
the property that these entities somehow can be said to have in common. However,
both extension and intension are aspects of the denotation of the expressions, they
refer to things in the world and are to be distinguished from ‘systematic’ linguistic
meaning (Cann 1993: 267).
Many contexts require that we take into account the intensional values of expres-
sions, i.a. the formalisation of tense and aspect markers and the formalisation of
alethic, deontic or epistemic modalities. To describe the interpetation of the different
modalities, model-theoretic semantics has taken up Leibniz’s idea of Â�possible worlds.
Take for instance a sentence such as (18), which contains a necessity operator:
(18) Necessarily, 4 + 5 = 9.
The proposition this sentence expresses is necessarily true, and this means that we can
find no world in which it would be false or, positively, that it is true under all circum-
stances we can think of. The interpretation of the necessity operator, symbolised by ☐,
can thus be defined as follows:
(19) If φ is a formula, then [☐φ]M,g,wn = 1 iff [φ]M,g,wm is 1 for all wm in W.
In this formula, W is a set of possible worlds and w is an element of this set. The for-
mula thus says that for each formula φ, if this formula is within the scope of a necessity
operator, it is true with respect to a model M, a variable assignment function g and a
world wn if it is true with respect to the model M, the variable assignment function
g and all worlds wm that the model provides. It is easy, then, to find an alternative
Â�formula for the possibility operator, symbolised by ◊.
(20) If φ is a formula, then [◊φ]M,g,wn = 1 iff [φ]M,g,wm is 1 for some wm in W.
Let us add here that not all sentences expressing possibility (or necessity) are of equal
possibility (necessity). For instance, (21) is slightly more possible that (22):
(21) My father might come to dinner.
(22) John Lennon might come to dinner.
To mark this difference in possibility, the world in which my father might come can be
said to be more accessible than the one in which John Lennon might come. This sug-
gests that it is interesting to have different accessibility relations R between worlds in
the model and to change (19) and (20) accordingly (see Cann 1993: 269–281):
(19bis) If φ is a formula, then [☐φ]M,g,wn = 1 iff [φ]M,g,wm is 1 for all wm ∈ W
where wnRwm.
(20bis) If φ is a formula, then [◊φ]M,g,wn = 1 iff [φ]M,g,wm is 1 for some wm ∈ W
where wnRwm.
The operators ☐ and ◊ allow analyses of other modal expressions, such as can and must
(see Kratzer 1981), and the introduction of operators such as O and P, associated with
the English readings ‘It is obligatory that’ and ‘It is permitted that’ respectively.
Tense is also a factor that introduces intensionality. Compare, for instance (14)
to (14b):
(14) b. A soldier died.
To find out whether (14) is true, it is sufficient to look at the world as it is now, to know
the extension of die’ now; on the other hand, to find out whether (14b) is true, we must
inspect the extension of die’ at other moments of time. Thus, we must take into account
the intension of die’.
It is not surprising, then, that the formalisation of tense follows roughly the same
lines as the formalisation of modality. It is necessary to make the semantic value of an
expression, noted as …, relative not only to a model, an assignment function and a
world, but also to a set of temporal instants (or intervals, as Cann does) ordered by the
precedence relation. When this is done, the PAST operator can be defined as follows
(Chierchia & McConnell-Ginet 1990: 215):
(23) If φ is a formula, then PAST φM,w,i,g = 1 iff for some i’ ∈ I such that i’ < i, φM,w,i’,g = 1
This formula says that a proposition expressed by a sentence with a verb in the past
time is true at a moment of time i – most frequently the time of utterance iu – when the
proposition was true at a moment of time i’ that preceded i. It is not difficult to devise
formulas for the present and the future, by changing the ordering relation between the
instants of time.
In view of the frequency of contexts that require that we take into account of the
intension of expressions, Montague finally decided to introduce an intensional opera-
tor and to consider all expressions first and foremost as intensional. This intensional
operator is symbolised as ^ written to the left of the notation for the extensional expres-
sion. However, since all expressions are considered to be intensional, Montague also
introduced the converse of the intensional operator, the extensional operator, noted
as v. Intensional expressions such as (14b) thus get translated as ^(die’(soldier’))(x)
and denote the intension of their extensional counterpart, noted sometimes as ℑ[(die’
(a soldier’)) (x)]. ℑ[(die’(a soldier’)) (x)]M,g (<w,i>) is then the value of the intension
of [(die’(a soldier’)) (x)] at index <w,i>. Intensional expressions are of a special type,
symbolised as <s,a> if a is a type, where s is not a basic type like e or t, but just a means
of indicating that the type is intensional.
The interpretation of intensional expressions was already described when we
discussed the tense and modality operator above: it is apparent there that intensions
are functions from world-time pairs, or indices, to extensions – which should not
cause a big surprise, since we defined intensions as properties that, at a time and in
a possible world, pick out the extension of the expression at that time and in that
possible world. This means that the denotation of an intensional expression <s,a> is
a function from the set of ordered pairs {<w,i>|w ∈ W & i ∈ I}, where W is the set of
worlds and I is the set of instants, to the set of denotations of type a. A set of ordered
pairs like the above is called the cartesian product of the set W and the set I and
noted as WxI, and the denotation of an intensional expression <s,a> is consequently
defined as follows:
(24) D<s,a> = DaWxI
This allows the following somewhat more formal notation for the denotation of an
intensional expression:
(25) If α is an extensional expression of type, then [^α]M,g,wn,i is a function f with domain
WxI such that for all <wm,i> in WxI, F(<wm,i>) = φM,g,wm,i’ (Cann 1993: 297)
8.â•… Contexts
So far, mainly following Cann (1993), we have introduced the most essential nota-
tional devices used in traditional model-theoretic semantics. It is not possible here
to introduce all notations used in so-called formal semantics – a misleading name
since, as should be clear by now, simply to formalise an expression does not furnish its
semantic value. Therefore, we set out to answer the following question: given a natu-
ral language sentence, is there an ‘automatic’ translation procedure that delivers the
translation of this sentence into a logical language and that thus, via the interpretation
of this logical language, delivers the essential interpretation of this natural language
sentence? We have tried to point out, following Cann, what such a procedure would
look like. Nevertheless, it would be an illusion to think that the procedure is wholly
mechanical. For this, the influence of context on the interpretation of natural language
sentences is too profound. Consider, for instance, the well-known sentence (26).
(26) The king of France is bald.
It is clear that such a sentence has to be interpreted with respect to a world-time pair:
in 1600, this sentence could have been true, whereas at the present time, it is false
(according to Russell 1905) or expresses no proposition at all (according to Strawson
1950). As Kaplan (1989) pointed out, this means that the definite description expresses
a function from a circumstance of evaluation, say a world-time pair, to a truth value.
However, this is not the case when the NP is introduced by a demonstrative. For the
demonstrative picks out its referent directly and it is this referent itself that is part of
the proposition that gets evaluated with respect to world-time pairs. If this is true,
there must be a distinction between the context of utterance, which delivers the refer-
ent of the demonstrative, and the circumstances of evaluation in which the propositions
that contain these referents get evaluated. This would mean, then, that an additional
level of meaning has to be added to the model, where it would be specified how the
context of utterance contributes to the determination of the proposition expressed.
This level of meaning is called character by Kaplan (1989). Meaning would thus have
at least two levels:
Character
Context Proposition (intension)
Intension
Circumstances Extension
Kaplan (1989) holds that character is a function from context to propositional

content. In his view, the linguistic content of a deictic expression is able to deliver, by
itself, the referent. A good example would be the first-person pronoun ‘I’, whose lin-
guistic meaning dictates that its referent is the speaker of the context of utterance. It
has progressively become clear, however, that both positions of Kaplan are untenable.
First, contrary to what Kaplan holds, the character or linguistic meaning of I does
not directly deliver a referent that is part of the proposition that gets evaluated with
respect to a world-time pair. As was pointed out by Nunberg (1993), some uses of ‘I’
do not refer to the person him- or herself, but to the person as far as he or she fulfills a
certain role with respect to a world and a time. This is, e.g. the case in (27):
(27) (Uttered by a prisoner to be executed the day after the utterance of this sentence:)
I am traditionally allowed to choose whatever I like for my last meal.
Here, it is not the speaker as such who is part of the proposition, but the speaker in as
far as he is a prisoner on death row.
Second, there are examples of uses of ‘I’, where ‘I’ does not refer to the speaker
but where the speaker uses ‘I’ because he takes up the role of another person as it
were, as in some echo-sentences, etc. This means that in any case we have to take into
account the referential intentions of speakers, and thus that character or linguistic
meaning as such can only be functional against the background of the intentions with
which speakers use them. Such intentions, however, cannot be described as functions
(Bach 1987), as Kaplan (1989) himself admits.
This suggests that there is no way of escaping the conclusion that the Â�translation
of natural language sentences into a logical language is sensitive to factors of a
Â�fundamentally pragmatic nature and is, therefore, not as ‘automatic’ as some seem to
hold. This is one of the reasons why certain recent theories try to incorporate these
Â�contextual and pragmatic factors from the outset and recognize that the determination
of the proposition expressed is sensitive to interpretation strategies of a fundamentally
pragmatic nature (cf. Kratzer’s theories, situation semantics, discourse representation
theory).
9.â•… List of some frequently used symbols
This list simply cannot be exhaustive, since every author in the field of model-theoretic
semantics seems to have his own preferences and, sometimes, even his own notations.
Consequently, this list is only instrumental, aimed at helping the reader to work his
or her way through the logical entries in this Handbook. We will first present the
notation, followed by the technical term that denotes it, alternative notations where
applicable, and finally an informal English paraphrase or definition.
p ∨ q Disjunction (alternative notation: Apq); p or q
p ∧ q Conjunction (alternative notations: Kpq, p.q, p&q); p and q
¬p Negation (alternative notations: Np, ~p, !p); not p
p → q Conditional (alternative notations: Cpq; p⊃q); if p, then q
p ↔ q Biconditional (Alternative notations: Epq, p≡q); p if and only if q
A ⇒ B A logically implies B (the conditional statement is a tautology)
A ⇔ B A is logically equivalent to B (the biconditional statement is a tautology)
A B Strict implication; Necessarily, if A, then B

(∃x)(Fx) Existential quantifier; There is an x, such that…

(∀x)(Fx) Universal quantifier; For all x, …
(ιx) (Fx) Iota-quantifier (alternative notation: E!); There is only one x such that (or: the x such
that)
A ⊆ B The set A is a subset of the set B
A ⊂ B The set A is a proper subset of B (A cannot be identical to B)

A ⊇ B The set B is a subset of set A
A ⊃ B The set B is a proper subset of set A (B cannot be identical to A)
A ∪ B The union of sets A and B
A ∩ B The intersection of A and B
A × B The Cartesian product of A and B, i.e. the sets of all ordered pairs <x,y> where x is a
member of A and y a member of B
A − B The set of all members of set A that are not also members of set B
a ∈ B a is an element of set B
{a,b,…} The set consisting of the members a,b,..
{x | f(x)} The set consisting of all the items that meet the condition f
ĈA A is provable by the rules of inference (syntactic validity)
|=A A is valid, i.e. true in all states of affairs
A |= B B is a logical consequence of A (every model of A is also a model of B)
A/B A syntactic category of expressions that combine with expressions of category B to
yield expressions of category A (cf. the mathematical formula A/B x B = A)
<a,b> The ordered pair of which a is the first member and b the second; in Montague
semantics the logical type of expressions that combine with an extension of type a to expres-
sions of type b
Da The set of possible denotations for expressions of type a
DaDb The function from the set of possible denotations for expressions of type b to the set
of denotations for expressions of type a
 φM,g,w,c,t The semantic value of an expression φ with respect to a model M, a variable as-
signment function g, a possible world w, a context c and a time point t (this set of indices can
be extended; moreover, the letters used to denote each of the indices may vary) (alternative
notation: INT<i,j>, a function which assigns denotations to expressions with respect to a world
i and a context j)
∨A The extension of A
∧A The intension of A
ℑ[φ]M,g The intension of the extensional expressions φ with respect to M and g
w1Rw2 The possible world w2 is accessible from the possible world w1
☐ The necessity operator
◊ The possibility operator
O It is obligatory that…
P It is permitted that…
G From now on, it will always be the case that (Going to be)
F rom now on, it will at least once be the case that (Future)
H Up to now, it has always been the case (Has been)
P At least once before now, it has been the case that (Past)
(It is clear that these temporal operators can also be noted in other ways, and that
more distinctions are possible; the four operators given here are proposed in Gamut 1991: 54.)
Kiφ Epistemic (knowledge-)operator: person i knows that φ
λ The lambda operator (cf. supra)
References
Bach, K. (1987). Thought and reference. Oxford University Press.

Barwise, J. & R. Cooper (1981). Generalized quantifiers and natural language. Linguistics and
Â�Philosophy 4(1): 159–219.
Barwise, J. & J. Etchemendy (1989). Model-theoretic semantics. In M. Posner (ed.) Foundations of
cognitive science: 207–243. MIT Press.
Barwise, J. & J. Perry (1983). Situations and attitudes. MIT Press.
Cann, R. (1993). Formal semantics. Cambridge University Press.
Carnap, R. (1947). Meaning and necessity. University of Chicago Press.
Chierchia, G. & S. Mcconnell-Ginet (1990). Meaning and grammar. MIT Press.
Devlin, K. (1991). Logic and information. Cambridge University Press.
Dowty, D., R. Wall & S. Peters (1981). Introduction to Montague semantics. Reidel.
Galmiche, M. (1991). Sémantique linguistique et logique. Presses University de France.
Gamut, L.T.F. (1991). Logic, language & meaning. The University of Chicago Press.
Gärdenfors, P. (1987). Generalized quantifiers. Kluwer.
De Jong, F., L. Oversteegen & H. Verkuyl (1988). Betekenis en taalstructuur. Foris.
Kamp, H. & U. Reyle (1993). From discourse to logic. Kluwer.
Kaplan, D. (1989). Demonstratives. In J. Almog, J. Perry & H. Wettstein (eds.) Themes from Kaplan:
Kratzer, A. (1981). The notional category of modality. In H. Eikmeyer & H. Rieser (eds.) Words,
worlds and contexts. De Gruyter.
McCawley, J.D. (1981). Everything that linguists have always wanted to know about logic (but were
ashamed to ask). University of Chicago Press.
Nunberg, G. (1993). Indexicality and deixis. Linguistics and Philosophy 16(1): 1–43.
Partee, B. (1986). Noun phrase interpretation and type-shifting principles. In J. Groenendijk, D. De
Jongh & M. Stokhof (eds.) Studies in discourse representation theory and the theory of generalized
quantifiers. Foris.
Partee, B., A. Ter Meulen & R. Wall (1987). Mathematical methods in linguistics. Kluwer.
Russell, B. (1905). On denoting. Mind 14: 479–493.
Von Stechow, A. & D. Wunderlich (eds.) (1991). Semantik/Semantics. De Gruyter.
Strawson, P. (1950). On referring. Mind 59: 320–344.
Phenomenology
Peter Reynaert & Jef Verschueren

1.â•… The study of ‘phenomena’
Phenomenology in its present-day sense refers to a philosophical method for the

Â�rigorous investigation of subjectively experienced ‘phenomena’. Every object of direct
intuition or original experience counts as a genuine phenomenon to be investigated.
Although phenomenology has a profound respect for the data of experience, it is not
another positivism or empiricism. Since direct intuition or original experience are not
confined to sense-perception, phenomenology accepts a reality beyond the realm of
sensory experience, such as ideal logical and mathematical objects and ethical values.
Phenomenology takes the entire field of intuitive, conscious acts with their correlated
experienced objects as its starting point.
Phenomenology, as discussed in this entry, does not refer then to so-called
Â�phenomenal consciousness whose nature is hotly debated in contemporary Philoso-
phy of Mind. Phenomenology is not to be confused with the epistemological position
of ‘phenomenalism’ either, which makes a rigid distinction between appearance and
reality and which cherishes the view that sensations of phenomena are all that exists.
Phenomenology is not a sense-datum theory in this sense at all. On the contrary, the
main epistemological thesis is that reality can only be ascribed to what is experienced
or intuited in a certain a priori determined way. Experiencing determines the reality of
the transcendent object, and is not consciousness of some intra-mental Â�representation
of the object.
2.â•… History and basic tenets of the phenomenological movement
Although the term already appeared in 18th century philosophical texts and gained
prominence through Hegel’s Phenomenology of mind, contemporary phenomenol-
ogy basically derives from the philosophy of Edmund Husserl (1984, 1976, 1965).
Â�Husserl’s main source of inspiration was Franz Brentano’s empirical psychology
(Brentano, 1874). The content of this psychology would certainly surprise today’s
empirical psychologist. Brentano developed a descriptive psychology that was to be
the basis for the study of cognitive processes and hence in his mind for a scientific
philosophy. Description is based on careful intuitive consideration of the structural
218 Peter Reynaert & Jef Verschueren
Â�
properties of psychological phenomena, and leads to a categorization of these processes
(f.i. qua sensations, feelings, perception, etc.). In connection with his attempt to sepa-
rate physical from psychological phenomena, Brentano developed his famous doctrine
of intentionality as the latter’s fundamental property. Every psychical phenomenon is
characterized by the directedness toward an object. Brentano’s psychology thus pre-
pared one of the basic concepts of Husserl’s philosophy, viz. intentionality. But Husserl
developed his own notion of intentionality in response to the problems generated by
Brentano’s alledgedly immanentist comprehension of intentionality.
Phenomenology is a science of the intentional correlations between acts of con-
sciousness and their objects. It studies the way in which specific kinds of objects cor-
relate with specific kinds of conscious acts for their intuitive givenness to be realised.
Husserl understands this intentional correlation as a constitution of the experienced
object. Constitution means that the experience contains an a priori rule, which pre-
scribes when it is legitimate to posit the reality of the experienced object. This is the
problem of the intuitive fullfilment of the act. Adequate fullfilment, which Husserl
comprehends as the truth of the act of consciousness, is realized when the object is
intuitively given exactly as it was intended. This results from a synthesis of the differ-
ent acts through which the object is experienced. Husserl originally developed this
central idea of intentional constitution by overcoming, under the influence of Frege’s
criticism, his own psychologism in arithmetics, which reduces numbers to the acts of
calculation. Numbers are constituted by but not reducible to particular acts of math-
ematical reasoning. To get to the things themselves (Zu den Sachen selbst), one has
to study the interaction between consciousness and its objects. The “phenomenon”
Â�phenomenology studies is thus this correlation between act and object.
The emphasis on the role of consciousness in the constitution of phenomena does
not lead to relativism. The correlation between act and object is subject to scientific
rigor in philosophy and leads to universally valid descriptions of the various categories
of objects and of their constitutive acts. Phenomenology secures the pure descriptive
investigation of this field of phenomena by a specific method, the transcendental-Â�
phenomenological reduction. Reduction or ‘bracketing’ suspends the normally
accepted existence of the world, not because it should be doubted, but because it is
important to avoid any metaphysical or physical presuppositions concerning the nature
of constitutive consciousness and its relation to the experienced world. The reduction is
not motivated by any Cartesian doubt, but by the concern to avoid a petitio principii. If
consciousness were understood as mundane, it would be logically inconsistent to assign
it a constitutive role in the appearance of the world. Husserl understands consciousness
as non-mundane or pure, i.e. transcendental. ‘Pure’ phenomenology tries to detail the
‘essences’ of acts of consciousness, i.e. their universal and eidetic structures. Phenom-
enology also analyzes the so-called correlative a priori of the constituted objects, which
results in the formulation of formal and material ontologies. However, those ‘essences’
Phenomenology 219
can only be studied on the basis of a second important methodological constraint,

which Husserl calls eidetic variation. Starting from the particular, individual and sub-
jective experiences, one has to modify them in order to identify their inalterable essen-
tial properties. Phenomena have to be described on the basis of this double method
rather than through the grid of preconceived theoretical presuppositions and causal
explanations. The eidetic description of the correlation of act and object is the subject of
static phenomenology. Transcendental consciousness is only fully understood in terms
of a genetic phenomenology, which describes the genesis of the acts of consciousness.
Finally, phenomenology is also an analysis of the logical principles of consciousness,
such as its egological structure, the noesis-noema correlation and its fundamental tem-
poral characteristics. In that sense, phenomenological method led Husserl to a full-
fledged transcendental philosophy of an idealistic vein.
Husserl’s phenomenology covered a broad range of topics. Thanks to the ongoing
publication of his writings in the Husserliana-series, we now have a clearer insight
in the scope of his thinking, which includes a psychology, a philosophy of language,
an ethics and a metaphysics. As one of the first important phenomenologists, Max
Scheler developed his original ethical theory of material values (materiale Wertethik).
Husserl’s phenomenological method greatly influenced Martin Heidegger’s Sein und
Zeit (dedicated to Husserl), who nevertheless rejected Husserl’s philosophy of con-
sciousness and returned to basic ontological questions. Jean-Paul Sartre’s existentialist
philosophy equally has its roots in the phenomenological method. Maurice Merleau-
Ponty’s philosophy of the embodied perception of space and time and Paul Ricœur’s
studies of volitional experience also refer to Husserl but they both take a critical dis-
tance from his idealist philosophy. Being a real philosophical movement, where par-
ticipants agree on the rejection of Husserlian idealism, phenomenology defies brief
characterizations and attempts at providing succinct overviews of its philosophical
ramifications. A good introduction is Spiegelberg (1976).
3.â•… Phenomenology, linguistics, and the social sciences
Outside of philosophy, phenomenology had a significant impact on the field of

psychology. Karl Jaspers adopted phenomenology as a basic method of research in
psychopathology. He assigned a major role to the empathic description of patients’
pathological subjective experiences (see Jaspers 1957). One of the most lasting prod-
ucts of this influence, with effects far beyond psychotherapy, was his renewed use
of the notion of Verstehen, the empathic understanding of mental phenomena from
within. This knowledge is based on the intuitional understanding of mental processes,
and leads to the comprehension of meaningful psychological relations showing an
inner motivational nexus (as opposed to the outer causality found by explanation, or
220 Peter Reynaert & Jef Verschueren
Erklären). Aron Gurwitsch integrated in a phenomenological framework insights of

James, Piaget, the analytical psychologists and the gestaltists (1957). Today various
cognitive scientists turn to a revived phenomenological description for clarification of
the structural features of embodied consciousness and cognitive processes (see for a
contemporary contribution Petitot et al. 1999).
Husserl’s thinking influenced linguistics in a variety of ways (as described briefly
by Jakobson 1970). Husserl developed from early on, in his Logische Untersuchungen, a
theory of meaning (Bedeutungslehre), in line with his notion of intentional conscious-
ness, which came close to the Fregean approach of sense and reference. Subjective
meaning-intentions (Bedeutungsintentionen) animate language and found its mean-
ingfulness. Husserl’s thinking supported the belief that linguistics did not have to be
restricted to an exclusively empirical enterprise. The phenomenological emphasis on
the importance of primary subjectivity reinforced faith in the speaker’s intuitions as
a basic source of linguistic data. Belief in the possibility of a universal grammar was
revived. In general, the structuralist concern with the essence of langue as opposed to
the more contingent features of parole was legitimated. Phenomenology also influ-
enced the functionalism of the Prague school (see, e.g. Mathesius 1964). Moreover, a
‘phenomenology of language’ was programmatically anticipated by most of the phi-
losophers in the phenomenological tradition. Indeed, language becomes one of the
central phenomena in human experience as soon as one crosses the boundaries of
individual consciousness into the domain of intersubjectivity, where awareness of
other conscious subjects is part of the experienced world. Merleau-Ponty repeatedly
reflected on this problem of intersubjective linguistic communication in texts on
expressive, meaningful behaviour (1969).
The experience of intersubjectivity, the awareness that the experienced life-world
is not a private world but one accessible to everyone, is the basis for phenomenologi-
cal thinking about society. The implications of phenomenology for the social sciences
were developed in particular by Alfred Schuetz (1940, 1967). In Schuetz’s view, social
life is mainly structured around shared meanings:
I posit meaningful acts in the expectation that Others will interpret them meaningfully,
and my schema of positing is oriented with respect to the Others’ schema of
interpretation. On the other hand, I can examine everything which, as a product of
Others, presents itself to me for meaningful interpretation as to the meaning which the
Other who has produced it may have connected with it. Thus, in these reciprocal acts
of positing meaning, and of interpretation of meaning, my social world of mundane
intersubjectivity is built; it is also the social world of Others, and all other social and
cultural phenomena are founded upon it. (1940: 468)
Usually, the meaningful structure of the life-world is not questioned. But the process
of meaning construction can always be reactivated, and it is precisely the task of the
social and cultural sciences to do so. Although the social scientist will always have to
Phenomenology 221
refer back to his or her own experience of the life-world, he or she has to try to take the
attitude of a disinterested observer, thereby modifying the framework of orientation so
that the formulation and testing of ‘second-level constructs’ becomes possible.
4.â•… Implications for pragmatics
The phenomenological emphasis on the role of consciousness and the importance of

immediate experiences could be invoked to motivate and inspire a straightforward
pragmatic perspective centered around the actual functioning of language. In addition
to this general and partly speculative implication for the field of linguistic pragmat-
ics, there is also a direct line from phenomenology to pragmatics via the ethnometh-
odological tradition in sociology from which the field of conversation analysis has
emerged. Ethnomethodology (see Garfinkel 1967) takes commonplace activities and
experiences of everyday life, such as face-to-face conversations, as primary phenom-
ena to be studied. Applying a kind of eidetic variation to them, ethnomethodology
tries to understand these phenomena as instantiations of more general ‘activity types’.
Through the analysis of these original experiences, it seeks to elucidate the organiza-
tion of social life.
References
Brentano, F. (1978). [1874] Psychologie vom empirischen Standpunkt. Meiner.

Garfinkel, H. (1967). Studies in ethnomethodology. Prentice-Hall.
Gurwitsch, A. (1957). Théorie du champ de la conscience. Desclée de Brouwer.
Husserl, E. (1984). [1901] Logische Untersuchungen, zweites Buch. Martinus Nijhoff.
——— (1976). [1907] Ideen zu einer reinen Phänomenologie und phänomenologischen Philosophie.
Martinus Nijhoff.
——— (1965). [1925] Phänomenologische Psychologie. Martinus Nijhoff.
Jakobson, R. (1970). Main trends in the science of language. Harper & Row.
Jaspers, K. (1957). Philosophical autobiography. In P.A. Schilpp (ed.) The philosophy of Karl Jaspers:
1–94. Open Court Publ. Comp.
Mathesius, V. (1964). On the potentiality of the phenomena of language. In J. Vachek (ed.) Prague
school reader in linguistics:1–32. Indiana University Press.
Merleau-Ponty, M. (1969). La Prose du Monde. Gallimard.
Petitot, J., F. Varela, J.-M. Roy & B. Pachoud (1999). Naturalizing Phenomenology. Stanford Univer-
sity Press.
Schuetz, A. (1940). Phenomenology and the social sciences. In M. Farber (ed.) Philosophical essays in
memory of Edmund Husserl: 164–186. Harvard University Press.
——— (1967). Concept and theory formation in the social sciences. In Collected papers, vol. 1: 48–66.
Martinus Nijhoff.
Spiegelberg, H. (1978). The Phenomenological Movement. Martinus Nijhoff.
Philosophy of action
Filip Buekens
Tilburg University
Philosophers since Aristotle have always been concerned with human action, but to a
very significant degree this concern has been subservient to problems in other philo-
sophical disciplines such as ethics, philosophy of law, metaphysics, logic, philosophy
of language and even artificial intelligence. This exposé on the philosophy of action
more or less reflects this tradition, without denying that in the latter half of this cen-
tury influential books and articles in the philosophy of action have shaped discussions
in ethics, philosophy of mind, speech act theory and pragmatics.
The philosophy of action has to a great extent been focused on how the mind is
related to action. Any attempt to analyse the notion of ‘action’ as simply that which is
expressed or denoted by a verb connected to a noun phrase that refers to a human being
is obviously mistaken. Verbs may signify states (‘believe’, ‘doubt’), tendencies (‘want’)
or events (‘fall’, ‘walk’), but no use of these verbs logically entails that what is described
can be considered an action. Linguistic differences between active and passive phrases
(‘I raise my arm’/‘my arm was raised’) may provide clues as to the existence of the
difference between action and mere happenings, but they are by no means infallible:
the description ‘my arm rises’ (passive mode) is true if I raise my arm (active mode),
although the former does not entail that what I did was intentional (see Kenny 1963).
To act is to bring about something. An agent is that which brings the action about. This
invites us to look at the mental antecedents of what happens in order to understand the
bodily movements. The classic move is to say that an agent’s reasons play a central role
in identifying and explaining action. This idea goes back to Aristotle.
Aristotle’s Ethica Nicomachea is often interpreted as claiming that an action is the
conclusion of a practical syllogism (the conclusion of a theoretical syllogism being a
proposition). The premisses of a practical syllogism are a combination of cognitive
and conative states (beliefs and desires) which explain and justify the action. The rela-
tion between one’s beliefs and desires and the action is a justificatory one: we under-
stand a person insofar as we can reconstruct the premisses on which he acts. They
show us how the agent saw what he did, what the rationale of his action was.
Deliberation is the first stage of practical reasoning. Interaction between beliefs
and desires leads to the weighing of different options. A person acts rationally (and is
a rational agent) insofar as this process of deliberation is not disturbed by ignorance,
weakness of the will or irrational desires. Many philosophers believe that Bayesian
decision theory, in which an action type is selected on the basis of expected utility
(desirability) and probability of success (beliefs about one’s capacities, competence and
the environment), is a latter-day formulation of what Aristotle must have had in mind
when he wrote about the relation between deliberation and action. Practical delibera-
tion leading to the conclusion that a particular type of action is desirable is, further-
more, a form of non-monotonic reasoning: given new or additional information, the
agent may reject a conclusion he arrived at on the basis of what he knew. Part of what
it means to be a rational, experienced agent (and what it means to act rationally) is
to find the middle ground between knowing that all action, everything we can do (or
refrain from doing) can go wrong (because we didn’t have enough information, or
acted on false beliefs) and the fact that beyond a certain point, prolonged deliberation
makes action less inefficient (see Raz ed. 1978).
A related theme in Aristotle’s view of action is that the presence of a mental ante-
cedent is a necessary condition for something’s being an action. Bodily movements are
preceded by wishes, desires, volitions or intentions. In one form or another, this view
has been defended by Locke, Hume, Mill, Prichard, Wittgenstein and Ryle. A central
problem, however, remained: is the mental antecedent we point to when we explain
an action also the cause of that action? Aristotle accepted that the reasons one has for
acting are also the causes of the action, and Hume, Locke and Mill sided with him on
this issue. Under the strong influence of the writings of Ludwig Wittgenstein, however,
twentieth-century philosophers are deeply divided over this issue.
A major obstacle to the acceptance of the causal view was the famous logical con-
nection argument defended by Wittgenstein (1953). The argument relied on the premise
that cause and effect must be logically distinct entities. Since there is a logical or concep-
tual relation between a person’s wanting (willing, intending) to A and her A-ing, the lat-
ter (an action) cannot be an effect of the former. The argument can be formulated along
the following lines: if a person is intending to do something, then if she does not forget
to do it, if no countervailing circumstances arise or if she has not given up her intention
for some reason or another, and then if she doesn’t do the thing, we will conclude that
she does not really intend to do it (see Malcolm 1968). There is a conceptual connection
between intention and doing, strong enough to rule out the possibility of a causal (and
therefore contingent) connection between intending and acting.
The anti-causalist view does not necessarily deny that actions are caused. What it
rejects is the view that what explains and justifies (in the light of the agent, at least) an
action (its reasons) are those causes. A proponent of that view may or may not accept
the view that there are actually two different kinds of causal relations: event causation
and agent causation (a distinction that goes back to Plato, Timaeus 46c–48a). Event
causation is what we invoke when we account for causal relations in which no content-
ful events are involved. When content enters the scene, we invoke agent causation. If
A kills B, then A caused it to be or brought it about that B was dead (Chisholm 1976).
The agent himself is the cause of his action. Chisholm, among others, has defended the
224 Filip Buekens
view that actions are caused by their agents (see Chisholm 1976). Philosophers who
deny that reasons are causes tend to adopt the agent-causality-thesis (given the fact
that it is difficult to deny that actions – which are, after all, bodily movements – are
both caused and themselves causes).
The usefulness of the distinction between event causation and agent causation is
reinforced by other considerations. Agents are free to do what they want. A man is free
to deliberate, decide and act, and he is to be given credit for his successes and blamed
for his failures. What he does is his own achievement and he is (morally and legally)
responsible for it. How can this deeply-rooted idea of human freedom be accounted
for if human actions fall under event causation? Event causation states that there are
regular relations between cause and effect: they fall under a deterministic law. How
can this be reconciled with the alleged freedom that characterizes human action and
that has to be accounted for when we hold agents responsible for their actions?
To the now implausible answers inspired by or based on one or another form of
Cartesianism (including those that introduce a mental substance that falls outside the
realm of deterministic laws), agent causality seemed to offer a reasonable alternative:
without denying that causal relations hold between bodily movements and their physi-
ological causes (movements of the limbs, muscles, etc.), we may accept that intentional
actions are causally connected only to the agent that brings them about. This view,
however, was eventually rejected on the basis of strong ontological considerations.
Introducing two radically different kinds of causality makes intentional action into
something entirely magical. Referring to an agent in the explanation of an action is
simply a totum pro parte-explanation of that event. When we say that X intentionally
killed the goose that laid the golden egg, we say that the action can be explained by
reference to what X believes and desires. Referring to the agent is often elliptic for a
reference to what he believes and desires (Davidson 1980: 13).
This, however, is not sufficient as an argument against the Wittgenstein-inspired
‘logical-connection argument’. The argument can be rejected on the following
grounds: rational agents act on the outcome of a practical syllogism. Aristotle defined
the outcome of such a syllogism as the action itself, but this view came under pres-
sure because actions (non-semantical entities) can never have logical connections
with sentences (semantic entities) (see Davidson 1980). Only their descriptions can
be logically related to the contents of an agent’s beliefs and desires (which are semantic
entities). Events that verify a description under which they are explained by an agent’s
beliefs and desires are actions and they are, under that particular description, inten-
tional actions. Hence, the distinction between what persons do and what happens to
them is drawn in terms of how we explain what they do. Some movements of A’s arm
can be explained as intentional arm-movements, others not. If A moves his arm in
order to kill Mary Morningstar, he had the inten-tion to kill Mary Morningstar and
under that description his action was intentional. The notion ‘intentional action’ is
an intensional notion: under the description ‘killing Mary Morningstar’ the action
was intentional, under the description ‘killing Mary Eveningstar’, it was not (obviously
assuming that Mary Morningstar is identical with Mary Eveningstar).
The ‘multiple-description’ thesis underlying this view (actions are particular
events verifying different descriptions) is a central issue in contemporary philosophy
of action (see Anscombe 1957). It reveals that the old distinction between what an
agent intentionally does and what happens to him is an intensional distinction. Under
certain descriptions, he acts intentionally, under other descriptions, things befall him.
The ‘logical connection argument’ is based on a confusion between the events and
the way we describe them. The descriptions under which our actions are intentional
are derived from our reasons for doing them. Reasons and actions are related by the
ordinary causal relation that holds between events. This relation is purely extensional.
Their descriptions, however, are logically related.
This solution (based on Davidson 1980 and going back to Anscombe 1957) is now
generally accepted as the best counter-argument against agent causality. It can shed an
interesting light on the relation of freedom and determinism. Hume drew attention
to the following seemingly paradoxical situation that arises if we accept the view that
reasons for actions are also their causes: given the fact that human action is part of the
causal order, then they are also necessitated. How can this be reconciled with that other
important aspect of human life, that we are free to do what we want? Hume’s writings
provide one of the main sources of today’s widely held view that an action being caused
or necessitated by its antecedents is compatible with its being a free action. He draws a
distinction between the liberty of spontaneity, which, in his Treatise on human under-
standing, Hume regards as liberty in ‘the most common sense of the word’, and the kind
of liberty ‘which concerns us to preserve’. The liberty we want to preserve is the liberty,
for instance, to buy an album by the Beatles (instead of by the Rolling Stones): under
normal circumstances, there is no constraint or control on my action, as: I cannot think
of having no alternative. That is not the case when I am forced by violence or the threat
of violence by someone else, to do something. In that case, liberty of spontaneity is lost
and I am not free to do what I want. We tend to confuse this liberty of spontaneity (the
only kind of liberty we need to explain human freedom) with something that is quite
different, viz. ‘the liberty of indifference’. This latter kind of liberty is incompatible with
necessity and causes. Our failure to distinguish these two kinds of liberty leads us to
mistakenly deny that human actions are caused.
The debate over freedom vs. determinism in action is connected to a problem in
the philosophy of mind: given the fact that actions are bodily movements and that they
have causal ancestors (beliefs and desires) and external effects, all these events verify
physical descriptions. Is it possible to replace an explanation of what one does in terms
of cognitive and conative states by explanations of a totally different kind, viz. explana-
tions in terms of the physical properties of those causes and effects? After all, when we
226 Filip Buekens
say that causal relations hold between actions, beliefs and other attitudes, are we then
not committed to the view that nomological relations hold between those events? And
why should we rely on mental or intentional explanations if lawful explanations are
(in principle) possible? This is the problem of reductionism.
Even when we accept the view that causal relations fall under laws, this state-
ment of the causal-nomological view of causation is misleading. That singular causal
relations fall under laws can mean that (a) the predicates we use to describe those
relations must figure in those laws, or (b) that there must be laws that cover those
singular causal relations, although we do not possess those laws (to be spelled out in
physics). ‘Falls under a law’ does not entail that every generalization we use in explain-
ing actions is (or should be) a lawlike generalization. One important consequence of
this view is that the vocabulary we use to describe and explain intentional actions (and
their causal ancestors: beliefs and desires) is not reducible to or cannot be replaced by
a law-governed physical vocabulary. The descriptions under which actions are inten-
tional are not equivalent to physical descriptions. Application of intentional descrip-
tions is based on constitutive principles such as that a person is rational (most of the
time), that he or she does what he or she thinks it is rational to do, etc. Those consti-
tutive principles have a normative status. Physical concepts do not respond to them.
When we want to see men as rational agents performing intentional actions, realizing
goals and purposes with those actions and when we subject them to moral evaluation,
we have no other option than to describe them in the intentional vocabulary. The
intentional stance (see Dennett 1987) or the point of view of the interpreter (Davidson
1984) is thus necessary to see someone perform actions (and not merely to see his
body moving in this or that way).
This brings us to a last point, viz. the relation between action and responsibility.
On the basis of the fact that actions are particulars and thus subjected to different
descriptions, we can explain the relationship between being declarations of respon-
sibility and how one finds apologies for what one does. Persons are responsible for
what they do under the description that makes the action intentional. Persons are not
responsible for all the consequences of their actions, but only for those they intended
(or could reasonably foresee). I may apologize for what I did (killing the goose that
laid the golden egg) by giving the description under which the action was intentional
(killing the goose I want to eat for dinner). The excuse works under the assumption
that ‘killing the goose that laid the golden egg’ and ‘killing the goose I want for dinner’
refer to the same action and that the former description was not the one under which
the action was intentional. It follows that there is no description which is the descrip-
tion of a given action (any more than there is the description of an object or event
(see Austin 1961)). Ascriptions of responsibility (and by extension: rights, duties,
praise and blame) presuppose the non-eliminability of intentional descriptions or theÂ�
intentional vocabulary.
Many technical insights in philosophy of action were developed and/or used by

J.L. Austin, whose seminal work How to do things with words (Austin 1962) had a
major influence on the emerging field of pragmatics. Austin, who was well aware of
the importance of insights in the philosophy of action for speech act theory, used in
fact a great many insights in his work: the distinction between the locutionary, illo-
cutionary and perlocutionary actions is a distinction on the level of descriptions of
actions; perlocutionary actions are distinguished from illocutionary actions because
they are described in terms of the (intended) effects the speaker wants to bring about
in the hearer. J.R. Searle’s Speech acts (Searle 1969) made speech act theory a semi-
autonomous discipline whose main sources were no longer the philosophy of action
but the philosophy of language and linguistics.
References
Anscombre, G.E.M. (1957). Intention. Oxford University Press.

——— (1979). Under a description. Noûs 13: 219–233.
Austin, J.L. (1961). A plea for excuses. In J.L. Austin Philosophical papers. Clarendon Press.
——— (1962). How to do things with words. Clarendon Press.
Brand, M. & D. Walton (1976). Action theory. Reidel.
Chisholm, R. (1976). Person and object. Allen & Unwin.
Davidson, D. (1980). Essays on actions and events. Clarendon Press.
——— (1984). Inquiries into truth and interpretation. Clarendon Press.
Dennett, D. (1987). The intentional stance. MIT Press.
Kenny, A. (1963). Will, freedom and power. Oxford University Press.
Malcolm, N. (1968). The conceivability of mechanism. Philosophical Review 77: 45–72.
Moya, C. (1990). The philosophy of action. Polity Press.
Raz, J. (ed.) (1978). Practical reasoning. Oxford University Press.
Philosophy of language
Asa Kasher
Tel Aviv University
Philosophical studies of natural language serve two purposes: (a) They enhance our
understanding of the role played by language within the framework of major facets of
human life, such as thought, knowledge or scientific explanation, and (b) they enhance
our understanding of language itself. Pragmatics, in some sense of the term, has been
involved in the pursuit of each of these purposes. A particularly illuminating example
of how pragmatics, in a sense, appears in a philosophical study of language of type (a)
is Bas van Fraassen’s philosophical theory of scientific explanation (van Fraassen 1980).
Van Fraassen’s starting point is the threefold division of properties and relations into
syntactic, semantic and pragmatic, as introduced by Charles Morris and used in Morris
(1955). Applying this division to theories in science, an interesting distinction emerges
between syntactic aspects of a theory, such as the property of consistency, semantic
aspects thereof, such as empirical adequacy and verisimilitude, and presently most
interestingly, its pragmatic aspects, such as context-dependency. Van Fraassen’s view
is that “the language of theory appraisal, and specifically the term ‘explains’ is radically
context-dependent” (1980: 91). On van Fraassen’s view, an explanation is an answer to
a ‘why?’-question about certain facts, as compared to certain contextually determined
alternatives that are not the case: ‘why does this material burn yellow, rather than …?’.
Moreover, an explanation involves not only certain facts and certain contrastive alter-
natives, but also a contextually determined respect in which an answer is sought to the
‘why?’-question. In one respect, the question ‘why does the blood circulate through
the body?’ is related to causal answers in terms of the heart pumping the blood through
the arteries, while in another respect, the question is related to functional answers in
terms of bringing oxygen to every part of the body tissue.
Views of the nature of scientific explanation seemingly belong to a branch of
philosophy that studies science, namely the philosophy of science. However, as the
example of van Fraassen’s pragmatic theory of explanation shows, a view within the
philosophy of science could rest on an application of a view of language to the field of
science, to the activity of posing ‘why?’-questions under certain context-dependent
conditions. Thus, a philosophical theory of scientific explanation shows one way in
which science rests on natural language, as well as one way in which deeper under-
standing of a pragmatic facet of science, namely the activity of explaining, rests on
deeper understanding of a pragmatic facet of language, namely the knowledge of

Â�posing and answering context-dependent ‘why’-questions.
The analysis of scientific explanation in terms of using ‘why?′- questions is not a
singular case of a link between an interesting concept and knowledge of some Â�language
use. Various links have been exposed, suggested and discussed.
One of the most important and profound among those links involves language
and thought. For many proponents of analytical philosophy, it has been a basic tenet
of their philosophical view, “that the philosophy of thought can be approached only
through the philosophy of language […] the purpose of the philosophy of thought can
be achieved by an explanation of what it is for the words and sentences of a language
to have the meanings that they bear, an explanation making no appeal to an antecedent
conception of the thoughts those sentences express” (Dummett 1991: 3). This ingredi-
ent of analytical philosophy lend itself to generalization in a way that has underlined
much of twentieth century philosophy. For ancient philosophers, ‘first philosophy’ was
the study of the principles, first causes and essential attributes of being as such. For
philosophers of early modern times, philosophy of knowledge that dissolves all forms
of skepticism usually served as first philosophy. For most philosophers in the vari-
ous manifestations of the modern analytical tradition, philosophy of language is, in a
sense, first philosophy. Without an understanding of natural language, its knowledge
and its use, it is impossible to understand anything in which language plays an essen-
tial role. On this view, since it is impossible to understand belief and desire, knowl-
edge and intention, understanding and decision without understanding thought, it is
also impossible to understand any of these basic elements of human existence without
understanding language. Accordingly, all major branches of philosophy, such as phi-
losophy of mind, epistemology (philosophy of knowledge and belief), philosophy of
science, metaphysics, ontology, philosophy of religion, philosophy of art, philosophy
of history, philosophy of action, and others, are grounded on some theory of natural
language, as developed within the philosophy of language.
The view that deems philosophical studies of language to be the very foundations
of philosophy in general should not be confused with two other general approaches
that were used during earlier stages of the same philosophical tradition. First, logi-
cal positivists and others thought in terms of dissolving most of philosophy, rather
than grounding it on firm foundations provided by a philosophy of language. Rudolf
Â�Carnap wrote in such a vein on “the elimination of metaphysics through logical analy-
sis of language” (Carnap 1932). Analytical philosophy of the second half of the twen-
tieth century is not committed to the view that a proper philosophical analysis of
natural language, as well as its knowledge and use, is bound to result in elimination of
philosophical problems or elimination of some branches of philosophy.
Secondly, supporters of ordinary language philosophy and others put at the center
of their philosophical discussions the related expressions used by ordinary speakers of
230 Asa Kasher
a certain natural language. Thus, such a philosopher’s interest in the human mind “lies
in the words we use when we speak – or think we do – of the qualities, faculties and
performances of men’s minds, and in the persistent misunderstanding of such words by
theorists” (Austin 1970 [1950]: 45). Again, analytical philosophy of the second half of
the twentieth century is not committed to the view that proper philosophical analysis
proceeds by most careful attention being paid to the ordinary and extraordinary ways of
putting to use certain words, phrases and sentences under certain contexts of utterance.
For many years now, even philosophical studies of language itself have not been
confined to linguistic minutiae, philosophically illuminating as these might and some-
times do turn out to be. Almost all of what has been done during the last four decades
of the twentieth century under the label of philosophy of language has taken an utterly
different form, namely that of developing theories of certain facets of natural language,
mostly in a programmatic way. In order to understand the unique nature of this philo-
sophical enterprise, some of the components of that form should be briefly clarified,
namely ‘natural language’, ‘certain aspects of natural language’, ‘theories’ and theirÂ�
‘programmatic’ nature.
2.â•… Natural language
The philosophical standards of rigor and explicitness that have been some of the
emblems of analytical philosophy, have been applied to the problem of what the object
of study is, or, what should be investigated when natural language is under consider-
ation? Discussions of this problem have had two foci. One problem is that of the iden-
tification of natural language. Noam Chomsky has introduced the distinction between
e-languages and i-languages, where the former are exemplified by, say, English, Dutch,
German, Italian or Modern Hebrew, which are of historical, cultural and sometimes
political identity, and the latter, which are human cognitive systems of knowledge
usable by every human being under normal neuro-psychological conditions Â�(Chomsky
1986). Chomsky’s own revolutionary work in the study of syntax has taken the form of
a theoretical study of certain cognitive systems of the human mind/brain (for a simi-
lar approach to pragmatics, see Kasher 1991). Much work, both within philosophy of
language and within linguistics, rests on the assumption that its object of study is some
historical entity of the kind like English, as spoken in various places during long periods
of time. David Kaplan, for example, is of the view that “[t]o use language as language,
to express something, requires an intentional act. But the intention that is required
involves the typical consumer’s attitude of compliance” (Kaplan 1989b: 602).
The second problem is that of the basic nature of natural language. It has been
usually taken for granted that language is an instrument of communication and much
of what philosophers of language and linguists said about natural language has been
shaped by this assumption. For example, Paul Grice’s theory of meaning is couched in
terms of certain speakers’ intentions with respect to their audience: ‘Speaker S meant
something by uttering x’ is true only if S uttered x intending one’s audience to pro-
duce a certain effect, such as a belief that what has been said is true (Grice 1989: 99).
Â�Chomsky has argued against the view of language as an instrument of communica-
tion, on the basis of his view that systems of knowledge of language are autonomous
and can be put to numerous uses besides communication (Chomsky 1975; for a simi-
lar approach to pragmatics see Kasher 1991). Dummett has recently defended a view
of natural language that takes both thought and communication to play major roles
(Dummett 1989; see also Pateman 1987).
3.â•… Certain aspects of natural language
Unlike linguists, who supposedly are in pursuit of understanding all aspects of natural
language, philosophers have had only a few of these aspects under intensive consider-
ation. Although some philosophical work has been done in areas such as phonology
(e.g. Bromberger & Halle 1991), most of the philosophical work on natural language
has pertained to semantics and pragmatics, and to the former more than to the latter.
Philosophers have been attracted to semantics mainly for two reasons: first, the notion
of ‘meaning’ has been taken to play a major role in a variety of theories of much philo-
sophical interest, and secondly, developments within mathematical logic have pro-
vided philosophers with powerful tools for presenting, discussing, and often solving
philosophical problems of certain types, first and foremost, problems related to mean-
ing relations that can be reduced to problems related to logical forms and captured by
the logical devices of philosophical logic (for an early example of such an approach to
meaning problems of normative discourse, see Hilpinen 1971; for an updated survey
of related developments, see Burge 1992). Philosophical attraction to pragmatics was
created as a natural result of dissatisfaction with theories of meaning that had been,
in a sense, purely semantic, particularly under the influence of Wittgenstein’s later
philosophy of ‘meaning as use’ (Wittgenstein 1953; for an analytical commentary, see
Baker & Hacker 1980, 1985.) We return in the sequel to philosophical contributions
to pragmatics.
4.â•… Programmatic theories
The products of philosophical discussions of certain aspects of natural language are

often different from the products of linguistic investigations, even if a similar fam-
ily of facts is under consideration on both occasions. The distinction between those
232 Asa Kasher
products applies not only to the obvious cases, such as a project of describing a certain
corpus of utterances or a project of specifying the rules governing certain linguistic
phenomena in some natural language, be it in the cultural or the cognitive sense of the
term. The distinction applies also to the most general, abstract and ambitious projects
of understanding the basic elements of major linguistic systems. The objective of a
theoretical pursuit of understanding knowledge and use of natural language, within
the cognitive, generative approach, for example, is to reach an adequate understanding
of ‘universal grammar’, of the constraints imposed on any system of knowledge and
use of language that can be naturally acquired and used by any neuro-psychologically
normal human being. A linguistic theory that would specify these constraints in an
adequate way will be comprehensive, in the sense of providing full details of the system
of constraints, on a certain level of abstract representation of language. A philosophi-
cal theory of some aspect of natural language is often of an utterly different nature. For
instance, a semantic theory that analyses meaning relations in terms of truth condi-
tions (Davidson 1984; LePore 1986) is not comprehensive on any level, in the above
mentioned sense. It proposes a feature of semantic systems of natural language, it obvi-
ates all kinds of apparent difficulties (often relevant to pragmatics, such as Davidson
1978, 1979, 1986), but it remains of the nature of a purely programmatic suggestion,
deep and fruitful as it manifestly is or may turn out to be.
Programmatic theories that have emerged within philosophy of language often
lend themselves to use for appropriate development of theories within linguistics.
Philosophical theories of ‘logical form’, of ‘possible worlds’ and of ‘speech acts’, for
example, have been used for theoretical purposes within linguistic frameworks with-
out severe interface problems (for a discussion of ‘logical form’, see Lappin 1991; for
formal semantics, see Dowty, Wall & Peters 1981; for a philosophical theory of speech
acts and its application to rather linguistic purposes, see Vanderveken 1990, 1991).
Philosophical theories, though of a programmatic nature, have made important
contributions to the development of pragmatics, mostly by creating conceptual frame-
works for theoretical discussions of highly important families of linguistic facts and by
putting forward and examining in a critical way general theoretical claims about ways
of language use. Major topics in this endeavor have been general work on pragmat-
ics (e.g. Morris 1946; Bar-Hillel 1954; Martin 1959); issues related to the delineation
of pragmatics (e.g. Stalnaker 1972; Kasher 1979; Davis 1979); and speech act theory
(e.g. Austin 1962; Searle 1969; Bach & Harnish 1979; Vanderveken 1990). Much philo-
sophical work has also been done on particular speech acts, such as asserting, promis-
ing, asking, commanding, and numerous other ones. Other topics that have received
major treatments include reference (e.g. Donnellan 1966; Kripke 1977), indexicals
(e.g. Â�Bar-Hillel 1954; Kaplan 1979; Wettstein 1984; Kaplan 1989a), presuppositions
(e.g. Strawson 1952; Soams 1982), and implicatures (e.g. Grice 1989; Kasher 1976;
Hintikka 1986; Récanati 1993).
References
Austin, J.L. (1962). How to do things with words. Oxford University Press.
——— (1970). [1950] ‘Intelligent behaviour’, a critical review of Gilbert Ryle’s ‘The concept of mind’.
In O.P. Wood & G. Pitcher (eds.) Ryle: 45–51. Macmillan.
Bach, K. & R.M. Harnish (1979). Linguistic communication and speech acts. MIT Press.
Baker, G.P. & P.M.S. Hacker (1980). Wittgenstein, understanding and meaning. Blackwell.
——— (1985). Wittgenstein, rules, grammar and necessity. Blackwell.
Bar-Hillel, Y. (1954). Indexical expressions. Mind 63: 359–379.
Bromberger, S. & M. Halle (1991). Why phonology is different. In A. Kasher (ed.): 56–77.
Burge, T. (1992). Philosophy of language and mind, 1950–1990. Philosophical Review 101: 3–51.
Carnap, R. (1932). Ueberwindung der Metaphysik durch logische Analyse der Sprache. Erkenntnis
2(4): 219–241.
Chomsky, N. (1975). Reflections on language. Pantheon Books.
——— (1986). Knowledge of language. Praeger.
Davidson, D. (1978). What metaphors mean. Critical Inquiry 5: 31–47.
——— (1979). Moods and performances. In A. Margalit (ed.): 9–20.
——— (1984). Inquiries into truth and interpretation. Clarendon Press.
——— (1986). A nice derangement of epitaphs. In LePore (ed.): 433–446.
Davis, S. (1979). Speech acts, performance and competence. Journal of Pragmatics 3: 497–505.
Donnellan, K. (1966). Reference and definite descriptions. Philosophical Review 65: 281–304.
Dowty, D.R., R.E. Wall & S. Peters (1981). Introduction to Montague semantics. Reidel.
Dummett, M. (1989). Language and communication. In A. George (ed.) Reflections on Chomsky:
192–212. Blackwell.
——— (1991). The logical basis of metaphysics. Duckworth.
van Fraassen, B.C. (1980). The scientific image. Clarendon Press.
Grice, H.P. (1989). Studies in the way of words. Harvard University Press.
Hilpinen, R. (ed.) (1971). Deontic logic. Reidel.
Hintikka, J. (1986). Logic of conversation as a logic of dialogue. In R.E. Grandy & R. Warner (eds.)
Philosophical grounds of rationality: 259–276. Oxford University Press.
Kaplan, D. (1979). On the logic of demonstratives. In P.A. French, T.E. Uehling & H.K. Wet-
tstein (eds.) Contemporary perspectives in the philosophy of language: 401–412. University of
Â�Minnesota Press.
——— (1989a). Demonstratives. In J. Almog, J. Perry & H. Wettstein (eds.): 481–563.
——— (1989b). Afterthoughts. In J. Almog, J. Perry & H. Wettstein (eds.): 565–614.
Kasher, A. (1976). Conversational maxims and rationality. In A. Kasher (ed.) Language in focus:
197–216. Reidel.
——— (1979). What is a theory of use? In A. Margalit (ed.): 37–55.
——— (1991). Pragmatics and Chomsky’s research program. In A. Kasher (ed.): 122–149.
——— (ed.) (1991). The Chomskyan turn. Blackwell.
Kripke, S. (1977). Speaker’s reference and semantic reference. Midwest Studies in Philosophy 2:
255–276.
Lappin, S. (1991). Concepts of logical form in linguistics and philosophy. In A. Kasher (ed.): 300–333.
LePore, E. (ed.) (1986). Truth and interpretation. Blackwell.
Margalit, A. (ed.) (1979). Meaning and use. Reidel.
234 Asa Kasher
Martin, R.M. (1959). Towards a systematic pragmatics. North-Holland.

Morris, C. (1946). Signs, language and behavior. Prentice Hall.
——— (1955). Foundations of the theory of signs. In O. Neurath, R. Carnap & C. Morris (eds.)
Â�Foundations of the unity of science, vol. 1: 63–137. University of Chicago Press.
Pateman, T. (1987). Language in mind and language in society. Clarendon Press.
Récanati, F. (1993). Direct reference. Blackwell.
Soams, S. (1982). How presuppositions are inherited. Linguistic Inquiry 13: 483–545.
Stalnaker, R.C. (1972). Pragmatics. In G. Harman & D. Davidson (eds.) Semantics of natural Â�language:
380–397. Reidel.
Strawson, P.F. (1952). Introduction to logical theory. Methuen.
Vanderveken, D. (1990/1991). Meaning and speech acts, 2 vols. Cambridge University Press.
Wettstein, H. (1984). How to bridge the gap between meaning and reference. Synthese 58: 63–84.
Philosophy of mind
Stefaan E. Cuypers
1.â•… Philosophy of mind naturalized
In contradistinction to both classical, armchair philosophy of mind and a priori,

Â�ordinary language philosophy of mind, and in spite of some forceful attempts to
reverse this trend (e.g. Searle 1992; Hasker 1999), contemporary philosophy of mind
is largely naturalized. Whereas the older forms of philosophy of mind were based upon
the methods of introspection and conceptual analysis, recent philosophy of mind
draws heavily upon the empirical methods and objective findings of the sciences.
The present link of philosophy of mind especially to Â�psychology, Â�neurophysiology,
artificial intelligence and linguistics establishes its position among the other cogni-
tive sciences. Adopting a general scientistic outlook and, in particular, the natural-
istic ontology this entails, excludes spiritualistic dualism – of whatever kind – as a
possible theory of the mind from the start. The only viable frame of research for an
investigation of the mind compatible with a scientific and naturalistic worldview is
materialistic monism or – to use the current term – Â�physicalism (Gillett & Loewer
2001). It follows that most contemporary studies of the nature and typical aspects of
the mind – consciousness and intentionality – are constrained by a minimal onto-
logical commitment to the physical stuff of physics, chemistry and neurophysiology.
It is this trend that I will focus on.
In this survey, I start with outlining the basic solutions of a physicalistic theory
of the mind to three fundamental problems: the mind–body problem, the problem
of intentionality or the status of folk psychology and the debate on Â�psychosemantics.
I will end with a note on cognitive pragmatics.
2.â•… The mind–body problem
Given naturalism, how are mental phenomena possible? Physicalism normally tries
to save these phenomena by reconciling the common-sense view – there are mental
events and states such as pains and beliefs – with the scientific view that man is noth-
ing but a physico-chemical mechanism. At the end of the sixties functionalism offered
a way out of the dead-end of the materialistic research programme. Neither philo-
sophical behaviorism nor the identity-theory – the two early alternatives for dualism –
could cope with the demands of modern cognitive psychology. The former alternative
236 Stefaan E. Cuypers
could not account for the causal character of the mental, while the latter could not do
justice to its relational character.
Logical behaviorism is a peripheralist view claiming that the organism can be
treated as a stimulus-reaction mechanism without intervening variables and that
its mental states are reducible to bodily behaviors or at least behavioral dispositions
(Ryle 1949; compare also Hempel 1949). These dispositions are modelled upon physi-
cal dispositions such as the brittleness of glas and are analysed in terms of hypotheti-
cals expressing counterfactual relations between environmental input and behavioral
output, e.g. ‘if there were aspirins around, then X would take some’ in the case of a
headache. Behavioral dispositions are called ‘pure’ dispositions because they are not
inner states of the organism and consequently not causes of its outer behavior either.
The identity-theory is a centralist view holding that the brain is a necessary as well
as a sufficient condition for the mental and that mental states are fully reducible to the
physical states of human brains (Place 1956; Smart 1959). Since mental states are identi-
cal to inner brainstates, mental causality is unproblematically a form of physical causal-
ity. Unfortunately, however, the principle of neurophysiological sufficiency implies that
the relations between sensory input and behavioral output are completely inessential to
the individuation and constitution of mental states. In sum, while behaviorism inade-
quately defines the mind as a causally impotent black box, the identity-theory identifies
it equally inadequately with a relationally truncated brain in a vat.
Functionalism resolves the difficulties of both behaviorism and the Â�identity-theory
at one stroke. On a functionalist view, mental states are topic-Â�neutrally definied as
abstract inner states of the organism which fulfil a certain function or causal role in
the bringing about of behaviors – as states that are apt to be the causes of certain
behavioral effects (or the effects of certain sensory causes) (Lewis 1966; Putnam 1967;
Armstrong 1968; Block 1980 gives a survey). A mental state kind is not a natural kind
(like ‘water’), but a functional kind (like ‘table’) which is defined in terms of its mac-
roscopic causal relations to sensory input kinds, behavioral output kinds and other
intervening mental state kinds. Thus, for example, what makes a state ‘the belief that it
is raining’ is that it has certain sensory causes such as ‘experiences of rain’ and that in
combination with certain other mental states – including ‘the desire to keep dry’ – it
has certain behavioral effects such as ‘taking an umbrella’. This functional conception
of the mental obviously incorporates its causal as well as its relational character.
Although functionalism thus defined is a non-reductive theory of the mind, it still
remains straightforwardly compatible with naturalism. For, employing a well-known
distinction made by computer science between software and hardware, functionalism
plausibly holds that ‘software’ mental states which play certain causal roles can be imple-
mented or realized in ‘hardware’ brainstates which physically execute those causal roles.
The mental supervenes on the physical without the former being reducible to the Â�latter
(Kim 1993). Moreover, functionalism is radically liberal regarding the Â�allocation of
mindedness in the physical universe because mental states as functional states are not
only realizable in human brains, but also multiply realizable in the brains – or anaÂ�logous
mechanisms – of animals, computers, Martians and other ET-s. Hence, apart from doing
justice to both the causal and the relational character of the mind, functionalism can
also meet other important requirements of a materialistic research programme, viz. the
autonomy of cognitive psychology as well as the integrity of a naturalistic worldview.
Despite the virtues of a functional analysis of the nature of mental states, func-
tionalism is beset with difficulties of its own when confronted with the problems of
intentionality and consciousness. The causal role of a propositional attitude such as
the belief that it rains, does not seem to constitute its intentional content (i.e. that it
rains), while the causal role of a sensation such as a headache, equally does not seem
to determine its phenomenological quality (i.e. what it is like to feel pain) (Nagel 1974;
Block 1978; Searle 1980). This latter Cartesian problem of consciousness – the qualia
problem – and its physicalistic solution (e.g. Dennett 1991) can, however, be put aside
in the present context. Also, the problems of supervenience, mental causation and
physical realization are not immediately relevant here (Kim 1998; Shoemaker 2007).
3.â•… The problem of intentionality or the status of folk psychology
Given naturalism, how is intentionality or aboutness possible? How do some things

manage to be about or to represent other things if physical things in themselves
are not about or do not represent anything whatsoever? The physicalistic solution
to this problem – Brentano’s problem (Dennett & Haugeland 1987) – is in the first
instance intimately bound up with the fate of three different physicalistic views
on the status of folk psychology. Folk psychology is the ubiquitous common sense
strategy to explain and predict behavior in intentional terms – in terms of beliefs,
desires and other so-called ‘propositional attitudes’. In light of the tension between
this intentional psychology and the naturalistic worldview, a physicalist must either
eliminate or reduce propositional attitude psychology. If a physicalist opts for the
project of reduction or reconciliation, then he can subsequently espouse either a
realistic or an instrumentalistic construal of the ordinary conceptual framework in
terms of which we make sense of each other. All three views – intentional elimina-
tivism, intentional realism and intentional instrumentalism – agree on the (proto)-
theoretical character of folk psychology but differ in opinion about its empirical
and ontological virtues.
The most radical form of intentional eliminativism is Patricia and PaulÂ�
Churchland’s eliminative materialism (Churchland Smith 1986; Churchland 1989,
2007). According to this sceptical attitude towards folk psychology, it is doubtful that
propositional attitudes and their intentionality can ever be compatible or Â�reconcilable
with naturalism. In the Churchlands’ opinion, because of the empirical failings of

folk Â�psychology, it had better be eliminated and should subsequently be replaced by
computational neuroscience. Propositional attitude psychology suffers not only from
explanatory impotence with regard to central psychological phenomena such as mem-
ory and learning, but also from historical stagnation as well as theoretical isolation.
Folk psychology has not made any significant progression since Greek antiquity and
cannot be integrated into the emerging synthesis of the several empirical sciences.
Thus, since folk psychology has no empirical integrity, it is quite probably a false
Â�Â�theory. Â�According to the eliminativists, one should therefore resolutely take an attitude
towards folk psychology analogous to the already widely accepted rational attitude
towards Aristotelian cosmology and seventeenth century alchemy.
The most notorious form of intentional realism is Jerry Fodor’s representa-
tional theory of the mind (RTM) (Fodor 1981; compare also Field 1978). This view
involves a basically positive attitude towards folk psychology; this intentional psy-
chology is not only practically indispensable, but also scientifically respectable –
it even employs intentional ceteris paribus laws to explain and predict behavior. In
contrast to sceptical eliminativism, Fodor’s view is that folk psychology is roughly true
because it is an empirically adequate, predictively powerful and successful strategy.
The success of propositional attitude psychology is explained by its objective reality.
Given naturalism, intentional realism implies reductionism: if the mental is real then
it really has to be something else, namely the physical.
In order to account for the intentionality of the propositional attitudes in such a
realistic fashion, the functionalistic theory of the mind must, according to Fodor, be
enlarged with a theory of mental representations. Calling upon the computer model of
the mind (Haugeland 1985; Block 1995), the thinking of thoughts is construed as the
manipulation of mental symbols which are systematically organized in the language
of thought (LOT) (Fodor 1975, 2008). In short, the intentional properties of a propo-
sitional attitude are not constituted by its functional role, but they are derived from
the intentional properties of a mental symbol to which the attitude is computation-
ally related. Intentionality, then, is mechanically possible because mental symbols have
both intentional (or semantic) and formal (or syntactic) properties and they play their
causal roles in virtue of these latter properties.
The best-known version of intentional instrumentalism is Daniel Dennett’s inten-
tional stance theory of the mind (IST) (Dennett 1978, 1987). Although this view also
has a favorable attitude towards folk psychology, it argues that this Â�intentional psy-
chology is not literally an empirical theory and consequently not really true (Cuypers
1995). In light of the truth of naturalism and the improbability of reductionism as well
as the highly problematic postulate of mental representations, there are, strictly speak-
ing, no such things as propositional attitudes. However, it still remains very handy to
keep talking as if there were propositional attitudes to explain and predict behaviors.
The success of propositional attitude psychology can sufficiently be explained by its

instrumental reliability.
To account for the intentionality of the propositional attitudes in such an instru-
mentalistic fashion the (neo-)behavioristic theory of the mind must, according to
Dennett, be enlarged with a theory of interpretation. Belief-desire profiles are attrib-
uted to an intentional system from the intentional stance in order to rationally inter-
pret its behaviors. Folk psychology is a ‘dramatic’ interpretation game (Quine 1960)
played according to rational and normative principles by players who adopt the inten-
tional stance. It follows that the intentional properties of the behavioral patterns and
the attributed propositional attitudes are derived from the intentional properties of
the interpreter who adopts the intentional stance. Intentionality, then, is reconcilable
with naturalism because propositional attitudes can be attributed to configurations of
bodily movements by way of a rational interpretation of such physical configurations
from the intentional stance.
Neither intentional realism nor intentional instrumentalism can be the whole
story about the naturalization of intentionality, because the intentional properties
of mental representations and rational interpreters themselves remain unexplained.
Both RTM and IST presuppose a more original source of intentionality. What kind of
original intentionality constitutes these derivative forms of intentionality? Following
Â�Dennett (1990), one can, of course, deny the plausibility of such a distinction between
original and derivative intentionality, but if one sticks to common sense, then one is
involved in the physicalistic project of developing a psychosemantics (Fodor 1987).
4.â•… Psychosemantics
The philosophy of mind merges with the philosophy of language and meaning as a
consequence of its commitment to investigate the constitution of original intentional-
ity or semanticity. The psychological theory of the intentional content of propositional
attitudes runs parallel to the linguistic theory of the semantical content of language-
expressions, for only two things in the universe have the power to represent or mean
other things, viz. thoughts and symbols. What, then, are the metaphysically necessary
and sufficient conditions for these things to have content or meaning? With regard
to this basic question in (psycho)semantics, two opposing doctrines offer conflict-
ing answers: semantic holism as against semantic atomism. These semantic doctrines
regarding the meaning of words and sentences (symbols) mutatis mutandis apply to
the content of representations and propositional attitudes (mental symbols).
Crucial to semantic holism is the principle of contextuality: ‘Only in the context
of the language does a sentence (and therefore a word) have meaning’. The doctrine
that a symbol does not have meaning unless it is part of a larger context or whole
derives from Frege, Wittgenstein and structuralism. Only whole languages and whole
Â�cognitive systems have original meanings, whereas the meanings of linguistic and
psychological parts are merely derivative. It follows that the constitution of mean-
ing primarily depends upon the internal relation between (mental) symbols. The doc-
trine of semantic holism has two major strands in contemporary philosophy of mind
and meaning: interpretationism (Davidson 1984; Putnam 1988) and conceptual role
semantics (Block 1986; Harman 1987). Apart from facing serious linguistic difficul-
ties, semantic holism has to deal with the epistemological and ontological worries of
relativism, idealism and even solipsism (Fodor & Lepore 1992).
Central to semantic atomism is the principle of compositionality: ‘The meaning
of a sentence is a function of the meanings of its composing words (together with its
syntactic structure)’. It is the doctrine – originating in British empiricism, behaviorist
psychology and logical positivism – that an individual symbol can have meaning in iso-
lation, even if cut off from everything else. As a consequence, the constitution of mean-
ing cannot hinge on an internal symbol–symbol relation, but must depend upon an
external symbol–world relation. The World and its Natural Evolution are therefore the
original sources of meaning, totally independent from human interpretation, strategy
(Grice 1989) or activity (Wittgenstein 1953). The doctrine of semantic atomism also has
two major strands in contemporary Â�philosophy of mind and meaning: informational
semantics (Dretske 1981; Fodor 1990, 1994) and teleological semantics or biosemantics
(Millikan 1984, 1993; Papineau 1987). Besides confronting serious linguistic difficul-
ties, semantic atomism is beset with the epistemological and ontological problems that
arise for metaphysical realism and objectivism in general (Putnam 1983).
The opposition between semantic holism and atomism turns out to be the most
exciting and hotly debated issue in the philosophy of mind and meaning of the last two
decades (Fodor & Lepore 2002). It remains to be seen whether the different Â�proposals
to account for the phenomenon of original semanticity will in the end be compatible
with the general project of a naturalized philosophy of mind and meaning.
5.â•… Cognitive pragmatics
Notwithstanding some frail attempts to introduce pragmatic factors into the

Â�constitution of original semanticity (Haugeland 1990), pragmatics untill recently was a
very much neglected element in the constitution of meaning and reference within the
context of contemporary philosophy of mind and language (one of the few direct con-
tributions being Dascal 1983, in which the question is approached how language use,
rather than language structure, is connected to and reflects the workings of the mind).
One assumption behind this neglect is that the linguistic theory of speech acts is con-
sidered as secondary to and derived from the psychological theory of intentionality
and semanticity (Searle 1983). The capacity of speech acts to represent reality is merely
seen as an extension of the more fundamental capacity of the mind to represent reality.
Another, more basic assumption behind the inattention to the relation between psy-
chosemantics and pragmatics has to do with the division of labour in the philosophy
of language. Mainly since H.P. Grice’s landmark paper Logic and conversation (1975)
the divide between ‘what is said’ and ‘what is implied’ has, paradoxically perhaps,
ascertained, and even reinforced the central place for truth-conditional semantics at
the expense of the pragmatics of communication and conversation (Taylor 1998).
Quite recently, however, this situation in contemporary philosophy of mind
and cognitive science very much improved due to the advent of cognitive pragmatics
(Kasher 1988; Carston, Guttenplan & Wilson 2002). This is the interdisciplinary study
of language use and verbal communication within the methodological and Â�theoretical
framework of the cognitive sciences. Against the background of the cooperative prin-
ciple and conversational maxims Grice offered a philosophical rational reconstruction
of the phenomenon of conversational implicature without any concern for empirical
cognitive processes. Cognitive pragmatics precisely investigates the pragmatic capacity
of our mind in conversational contexts. A hearer’s attribution of a speaker’s intention is
a specific sort of mental state attribution, accounted for in theories of mind-reading –
theory-theory or simulation-theory (Carruthers & Smith 1996). A hearer’s inference
of a speaker’s intention is a specific sort of information processing, accounted for in
views on the architecture of the mind – modular or general (Carruthers 2006). Con-
versational implicature and other communicative phenomena, such as indexicality
and context-sensitivity, are studied evidenced-based, i.e. founded on empirical data
and clinical cases (autism, schizophrenia).
One important, specific approach in cognitive pragmatics is relevance theory
Â�(Sperber & Wilson 1995; Carston 2002; Carston & Powell 2006) Relevance, as a
Â�property of inputs to cognitive processes, is a trade-off between cognitive effects and
processing effort: the greater the ratio of effects to effort the greater the relevance of an
input. Our evolved minds are, under selection pressure, geared towards the maximiza-
tion of relevance. In the interpretation of utterances, then, the hearer, who interprets on
the presumption of optimal relevance, will reduce the set of possible interpretations to
the one that matches best his expectation of relevance. In contrast to the more classical
view on pragmatic interpretation as a personal level accomplishment (Récanati 2004),
relevance theory holds that the cognitive processes, which underlie utterance inter-
pretation, operate on a sub-personal level. Relevance theory is a potentially disturbing
theory in that it radically questions the validity of traditional views on the seman-
tics/pragmatics distinction (as well as the explicit/implicit distinction) (Szabó 2006).
Not only is there, in principle, no limit to the pragmatic contributions to the truth-
conditional content of utterances, there is also only a single inferential Â�mechanism
of mutual adjustment that processes (explicit) linguistically encoded meanings and
(implicit) communicated meanings. On this fascinating line of research in Â�cognitive

science, pragmatics is not an adjunct to semantics any more but the proper basis or
foundation for a philosophy of language (compare also Szabó 2005).
References
Armstrong, D.M. (1968). A materialist theory of the mind. Routledge.

Block, N. (1978). Troubles with functionalism. In N. Block (ed.) (1980): 268–305.
———â•–(1980). What is functionalism? In N. Block (ed.) (1980): 171–184.
———â•–(1986). Advertisement for a semantics for psychology. Midwest Studies in Philosophy 10:
615–678.
———â•–(1995). The mind as the software of the brain. In E.E. Smith & D.N. Osherson (eds.) Thinking.
Second edition: 377–425. MIT Press.
Block, N. (ed.) (1980). Readings in philosophy of psychology, vol. 1. Methuen.
Carruthers, P. (2006). The architecture of mind. Oxford University Press.
Carruthers, P. & Smith, P.K. (eds.) (1996). Theories of theories of mind. Cambridge University Press.
Carston, R. (2002). Thoughts and utterances. Blackwell.
Carston, R., S. Guttenplan & D. Wilson (eds.) (2002). Pragmatics and cognitive science. Special issue
of Mind & Language 17: 1–206.
Carston, R. & G. Powell (2006). Relevance theory – New directions and developments. In E. Lepore
& B.C. Smith (eds.) The Oxford handbook of philosophy of language: 341–360. Clarendon Press.
Churchland Smith, P. (1986). Neurophilosophy. MIT Press.
Churchland, P.M. (1989). A neurocomputational perspective. MIT Press.
———â•–(2007). Neurophilosophy at work. Cambridge University Press.
Cuypers, S.E. (1995). Dennett’s mild realism in the philosophy of mind. In P. Cortois (ed.) The many
problems of realism: 189–207. Tilburg University Press.
Dascal, M. (1983). Pragmatics and the philosophy of mind. John Benjamins.
Davidson, D. (1984). Inquiries into truth and interpretation. Clarendon Press.
Dennett, D.C. (1978). Brainstorms. Bradford Books.
———â•–(1987). The intentional stance. MIT Press.
———â•–(1990). The myth of original intentionality. In K.A. Mohyeldin Said, W.H. Newton-Smith,
R. Viale & K.V. Wilkes (eds.) Modelling the mind: 43–62. Clarendon Press.
———â•–â•–(1991). Consciousness explained. Little, Brown.
Dennett, D.C. & J.C. Haugeland (1987). Intentionality. In R.L. Gregory (ed.) The Oxford companion
to the mind: 383–386. Oxford University Press.
Dretske, F.I. (1981). Knowledge and the flow of information. MIT Press.
Field, H.H. (1978). Mental representations. In N. Block (ed.) (1981). Readings in philosophy of
Â�psychology, vol. 2: 78–114. Harvard University Press.
Fodor, J.A. (1975). The language of thought. Harvard University Press.
———â•–(1981). Representations. MIT Press.
———â•–(1987). Psychosemantics. MIT Press.
———â•–(1990). A theory of content and other essays. MIT Press.
———â•–(1994). The Elm and the expert. MIT Press.
———â•–(2008). LOT 2. Oxford University Press.
Fodor, J.A. & E. Lepore (1992). Holism. Blackwell.
———â•–(2002). The compositionality papers. Oxford University Press.

Gillett, C. & B. Loewer (2001). Physicalism and its discontents. Cambridge University Press.
Grice, H.P. (1975). Logic and Conversation. In P. Cole & J.L. Morgan (eds.) Syntax and Semantics,
vol. 3: 41–58. Academic Press.
———â•–(1989). Studies in the ways of words. Harvard University Press.
Harman, G. (1987). (Nonsolipsistic) conceptual role semantics. In E. Lepore (ed.) New directions in
semantics: 55–81. Academic Press.
Hasker, W. (1999). The emergent self. Cornell University Press.
Haugeland, J.C. (1985). Artificial intelligence: the very idea. MIT Press.
———â•–â•–(1990). The intentionality all-stars. In J.E. Tomberlin (ed.) Philosophical Perspectives,
vol. 4: 383–427. Ridgeview Publ. Co.
Hempel, C.G. (1949). [1935] The logical analysis of psychology. In C.G. Hempel (2000). Selected
Philosophical Essays: 165–180. Cambridge University Press.
Kasher, A. (ed.) (1988). Cognitive aspects of language use. Special issue of Journal of Pragmatics 12:
505–804.
Kim, J. (1993). Supervenience and mind. Cambridge University Press.
———â•–(1998). Mind in a physical world. MIT Press.
Lewis, D. (1966). An argument for the identity theory. In D. Lewis (1983). Philosophical papers,
vol. 1: 99–107. Oxford University Press.
Millikan, R.G. (1984). Language, thought, and other biological categories. MIT Press.
———â•–(1993). White queen psychology and other essays for Alice. MIT Press.
Nagel, T. (1974). What is it like to be a bat? In D.R. Hofstadter & D.C. Dennett (eds.) (1982). The
mind’s I: 391–403. Bantam Books.
Papineau, D. (1987). Reality and representation. Blackwell.
Place, U.T. (1956). Is consciousness a brain process? In C.V. Borst (ed.) (1970). The mind-brain
Â�identity theory: 42–51. Macmillan.
Putnam, H. (1967). The nature of mental states. In H. Putnam (1975). Language and reality: 429–440.
Cambridge University Press.
———â•–(1983). Realism and reason. Cambridge University Press.
———â•–(1988). Representation and reality. MIT Press.
Quine, W.V.O. (1960). Word and object. MIT Press.
RéCanati, F. (2004). Literal meaning. Cambridge University Press.
Ryle, G. (1949). The concept of mind. Hutchinson.
Shoemaker, S. (2007). Physical realization. Oxford University Press.
Szabó, Z.G. (2006). The distinction between semantics and pragmatics. In E. Lepore & B.C. Smith
(eds.) The Oxford handbook of philosophy of language: 361–389. Clarendon Press.
Szabó, Z.G. (ed.) (2005). Semantics versus pragmatics. Oxford University Press.
Searle, J.R. (1980). Minds, brains, and programs. In D.R. Hofstadter & D.C. Dennett (eds.) (1982).
The mind’s I: 353–373. Bantam Books.
———â•–(1983). Intentionality. Cambridge University Press.
———â•–(1992). The rediscovery of the mind. MIT Press.
Smart, J.J.C. (1959). Sensations and brain processes. In C.V. Borst (ed.) (1970). The mind-brain iden-
tity theory: 52–66. Macmillan.
Sperber, D. & D. Wilson (1995). Relevance: communication and cognition, 2nd ed. Blackwell.
Taylor, K. (1998). Truth and Meaning. Blackwell.
Possible worlds semantics
Paul Gochet
1.â•… Some logical problems
In 1919, Irving Lewis axiomatically introduced the connective , rendered as

Â�‘Necessarily if … then …’ or ‘It is impossible for … to be the case and for … not
to be the case’, and called strict implication. It was explained as follows: “the rela-
tion of strict implication expresses precisely that relation which holds when valid
Â�deduction is possible and fails to hold when valid deduction is not possible” ( Lewis &
Langford 1932: 247).
In other words, ‘A’ strictly implies ‘B’ if and only if ‘B’ is deducible from ‘A’. Lewis &
Langford (1932: 246) state that q is deducible from p if p → q is a tautology or neces-
sary truth. For instance, ‘Some men have beards’ is deducible from the two premises
‘Vinegar is sour’ and ‘If vinegar is sour, then some men have beards’, since the condi-
tional statement ‘(Vinegar is sour & (Vinegar is sour → some men have beards) → some
men have beards’ is a tautology. The statement ‘Some men have beards’, however, is not
deducible from ‘Vinegar is sour’ since the conditional statement ‘Vinegar is sour →
some men have beards’ is not a tautology.
To equate (as Lewis & Langford implicitly do) ‘tautology’ with ‘necessary truth’
is misleading. All tautologies are necessary truths, i.e. all tautological conditionals
satisfy the predicate ‘X is a necessary statement’, but some necessary truths fail to be
tautologies. The distinction between tautologies and non-tautologies is absolute
(tautologies are formulas true under all possible assignments of truth values to
the propositional variables occurring in them). On the other hand, the distinction
between necessary truths and non-necessary truths is relative to the axiomatic Â�system
under consideration.
For Lewis & Langford, the intended meaning of the strict conditional p 

(p → q) was ‘q is deducible from p’. If, however, several non-equivalent axiomatic sys-
tems can be designed to capture the strict conditional, it is no longer true that the lat-
ter expresses the relation of deducibility. In 1959, Lemmon raised the crucial question
‘Is there only one correct system of modal logic?’ and offered different readings of ,
each conferring validity on the theses of the systems considered, thus making them
sound (i.e. turning their axioms into valid statements and conferring the property of
preserving validity on their rules of inference). Consider, for instance, the axiomatic
system whose axioms are  (p → q) → ( p →  q) and ( p → p) and whose rules
of inference are modus ponens and ‘If A is a tautology, then  A is a theorem’. That
system is sound if the box operator is interpreted as ‘It is a tautology that …’. Under
the latter interpretation, the axiom ( p → p) means that whatever is tautologous
is true. Clearly this axiom is valid under that interpretation. If, however, we substi-
tute for the second rule of inference a new one, called rule of necessitation (‘If A is a
theorem, then  A is a theorem’), we obtain a system in which we are able to derive
 ( p → p) as a theorem. This theorem, however, fails to be valid when the box
operator is interpreted as before. As Lemmon puts it, “[…] though it may be a logical
truth that what is tautologous is true, it is not a tautology that what is tautologous is
true” (Lemmon 1959: 32).
Lemmon succeeded in providing the most important axiomatic systems of modal
logic with an interpretation under which they are sound. In other words he proved the
soundness of those systems. For all its intrinsic importance, Lemmon’s interpretation
satisfied only the first of the two adequacy conditions which an axiomatic system has
to fulfill. Logicians insist on getting not only a soundness proof but also a completeness
proof, i.e. a proof showing that all theses which are valid under a given interpreta-
tion are provable. For non-modal propositional logic, that challenging requirement
had already been met in 1922 by Post. Completeness proofs for modal systems were
only discovered in the late 1950s. Most of them make use of a possible worlds semantics.
This new kind of semantics proved extremely successful even outside logic.
Â�Richard Montague (1970; reprinted in Montague 1974), M.J. Cresswell (1973, 1985,
1990) and David Kaplan (1977) applied possible worlds semantics to fragments of
natural language. They developed a framework in which they could tackle problems
of major importance for pragmatics (the semantics of indexicals, the speaker-hearer
relationship, context-dependent interpretation).
2.â•… The emergence of possible worlds semantics
The idea of a plurality of possible worlds originated with Leibniz who argued that
God, in creating the world, chose the best among many possible worlds. Possible
worlds semantics results from the combination of the idea of possible worlds with that
of truth. According to Carnap, we owe to Leibniz the view that necessary proposi-
tions are propositions which are true in all possible worlds. Yet when reading Leibniz’
Â�writings, we constantly come across the traditional definition which says that neces-
sary propositions are propositions the denial of which entails contradiction. There is
only one place where Leibniz is getting close to the modern definition, i.e., where he
says that necessary propositions enjoy “eternal truth in so far as they hold not only
as long as the world will last, but would still hold if God had created the world in
Â�accordance with another plan” ( Leibniz in Couturat 1903: 18).
246 Paul Gochet
Although Leibniz paved the way toward the modern definition of necessary truth
as a proposition which is true in every possible world, his own definition remains
faithful to Aristotle’s correspondence theory of truth which applies to what actually
exists in contradistinction to what exists potentially. Leibniz does not say that a neces-
sary proposition is true in all possible worlds, but that it would still hold if the world
were different. Leibniz assumes that we understand counterfactual Â�conditionals. His
account of possible worlds and necessary truth rests upon them.
Carnap (1956) recasts the Leibnizian concept of possible worlds in linguistic
terms. Two auxiliary concepts are introduced: state-description and the range of a
Â�sentence. The definition of the former reads as follows:
A class of sentences in S1 which contains for every atomic sentence either this sentence
or its negation, but not both, and no other sentences, is called a state-description in S1,
because it obviously gives a complete description of a possible state of the universe
of individuals with respect to all properties and relations expressed by predicates
of the system. Thus the state-descriptions represent Leibniz’ possible worlds […].
(Carnap 1956: 9)
Next comes the definition of the range of a sentence as the class of all state-Â�descriptions
in which a given sentence S1 holds. He then embraces Wittgenstein’s account of semantic
competence, an account which anticipates what is advocated by possible worlds seman-
ticists: “[…] to know the meaning of a sentence is to know in which of the Â�possible cases
it would be true and in which not” (Carnap 1956: 10).
Anticipating the possible worlds semanticists who define necessary propositions as
propositions which are true in all possible worlds, Carnap (1956: 10) writes: “A sentence
Si is L-true (in system S1) = Df Si holds in every state-Â�description”. Having a definition
of the metalinguistic concept of truth (logical or necessary truth), Carnap proceeds to
define the modal operator ‘N’ as follows: “For any sentence ‘…’, ‘N(…)’ is true if and
only if ‘…’ is L-true” ( Carnap 1956: 174). Later, Kripke (1963a: 70) replaced the ‘abso-
lute’ notion of possible world with a relative notion of one world being possible relative
to another.
In the standard version of possible worlds semantics, a set of possible worlds (W)
is given together with an accessibility relation (R) defined on that set. The box and
the diamond operators are defined as follows ‘ A’ is true at world w if for all w’ such
that w R w’, ‘A’ is true at world w’. (‘Necessarily A’ is true at a given world if ‘A’ is true
at every world accessible to it), ‘A’ is true at w if for some w’ such that w R w’, ‘A’ is true
at w’. The necessity operator appears in a new guise, as a universal quantifier binding
world variables. Correlatively the possibility operator shows up as a particular quanti-
fier binding world variables. We should not conclude that modal propositional logic
is just a notational variant of first order predicate logic. Modal operators are not just
any quantifiers you may happen to think of. They are Â�quantifiers which bind variables
whose domain is restricted by the accessibility relation.
The interpretation of the box operator will vary in accordance with properties we
ascribe to the accessibility relation. To turn the axiom  p → p into a valid formula,
it is necessary and sufficient to require that R be reflexive (i.e. to require that every
world be accessible to itself). To turn the axiom  p →   p into a valid state-
ment, it is necessary and sufficient to require that R be transitive. Thanks to possible
worlds semantics, different interpretations have been associated with the box opera-
tor depending on whether it occurs within the axiomatic systems of different modal
logics. By the same token, different interpretations have been given to the strict con-
ditionals ( ) which are defined in terms of the box operator and the truth-functional

conditional (p p) = Df  (p → q))

3.â•… Key concepts of possible worlds semantics
Possible worlds semantics is a type of model-theoretic semantics. Spelling out the

semantics of a formal language amounts to supplying a systematic way of attaching
meanings to every well-formed formula of the language under consideration. As
the set of well-formed formulas is generally infinite, this mapping of meanings onto
Â�formulas is done recursively.
The semantics of a modal propositional logic consists of a frame (F), i.e. a pair
〈W,R〉, whose members are a set of possible worlds and an accessibility relation. We
build a model (M) by adding a valuation (V) to the frame, i.e. a function which assigns a
truth value to every atomic sentence in every world. In addition, we have to extend the
valuation function to assign a truth value to formulas of any complexity. This is done
by supplying a recursive definition of truth relative to the model at hand. Once we have
reached that stage, we are in a position to define the notion of logical consequence and of
validity (B is valid if it is true in all models). We are also in a position to raise the funda-
mental questions of the soundness and completeness of the axiomatic system considered.
For the basic axiomatic system K, the semantics consists of M = 〈W, R, V〉 with
no constraint upon R. The valuation function V is extended to complex formulas by a
recursive definition of truth which reads as follows:
–â•fi For all atomic sentences, ‘p’ is true in model M at world w if and only if it is
assigned the value true by the valuation function V i.e. if V(p) = 1
–â•fi ‘¬A’ is true in M, w if and only if ‘A’ is not true in M, w
–â•fi ‘A ∨ B’ is true in M, w if and only if either ‘A’ or ‘B’ is true in M, w.
–â•fi ‘ A’ is true in M, w if and only if for all w’ such that wRw’, ‘A’ is true in M, w’.
(Kripke 1963)
248 Paul Gochet
One might object that explaining the meaning of ‘Possibly A’ by means of a clause
which involves the notion of possible world smacks of circularity. This objection is
misguided. The clause under examination achieves what it is meant to achieve. As
Hintikka puts it, “[…] by stepping from a world to its alternatives, we can reduce the
truth conditions of modal statements to truth conditions of non-modal statements”
(Hintikka 1973: 198). The clause which operates the reduction should not be seen as
a definition of ‘necessarily’. It is a clause in the recursive definition of ‘truth’ for the
modal system under examination.
What is the semantic significance of a definition of truth for a formal or for
a Â�natural language? How can such a definition fit the task traditionally ascribed to
semantics, (i.e. associate a meaning with each of the syntactically well-formed expres-
sions of the language)? Cresswell argued that knowing the meaning of, e.g. ‘There is
life on Mars in 2006’, does not consist of knowing the truth value of that sentence. It
consists of Â�knowing its truth conditions, an altogether different kind of knowledge. As
Â�Cresswell puts it, “knowing the meaning of (a given sentence) is knowing, given any
possible world, whether it is true in that world” (Cresswell 1978: 12). The latter kind
of Â�knowledge does not entail the former. Possible worlds semantics shows quite clearly
where the difference lies:
Since we do not know in general which of the many possible worlds is the actual one,
we can know whether (a given sentence) would be true in this or that possible world,
because we know whether it would be true if that world were actual. Since we are not
omniscient, we will not know whether that world is actual. (Cresswell 1978: 12)
In several papers written around 1970, Richard Montague showed that large portions
of natural language could be translated into a higher-order intensional logic for which
a possible worlds semantics could be spelled out (Montague 1974; Thayse et al. 1989).
4.â•… From possible worlds semantics to pragmatics
The truth-conditional semantics is appropriate for the interpretation of sentences in

isolation. Problems arise however when we try to apply it to discourses. Consider the
following sequence:
(1) A man walks in the street. He meets Mary
In order to capture the anaphoric link that connects the pronoun ‘he’ to its antecedent,
we have to give the quantifier wide scope: ‘∃x(x is a man who walks in the street ∧ x
meets Mary)’ and to turn a sequence of sentences into a single sentence. A static formal-
ization of that kind fails, however, to cater for the possibility of appending new items to
the sequence (1), such as
(2) he smiles.
Moreover, reducing a discourse to a single sentence can alter the meaning. If we turn
the discourse: ‘Exactly one boy walks in the park. He whistles’ into the single sentence
‘Exactly one boy walks in the park and whistles’, we change the truth-conditions of
the initial sequence of sentences. The single sentence is true if there is one and only
one individual who has the properties of being a boy, of walking in the park and whis-
tling. This does not exclude the possibility of other boys walking in the park without
whistling. On the contrary the initial discourse excludes that possibility (Gamut 1991:
268–269).
Several dynamic approaches were developed to take discourse into account:
Kamp’s Discourse Representation Theory, Groenendijk’s & Stokhof ’s Dynamic Predi-
cate Calculus, Veltman’s Update Semantics. A common feature of all the versions of
dynamic semantics is the important role granted to context. This forces us to recon-
sider the border-line between semantics and pragmatics. In dynamic semantics the
two fields overlap.
The role of the context is needed to identify the antecedent of the anaphoric ‘his’ in
the following discourse: ‘I collect Elvis Presley’s belongings. A man I met on the train
sold me his hat’ (Hodges 2008: 523).
Groenendijk & Stokhof set off their new account of meaning against the truth-
conditional account in this passage: “The meaning of a sentence does not lie in its
truth conditions, but rather in the way it changes (the representation of) the inforÂ�
mation of the interpreter” (Groenendijk & Stokhof 1991: 43).
In the same vein, Veltman defines meaning as context change potential. Compare
the following two sequences in which ‘it might be’ is understood as ‘given the inforÂ�
mation state of the speaker, it is possible that’:
(3) It might be raining. It does not rain.
(4) It does not rain. It might be raining.
As opposed to truth-conditional semantics, update semantics enables us to explain the

acceptability of discourse (3) and the non-acceptability of discourse (4). After an infor-
mation state has been updated with the information that it is raining, it is no longer con-
sistent with the information that it might be raining. If, as in (3), things are presented
in the opposite order, there is no problem (Groenendijk, Stokhof & Veltman 1996: 195).
Updating an information state by making an assertion may either reduce the size
of the state or increase it. As Maria Aloni observes: “atoms and negations narrow
down the alternatives under consideration by eliminating the possibilities that do not
satisfy the information content of these sentences”. The second sentence in (3) is a
case in point. In contrast, “existentially quantified sentences add structure to the state
by setting up new items as potential topics for further discourse” (Aloni 2001: 23).
250 Paul Gochet
An assertion may neither reduce nor expand the set of possibilities. This third case
is exemplified by the first assertion in (3). On Veltman’s account, the might-sentence
acts as a test which has to be passed successfully for the sentence ‘It does not rain’ to
perform the expected update. Recent research shows, however, that the role of might
involves more than a simple test.
The notions of update and context used in dynamic semantics shed new
light on old logical puzzles in epistemic logic which have kept on intriguing logi-
cians and linguists (Gochet & Gribomont 2006). Let me revisit an old puzzle due
to Quine (1976 [1956]). Quine’s scenario is this: There is a certain man in a brown
hat whom Ralph suspects to be a spy. There is also a man whom Ralph has seen at
the beach whom he does not suspect at all of being a spy. Unknown to Ralph, how-
ever they are the same man. Quine raises this question “Can we say of this man
Â�(Bernard J. Ortcutt, to give him a name) that Ralph believes him to be a spy?”
(Quine 1976: 187). It looks as though from the first part of the story we could infer
(i) ‘∃ x (x = Ortcutt ∧ Ralph believes x to be a spy)’ and from the second, (ii) ‘∃ x (x =
Ortcutt ∧ ¬Ralph believes x to be a spy)’. But if we Â�substitute ‘Ralph believes’ for the
box operator in the principle of modal logic x = y ⊃  x = y, we can derive (iii) ‘ ∃ x(x =
Ortcutt ∧ Ralph believes x to be a spy and not to be a spy)’ from (i) and (ii). (Interme-
diate steps are omitted here.) Sentence (iii), however, is counter-Â�intuitive. Ralph is ill-
informed but not inconsistent. The challenge is how to account for the compatibility
of (i) and (ii), without ascribing logical inconsistency to Ralph.
In order to represent the puzzle of mistaken identity which is exemplified by
Ralph’s predicament, the speaker must assume two different ways of identifying
Â�Ortcutt and a shift from one identifier to the other (Aloni 2001).
Interpreted in isolation, conclusion (i) is as acceptable as conclusion (ii). But this
is not so any more if (i) and (ii) are provided with a context such as the set of sentences
(a) or the set of sentences (b):
a. Ralph believes that the man with the brown hat is a spy. The man with the brown
hat is Ortcutt.
b. Ralph believes that the man seen at the beach is not a spy. The man seen at the
beach is Ortcutt.
The information state resulting from an update with the set of sentences (a) and
(b), respectively, makes up a context which calls for sentence (i) and sentence (ii),
respectively.
Maria Aloni mentions several pragmatic principles which are at work in the selec-
tion of one of the two sentences (i) and (ii): the anchor principle (Interpretation should
be anchored in the context), Grice’s conversational maxims and the prohibition of
conceptualisation shift.
The semantics of wh-questions is a shared concern of both logicians and

Â�linguists. Already in the sixties, Hintikka raised the question of how a sentence such
as ‘a knows who b is’ could be formalized. He proposed the following formalization:
‘∃ x (a Knows x = b)’ which can be rendered as ‘there is an individual x such that
in all the worlds w compatible with what a knows, x = b’. On this account, know-
ing-who is a relation between the knower and the known. Gerbrandy offered a new
account of knowing-who which stresses the role of the context. The question ‘Who is
Bill Â�Clinton?’ he observes, can be answered by (1) ‘Bill Clinton is the president of the
United States’, by (2) ‘That is Bill Clinton’ (pointing to him), or by (3) ‘Bill Clinton was
born on august 19, 1946’. The first answer would fit the context of a high-school exam.
The second would fit a fund-raising party at which Bill Clinton is present and the last
one a conference of astrologers (Gerbrandy 1997: 137). A possible worlds semantics
for knowing-who in which the pragmatic parameter of context is taken into account is
spelled out in Gerbrandy’s essay ‘Identity in Epistemic Semantics’ (Gerbrandy 2000).
References
Aloni, M. (2001). Quantification under Conceptual Covers. ILLC Dissertation Series.

Carnap, R. (1956). Meaning and necessity. Phoenix Books.
Couturat, L. (ed.) (1903). Opuscules et fragments de Leibniz. Alcan.
Cresswell, M.J. (1973). Logics and language. Methuen.
——— (1978). Semantic competence. In F. Guenthner & M. Guenthner-Reutter (eds.) Meaning and
translation: 9–27. Duckworth.
——— (1985). Structured meanings. MIT Press.
——— (1990). Entities and indices. Kluwer.
Gamut, L.T.F. (1991). Logic, Language and Meaning, vol.2. Intensional Logic and Logical Grammar.
The University of Chicago Press.
Gerbrandy, J. (1997). Questions of identity. In P. Dekker, M. Stokhof & Y. Venema (eds.) Proceedings
of the 11th Amsterdam Colloquium: 133–138. ILLC Department.
——— (2000). Identity in Epistemic Semantics. In L. Cavendon, P. Blackburn, N. Braisy &
A. Shimojima (eds.) Logic, Language and Computation, vol. 3[s1]: 147–159. CSLI.
Gochet, P., E. Gregoire, P. Gribomont, G. Hulin, A. Pirotte, D. Roelants, D. Snyers, A. Thayse,
M. Vauclair & P. Wolper, with the collaboration of Ph. Delsarte (1989). Montague’s Semantics.
In [s2] A. Thayse (ed.) From Modal Logic to Deductive Database: 115–163. John Wiley & Sons.
Gochet, P. & P. Gribomont (2006). Epistemic Logic. In D. Gabbay & J. Woods (eds.) Handbook of the
History of Logic, Twentieth Century Modalities, vol.7: 99–195. Elsevier.
Groenendijk, J. & M. Stokhof (1991). Dynamic Predicate Logic. Linguistics and Philosophy 14:
39–100.
Groenendijk, J., M. Stokhof & F. Veltman (1996). Coreference and Modality. In S. Lappin (ed.)
The Handbook of Contemporary Semantic Theory: 179–213. [s3]Blackwell.
Hughes, G.E. & M.J. Cresswell (1996). New Introduction to Modal Logic. Methuen.
Hintikka, J. (1973). Grammar and logic. In J. Hintikka et al. (eds.) Approaches to natural language:
197–214. Reidel.
252 Paul Gochet
Hodges, W. (2008). The logic of Quantifiers. In R.E. Auxier & L.E. Hahn (eds.) The Philosophy of
Jaakko Hintikka: 521–534. Open Court.
Kaplan, D. (1977). On the Logic of demonstratives. In P. French et al. (eds.) Contemporary perspec-
tives in the philosophy of language: 401–412. University of Open Court. Minnesota Press.
Kripke, S. (1963). Semantic considerations on modal logic. Acta Philosophica Fennica: 83–94.
Lemmon, E.J. (1959). Is there only one correct system of modal logic? The Aristotelian Society Supple-
mentary Volume: 23–40. Harrisson.
Lewis, I. & C. Langford (1932). Symbolic logic. Dover.
Montague, R. (1974) [1970]. Formal philosophy. Yale University Press.
Quine, W.V. (1956). Quantifiers and Propositional Attitudes. Journal of Philosophy 53: 177–187.
[reprinted in The Ways of Paradox and Other Essays, revised and enlarged ed.: 185–196.Â�
Harvard University Press.].
Reference and descriptions
Andrea Bianchi
University of Parma
I am Andrea Bianchi, the author of the article you are reading. If I am telling the truth,
as I invite you to believe, I am related in a peculiar way to two different expressions,
a proper name (“Andrea”) and a definite description (“the author of the article you are
reading”). Let us say that the name refers to, and the description, as used in this con-
text, describes, or denotes, myself.1 But what exactly do the relations of referring and
describing (denoting) consist in? The issue was, and still is, at the center of an intense
debate in contemporary philosophy of language. As we shall see, different answers to
this question reveal different understandings of how natural languages work.
1.â•… Do proper names describe?
1.1â•… Gottlob Frege

The German mathematician Gottlob Frege is commonly acknowledged as the father
of contemporary philosophy of language. In his very influential works, he treated
proper names and definite descriptions on a par.
From a semantic point of view, according to Frege, there are three basic types of
meaningful expressions: sentences (Sätze), concept words (Begriffswörter), and what
he called “Eigennamen” (literally, “proper names”). Expressions of any of these types
have a sense (Sinn) and, if not empty, what he called a “Bedeutung”.2 The Bedeutung of
ïœ±.â•… In this article, I shall focus on singular definite descriptions and say nothing about plural defi-
nite descriptions (“the authors of the article you are reading”) and indefinite descriptions (“an
Italian philosopher”). So, in what follows I shall often use “definite description” or even “description”
as shorthand for “singular definite description” (as well as “name” for “proper name”). I shall use the
verbs “describe” and “denote” accordingly, to deal with the relation that singular definite descrip-
tions bear to their denotatum, without taking any stand on what other kinds of descriptions do.
ïœ².â•… The translation of the word “Bedeutung” as it occurs in Frege’s texts is a vexata quaestio. The
most literal is certainly “meaning”. However, Frege explicitly takes some expressions without a
Bedeutung (e.g. “Odysseus”) to be meaningful. For this and other reasons, many opt for “refer-
ence” instead. Unfortunately, while this would work in the case of Eigennamen, as we shall see in
a moment, it sounds quite odd when sentences are at stake. So, I prefer to keep the German word
in the text. For a more detailed discussion of these translation issues, see Beaney 1997a: 36–46.
254 Andrea Bianchi
a sentence is a truth-value, that of a concept word is a concept, i.e. a function whose

values are truth-values, and, finally, that of an Eigenname is an object. The sense, in
each of the three cases, is a mode of presentation of the Bedeutung. Expressions with the
same Bedeutung may have different senses. This is quite unobjectionable in the case of
sentences, once it is assumed that they have as their Bedeutung one of the two truth-
values. “Andrea Bianchi is from Venice” and “Gottlob Frege is German” are both true,
but clearly differ in sense: they present the True, so to speak, in two different ways.
Frege called the sense of a sentence “thought” (Gedanke). So, we may say that the two
sentences express different thoughts. Given that surely no one would consider the two
sentences as synonymous, identity of Bedeutung, while necessary, is not sufficient for
synonymy; only identity of sense is. If two expressions have the same sense, they are
cognitively equivalent: linguistic competence alone ensures that they do not have a
different Bedeutung.
Now, let us have a look at the category of Eigennamen, which is the most interest-
ing for our purposes. Frege does not characterize it from a syntactic point of view. He
limits himself to saying that an Eigenname is an expression (which “can also consist of
several words or other signs”) that “has as its Bedeutung a definite object (this word
taken in the widest range), but not a concept or a relation” (Frege 1892a: 27).3 However,
it is quite clear from his examples (“the point of intersection of a and b”, “Aristotle”, “the
celestial body most distant from the Earth”, “the Moon”, and “Bucephalus”, just to men-
tion some), that he considers both proper names and singular definite descriptions to
be Eigennamen.
Of an Eigenname, be it a “word, sign, combination of signs, expression”, Frege writes
that it “expresses its sense, stands for [bedeutet] or designates [bezeichnet] its Bedeutung”;
and he adds that by “employing” it, “we express its sense and designate its Bedeutung”
(31). So, to come back to our examples, it seems that in Frege’s opinion the proper name
“Andrea” and the description “the author of the article you are reading” have the same
kind of relation to myself. To use the terminology we introduced at the beginning, we
may conclude that Frege did not distinguish referring and describing (denoting).
Well, then, but what does this relation consist in, according to him? To answer,
we must take a closer look at the notion of sense. As we have already said, a sense is
a mode of presentation of a Bedeutung; in this case, of the object that is referred to by
“Andrea” and described by “the author of the article you are reading”. Furthermore,
ïœ³.â•… As a matter of fact, this is not quite accurate, for two reasons. On the one hand, as already noted,
according to Frege there are Eigennamen that do not have a Bedeutung. On the other, in his account
sentences have definite objects as their Bedeutung (though very peculiar ones: the True and the
False) but do not count as Eigennamen. Just for the record, I have modified all Michael Beaney’s
translations so as to leave “Eigenname” untranslated (in this context, “proper name” is misleading,
as we shall see in a moment).
“[t]he sense of an Eigenname is grasped by everybody who is sufficiently familiar with

the language or totality of designations to which it belongs” (27). Since our linguis-
tic competence is insufficient to settle the truth value of “Andrea is the author of the
article you are reading”, we must recognize that the name and the description have
different senses (which, given the way the world is, ‘present’ or determine the same
object, me). It remains to be seen, though, which these two senses are. The case of the
description is easier: it wears its sense on its sleeve, so to speak. In fact, the expression
is compound, and one of its constituents is a complex concept word, i.e. a predicate
(“author of the article you are reading”), which individuates a condition (authoring
the article you are reading). It is the fact that I am the unique satisfier of this condition
that makes me the Bedeutung of the description. What does the real work here is, then,
the overt semantic structure of the expression. If this is true, saying that the description
expresses a sense, i.e. a mode of presentation, which presents an object (me), amounts
to not much more than registering a triviality, namely that someone (in fact, me) is
presented by the description as the one who authored the article you are reading.
Things become much more complicated when we come to proper names. The
problem, obviously, is that they are not compound expressions: they do not have syn-
tactic structure, hence they do not seem to have semantic structure. What is, then,
the mode of presentation that they express? Frege was mainly interested in scientific
languages, where proper names are not so common, so he did not discuss the issue
extensively. However, in the few places where, quite incidentally, he did, his solution
was overwhelmingly simple: they work precisely as descriptions. Indeed, they have
a (covert) semantic structure identical to that of descriptions, and so, are associated
with a condition whose unique satisfier is their Bedeutung. Here, for example, is a
Â�well-known passage:
In the case of an actual Eigenname such as ‘Aristotle’ opinions as to the sense may differ.
It might, for instance, be taken to be the following: the pupil of Plato and teacher of
Alexander the Great. Anybody who does this will attach another sense to the sentence
‘Aristotle was born in Stagira’ than will someone who takes as the sense of the name:
the teacher of Alexander the Great who was born in Stagira. So long as the Bedeutung
remains the same, such variations of sense may be tolerated, although they are to be
avoided in the theoretical structure of a demonstrative science and ought not to occur
in a perfect language (27n; for similar considerations, see also Frege 1918–19: 65–6)
So, it seems that according to Frege for every proper name there is a description which
expresses the same sense (even though we should perhaps relativize this to speak-
ers). If this is true, the name (e.g. “Aristotle”), and the description (e.g. “the pupil of
Plato and teacher of Alexander the Great”) are synonymous (again, perhaps with the
due relativization). The description wears the sense of the name on its sleeve. What
makes it the case that “Aristotle” refers to Aristotle is the fact that Aristotle uniquely
satisfies the condition individuated by the predicate occurring in “the pupil of Plato
256 Andrea Bianchi
and teacher of Alexander the Great”, namely that of being a pupil of Plato and teacher
of Alexander the Great. In the end, we see that Frege thought that proper names refer
by describing. The notion of reference is explained in terms of satisfying a condition.
Frege’s account of reference has been immensely popular. Moreover, after the
Â�Second World War Rudolf Carnap (1947) offered a framework for formal semantics that
is congenial to it. To each expression of a language, compositionally, an intension and an
extension are associated. Proper names and definite descriptions are treated on a par. In
both cases, their extension is an object, while their intension can be seen as a function
from possible worlds to objects. The intension ‘represents’ the meaning (the sense, in
Frege’s terms) of the expression: what the “semantical rules” associate to the expression,
and so “what is grasped by everybody who is sufficiently familiar with the language”.
Given the way the world is, i.e. given the relevant non-linguistic facts (for example,
that I, rather than someone else, authored the article you are reading), the inten-
sion determines an object (the extension of the expression), which is the one that the
Â�expression refers to. Again, reference is explained in terms of satisfying a condition.4
1.2â•… Bertrand Russell

Contrary to Frege, Bertrand Russell clearly distinguished referring and describing
(denoting): if an expression refers, it does not describe; if it describes, it does not refer.
On the one hand, in an article whose importance in twentieth century philosophy
can hardly be overrated, he offered an account of definite descriptions that is different
from Frege’s in that it does not take them to be referential expressions (Russell 1905).
We shall examine it in Section 2.1. On the other hand, he maintained that there would
be no describing if there were no referring.
Russell’s reasoning can be reconstructed in the following way. In order for some-
thing to describe, it must have semantic (Russell would have said logical) structure, and
in order for something to have semantic structure, there must be simple (i.e. unstruc-
tured) symbols that compose it. These simple symbols are what allow us to describe
whatever we may want to describe. Now, what is important to notice is that they can-
not but bear a relation to mundane entities (particulars as well as universals) which is
radically different from that of describing (they do not have semantic structure!). They
must simply stand for them. Among the simple symbols, Russell called those Â�standing
ïœ´.â•… Here, for example, is a passage where the notion of reference is explicitly connected with that
of satisfaction of a condition: “An (individual) description is an expression of the form ‘(ιx)(..x..)’;
it means ‘the one individual such that ..x..’. If there is one and only one individual such that ..x..,
we say that the description satisfies the uniqueness condition. In this case the descriptum, i.e., the
entity to which the description refers, is that one individual” (Carnap 1947: 32).
for particulars “proper names”.5 As far as I know, he did not use any special word for
the relation they bear to the particulars they stand for, but I think it is fair to call it
“reference”. If so, we have found something like a transcendental argument for the
existence of referring, a relation different from that of describing (denoting). In short:
in order for something to describe, something must refer; something describes; hence,
something refers.6
It remains to be seen what the relation of referring consists in. According to Â�Russell,
“it seems scarcely possible to believe that we can make a judgment or entertain a sup-
position without knowing what it is that we are judging or supposing about” (Russell
1911: 173). As a consequence, we can refer to something only if we know what it is:
the relation of reference is epistemically constrained. Moreover, Russell thought that
all kinds of knowledge may be accounted for in terms of knowledge by acquaintance,
where acquaintance is a “direct cognitive relation” by which one is “directly aware”
of the known entity (165). This is true, in particular, of knowledge by description: to
know some object by description, one has to know by acquaintance the entities which
are needed to describe it. This led Russell to defend what has been called “the Prin-
ciple of Acquaintance”: “Every proposition which we can understand must be composed
wholly of constituents with which we are acquainted” (173).7 Given that the constituents
of a proposition are those which the simple symbols in the sentences that express it
stand for, we must conclude that reference to something requires acquaintance with it:
we may refer only to what we are directly aware of.
For reasons I shall not touch upon, Russell ended up believing that we cannot be
acquainted with physical objects; the only particulars we may be acquainted with are
sense-data. This has curious consequences for the issue of proper names. In fact, no
actual proper name can be said to refer to a sense-datum – certainly, neither “Aristotle”
ïœµ.â•… “The only kind of word that is theoretically capable of standing for a particular is a proper
name, and the whole matter of proper names is rather curious” (Russell 1918–19: 200). Why the
matter is curious, we shall see in a moment.
ïœ¶.â•… As a matter of fact, in this form the argument is not valid, because, while it is true that to
Â�describe we need simple symbols, it is false that to do so we need proper names, i.e. simple symbols
that refer to particulars. To make it valid, one can also, as I am inclined to do, call “reference” the
relation that simple symbols bear to the universals they stand for. Alternatively, one can supple-
ment the argument with another showing that no complete description of the world is possible
that is purely general, i.e. that does not require the mention of any particular.
ïœ·.â•… By “proposition” Russell means here what may be expressed by sentences (Frege’s Gedanken).
Just for the record, the Principle of Acquaintance was already formulated in ‘On denoting’: “in
every proposition that we can apprehend (i.e. not only in those whose truth or falsity we can judge
of, but in all that we can think about), all the constituents are really entities with which we have
immediate acquaintance” (Russell 1905: 492).
258 Andrea Bianchi
nor “Andrea” do. If so, they are not simple symbols that refer to particulars. But then,
what are they, and how do they work? Russell’s answer is surprising:
The names that we commonly use, like ‘Socrates’, are really abbreviations for
descriptions …. A name, in the narrow logical sense of a word whose meaning is a
particular, can only be applied to a particular with which the speaker is acquainted,
because you cannot name anything you are not acquainted with…. We are not
acquainted with Socrates, and therefore cannot name him. When we use the word
‘Socrates’, we are really using a description. Our thought may be rendered by some
such phrase as, ‘The Master of Plato’, or ‘The philosopher who drank the hemlock’, or
‘The person whom logicians assert to be mortal’, but we certainly do not use the name
as a name in the proper sense of the word (Russell 1918–19: 200–1)
Indeed, it is “very difficult to get any instance of a name at all in the proper strict logi-
cal sense of the word”: “[t]he only words one does use as names in the logical sense are
words like ‘this’ or ‘that’ ”, since “[o]ne can use ‘this’ as a name to stand for a particular
[an actual object of sense] with which one is acquainted at the moment” (201).
It is fair to say, then, that starting from a semantic framework quite different from
that of Frege, driven by his epistemology Russell ended up making similar claims con-
cerning actual proper names: for both of them, they describe, as definite descriptions
do.8 However, whilst stressing this, one should not forget that they had very different
ideas about what describing amounts to, as we shall see in Section 2.1.
1.3â•… John Searle

We have seen that according to Frege proper names are synonymous with definite
descriptions, and according to Russell they are definite descriptions in disguise. In
a short article published in 1958, John Searle distanced himself from them. Indeed,
like Russell, he distinguished “the referring function from the describing function of
language” (Searle 1958: 172). But he claimed, contrary to Russell, that actual proper
names do refer, and, contrary to Frege, that while definite descriptions refer by describ-
ing, actual proper names refer in their own peculiar way. However, as we shall see, he
still maintained that they have a strong logical connection to descriptions: referring
Â�somehow presupposes describing.
Searle’s starting point is an apparent tension (or, even, to use his word, a “paradox”
(173)). On the one hand, he says, “we do not ordinarily think of proper names as hav-
ing a sense at all in the way that predicates do; we do not, e.g. give definitions of proper
names” (166). On the other hand, though,
ïœ¸.â•… See Kaplan 2005: 973–1000 for a careful discussion of the entanglement of semantic and epis-
temological issues in Russell’s philosophy, and of his “allowing the demands of his epistemology to
override his semantic convictions” (990).
let us ask how it comes about that we are able to refer to a particular object by using its
name. How, for example, do we learn and teach the use of proper names? This seems
quite simple – we identify the object, and, assuming that our student understands the
general conventions governing proper names, we explain that this word is the name of
that object. But unless our student already knows another proper name of the object,
we can only identify the object (the necessary preliminary to teaching the name) by
ostension or description; and, in both cases, we identify the object in virtue of certain
of its characteristics. So now it seems as if the rules for a proper name must somehow
be logically tied to particular characteristics of the object in such a way that the name
has a sense as well as a reference; indeed, it seems it could not have a reference unless
it did have a sense, for how, unless the name has a sense, is it to be correlated with
the object? (167–8)
To solve the paradox, Searle looks for “the unique function of proper names in our
language”:
To begin with, they mostly refer or purport to refer to particular objects; but of course
other expressions, definite descriptions and demonstratives, perform this function as
well. What then is the difference between proper names and other singular referring
expressions? Unlike demonstratives, a proper name refers without presupposing
any stage settings or any special contextual conditions surrounding the utterance
of the expression. Unlike definite descriptions, they do not in general specify any
characteristics at all of the objects to which they refer. “Scott” refers to the same object
as does “the author of Waverley”, but “Scott” specifies none of its characteristics,
whereas “the author of Waverley” refers only in virtue of the fact that it does specify
a characteristic (170)
Hence, properly speaking, proper names do not describe. However, as anticipated,
there is, even in the case of proper names, a close connection between referring and
describing, according to Searle. Indeed, he writes,
as a proper name does not in general specify any characteristics of the object referred
to, how then does it bring the reference off? How is a connection between name and
object ever set up? This, which seems the crucial question, I want to answer by saying
that though proper names do not normally assert or specify any characteristics, their
referring uses nonetheless presuppose that the object to which they purport to refer has
certain characteristics…. To use a proper name referringly is to presuppose the truth of
certain uniquely referring descriptive statements, but it is not ordinarily to assert these
statements or even to indicate which exactly are presupposed (170–1)9
ïœ¹.â•… Note that this has as a consequence something that makes Searle’s position vulnerable to at
least some of the criticisms to description theories of proper names that we shall examine in
Section 1.5: “it is a necessary fact that Aristotle has the logical sum, inclusive disjunction, of prop-
erties commonly attributed to him: any individual not having at least some of these properties
could not be Aristotle” (172).
260 Andrea Bianchi
So, in conclusion, here is Searle’s solution of the paradox:

does a proper name have a sense? If this asks whether or not proper names are used to
describe or specify characteristics of objects, the answer is “no”. But if it asks whether
or not proper names are logically connected with characteristics of the object to which
they refer, the answer is “yes, in a loose sort of way” (173)
But, loose as it may be, this would still be a way for proper names to have a (descrip-
tive) sense. Thus, I think it is fair to conclude that even according to Searle the relation
that a proper name bears to an object, if it refers to it, is mediated by some descrip-
tion. Basically, the idea that reference has to be explained in terms of satisfaction of a
(perhaps only loosely individuated, and perhaps not part of what is asserted when the
name is used) condition is not questioned by him.
1.4â•… Taking stock

Let us take stock. We have so far examined three historically important positions on
our question. Notwithstanding certain considerable differences between them, con-
cerning (actual) proper names Frege, Russell and Searle advanced description theories,
as Saul Kripke has claimed (though with regard to Searle some caution is required):
they all maintained that they are semantically equivalent to definite descriptions,
or to clusters of definite descriptions.10 And, certainly, they all subscribed to what
Keith Donnellan called the “principle of identifying descriptions”, according to which
“[a] name is worthless without a backing of descriptions which can be produced on
demand to explain the application”, where the backing “functions as the criterion for
identifying the referent” (Donnellan 1970: 335). In Section 1.5 we shall consider some
of the arguments that have been leveled against such theories and such a principle.
Concerning reference, Frege, Russell and Searle had different views. Russell saw it
as a direct relation with a particular that is grounded in acquaintance, while Frege and,
less neatly, Searle, explained it in terms of satisfying a condition. But, even here, there
is an important aspect on which they converge. Indeed, they all assumed that reference
is epistemically constrained: in order to refer, we have to have cognitive command of
what we are referring to. Frege and, again less neatly, Searle, thought that this com-
mand consists in knowing a condition that is uniquely satisfied by the entity referred
to, Russell that it consists in a direct awareness of it.
ïœ±ïœ°.â•… Kripke also claims that they all maintained that proper names have their reference fixed by a
definite description (where this means that the object referred to is the one that uniquely satisfies
the condition individuated by the predicate occurring in the description), or by a cluster of definite
descriptions. Strictly speaking, this is not true of Russell, since according to Russell actual proper
names do not refer, while proper names in the strict logical sense do not refer by describing.
One way to make this point is to say, with Howard Wettstein, that they were all
driven by the “intentionality intuition”, that is by “the traditional idea … that if one is
to speak or think about a thing, one must possess a discriminating cognitive fix on the
thing, that something about one’s cognitive state must distinguish the relevant item
from everything else in the universe” (Wettstein 2004: 57).
Another way to make more or less the same point is to claim, with David Kaplan,
that Frege’s, Russell’s and Searle’s are subjectivist semantics, at least insofar as proper
names are concerned. According to a subjectivist semantics,
everyone runs their own language. When we speak, we assign meanings to our
words; the words themselves do not have meanings. These assignments are, in theory,
unconstrained (except by whatever limitations our epistemic situation places on
what we can apprehend). In practice, it may be prudent to try to coordinate with the
meanings others have assigned, but this is only a practical matter.
… Although the entities which serve as possible meanings may be regarded as
objective, in the sense that the same possible meanings are accessible to more than one
person, the assignment of meanings is subjective, and thus the semantics is subjective.
Since each individual user must assign meanings rather than receiving them with the
words, each user’s semantics is autonomous. What the language community does
make available to each of its member is a syntax, an empty syntax to which each user
must add his own semantics.
… In this sense there is no semantic sharing. What each user can express is independent
of the resources of other members of the language community (Kaplan 1989b: 600–1)
Perhaps with less emphasis, the charge of subjectivism was made by Kripke as well:
The picture which leads to the cluster-of-descriptions theory is something like this:
One is isolated in a room; the entire community of other speakers, everything else,
could disappear; and one determines the reference for himself by saying – ‘By “Gödel”
I shall mean the man, whoever he is, who proved the incompleteness of arithmetic’
(Kripke 1972: 91)
As we shall soon see, this is by no means the only way of conceiving of semantics,
and in particular of reference. Perhaps, the intentionality intuition is not so good as an
intuition.
1.5â•… Saul Kripke

In the late Sixties and early Seventies, a powerful attack was launched, on various
fronts, against traditional accounts of the working of natural languages. As we shall
see in Section 2.3, Donnellan (1966) argued against them, and offered an alternative
one, with regard to definite descriptions. Kripke (1972) and Hilary Putnam (1975a)
did the same with regard to at least some general terms (terms for natural kinds,
262 Andrea Bianchi
substances and artifacts, and theoretical terms), and Kaplan (1978 and 1989a) with
regard to demonstratives and indexicals. Concerning proper names, Donnellan (1970)
Â�produced some counterexamples to the principle of identifying descriptions which we
mentioned in Section 1.4. The result of all this was the emergence of something that
has been called “new theory of reference” or “direct reference theory”. A major role
here was undoubtedly played by a series of three lectures that Kripke gave at Â�Princeton
University in January 1970, which were subsequently published under the title of
“Naming and Necessity”. There, among other things, Kripke offered devastating criti-
cisms of description theories of proper names and a new picture of how these refer.
Before looking briefly at the criticisms, let us make one consequence of descrip-
tion theories of names explicit. If a name is synonymous with a definite description, or
is an abbreviation of it, then certain sentences should turn out to be analytically true,
hence necessarily and a priori true. The most obvious example is that of a conditional
whose antecedent is an existential clause and the consequent a clause where the name
precedes and the description follows the copula (“If N exists, then N is DD”).11 Even if
one weakens the theory and limits oneself to saying that the description fixes the refer-
ence of the name, without giving its sense, i.e. without being synonymous with it, one
is committed at least to saying that these sentences are a priori true.
Now, let us take a name and the description that is purportedly synonymous
with it, or at least fixes its reference. For example, to stay with Frege: “Aristotle” and
“the pupil of Plato and teacher of Alexander the Great”. Is the sentence “If Aristotle
exist(ed), Aristotle is (was) the pupil of Plato and teacher of Alexander the Great”
necessarily true? No, we should be ready to say: surely, Aristotle’s life might have been
different, and he might not have studied with Plato, nor taught Alexander the Great.
And is the sentence a priori true? Again, no: we know only a posteriori, if we know it at
all, that Aristotle studied with Plato and taught Alexander. Indeed, we are all ready to
admit that it would have been possible in principle to discover that he did not do any
of this. But, if this is so, “Aristotle” and “the pupil of Plato and teacher of Alexander the
Great” are not synonymous, nor does the second fix the reference of the first.
We have just presented, in a nutshell, two of the three arguments that are usually
ascribed to Kripke: the modal argument and the epistemic argument. Of course, we
have only shown that “Aristotle” is not synonymous with, nor has its reference fixed
by, “the pupil of Plato and teacher of Alexander the Great”, not that there is no definite
description that is synonymous with it or at least fixes its reference. But the arguments
may be repeated for many other descriptions (for example, for all those mentioned by
ïœ±ïœ±.â•… There are some complications here that I shall ignore. Moreover, in the case of Searle one
should talk of a cluster of descriptions rather than of a single description. For reasons of space,
I shall present Kripke’s arguments in the simplest form.
Frege and Russell in the passages quoted). In fact, the modal argument is backed by
something that is difficult to deny, that is that proper names and definite descriptions
behave differently when modality is at stake: while a proper name is a rigid designa-
tor, in that it refers to the same individual with respect to any counterfactual situation
(technically, a possible world), a description (but there are exceptions, which have
sometimes been used to resist the argument) is not usually so, in that it describes
different individuals with respect to different counterfactual situations. Just to give
an example: while relative to the actual situation “the author of the article you are
reading” describes who that “Andrea” refers to, relative to the counterfactual situation
where Paolo rather than me wrote the article you are reading “Andrea” still refers to me
but “the author of the article you are reading” describes Paolo.
The third and last of Kripke’s arguments is that from ignorance and error (sometimes
also called semantic argument). It consists of two simple observations. First, sometimes
we do refer to something by a proper name even though we are not able to associate
to the name any condition that we believe to be uniquely satisfied by that something:
Consider Richard Feynman, to whom many of us are able to refer. He is a leading
contemporary theoretical physicist. Everyone here (I’m sure!) can state the contents
of one of Feynman’s theories so as to differentiate him from Gell-Mann. However,
the man in the street, not possessing these abilities, may still use the name ‘Feynman’.
When asked he will say: well he’s a physicist or something. He may not think that
this picks out anyone uniquely. I still think he uses the name ‘Feynman’ as a name for
Feynman (Kripke 1972: 81; for a similar observation, see Donnellan 1970: 343)
Second, sometimes we do refer to something by a proper name even though we only

associate to the name a condition that is uniquely satisfied by something else:
In the case of Gödel that’s practically the only thing many people have heard about
him – that he discovered the incompleteness of arithmetic. Does it follow that whoever
discovered the incompleteness of arithmetic is the referent of ‘Gödel’?
Imagine the following blatantly fictional situation…. Suppose that Gödel was
not in fact the author of this theorem. A man named ‘Schmidt’ whose body was
found in Vienna under mysterious circumstances many years ago, actually did
the work in question. His friend Gödel somehow got hold of the manuscript and
it was thereafter attributed to Gödel…. So, since the man who discovered the
incompleteness of arithmetic is in fact Schmidt, we, when we talk about ‘Gödel’, are in
fact always referring to Schmidt. But it seems to me that we are not. We simply are not
(Kripke 1972: 83–4; see also Donnellan 1970: 347–9)
These two observations are extremely powerful. They can be seen as counterexamples
to the principle of identifying descriptions, and suggest that proper names do not refer
by describing: what makes it the case that “Feynman” refers to Feynman and “Gödel”
to Gödel is not the fact that Feynman and Gödel uniquely satisfy a condition that is
264 Andrea Bianchi
somehow associated with their name. Moreover, the two observations show that the
intentionality intuition is not a good intuition, and that subjectivist semantics is not
good semantics.
But what, then, makes it the case that “Feynman” refers to Feynman, “Gödel” to
Gödel, or, for that matter, “Andrea” to me? Here is Kripke’s answer:
Someone, let’s say, a baby, is born; his parents call him by a certain name. They
talk about him to their friends. Other people meet him. Through various sorts of
talk the name is spread from link to link as if by a chain. A speaker who is on the
far end of this chain, who has heard about, say Richard Feynman, in the market
place or elsewhere, may be referring to Richard Feynman even though he can’t
remember from whom he first heard of Feynman or from whom he ever heard
of Feynman. He knows that Feynman is a famous physicist. A certain passage of
communication reaching ultimately to the man himself does reach the speaker.
He then is referring to Feynman even though he can’t identify him uniquely. He
doesn’t know what a Feynman diagram is, he doesn’t know what the Feynman
theory of pair production and annihilation is. Not only that: he’d have trouble
distinguishing between Gell-Mann and Feynman. So he doesn’t have to know
these things, but, instead, a chain of communication going back to Feynman
himself has been established, by virtue of his membership in a community which
passed the name on from link to link, not by a ceremony that he makes in private
in his study: ‘By “Feynman” I shall mean the man who did such and such and such
and such’ (Kripke 1972: 91–2)
This may seem naïve, and certainly there are details to be fixed. Indeed, Kripke’s aim
is to offer not a new theory, but only “a better picture than the picture presented by
the received views” (93), “a picture which, if more details were to be filled in, might
be refined so as to give more exact conditions for reference to take place” (94). Actu-
ally, it is not so easy to fill in all the details, and it must be admitted that even today,
forty years later, we do not possess a fully blown theory built on this picture.12 But, as
a picture, it is certainly very attractive. What makes it the case that a proper name that
we use refers to something is just that it was introduced, it does not matter where and
when, for that something, and, “from link to link”, reached us who are using it. We can
say, then, that according to the picture that Kripke offered and many others have tried
ïœ±ïœ².â•… Moreover, even leaving aside the details, there are some specific semantic problems that need
to be addressed. If proper names do not have a descriptive meaning, it is quite natural to think
that their semantic contribution is exhausted by the objects they refer to. This, however, is trou-
bling both because of empty names and of the apparent non substitutability of co-referential proper
names in propositional attitude contexts. For two different attempts to solve the first problem, see
Donnellan 1974 and Braun 1993. Kripke (1979) somehow deflates the second problem.
to develop, reference, at least when proper names are concerned, is basically a histori-
cal relation, which usually originates in a dubbing.13
A way to contrast Kripke’s picture with Frege’s, Russell’s and Searle’s is to claim,
again with Kaplan, that his is a consumerist semantics, at least insofar as proper names
are concerned. According to a consumerist semantics,
we are, for the most part, language consumers. Words come to us prepackaged with a
semantic value. If we are to use those words, the words we have received, the words of
our linguistic community, then we must defer to their meaning (Kaplan 1989b: 602)
Finally, what about the intentionality intuition? To it, Wettstein (2004: 75) opposes a
“motto”, “Linguistic Contact Without Cognitive Contact”. This is perhaps too strong,
because one may reject the intentionality intuition but claim (as I did in Bianchi 2005
and 2007) that words themselves put us in cognitive contact with things. But if we
replace it with this other, “Linguistic Contact Without Cognitive Command”, then
it is fair to say that Kripke showed that it might well be true. If he is right, to refer to
something we do not need to have cognitive command of what we are referring to:
acquiring a name that was introduced for it is enough.
2.â•… Do definite descriptions refer?
2.1â•… Russell again

We have seen (Section 1.1) that Frege took definite descriptions to be Eigennamen, i.e.
expressions having as their Bedeutung a definite object. In doing this, he treated them
differently from indefinite descriptions (“an author”) and quantificational phrases
(“some authors”, “all authors”), whose Bedeutung is, in his account, a second-level con-
cept, i.e. a function from concepts to truth-values (see Frege 1892b). On the contrary,
Russell (1905) offered a uniform analysis of all these “denoting phrases”:14 they all are
incomplete symbols that do not have “any meaning in isolation” (480) but make the
ïœ±ïœ³.â•… The picture has been variously named in the subsequent literature. Donnellan (1974)
Â�advocates a version of it that he calls “the historical explanation view”. It is more common, though
perhaps misleading, to refer to it as “the causal theory of (proper) names”, as in Evans 1973 (where
an attempt is made to articulate a view coupling some aspects of it with some aspects of the
Â�description theory of names) and Devitt 1974 (where it is developed to a certain extent).
ïœ±ïœ´.â•… Russell’s article begins with the following list: “By a ‘denoting phrase’ I mean a phrase such
as any one of the following: a man, some man, any man, every man, all men, the present King of
England, the present King of France, the centre of mass of the Solar System at the first instant of
the twentieth century, the revolution of the earth round the sun, the revolution of the sun round
the earth” (479).
266 Andrea Bianchi
sentences in which they occur express a general proposition, i.e. a proposition that
does not depend for its existence on the existence of, and so is not about, any particular
object described by them.
To understand Russell’s point, consider first quantificational phrases, such as
those which occur in (1) and (2):
(1) All authors are male.
(2) Some authors are male.
As is well known, their paraphrases in a first-order language are respectively (3)

and (4):
(3) ∀x(Author x → Male x).

(4) ∃x(Author x ∧ Male x).
Russell, like Frege, believed that (3) and (4) reveal the logical forms of (1) and (2): they
show what their truth-conditions are.
Now, note that no constituent of (3) and (4) can be seen as paraphrasing the
denoting phrases in (1) (“all authors”) and (2) (“some authors”). This is basically what
Â�Russell wanted to direct our attention to when he claimed that denoting phrases have
no meaning in isolation. In fact, (3) and (4), the analyses of (1) and (2), are consti-
tuted by a quantifier and an open formula. The open formula individuates a condition
(in Russell’s terminology, a propositional function, i.e. a function from individuals to
Â�propositions). When a quantifier is prefixed to it, something is said about the con-
dition itself: in the case of the universal quantifier, that the condition is always true
(i.e. true for all values of x); in the case of the existential one, that the condition is
sometimes true (i.e. true for some values of x). No particular object is mentioned in (3)
and (4), hence in (1) and (2): all these sentences express general propositions.
Russell extended this account to indefinite descriptions, as Frege had already
done. Indeed, he claimed,
(5) An author is male,
is equivalent to (2), and so, its logical form and truth-conditions are given by (4).15
But what about singular definite descriptions and sentences containing them,
as (6)?
(6) The author of this is male.
ïœ±ïœµ.â•… “ ‘C (some men)’ will mean the same as ‘C (a man)’, and ‘C (a man)’ means ‘It is false that ‘C (x)
and x is human’ is always false’ ” (481).
Here, as anticipated, Russell’s and Frege’s ways part. The latter assimilated (6) to (7):
(7) Andrea is male.
In fact, according to Frege, “the author of this” and “Andrea” have the same object, me,
as their Bedeutung (though they may well have different senses, as we have seen): by
uttering (6) and (7) I refer to myself and ascribe to myself the property of being male;
my assertion is in both cases about a particular object, which is determined by the
sense of the name or the description. On the contrary, according to Russell, (6) is exis-
tential in form, just as (5), the only difference being that the definite article “involves
uniqueness” (481).16 Indeed, (6)’s truth-conditions are shown by (8):
(8) ∃x(x Authored this ∧ ∀y(y Authored this → y = x) ∧ (Male x),
where the universally quantified clause expresses the uniqueness requirement (no
more than one individual authored the article you are reading). Like (4), (8) is con-
stituted by an open formula, which individuates a (quite complex) condition or prop-
ositional function (being that who is male and uniquely authored this), to which an
existential quantifier is prefixed, to the effect that of the condition it is said that it is
sometimes true (i.e. true for some values of x). Given that the condition has unique-
ness built into it, if it is sometimes true, it is true of exactly one value of x, namely of
exactly one individual. If no individual authored this, or more than one individual
authored it, or the only individual that authored it is not male, (8), i.e. (6), is false. We
see, then, that according to Russell by uttering (6) I do not refer and ascribe maleness
to myself; that my assertion is not about me, but about a condition; and that what I
assert is not something particular, that is that a particular individual, me, is male, but
something general, that is that one and only one individual authored this and this
individual (whoever he or she is) is male.
Now, if there is one and only one individual that authored this, no matter whether
male or not, there is still a sense in which the definite description occurring in an
utterance of (6) bears the relation of describing to a particular individual, in fact, me,
as Russell recognizes:
Every proposition in which “the author of Waverley” occurs being explained as above,
the proposition “Scott was the author of Waverley” (i.e. “Scott was identical with the
author of Waverley”) becomes “One and only one entity wrote Waverley, and Scott
was identical with that one”; or, reverting to the wholly explicit form: “It is not always
false of x that x wrote Waverley, that it is always true of y that if y wrote Waverley y is
ïœ±ïœ¶.â•… As a matter of fact, if “Andrea” is an abbreviation of a definite description, as Russell began

to believe soon after writing ‘On denoting’ (see Section 1.2), even (7) is existential in form and
“involves uniqueness”. However, if the position of grammatical subject were occupied by a logical
proper name, the resulting sentence would express a singular rather than a general proposition.
268 Andrea Bianchi
identical with x, and that Scott is identical with x”. Thus if “C” is a denoting phrase, it
may happen that there is one entity x (there cannot be more than one) for which the
proposition “x is identical with C” is true, this proposition being interpreted as above.
We may then say that the entity x is the denotation of the phrase “C”. Thus Scott is the
denotation of “the author of Waverley” (488)
Note, however, that the relation of denotation is of no semantic import. In fact, accord-
ing to Russell’s analysis, by uttering (6) we would not say anything different if “the
author of this” denoted someone else, e.g. Paolo, or even if it did not denote anyone
(which would happen if no individual existed who authored this, or if there existed
more than one). As a consequence, “we must abandon the view that the denotation is
what is concerned in propositions which contain denoting phrases” (484).
To show the fertility of his theory, Russell tested it against some logical puzzles,
which “serve much the same purpose as is served by experiments in physical science”
(485). In particular, he dealt with the problem of true negative existential sentences
containing definite descriptions (e.g. “the revolution of the sun round the earth does
not exist”), and with that of the apparent violation of Leibniz’s law because of the
non substitutability salva veritate of a definite description denoting a particular indi-
vidual (e.g. “the author of the article you are reading”) with a proper name referring
to him (“Andrea”) in certain linguistic contexts.17 His solutions, which we cannot
examine here, are extremely elegant and powerful.18 This is, probably, one of the
main reasons why Russell’s theory soon came to be considered a “paradigm” of phi-
losophy. After more than a century, although serious criticisms have been leveled
against it and alternative accounts have been offered (see Sections 2.2 and 2.3) the
theory is still mainstream. To recap, according to this theory, definite descriptions
are not referential expressions: even when they describe something, they do not
refer to it. Indeed, the sentences where they occur are general (existentially quanti-
fied) rather than about any particular individual that may happen to be denoted by
the descriptions in them.
ïœ±ïœ·.â•… Russell focused on propositional attitude contexts (“George IV wished to know whether Scott
was the author of Waverley”), but his considerations are relevant for modal contexts as well (“It is
not necessary that Scott was the author of Waverley”), as was shown by Arthur Smullyan (1948) to
rebut some arguments by Willard Van Orman Quine against modal talk.
ïœ±ïœ¸.â•… It is also worth noting that if proper names were definite descriptions in disguise, Russell’s
solutions might be extended to the problems of true negative existential sentences containing
them (e.g. “Pegasus does not exist”), and of the apparent violation of Leibniz’s law because of the
non substitutability salva veritate of co-denoting names in certain linguistic contexts (e.g. “Baby-
lonians believed that Hesperus was not Phosphorous”). Unfortunately, as Kripke has shown (see
Section 1.5), proper names are not definite descriptions in disguise.
2.2â•… Peter Strawson

In an article published in 1950, Peter Strawson set out to show that Russell’s theory of
descriptions “embodies some fundamental mistakes” (Strawson 1950: 2).19
We started our investigation on reference by saying that proper names refer; that
is, we took reference as a relation that certain kinds of expressions (e.g. “Andrea”) bear
to objects (e.g. me), though sometimes we have also said that people refer by using
certain kinds of expressions. This way of putting things is common to all the philoso-
phers we have considered so far, and seems to be harmless, even if some caution is
required in dealing with indexical expressions (e.g. “I”, “this”). However, according to
Strawson, this is already a mistake: “ ‘Mentioning’, or ‘referring’, is not something an
expression does; it is something that someone can use an expression to do” (Strawson
1950: 6). We shall come back to this general issue in Section 3. As a consequence of
this, Â�Strawson distinguishes meaning and reference. In fact, meaning is something
that expressions themselves possess. Hence, “[t]he meaning of an expression cannot
be identified with the object it is used, on a particular occasion, to refer to” (7). On
the contrary, “to talk about the meaning of an expression … is … to talk … about the
rules, habits, conventions governing its correct use, on all occasions”. And, in particu-
lar, “to give the meaning of an expression [which has a uniquely referring use] is to
give Â�general directions for its use to refer to or mention particular objects or persons”.
Having said this, let us have a look at the expressions which, according toÂ�
Strawson, have a uniquely referring use. Here is the very first paragraph of his article:
We very commonly use expressions of certain kinds to mention or refer to some
individual person or single object or particular event or place or process, in the course
of doing what we should normally describe as making a statement about that person,
object, place, event, or process. I shall call this way of using expressions the ‘uniquely
referring use’. The classes of expressions which are most commonly used in this way
are: singular demonstrative pronouns (‘this’ and ‘that’); proper names (e.g. ‘Venice’,
‘Napoleon’, ‘John’); singular personal and impersonal pronouns (‘he’, ‘she’, ‘I’, ‘you’,
‘it’); and phrases beginning with the definite article followed by a noun, qualified or
unqualified, in the singular (e.g. ‘the table’, ‘the old man’, ‘the king of France’). Any
expression of any of these classes can occur as the subject of what would traditionally
be regarded as a singular subject-predicate sentence; and would, so occurring,
exemplify the use I wish to discuss (1)
Strawson’s list of expressions is quite similar to those which Frege and Searle had in
mind. In particular, for Strawson too both proper names and definite descriptions may
ïœ±ïœ¹.â•… Just for the record, Russell was not exactly impressed by Strawson’s criticisms. In a reply
to Strawson’s article, he wrote: “I am totally unable to see any validity whatever in any of
Mr. Â�Strawson’s arguments” (Russell 1957: 385).
270 Andrea Bianchi
be used to refer. So, with regard to definite descriptions, Strawson’s position is very dif-
ferent from that of Russell, according to whom they are used to assert the existence of
something satisfying some condition with uniqueness built into it. This does not mean
that the existence and uniqueness requirements play no role in Strawson’s account.
Indeed, when we use a definite description to refer to something, we presuppose that
there exists some thing, and no more than one thing, to which we are referring. But
we are not asserting what we are presupposing; on the contrary, we are referring to
Â�something, by using a significant expression, to assert something about it.
The difference between Strawson’s account and Russell’s becomes evident when a
definite description happens not to (be used to) describe anything, as in (a use by me
of) “The king of France is bald”. According to Russell, as we have seen, in such a case
we should say that the sentence is false. Instead, for Strawson, as for Frege, in using the
sentence I say something which lacks a truth-value, because what I presuppose when
using it is not the case:
when we utter the sentence without in fact mentioning anybody by the use of the
phrase, ‘The king of France’, the sentence does not cease to be significant: we simply
fail to say anything true or false because we simply fail to mention anybody by this
particular use of that perfectly significant phrase. It is, if you like, a spurious use
of the sentence, and a spurious use of the expression; though we may (or may not)
mistakenly think it a genuine use (9–10)
It remains to be seen what, according to Strawson, makes it the case that, when we
use a definite description to refer to something, we do refer to it. Strawson does not
address this issue directly. However, he writes:
In the case of phrases of the form ‘the so-and-so’ used referringly, the use of ‘the’
together with the position of the phrase in the sentence (i.e. at the beginning, or
following a transitive verb or preposition) acts as a signal that a unique reference is
being made; and the following noun, or noun and adjective, together with the context
of utterance, shows what unique reference is being made (16)
The last sentence of this passage strongly suggests that Strawson thinks that what
a definite description refers to is determined by the condition individuated by the
predicate following the article. If this is so, his account of reference ends by being
quite similar to that of Frege, at least as far as definite descriptions are concerned:
by using them, we refer by describing, in that the entity referred to, if any, is that
described by the description. However, Strawson introduces an important element
of novelty: in many cases, the context of utterance is seen to play a major role. Indeed,
quite often the condition individuated by the predicate occurring in the description
is insufficient to determine a unique reference. Consider, for example, the sentence
“The table is covered with books”. According to Strawson, “[i]t is quite certain that
in any normal use of this sentence the expression ‘the table’ would be used to make a
unique reference, i.e. to refer to some one table” (11). But it is even more certain that
the condition of being a table is satisfied by more than one entity. How is it, then, that
by using the description we succeed in referring to something in particular? Here,
the context in which the sentence is uttered cooperates with the predicate occurring
in the description to select a specific table: the one which is somehow salient in the
context itself.
Even if one doubts Strawson’s positive account, it is generally agreed that the
Â�phenomenon Strawson put his finger on in his last example, that of the quite widespread
use of incomplete (also called “ambiguous” (Devitt 1974: 195), “improper” (Kripke
1977: 255), “imperfect” (Devitt 1981: 517), “indefinite” (Donnellan 1968: 204n))
Â�definite descriptions, is difficult to accommodate in Russell’s theory. Kripke, for
Â�example, whose defense of it against Donnellan’s criticisms we shall briefly mention
in Section 3, writes:
If I were to be asked for a tentative stab about Russell, I would say that although
his theory does a far better job of handling ordinary discourse than many have
thought, and although many popular arguments against it are inconclusive, probably
it ultimately fails. The considerations I have in mind have to do with the existence
of “improper” definite descriptions, such as “the table,” where uniquely specifying
conditions are not contained in the description itself. Contrary to the Russellian
picture, I doubt that such descriptions can always be regarded as elliptical with some
uniquely specifying conditions added (Kripke 1977: 255)20
2.3â•… Keith Donnellan

In 1966, Donnellan published an article where he argued against both Russell and
Strawson. Neither of them, in fact, realized that definite descriptions have two distinct
functions or uses, which Donnellan calls the attributive and the referential. Accord-
ing to him, “one or the other of the two views, Russell’s or Strawson’s, may be correct
about the nonreferential [i.e. attributive] use of definite descriptions, but neither fits
the Â�referential use” (Donnellan 1966: 283).
ïœ²ïœ°.â•… At the end of his article, Kripke reiterates the point: “it seems to me to be likely that Â�‘indefinite’
definite descriptions such as ‘the table’ present difficulties for a Russellian analysis”; in fact, “[i]t
is somewhat tempting to assimilate such descriptions to the corresponding demonstratives (for
example, ‘that table’)” (271). Assimilation like this is urged in Wettstein 1981, where the phenoÂ�
menon of indefinite definite descriptions and the difficulties it presents for Russell’s theory are
discussed in more detail. See also Kaplan 1978, Devitt 1981 and 2004, and Wettstein 1983. Russell’s
theory is defended against such criticisms in Bach 1981: 237–40 and 2004: 220–3, Neale 1990: sec.
3.7, and Soames 2005.
272 Andrea Bianchi
Here is how Donnellan characterizes the two uses:

A speaker who uses a definite description attributively in an assertion states something
about whoever or whatever is the so-and-so. A speaker who uses a definite description
referentially in an assertion, on the other hand, uses the description to enable his
audience to pick out whom or what he is talking about and states something about
that person or thing. In the first case the definite description might be said to occur
essentially, for the speaker wishes to assert something about whatever or whoever fits
that description; but in the referential use the definite description is merely one tool
for doing a certain job – calling attention to a person or thing – and in general any
other device for doing the same job, another description or a name, would do as well.
In the attributive use, the attribute of being the so-and-so is all important, while it is
not in the referential use (285)
Donnellan illustrates the distinction with an example:
consider the sentence, “Smith’s murderer is insane”. Suppose first that we come upon
poor Smith foully murdered. From the brutal manner of the killing and the fact that
Smith was the most lovable person in the world, we might exclaim, “Smith’s murderer
is insane”. I will assume, to make it a simpler case, that in a quite ordinary sense we do
not know who murdered Smith (though this is not in the end essential to the case).
This, I shall say, is an attributive use of the definite description.
The contrast with such a use of the sentence is one of those situations in which we
expect and intend our audience to realize whom we have in mind when we speak of
Smith’s murderer and, most importantly, to know that it is this person about whom
we are going to say something.
For example, suppose that Jones has been charged with Smith’s murder and has
been placed on trial. Imagine that there is a discussion of Jones’s odd behavior
at his trial. We might sum up our impression of his behavior by saying, “Smith’s
murderer is insane”. If someone asks to whom we are referring, by using this
description, the answer here is “Jones”. This, I shall say, is a referential use of the
definite description (285–6)
Now, it is quite clear that Russell did not consider Donnellan’s referential uses, or at
least did not take them as semantically relevant: according to him, as we have seen,
by using a sentence containing a definite description we express a general propoÂ�
sition, and we do not say anything about any particular object or person that may
happen to be denoted by the description in it. But, one may ask, why did Donnellan
claim that Strawson did not account for them either? Didn’t Strawson (or, for that
matter, Frege) point out, as opposed to Russell, that by using a definite description,
we standardly refer to a particular object or person about which we assert some-
thing? Donnellan’s answer is that Strawson failed to realize that when a description
is used referentially, what it refers to is not determined by the condition Â�individuated
by the predicate occurring in it. Indeed, when the condition is made to play this
determinative role, the description can be used, at most, to talk about whatever fits
it (in the context of utterance), rather than the particular the speaker wishes to talk
about, and so, is used attributively.
To develop this point and contrast vividly Russell’s, Strawson’s and Donnellan’s
views, let us go back to the discussion of Jones’s odd behavior at his trial. Let us sup-
pose that Jones is actually insane, as is manifested by his behavior, but that he is not
Smith’s murderer. In situation a, Smith was murdered by Adams, a very astute and not
at all insane criminal who has not been exposed. In situation b, he was not murdered
at all (because, let us say, he committed suicide). Now, let us ask: Would we say some-
thing true or something false if we uttered “Smith’s murderer is insane” in one of the
two situations? According to Russell, we would say something false on both occasions,
because either there would be no one who murdered Smith (situation b), or the per-
son who murdered Smith (in fact, Adams) would not be insane (situation a). Accord-
ing to Strawson, we would say something false in situation a, because the person we
referred to by our use, that is the one (Adams again) who satisfied the condition indi-
viduated by the predicate occurring in the description, murdering Smith, would not be
insane; and we would say something that would be neither true nor false in situation b,
because our use would, in his terminology, be spurious (even though we would think
it genuine) – we would not be referring to anyone, given that no one would satisfy the
condition. According to Donnellan, on the contrary, we would say something true
both in situation a and in situation b, because the person we want to talk about and,
in fact, succeed in talking about, Jones, is precisely as we stated, insane. Indeed, “the
description is here merely a device for getting one’s audience to pick out or think of the
thing to be spoken about, a device which may serve its function even if the description
is incorrect” (303–4); and this is so because “in the referential use as opposed to the
attributive, there is a right thing to be picked out by the audience and it being the right
thing is not simply a function of its fitting the description” (304).
To sum up, a definite description in referential use may refer to something while
describing something else (situation a) or nothing at all (situation b). If this is true, we
cannot but conclude, first, that referring and describing are different relations, and,
second, that not only proper names, as Kripke has shown (see Section 1.5), but also
definite descriptions do not refer by describing: once again, the notion of reference
cannot be explained in terms of satisfying a condition. As for the first point, Â�Donnellan
is quite explicit:
Russell’s definition of denoting (a definite description denotes an entity if that
entity fits the description uniquely) is clearly applicable to either use of definite
descriptions. Thus whether or not a definite description is used referentially or
attributively, it may have a denotation. Hence, denoting and referring, as I have
explicated the latter notion, are distinct and Russell’s view recognizes only the former.
274 Andrea Bianchi
It seems to me, moreover, that this is a welcome result, that denoting and referring
should not be confused (293)
Indeed, definite descriptions in referential use seem to be more akin to Russell’s

logical proper names (see Section 1.2), which “refer to something without ascrib-
ing any properties to it” (302), than to Russell’s definite descriptions, which serve
to express general propositions. Donnellan, in fact, goes as far as to say that
“[m]any of the things said about proper names [in “the narrow logical sense”] by
Russell can … be said about the referential use of definite descriptions without strain-
ing senses unduly” (282).
But what, if not the lexical material contained in it, makes it the case that a definite
description in referential use refers to what it refers to? Unfortunately, Â�Donnellan’s
answer to this question is much less articulated than it should be. From some of his
considerations, it seems that according to him referring is a function of having in mind.
What having in mind amounts to, and how we get from having in mind to reference,
however, is left in the dark. Perhaps, we may say that having in mind is a cognitive rela-
tion, which, in Donnellan’s account, plays the role that acquaintance played in Russell’s
account of the reference of logical proper names. And perhaps, we may add that refer-
ence is a historical relation, because we can refer only to what has entered our mind (we
refer back to it, so to speak). Nonetheless, it is important to note the difference between
Donnellan’s views and Kripke’s picture of the reference of proper names. While in
both cases reference is taken to be a historical relation, the history that Â�Donnellan’s
directs our attention to is the cognitive history of the speaker, and does not involve
“other members of the language community”, to use Kaplan’s words. Because of this,
Â�Donnellan’s semantic views run the risk of being accused of subjectivism.21 This is
probably one of the reasons which drove Kripke to react to them in the way we shall
briefly examine in Section 3.
3.â•… One further issue: Semantics versus pragmatics
According to Kripke, “the distinction Donnellan brings out exists and is of fundamen-
tal importance” (Kripke 1977: 257): there are attributive uses and there are referential
uses of definite descriptions. Indeed, the distinction may even be generalized: there
are attributive uses and referential uses of proper names as well. But Donnellan is
wrong in claiming that his considerations refute Russell’s theory. In fact, Donnellan’s
ïœ²ïœ±.â•… Donnellan (1968) tries to defend himself from the accusation, which was actually brought
against him by Alfred MacKay (1968).
Â�
distinction is not a semantic one. It is, rather, a pragmatic one, and as such is fully com-
patible with Russell’s account of the semantics of definite descriptions.22
To show this, Kripke distinguishes, “following Grice” (Kripke 1977: 262), two dif-
ferent notions, which are those that give the title to his article: semantic reference and
speaker’s reference. He illustrates them by means of an example involving proper names:
Two people see Smith in the distance and mistake him for Jones. They have a brief
colloquy: “What is Jones doing?” “Raking the leaves”. “Jones”, in the common language
of both, is a name of Jones; it never names Smith. Yet, in some sense, on this occasion,
clearly both participants in the dialogue have referred to Smith, and the second
participant has said something true about the man he referred to if and only if Smith
was raking the leaves (whether or not Jones was) (263)
According to Kripke, in this example “Jones, the man named by the name, is the
semantic referent”, while “Smith is the speaker’s referent, the correct answer to the
question, ‘To whom were you referring?’ ” (264). Of semantic reference, Kripke says
only that “[if] a speaker has a designator in his idiolect, certain conventions of his
idiolect (given various facts about the world) determine the referent in the idiolect”
(263). In the case of proper names, we may simply assume that semantic reference is
the historical relation we discussed in Section 1.5.23 On the contrary,
the speaker’s referent of a designator [is] that object which the speaker wishes to talk
about, on a given occasion, and believes fulfills the conditions for being the semantic
referent of the designator. He uses the designator with the intention of making an
assertion about the object in question (which may not really be the semantic referent,
if the speaker’s belief that it fulfills the appropriate semantic conditions is in error).
The speaker’s referent is the thing the speaker referred to by the designator, though it
may not be the referent of the designator, in his idiolect (264)
ïœ²ïœ².â•… For a defense of the idea that the distinction is semantically significant, see Donnellan 1978,
Wettstein 1981 and 1983, Devitt 1981 and 2004, and Amaral 2008. Note, however, that Wettstein’s
and Devitt’s accounts are different from Donnellan’s, in that they do not allow a definite descrip-
tion in referential use to refer to something which does not fit it. Indeed, they tend to assimi-
late definite descriptions in referential uses to demonstrative descriptions (see Note 20). So, their
Â�accounts appear to be Strawsonian more than Donnellanian. On the contrary, Bach 1981 and 2004,
Salmon 1982, 1991 and 2005, Neale 1990: chap. 3, and Soames 1994 and 2005: 388–94 side with
Kripke and argue that the distinction is to be accounted in pragmatic terms. Kripke’s approach to
Donnellan’s distinction is somehow anticipated, in passing, by Paul Grice (1969: 141–4).
ïœ²ïœ³.â•… Indeed, Kripke writes: “If the views about proper names I have advocated in ‘Naming and
Necessity’ are correct …, the conventions regarding names in an idiolect usually involve the fact
that the idiolect is no mere idiolect, but part of a common language, in which reference may be
passed from link to link” (273 n. 20).
276 Andrea Bianchi
Now, notice, first, that the notion of semantic reference can easily be extended to
cover Russell’s definition of denotation, and, second, that the above characterization
makes speaker’s reference somehow parasitic on semantic reference. Given all this, we
can finally see how with the help of his two notions Kripke purports to account for
Â�Donnellan’s distinction in a way which is at least compatible with Russell’s semantic
theory of descriptions:24
In a given idiolect, the semantic referent of a designator (without indexicals) is
given by a general intention of the speaker to refer to a certain object whenever the
designator is used. The speaker’s referent is given by a specific intention, on a given
occasion, to refer to a certain object. If the speaker believes that the object he wants to
talk about, on a given occasion, fulfills the conditions for being the semantic referent,
then he believes that there is no clash between his general intentions and his specific
intentions. My hypothesis is that Donnellan’s referential-attributive distinction should
be generalized in this light. For the speaker, on a given occasion, may believe that his
specific intention coincides with his general intention for one of two reasons. In one
case (the “simple” case), his specific intention is simply to refer to the semantic referent:
that is, his specific intention is simply his general semantic intention. (For example, he
uses “Jones” as a name of Jones – elaborate this according to your favorite theory of
proper names – and, on this occasion, simply wishes to use “Jones” to refer to Jones.)
Alternatively – the “complex” case – he has a specific intention, which is distinct from
his general intention, but which he believes, as a matter of fact, to determine the same
object as the one determined by his general intention. (For example, he wishes to
refer to the man “over there” but believes that he is Jones.) In the “simple” case, the
speaker’s referent is, by definition, the semantic referent. In the “complex” case, they
may coincide, if the speaker’s belief is correct, but they need not. (The man “over
there” may be Smith and not Jones.) (Kripke 1977: 264)
Given all this, Kripke’s hypothesis is “that Donnellan’s ‘attributive’ use is nothing but
the ‘simple’ case, specialized to definite descriptions, and that the ‘referential’ use is,
similarly, the ‘complex’ case” (264).
Kripke’s hypothesis is certainly intriguing, also because it promises to account for
some important linguistic phenomena without making semantic reference a function
of the speaker’s mind. However, it does not come without problems. One may be wor-
ried, for example, by its conspicuous appeal to intentions (general as well as specific).
In any event, it is not within the scope of the present article to evaluate this. What is
ïœ²ïœ´.â•… Kripke substantiates the last point by considering some hypothetical languages which are
as similar as possible to English except that Russell’s theory is stipulated to be correct for them:
“[s]ince the phenomenon Donnellan cites would arise in all the Russell languages, if they were
spoken, the fact they do arise in English, as actually spoken, can be no argument that English
is not a Russell language” (Kripke 1977: 266). Unfortunately, I cannot go into the details here.
The argument, however, is challenged in Devitt 2004: 287–8.
important to note here is rather that, first, thanks to Kripke, the notions of semantic
reference and speaker’s reference, which were quite often confused in the literature
about reference before his work, are now there to be further investigated; and, second,
a claim about their relationship, which makes the latter parasitic on the former, has
been advanced by him. This claim can be rephrased as a claim about the relationship
between semantics and pragmatics, at least as far as the phenomenon of reference is
concerned, and it certainly deserves to be discussed.25
References
Acero, J.J. & P. Leonardi (eds.) (2005). Facets of concepts. Poligrafo.

Amaral, F.S. (2008). Definite descriptions are ambiguous. Analysis 68: 288–97.
Bach, K. (1981). Referential/attributive. Synthese 49: 219–44.
Bach, K. (2004). Descriptions: points of reference. In Reimer & Bezuidenhout 2004.
Beaney, M. (1997a). Introduction. In Beaney 1997b.
Beaney, M. (ed.) (1997b). The Frege reader. Blackwell.
Beaney, M., C. Penco & M. Vignolo (eds.) (2007). Explaining the mental. Naturalist and Â�non-Â�naturalist
approaches to mental acts and processes. Cambridge Scholars Publishing.
Bianchi, A. (2005). Words as concepts. In Acero & Leonardi 2005.
Bianchi, A. (2007). Speaking and thinking (or: a more Kaplanian way to a unified account ofÂ�
language and thought). In Beaney, Penco & Vignolo 2007.
Braun, D. (1993). Empty names. Noûs 27: 449–69.
Carnap, R. (1947). Meaning and necessity. A study in semantics and modal logic. University of Chicago
Press. Second edition 1956.
Cole, P. (ed.) (1978). Syntax and semantics, Vol. 9: Pragmatics. Academic Press.
Davidson, D. & G. Harman (eds.) (1972). Semantics of natural language. Reidel.
Davidson, D. & J. Hintikka (eds.) (1969). Words and objections. Essays on the work of W.V. Quine.
Reidel.
Devitt, M. (1974). Singular terms. Journal of Philosophy 71: 183–205.
Devitt, M. (1981). Donnellan’s distinction. Midwest Studies in Philosophy 6: 511–24.
Devitt, M. (2004). The case for referential descriptions. In Reimer & Bezuidenhout 2004.
Donnellan, K.S. (1966). Reference and definite descriptions. Philosophical Review 75: 281–304.
Donnellan, K.S. (1968). Putting Humpty Dumpty together again. Philosophical Review 77: 203–15.
Donnellan, K.S. (1970). Proper names and identifying descriptions. Synthese 21: 335–58.
Donnellan, K.S. (1974). Speaking of nothing. Philosophical Review 83: 3–31.
Donnellan, K.S. (1978). Speaker reference, descriptions and anaphora. In Cole 1978.
Evans, G. (1973). The causal theory of names. Aristotelian Society Supplementary Volume 47: 187–208.
Reprinted in Evans 1985.
Evans, G. (1985). Collected Papers. Clarendon Press.
ïœ²ïœµ.â•… I would like to thank Antonio Capuano, Andrea Marino, and especially Paolo Leonardi,
Ernesto Napoli, and Marco Santambrogio for their comments on previous versions of this work.
278 Andrea Bianchi
Frege, G. (1892a). Über Sinn und Bedeutung. Zeitschrift für Philosophie und philosophische Kritik
100: 25–50. Translated as On Sinn and Bedeutung in Beaney 1997b.
Frege, G. (1892b). Über Begriff und Gegenstand. Vierteljahrsschrift für wissenschaftliche Philosophie
16: 192–205. Translated as On concept and object in Beaney 1997b.
Frege, G. (1918–19). Der Gedanke. Eine logische Untersuchung. Beiträge zur Philosophie des
deutschen Idealismus 1: 58–77. Translated as Thought in Beaney 1997b.
Grice, H.P. (1969). Vacuous names. In Davidson & Hintikka 1969.
Gunderson, K. (ed.) (1975). Language, mind and knowledge. University of Minnesota Press.
Kaplan, D. (1978). Dthat. In Cole 1978.
Kaplan, D. (1989a). Demonstratives: An essay on the semantics, logic, metaphysics, and epistemol-
ogy of demonstratives and other indexicals. In Almog, Perry & Wettstein 1989.
Kaplan, D. (1989b). Afterthoughts. In Almog, Perry & Wettstein 1989.
Kaplan, D. (2005). Reading ‘On denoting’ on its centenary. Mind 114: 933–1003.
Kripke, S.A. (1972). Naming and necessity. In Davidson & Harman 1972. Reprinted with a new
introduction as Naming and necessity. Blackwell, 1980.
Kripke, S.A. (1977). Speaker’s reference and semantic reference. Midwest Studies in Philosophy 2:
255–76.
Kripke, S.A. (1979). A puzzle about belief. In Margalit 1979.
MacKay, A.F. (1968). Mr. Donnellan and Humpty Dumpty on referring. Philosophical Review 77:
197–202.
Margalit, A. (ed.) (1979). Meaning and use. Papers presented at the Second Jerusalem Philosophical
Encounter April 1976. Reidel.
Neale, S. (1990). Descriptions. MIT Press.
Putnam, H. (1975a). The meaning of ‘meaning’. In Gunderson 1975. Reprinted in Putnam 1975b.
Putnam, H. (1975b). Mind, language and reality. Philosophical papers, volume 2. Cambridge
Reimer, M. & A. Bezuidenhout (eds.) (2004). Descriptions and beyond. Clarendon Press.
Russell, B. (1905). On denoting. Mind 14: 479–93.
Russell, B. (1911). Knowledge by acquaintance and knowledge by description. Proceedings of the
Aristotelian Society 11: 108–28. Reprinted in Russell 1918.
Russell. B. (1918). Mysticism and logic. Longmans, Green e Co. Reprinted by Dover in 2004.
Russell, B. (1918–19). The philosophy of logical atomism. The Monist 28: 495–527 e 29: 32–63,
190–222 e 345–80. Reprinted in Russell 1956.
Russell, B. (1956). Logic and knowledge. Essays 1901–1950. Edited by R.C. Marsh. Allen & Unwin.
Reprinted by Routledge in 1992.
Russell, B. (1957). Mr. Strawson on referring. Mind 66: 385–9.
Salmon, N. (1982). Assertion and incomplete definite descriptions. Philosophical Studies 42: 37–45.
Reprinted in Salmon 2007.
Salmon, N. (1991). The pragmatic fallacy. Philosophical Studies 63: 83–97. Reprinted in Salmon 2007.
Salmon N. (2005). Two conceptions of semantics. In Szabo 2005. Reprinted in Salmon 2007.
Salmon, N. (2007). Content, cognition, and communication. Philosophical papers, volume 2. Oxford
University Press.
Searle, J.R. (1958). Proper names. Mind 67: 166–73.
Smullyan, A.F. (1948). Modality and description. Journal of Symbolic Logic 13: 31–7.
Soames, S. (1994). Donnellan’s referential/attributive distinction. Philosophical Studies 73: 149–68.
Reprinted in Soames 2009.
Soames, S. (2005). Why incomplete definite descriptions do not defeat Russell’s theory of descrip-
tions. Teorema 24: 7–30. Reprinted in Soames 2009.
Soames, S. (2009). Philosophical essays, volume 1. Natural language: what it means and how we use it.
Princeton University Press.
Strawson, P.F. (1950). On referring. Mind 59: 320–44. Reprinted in Strawson 1971.
Strawson, P.F. (1971). Logico-linguistic papers. Ashgate (Second edition 2004).
Szabó, Z.G. (ed.) (2005). Semantics versus pragmatics. Clarendon Press.
Wettstein, H.K. (1981). Demonstrative reference and definite descriptions. Philosophical Studies 40:
241–57. Reprinted in Wettstein 1991.
Wettstein, H.K. (1983). The semantic significance of the referential-attributive distinction.
Â�Philosophical Studies 44: 187–96. Reprinted in Wettstein 1991.
Wettstein, H.K. (1991). Has semantics rested on a mistake? And other essays. Stanford University
Press.
Wettstein, H.K. (2004). The magic prism. An essay in the philosophy of language. Oxford University
Press.
Truth-conditional semantics
Robyn Carston
University College London
1.â•… The basics
Truth-conditional semantics is a theory of the meaning of natural language sentences.

It takes the language–world relation as the basic concern of semantics rather than the
language–mind relation: language is about states of affairs in the world. The semantic
competence of a speaker–hearer is said to consist in his/her knowledge, for any sentence
of his/her language, of how the world would have to be for that sentence to be true.
The meaning of sentences is given by a theory of the truth-conditions of those
sentences. That is, the theory consists of a set of T-sentences in a metalanguage, one
for each sentence in the language under analysis (the object language). The general
schema for T-sentences is given in (1), where S is the name of a sentence in the
object language and P is the conditions which guarantee the truth of that sentence,
given by a sentence in the metalanguage. Particular instantiations of the schema are
given in (2):
(1) S is true if and only if P
(2) a. ‘Snow is white’ is true iff snow is white.
b. ‘La neige est verte’ is true iff snow is green.
c. ‘Ex(Fx)’ is true iff there exists something which has the property ‘F’.
In examples (2b) and (2c) the object language and the metalanguage (the language
in which the T-theory is couched) are different from each other. There is nothing
essential about this, but recognising it might help to prevent an important misunder-
standing that can arise from looking only at cases like (2a) where the object language
and the metaÂ�language are the same: T-sentences do not state a (trivial) relation that a
sentence, say ‘Snow is white’, has to itself, that it is true just when it is true. Rather, they
state a semantic property of the sentence in the object language by telling us under
what conditions it is true.
The sentences in quotation marks on the left hand side of the formulas are uninter-
preted strings of the object language and their interpretation is given on the right hand
side by a set of conditions minimally guaranteeing their truth. That is, the sentence
on the right hand side gives the meaning of the sentence in quotes on the left hand
side, as meaning in this theory is equated with truth-conditions. Because the object
language sentences are being mentioned (quoted) while the metalanguage expressions
are being used, there is no problem in a language being used metalinguistically in the
formulation of its own truth-theory, as is the case for (2a).
There is, however, an obvious inadequacy in the formulation so far. The following
are plainly true but do not go any way towards giving an account of the meaning of the
object language sentences:
(3) a. ‘Snow is white’ is true iff cows have four legs.
b. ‘La neige est verte’ is true iff worms can fly.
The theory must be so constrained that these random couplings of object language
sentences with metalanguage sentences that happen to have the same truth value do
not arise. The necessary constraints may arise in meeting the further requirement that
for any object language with the property of recursion, hence for any natural language,
the theory must account for an infinite set of sentences. The way to achieve this is
through a finite list of axioms, one for each word of the language and one for each of
the ways of combining words, and at least some of these combinatory axioms must be
recursive. Examples of axiom types are given in (4):
(4) a. Reference axioms for names:
e.g. ‘Sigmund Freud’ refers to Sigmund Freud.
b. Satisfaction axioms for predicates:
e.g. ‘Intelligent’ is satisfied by intelligent things.
c. Connection axioms for syntactic structures:
e.g. ‘a is b’ is true iff what ‘a’ refers to satisfies ‘b’.
e.g. ‘P and Q’ is true iff ‘P’ is true and ‘Q’ is true.
The first three axioms together yield as a theorem the following T-sentence:
(5) ‘Sigmund Freud is intelligent’ is true iff Sigmund Freud is intelligent.
The correct set of axioms will potentially deliver as theorems the full (infinite) set of the
T-sentences, each one of which mentions one member of the (infinite) set of sentences
of the language whose meaning is under analysis. We can now see how the second aim,
that of avoiding the absurd unrevealing T-sentences given in (3), may be realized by this
approach. The reference axioms for ‘snow’ and for ‘la neige’ and the satisfaction axioms
for ‘white’ and for ‘verte’ should ensure that these are not generated as theorems.
It remains to point out that truth must be relativized to a particular language since
it is, in principle, possible that two different languages could have elements, including
whole sentences, in common. Suppose, for instance, the (hypothetical) language Zega-
rese has a sentence ‘Snow is white’ which means, in that language, ‘Alcohol is essential’.
It is evident then that a sentence is not simply true or false, but is true or false in a
particular language and the T-sentences must reflect this:
(6) a. ‘Snow is white’ is true in English iff snow is white.

b. ‘Snow is white’ is true in Zegarese iff alcohol is essential.
282 Robyn Carston
Finally, let us return to the semantic competence of the native speaker of a language.
This competence, or knowledge, is often said to be reflected in his/her capacity to
make certain judgements about sentences in his/her native language. For instance,
English speakers know that there is some kind of identity relation between the pair of
sentences in (7) and that there is some kind of inclusion relation between the pair of
sentences in (8):
(7) a. Jane sold Sally a computer.
b. Sally bought a computer from Jane.
(8) a. Pat is a bachelor.
b. Pat is a man.
If knowledge of meaning is taken to be knowledge of truth-conditions, the ability

to make these particular judgements is easily explained: we judge two sentences
as synonymous (identical in meaning) if their truth-conditions are such that they
must be true in the same circumstances (and false in the same circumstances); we
judge one sentence to imply, or entail, another if their truth-conditions are such
that for every circumstance in which the first is true so is the second. Knowledge
of other semantic properties such as ambiguity, anomaly and contradiction can be
similarly accounted for in truth-conditional terms. (References which give general
introductions are Blackburn 1984; Devitt & Sterelny 1987; Kempson 1977; Larson
1990; Martin 1987).
2.â•… Short history and key names
The context in which the truth-theoretic approach to semantics arose was that of
the nineteenth century investigations into the interpretation of mathematical logic
and other formal systems. The work of Gottlob Frege was foundational. One of his
ideas was that these formal systems could be given an interpretation, that is, could be
assigned meaning, by providing each sentence of the system with truth-conditions. He
is also credited with the principle of compositionality which, as we have seen, is central
to the project of formulating a truth theory for a language with an infinite number of
sentences. The logician Alfred Tarski is celebrated for his definition of the concept
of truth for formal logical languages. The schema in (1) is standardly known as the
Tarskian formula. He was unsure, though, about the possibility of applying his truth
definition to natural languages. It was the philosopher of language Donald Davidson
who built on the work of Frege and Tarski in providing the sort of axiomatic truth-
conditional semantics for natural language outlined above. In parallel with Davidson’s
work, Montague developed the model-theoretic embodiment of the truth-conditional
approach to the semantics of natural language. This involves the construction of
abstract mathematical models of those things in the world which constitute the
semantic values of the language expressions.
A further development of the truth-conditional approach was the addition of
‘possible worlds’ into the theory. These were originally introduced by Saul Kripke to
investigate formal systems incorporating necessity and possibility operators. This can
be carried into natural language semantics:
(9) a. ‘Necessarily, horses are mammals’ is true iff ‘horses are mammals’ is true in
every possible world.
b. ‘Possibly, Mary is dancing’ is true iff ‘Mary is dancing’ is true in at least one
possible world.
Incorporating possible worlds into the semantics is a way of reflecting our knowledge
that language can express ways in which the world might be, as well as the way it is.
Montague, Lewis and others have used possible worlds in giving the truth-conditional
semantics of sentences which contain modals, counterfactual conditionals and propo-
sitional attitudes. They have made the further addition of a set of ordered times in
order to capture the semantics of the different tenses and aspects which sentences of
natural languages manifest. (Landmark references are Frege 1892, 1977 [1918]; Â�Tarski
1944; Davidson 1967; Lewis 1970; Montague 1974. Helpful secondary sources are
Dowty, Wall & Peters 1981; Chierchia & McConnell-Ginet 1990; Evnine 1991).
3.â•… Brief comparison with other approaches to linguistic meaning
Very broadly speaking, twentieth century philosophers working on the semantics of

natural language have taken one of two approaches: (a) the truth-conditional, or (b)
the intention-convention. While the language–world connection is fundamental on
the first approach, the language–mind connection is central on the second, in which
linguistic meaning is accounted for in terms of psychological entities such as inten-
tions and beliefs. Grice’s theory of meaning (to be distinguished from his pragmatic
‘theory of conversation’) is the primary exemplar of this second approach: the mean-
ing of a sentence here is analyzed in terms of regularities over the intentions with
which utterers produce the sentence on different occasions. The two approaches are
very different and yet, as Strawson (1971 [1969]) indicated, they are not incompatible.
They will probably both prove necessary to a picture of meaning which aspires to a
fully adequate account of the relations between public language systems, individual
thought and the external world.
Within early generative linguistics, a third approach to natural language semantics
developed. Generative linguistics takes as its subject matter a speaker–hearer’s linguis-
tic competence: phonological, syntactic, semantic. This knowledge is characterised as
284 Robyn Carston
systems of mental representations and rules for transforming one sort of representa-
tion into another. Hence, within this framework it has been natural to assume a level of
semantic representation into which (some level of) lexico-syntactic representation is
mapped or ‘translated’. A semantic representation consists of a structured set of seman-
tic markers or conceptual primitives (Katz 1972; J.A. Fodor 1975; Jackendoff 1983).
Like the truth-conditional approach this one meets the usual conditions placed
on a semantic theory: namely, that it should provide an account of the native speaker’s
knowledge of such semantic properties and relations of his/her language as (a) anom-
aly (b) ambiguity, (c) synonymy, (d) contradiction, (e) entailment, etc. However, the
fundamental failing of these ‘translational’ accounts, which Lewis (1970) and other
truth-conditionalists have pointed out, is the following:
one could have a competence thus based upon the mapping relation, and yet not know
what a single sentence of the language meant. A speaker–hearer would know that only
if he knew what sentences of the theory’s language [markerese, conceptual primitives]
meant; but this is knowledge of precisely the kind that was to be accounted for in the
first place. (Evans & McDowell 1976: ix).
Again, there are those who would say that there is a place for both accounts, the trans-
lational and the truth-conditional. For instance, a theory whose primary concern is
utterance understanding may well require a psychological level of logico-semantic
representation (in a language of thought, perhaps) into which natural language forms
are mapped, while also recognising that, ultimately, the representations must be inter-
preted, that is, related to the world via truth-conditions. This view is central to the
relevance-theoretic approach to natural language semantics (see Sperber & Wilson
1986; Blakemore 1987; Carston 1988; and Wilson & Sperber 1988a, 1993). It follows
from the richly articulated program of J.A. Fodor (1975, 1987, 1990) in the philosophy
of mind and language.
Finally, it should be noted that in recent years there has been important work on
incorporating a Davidsonian truth-conditional approach to natural language seman-
tics into the Chomskyan generative linguistic enterprise (see, in particular, HigginÂ�
botham 1985, 1986).
4.â•… Truth-conditional semantics and pragmatics
The dominant framework for modern pragmatics has been provided by Grice (1975,
1989 [1967]), with his crucial distinction between saying and implicating. ‘What is
said’ is equated with the truth-conditional content of the utterance and much of the
rest of ‘what is meant’ is to be taken as implicated, following from conversational
maxims that constrain and guide human communicative behavior. As a first approxiÂ�
mation, then, it may seem reasonable to suppose that the domains of truth-conditional
semantics and pragmatics are strictly complementary, as Gazdar (1979) maintained.

Let us assess this by considering two fundamental facts about natural languages: (a)
their indexicality; (b) the existence of meaningful linguistic elements that do not seem
to contribute to truth-conditional content. These present challenges to the truth-
conditional approach that did not arise in its application to formal systems.
4.1â•… Indexicality
The reference of many natural language expressions is entirely dependent on features
of the particular context in which they are used; examples are ‘today’, ‘I’, ‘here’, ‘that’,
‘the cat’. So the T-sentence for ‘I am happy now’ is not (10a) but (10b):
(10) a. ‘I am happy now’ is true iff I am happy now.
b. An utterance of ‘I am happy now’ is true iff the speaker is happy at the time of
utterance.
The sentence type ‘I am happy now’ is neither true nor false. In general, it is utterances
or propositions or sentence-context pairs that are either true or false. This point was
noted at least as far back as Frege (1977 [1918]) who emphasised the sentence/thought
distinction and insisted that it is thoughts, or propositions, which are the bearers of
truth (or falsity).
To accommodate indexicality, Montague (1974), Lewis (1970, 1979, 1981) and
others enriched the apparatus of truth-conditional semantics by allowing certain
details of context to be drawn into the account, in the form of an index of coordinates,
including the speaker, the addressee, the time of utterance, the place of utterance, indi-
cated objects, etc. This was generally known as ‘indexical semantics’ though Montague
referred to it as ‘pragmatics’ (in a formal sense of the term, clearly different from the
Gricean one above). The assumption behind this is that the context-sentence pair will
deliver the proposition expressed by the utterance (its truth-conditional content); that
is, that for each indexical element of the sentence there is a rule or convention for
picking out its referent in the context. An immediate issue here is what such a formal
notion of context amounts to and how it is to be delimited for particular sentences.
Even if we set this aside, there is the remaining problem that the assumption does not
seem justified for the main bulk of indexical terms. While it might seem reasonable
for cases like ‘I’ (the speaker), and ‘today’ (the day of utterance), it seems much less
so for third person pronouns, demonstratives and definite descriptions. Consider an
example taken from Blakemore (1987: 10):
(11) a. Have you heard Perahia’s recording of the ‘Moonlight Sonata’?
b. Yes, it made me realize I’d never be able to play it.
Even supposing that both the entities referred to by ‘Perahia’s recording of the ‘Moon-
light Sonata’’ and ‘Moonlight Sonata’ are included among the contextual coordinates,
286 Robyn Carston
there is no rule attached to the word ‘it’ that will determine which is referred to on
each of the two uses. That is, the sentence-context pair alone does not determine the
truth-conditional content of the sentence uttered by B.
Within pragmatic theory, construed as a cognitive theory of the processes involved
in understanding verbal utterances, assigning reference to indexical expressions is
an inferential, hence pragmatic, process. According to Sperber & Wilson (1986) and
Â�Wilson (1992), reference is assigned on the basis of the encoded content of the indexical
expression, accessible contextual assumptions and, crucially, the general presumption
that accompanies all utterances that they are optimally relevant. If this is right, then we
already have a departure from the neat dichotomy between truth-conditions and prag-
matics; rather, pragmatically derived content contributes to truth-conditional content.
4.2â•… Non-truth-conditional aspects of semantics

The truth-conditional account does not seem to apply to non-declarative sentences
such as imperatives, optatives, exclamatives and interrogatives, which cannot be true
or false. Some would argue that it is not even fully adequate in giving the semantics
of all declaratives since it does not capture their illocutionary force potential: e.g. to
promise, to warn, to apologise, to advise, etc.
There have been a number of responses to these apparent problems. On one
approach, advocated by Lewis (1970) among others, the idea is that every sentence
is underlyingly declarative, so the imperative ‘Go home now’ is really an ellipsed ver-
sion of ‘I order you to go home now’ which, as a declarative, is truth-evaluable, and
so can be characterised by means of truth-conditions. J.D. Fodor (1977: 49–58) and
Levinson (1983: Ch. 5) assess various versions of this ‘performative hypothesis’. On a
different view, although non-declaratives cannot be assigned truth-conditions, they
can be analyzed in terms of analogous notions, derivable from the notion of truth-
condition. McGinn (1977), for instance, treats imperatives as having fulfilment condi-
tions which are met if and only if it is made the case that the state of affairs described
by the imperative is true. On a third, and probably more promising approach, every
utterance is seen as having both a propositional (hence truth-conditional) content,
and a mood, indicated by such features as word order and verbal morphology, whose
semantics is not truth-conditional. There is considerable debate over the correct way
to characterise the semantics of mood: on Searle’s (1969, 1979) view, different moods
indicate different illocutionary forces, or speech acts, so, for example, an imperative
has the force of an order or a request to do something; Wilson & Sperber (1988b)
argue rather convincingly for the view that mood indicators encode information about
propositional attitude (belief, desire) and that if and when speech act information is
derived this is a function of pragmatic inference based on the encoded propositional
attitude information and context.
In fact there are many other elements in any natural language which, while clearly
meaningful, do not seem to contribute to the truth-conditional content of an utter-
ance. Expressions like ‘but’, ‘therefore’ and ‘moreover’, that Grice characterized as com-
municating conventional implicatures, are well-known. In addition, there are various
sentence adverbials and particles, whose function appears to be to indicate attitudinal
and speech act information:
(12) a. Frankly, Ian is a fool.
b. Fortunately, it didn’t rain.
c. You’re going, eh?
d. I have to go, alas.
In each of these, the split between the truth-conditional content and the non-truth-
conditional is quite evident: (12a), for instance, is true if and only if Ian is a fool.
There is considerable work at present on the best way to characterise the semantics
of these ‘extra’ elements and on their role in utterance interpretation (see Blakemore
1987; Â�Wilson & Sperber 1993).
The upshot of these last two sections is that the relationship between truth-
conditional semantics and pragmatics postulated above must be revised. Natural
Â�language sentences seldom encode a determinate set of truth-conditions. The sources
of this shortfall include indexicality, as discussed above, ambiguity, ellipsis, and vague-
ness, all of which require the intervention of pragmatics at the level of what is said. It
seems then that the truth-conditional content of an utterance is an amalgam of lin-
guistically encoded (conventional) meaning and pragmatically inferred material. On
the other hand, it is clear that there are aspects of communicated meaning which,
while not truth-conditional, are not pragmatic either, in that they are not derived on
the basis of maxims or principles of communication but come from the language itself.
In other words, the semantics of natural languages cannot be fully articulated in truth-
conditional terms, though there can be little doubt that truth-conditions will remain
central to any comprehensive semantic enterprise.
References
Blackburn, S. (1984). Spreading the word. Clarendon Press.

Blakemore, D. (1987). Semantic constraints on relevance. Blackwell.
Carston, R. (1988). Implicature, explicature and truth-theoretic semantics. In R. Kempson (ed.)
Mental representations: 155–181. Cambridge University Press.
Chierchia, G. & S. McConnell-Ginet (1990). Meaning and grammar. MIT Press.
Davidson, D. (1967). Truth and meaning. Synthese 17: 304–323.
Devitt, M. & K. Sterelny (1987). Language and reality. Blackwell.
Dowty, D., R. Wall & S. Peters (1981). Introduction to Montague semantics. Reidel.
Evans, G. & J. McDowell (1976). Truth and meaning. Clarendon Press.
288 Robyn Carston
Evnine, S. (1991). Donald Davidson. Polity Press.

Fodor, J.A. (1975). The language of thought. Crowell.
——— (1987). Psychosemantics. MIT Press.
——— (1990). A theory of content and other essays. MIT Press.
Fodor, J.D. (1977). Semantics. Harvester.
Frege, G. (1892). Über Sinn und Bedeutung. Zeitschrift fur Philosophie und philosophische Kritik 100:
25–50.
——— (1977) [1918]. The thought. In Logical investigations. Blackwell.
Gazdar, G. (1979). Pragmatics. Academic Press.
Grice, P. (1975). Logic and conversation. In P. Cole & J. Morgan (eds.) Syntax and semantics 3: 41–58.
Academic Press.
——— (1989) [1967]. The William James lectures. In Studies in the way of words: Part I. Harvard
University Press.
Higginbotham, J. (1985). On semantics. Linguistic Inquiry 16: 547–593.
——— (1986). Linguistic theory and Davidson’s program in semantics. In E. Lepore (ed.) Truth and
interpretation: 29–48. Blackwell.
Jackendoff, R. (1983). Semantics and cognition. MIT Press.
Katz, J. (1972). Semantic theory. Harper.
Kempson, R. (1977). Semantic theory. Cambridge University Press.
Larson, R. (1990). Semantics. In D. Osherson & H. Lasnik (eds.) An invitation to cognitive science:
23–42. MIT Press.
Levinson, S. (1983). Pragmatics. Cambridge University Press.
Lewis, D. (1970). General semantics. Synthese 22: 18–67.
——— (1979). Scorekeeping in a language game. Journal of Philosophical Logic 8: 339–359.
——— (1981). Index, context and content. In S. Kanger & D. Ohman (eds.) Philosophy and grammar:
79–100. Reidel.
——— (1983). Philosophical papers, vol. 1. Oxford University Press.
McGinn, C. (1977). Semantics for nonindicative sentences. Philosophical Studies 32: 301–311.
Martin, R. (1987). The meaning of language. MIT Press.
Searle, J. R. (1969). Speech acts. Cambridge University Press.
——— (1979). Expression and meaning. Cambridge University Press.
Sperber, D. & D. Wilson (1986). Relevance. Blackwell.
Strawson, P. (1971) [1969]. Meaning and truth. In Logico-linguistic papers: 170–189. Methuen.
Tarski, A. (1944). The semantic conception of truth. Philosophy and Phenomenological Research 4:
341–76.
Wilson, D. (1992). Reference and relevance. UCL Working Papers in Linguistics 4: 167–191.
Wilson, D. & D. Sperber (1988a). Representation and relevance. In R. Kempson (ed.) Mental repre-
sentations: 133–153. Cambridge University Press.
——— (1988b). Mood and the analysis of non-declarative sentences. In J. Dancy, J. Moravcsik &
C. Taylor (eds.) Human agency: 77–101. Stanford University Press.
——— (1993). Linguistic form and relevance. Lingua 90: 1–25.
Universal and transcendental pragmatics
Joachim Leilich
1.â•… Origins
The work of Jürgen Habermas is normally regarded as a continuation of the move-

ment of critical theory. With Horkheimer and Adorno, the founding fathers of Â�critical
theory, Habermas shares the interest in the combination of sociological theory with
philosophy and a strong attitude against positivistic tendencies in the social sci-
ences (Adorno 1969). More than other members of the movement of critical theory
Â�Habermas concentrated himself upon metatheoretical questions, an interest which is
especially manifest in Habermas (1968, 1970). In the development of these interests
a main focus of Habermas’ work came to lie on the notion of ‘interaction’. Starting
from the hypothesis that language and communication play an important role in the
coordination of actions of individuals, Habermas developed his program of ‘universal
pragmatics’. The first effort in this direction was Habermas (1971). The main ideas
of this paper were later, accompanied by metatheoretical reflections, worked out in
Was heißt Universalpragmatik (Habermas 1976). The connection between this idea of
universal pragmatics with sociological theory Habermas elaborated in critical discus-
sions with sociologists like Weber, Durkheim, Parsons and early critical theory in his
voluminous Theorie des kommunikativen Handelns (1981), and looking back upon his
intellectual development in the introduction to the first volume of his Philosophische
Texte (2009).
Habermas (1992) is an effort to make the insights of his 1981 work fruitful in the
context of the theory and philosophy of law.
The origins of the thought of Karl-Otto Apel the proponent of a Â�‘transcendental
pragmatics’ do not lie in critical theory, but in the work and influence of Martin
Â�Heidegger, a philosopher who is normally regarded as standing in sharp opposition to
critical theory. In his early years Apel developed his thoughts in the sphere of ‘herme-
neutical philosophy’, showing a special interest in the philosophy of language, where his
thinking was in the beginning inspired by Neo-Humboldtian trends such as the Sapir-
Whorf hypothesis, and parallel movements in German philosophy (Weisgerber). But
beside his interest in Heidegger, who dominated German post-war philosophy, Apel
soon played a very important role in reintroducing interest in analytical philosophy in
Germany after the break-down of the Third Reich. His reception of Wittgenstein and
Peirce led to a step by step transformation of his thinking. This is documented in the
290 Joachim Leilich
collection of essays published under the title Transformation der Philosophie (1973a),
which culminated in the idea of a transcendental pragmatics. Later Apel tried to make
use of his paradigm in the field of the explanation of action and the philosophical
foundations of the humanities (Apel 1979), but most of his efforts concentrated on the
foundations of ethics where he developed (together with Habermas) a ‘Diskursethik’
(discursive ethics) (Apel 1988; Habermas 1983).
Habermas and Apel roughly share the same paradigm. Relying on this close kin-
ship, the further elucidations of universal and transcendental pragmatics below take
Habermas’ thought as starting-point, letting aside the more biographical question of
in which manner these two thinkers influenced one another.
2.â•… Habermas’ universal pragmatics
Habermas introduced his idea of universal pragmatics as an effort to do for speech

(parole), what Chomsky did for language (langue). His thoughts concerning a theory
of communicative competence attempt to elucidate a system of rules with the help of
which possible situations of coming to understanding with someone (‘Verständigung’)
about some matter are generated. The triangular situation of someone aiming to reach
a consensus with somebody else about something serves as Habermas’ starting point
for the conceptualization of the phenomena of speech. This starting point is favored
by the so called double structure of speech acts. A speech act is composed of an illo-
cutionary role, which is or can in principle be expressed by a performative verb, and
a propositional content. The dominating performative part is used to constitute an
intersubjective relation between speaker and hearer, the propositional part is used in
order to communicate about states of affairs. Therefore – putting aside speech acts
without a propositional content – each communication happens simultaneously on
two levels: the level of intersubjectivity, where the subjects speak together, and the level
of objects or states of affairs about which they try to come to an understanding.
The intersubjective relation between speaker and hearer is elaborated by Habermas
in terms of validity-claims (‘Geltungsansprüche’). Habermas differentiates between
four validity-claims: understandability, truth, correctness (or legitimacy), and sincer-
ity. Such validity-claims are like an offer from the speaker to the hearer which can be
accepted or refuted. Understandability has a special status because it functions like a
precondition of all communication. The speaker claims that his utterance is under-
standable, that its propositional content is true, that the illocutionary component is
correct or legitimate and that s/he is sincere. The validity claim of correctness or legiti-
macy can best be illustrated by giving an order, because in such a context it is evident
that the question whether the order was legitimate can explicitly arise, in the sense
that we can ask whether the speaker had the right to give that order. But the question
whether a speaker had a right in this sense can be extendend to other cases: If for
example we have no reasons to believe that p, we have no right to assert p. If we
recommend to the speaker (to do) p, (doing) p must be better for the speaker than
not-p. Therefore the validity claim of correctness (or legitimacy) can best be under-
stood as the claim that the preparatory and essential rules for speech acts according
to Searle’s analysis in Speech acts are fulfilled. Though the notion of truth is standardly
used in the context of assertives only, there are analogies in other types of speech acts
as well. Thus, in a lot of speech acts there are requirements for existential presupposi-
tions to come out true: I can only successfully ask someone for a cigarette if it is true
that s/he has cigarettes, I can only warn someone about doing p, if it is true that p is
dangerous or disadvantageous for the hearer.
Though the four validity-claims are present in each propositionally differentiated
speech act, following Habermas’ analysis, in different types of speech acts one of these
validity claims is stressed. Thus, in assertives it is the validity claim of truth which
is stressed, but an order is dominated by the validity claim of correctness. On this
basis, Habermas illustrates the difference between two fundamental modes of speech,
namely a cognitive use of language (through assertives) and a communicative mode
concerning those uses of language which Austin originally posed as performative uses
against constative uses of language. Inspired by the work of Karl Bühler, Habermas
transformed this twofold distinction into a threefold distinction: In each speech act
we have to do with a question of the relation of propositional content to the state of
affairs in an ‘outer’ world. This relation corresponds in Bühler’s work to the representa-
tive function of speech and in Habermas to the validity claim of truth. Bühler’s appela-
tive function of speech corresponds to the validity claim of correctness which has to
do with the social-communicative relation between a speaker and his audience. The
third dimension – the expressive function of speech in Bühler – concerns the relation
between the speech act and the ‘inner’ world of the speaker, who may or may not be
sincere in the expression of his beliefs and intentions.
Habermas makes a strong and – in the architectonics of his social philosophy –
important difference between instrumental, or in social contexts strategical action,
which is shaped by the technical means-ends rationality, and the rationality of com-
munication where someone tries to come to an understanding with someone in a
cooperative manner through the acceptance of validity claims. In communicative
action those validity claims are accepted without further discussion, the validity of
those claims is not thematic, but they serve as means for information and the coordi-
nation of action. If the validity claims are not accepted and become problematic, the
speaker and the hearer must explicitly try to reach a consensus in a discourse, if they do
not want to break off communication and try to reach their aims through strategical
means such as for example threats or manipulations, which are from Habermas’ point
of view no real communication-acts at all, because they do not rely on validity-claims.
292 Joachim Leilich
From the four validity claims only two – truth and correctness – can be thema-
tized discursively. Habermas tried to formulate the rules for a fair discourse formally
by giving rules which an ideal speech community should have to follow. In theoretical
discourse it is the truth of statements which is at issue, in practical discourse the ques-
tion is whether a recommended moral norm is obligatory or not. In his views about
practical discourse, Habermas defends a cognitive ethics using a principle of univer-
salibility as guideline.
3.â•… Apel’s transcendental pragmatics
K.O. Apel situates his thinking in the realm of tension which has as poles (1) pragmatic
tendencies in analytic philosophy (as for example in Wittgenstein and Austin), (2)
classical semiotic theories (Peirce, Morris) and (3) transcendental philosophy (Kant).
Apel sketched several different ways to develop his transcendental pragmatics. One
goes along Habermas’ analysis of the double-structure of speech acts using the notion
of validity-claims in order to elucidate the ‘logic’ of human communication. But a
historically earlier and systematically more important way takes as point of reference
the classical semiotical theories as developed by Charles S. Peirce and Charles W. Mor-
ris. Apel uses the triadic sign-relation, which is constituted by the moments of the
(real) object, the sign, and the interpreter (or sign-user), in order to classify types of
first philosophy (philosophia prima) which can be found in the history of philosophy
(Apel 1978, 2011). He differentiates between three fundamental paradigms of first phi-
losophy. They can be characterized through the question of which of the three dimen-
sions of the sign-relation they focus on. For classical metaphysics (Greek and medieval
philosophy) neither the triadic sign-relation nor the dual subject–object relation is
viewed as a methodological relevant condition for the possibility of valid knowledge.
Classical ontology abstracts from the signs and the sign-interpreting mind and treats
philosophical questions only on the level of an object–object relation, without allow-
ing the possibility for the knowledge of objects to be mediated by signs or the know-
ing subject. The mind and signs are viewed as objects among other objects, but not
as conditions for knowledge of objects which can be grasped by philosophical reflec-
tion. This first paradigm abstracts from two of the three dimensions of the semiotic
sign-function.
With the second paradigm – classical philosophy of consciousness – which domi-
nated philosophy from Descartes to Husserl, the mediation of objective knowledge
through consciousness comes into play. But for this paradigm the signs are not rel-
evant, since they only serve as marks for ideas in the mind. So the full-blown triadic
relation remains reduced to a dual subject–object relation, where only solipsistic con-
sciousness but not signs are a theme of philosophical reflection.
The third (transcendental-semiotical) paradigm is constituted by a semiotical

transformation of classical transcendental philosophy. It replaces the twofold subject–
object relation by the triadic semiotic relation, viewing the last one as a transcendental
condition of the possibility and validity of knowledge. The two volumes of Transfor-
mation der Philosophie show Apel’s own way to such a transcendental semiotics but
also the transformation process philosophy has itself undergone through its history.
From the point of view of Apel’s ‘transcendental re-interpretation’ of the triadic
sign-relation, the pragmatic dimension should neither be constructed as an empirical-
behavioristic (Morris), nor as a formal-constructive (Carnap) discipline, where one
speaks in a meta-language about the phenomena of speech. The function of language
as a mediation of interpretation, and an intersubjectively valid representation of
objects cannot, in principle, be objectified. Hence it turns out to be a transcendental
condition of the possibility, rather than a possible object, of sign-mediated knowl-
edge. Apel claims that we must assign a transcendental interpretation not only to the
syntactico-semantical part of the sign-function but also to its pragmatic dimension,
i.e. to the dimension of our actual language-use, and in that context to the dimension
of our construction and successful interpretation of language-systems as semantical
frameworks of possible-world description. Philosophical talk about the problems of
language-construction and interpretation fulfills the reflective function of a transcen-
dental pragmatics within the frame of a transcendental semiotics. Insofar as seman-
tical frameworks are conditions for the possibility of world interpretation, speaking
about the construction and interpretation of those frameworks must be considered as
a transcendental reflection on the subjective-intersubjective conditions of the possibil-
ity of world-interpretation.
Apel’s much disputed thesis concerning the possibility of ultimate foundations
(‘Letzbegründung’) is closely related to the reflective character of a transcendental
interpretation of pragmatics (Apel 1975). The possibility of ultimate foundations or
‘fundamental grounding’ is of course excluded, if we consider the relation of justi-
fication as a deductive relation between propositions. In this context, Apel admits,
we would be inevitably confronted with a trilemma: (1) We would have to break off
justification at an arbitrary point (for example by claiming that a premise needs no
further justification), (2) we would have to go on with justifications ad infinitum, or
(3) our justification would take the form of a circular argument. But such a trilemma
only exists on the level of deductive relations between sentences. On this level, Apel
claims, certainty can never be reached; his argumentation here builds on Popper’s fal-
libilism. But on the level of a reflection on the ultimate presuppositions of intersubjec-
tive valid communication, we can reach the insight that some principles could not be
consistently denied. Apel resumes his argumentation strategy as follows: “If I cannot
challenge something without actual self-contradiction and cannot deductively ground
it without formal-logical petitio principii, then that thing belongs precisely to those
294 Joachim Leilich
transcendental-pragmatic presuppositions of argumentation which one must always

have accepted, if the language game of argumentation is to be expected to retain its
significance” (Apel 1975: 264). Therefore, from the transcendental-pragmatic point of
view, fallibilism is restricted to the empirical realm, but cannot be generalized so far
that even the presuppositions of critical argumentation (and with them the presup-
positions of fallibilism itself) would stand open to a revision in the light of empirical
discoveries.
As an illustration of such a transcendental-pragmatic argumentation strategy we
could consider someone who claims that s/he makes no truth-claims, or somebody
who claims that we should never aim at a consensus. Someone who claims never to
make truth-claims makes a truth-claim in the ‘performative’ part of his/her speech
act, but in the propositional content of the same speech act the very possibility of the
speech act itself is denied. The truth-claim belongs to the very possibility of the act of
assertion and it cannot be denied without a ‘performative self-contradiction’. Someone
who claims that we should not aim at a consensus, claims that his/her claim should be
accepted by everyone as convincing. The addressee of such validity-claims (in the sense
of Habermas) is an infinite community of communicators which is presupposed in all
communication. Among the insights we can reach in such a transcendental-Â�reflective
way, we can find the normative claims that the members of the infinite community
of communicators should respect one another as accountable subjects, that they
should treat one another as having equal status and equal rights, and they should act
in a cooperative manner. For Apel (and Habermas) the most important connecting
link with ethics is that in arguing and communicating we have to presuppose such
normative claims.
Apel (like Habermas) referred to Peirce for a consensus-theory of truth, where truth
could be identified with the ultimate opinion of an ideal community of communicators.
One of the main points of disagreement between Habermas and Apel concerns the
transcendental idiom used by Apel to formulate his standpoint and the Â�philosophical
ambitions which are connected with Apel’s defence of the possibility of ultimate foun-
dations. From Habermas’ point of view Apel’s transcendental pragmatics should be
like his universal pragmatics conceived as a ‘reconstructive science’ where a practical
ability which has the form of a ‘know how’ is transformed into a ‘know that’. Such a
reconstruction of pre-theoretical knowledge always contains hypothetical elements
which must and can be scrutinized as other empirical hypotheses. The main dif-
ferences between Habermas and Apel lie not so much on the level of the results, but
on the level of the meta-theoretical interpretation of such results. As the name
transcendental pragmatics already indicates, Apel elucidates the standards of his
theory in the light of a (semiotically) transformed transcendental philosophy, which
tries to discover the ultimate and necessary conditions which must be imposed upon
empirical science.
Apel’s Transformation der Philosophie ended with a paper which can be regarded
as a founding document of discursive ethics. After the publication of Transformation,
Apel spent a lot of attention on problems of ethics. The situation of ethics today can
according to Apel be characterized by a paradoxical problem situation: Under the
influence of positivistic and post-positivistic tendencies, philosophy has become very
pessimistic with regard to the possibility of rational foundations for ethics, whereas
the need for a universal ethics has never been so large, because of the technological
consequences of modern science. Philosophical thought on ethic matters tends toward
something like a complementary situation, where rationality is restricted to scientific
questions, whereas the questions of values is mostly viewed as a matter of subjective
decisions or culture-relative forms of life without the possibility of founding them
rationally.
In Das Apriori der Kommunikationsgemeinschaft und die Grundlagen der Ethik
(1973b) Apel argued that through a reflection on the possibility of rational argumenta-
tion we can discover ethical presuppositions. One of the results of his transcendental-
pragmatic reflections on the conditions of the possibility of communication was the
insight that in all rational argumentation we are obliged to presuppose the account-
ability of other subjects, their equal rights and status, and principles of cooperation
which cannot be reduced to strategical forms of interaction. In ethical discourse the
decision about practical questions has to be arranged in a manner so that rules of an
ideal community of communication (which corresponds to Habermas’ ideal speaker’s
situation) are respected. This foundational norm of ethical discourse is justified by Apel
through the reflective procedure of trying to demonstrate that those norms of equality
are already presupposed in all argumentation. The basic norm of discursive ethics is not
a material norm, but a procedural norm which tries to show that only such material
norms are well-founded, which could be accepted by such an ideal community. Because
of the requirement that moral norms should be acceptable by all (possible) participants,
there are strong similarities to Kant’s categorical imperative and more recent forms of
ethical thought as they can be found in the works of Rawls and Kohlberg. In Diskurs und
Verantwortung (1988), Apel concentrated himself on the question of the realizability of
such discursively founded norms under concrete historic and situational circumstances.
References
Adorno, T.W. et al. (1969). Der Positivismusstreit in der deutschen Soziologie. Luchterhand.
Apel, K.O. (1973a). Transformation der Philosophie, 2 vols. Suhrkamp.
——— (1973b). Das Apriori der Kommunikationsgemeinschaft und die Grundlagen der Ethik. In
K.O. Apel (1973a): 358–435.
——— (1975). The problem of philosophical fundamental grounding in the light of a transcendental
pragmatic of language. Man and World 8: 239–275.
296 Joachim Leilich
——— (ed.) (1976). Sprachpragmatik und Philosophie. Suhrkamp.

——— (1978). Transcendental semiotics and the paradigms of first philosophy. Philosophic Exchange
2: 3–22.
——— (1979). Die Erklären-Verstehen Kontroverse in transzendentalpragmatischer Sicht. Suhrkamp.
——— (1988). Diskurs und Verantwortung. Suhrkamp.
——— (2011). Paradigmen der Ersten Philosophie. Suhrkamp.
Habermas, J. (1968). Erkenntnis und Interesse. Suhrkamp.
——— (1970). Zur Logik der Sozialwissenschaften. Suhrkamp.
——— (1971). Vorbereitende Bemerkungen zu einer Theorie der Kommunikativen Kompetenz.
In J. Habermas & N. Luhmann (eds.): 101–142.
——— (1976). Was heißt Universalpragmatik? In K.O. Apel (ed.): 174–272.
——— (1981). Zur Theorie des kommunikativen Handelns. Suhrkamp.
——— (1983). Moralbewußtsein und kommunikatives Handeln. Suhrkamp.
——— (1992). Faktizität und Geltung. Suhrkamp.
——— (2009). Sprachtheoretische Grundlegung der Soziologie. Philosophische Texte Band1. Suhrkamp.
Habermas, J. & N. Luhmann (eds.) (1971). Theorie der Gesellschaft oder Sozialtechnologie. Suhrkamp.
Ludwig Wittgenstein
Joachim Leilich
1.â•… Biographical sketch
Ludwig Wittgenstein was born in Vienna in 1889, as the youngest child of a well-
known and extraordinarily rich family. His youth must have been overshadowed
by the suicide of two of his brothers; a third one took the same decision at the end
of the First World War. We know from Wittgenstein’s correspondence and some of
his diaries which were not destroyed, that Ludwig too suffered from depression and
Â�considered suicide himself.
He started studies of engineering in Berlin and continued them some years later
in Manchester. So it seemed at least as if he would be able to succeed his father in the
important steel company he directed. But during the Manchester period (1908–11)
Wittgenstein became interested in the foundations of mathematics. He went to Jena, to
meet Gottlob Frege, who had published the Laws of Arithmetic, but Frege advised him
to study with Bertrand Russell in Cambridge, who had just finished the first volume of
his Principia Mathematica in collaboration with Whitehead. Russell was quickly con-
vinced by the extraordinary intensity and quality of Wittgenstein’s thought. Most of
his student years Wittgenstein spent outside Cambridge in Norway, as we know from
his extensive correspondence with his ‘teachers’ in Cambridge. At the outbreak of the
First World War, Wittgenstein went back to Austria, serving as a volunteer in the army
of what was then still the Austrio-Hungarian monarchy.
During the war Wittgenstein finished his Logisch-Philosophische Abhandlung,
now famous as the Tractatus and the only philosophical book he published Â�during
his lifetime. But when the Tractatus made him famous among Cambridge
Â�philosophers and members of the Vienna Circle in the 1920’s, Wittgenstein had
already retired from philosophy. After the war he committed ‘financial suicide’, as
his lawyer called it, by giving away to two of his sisters the total of the fortune he
had inherited after the death of his father in 1913. From then on he lived extremely
modestly and sometimes even in need of financial support by friends. He fol-
lowed a course as a primary school teacher and became a simple schoolmaster in
rural districts of Austria. Conflicts with parents, often caused by his temperament
which lacked self-control, led after some years to his resignation. Back in Vienna he
became involved in the construction of a house for his sister Margarethe. The house
is sometimes considered as an Â�architectural counterpart to his early philosophy as
laid down in the Tractatus.
298 Joachim Leilich
In 1929, Wittgenstein, forty years old and meanwhile a famous philosopher,

went back to Cambridge as a researcher. His Tractatus was accepted as a doctoral
Â�dissertation, and from then on dr. Wittgenstein developed what was to become his
later philosophy which deviated stylistically and substantively from his earlier views.
When the Second World War broke out, Wittgenstein, with the support of John May-
nard Keynes, became a British citizen. Having three jewish grandparents Wittgenstein,
after the annexion of Austria, would have become a victim of the racial laws had he
returned to Vienna. At the same time G.E. Moore’s chair as a professor of philoso-
phy became vacant, and Wittgenstein, although since years without any publications,
became his successor. From this function he retired shortly after the war, spending
the rest of his life most of the time in loneliness at the Irish west-coast. His attempts
to collect his thoughts in a definitive manner were unsuccesful and he was always
busy revising the materials he had collected in lots of notebooks. Shortly after his
death in 1951, Wittgenstein’s trustees published the Philosophische Untersuchungen/
Philosophical Investigations – a collection of remarks written in German. All of the
publications of the later Wittgenstein are compilations from Wittgenstein’s Nachlass
and were probably never intended for publication in the form in which they appeared.
Among the latest editions of Wittgenstein’s works, there are Michael Nedo’s Â�Wiener
Ausgabe, started in 1993, and the edition of the complete Nachlass on CD-Rom, pre-
pared by the Wittgenstein Archive in Norway, in which the development of his thought
is presented in a strictly chronological order, paying the price of frequent overlaps
between the texts and lots of stylistic variants.
2.â•… General characteristics of Wittgenstein’s conception of philosophy
It is common fare to speak about the two philosophies which Wittgenstein developed,
a first one in which the ideal of logical analysis is predominant, and a later one which
is often considered as a contribution to the movement of ordinary language philoso-
phy. In fact there is no sharp dividing line between Wittgenstein’s early and his later
philosophy, but a process of continuous rethinking of earlier points of view turning
gradually into a form of radical self-criticism. Wittgenstein’s philosophy in general can
be characterized as the view that language is misleading as we do philoÂ�sophy. In his
early period he thought – inspired by Frege and Russell – that language hides the true
logical form of thought and thus contributes to the production of pseudoproblems as
a result of logical confusion. In his later philosophy it is not so much surface grammar
which misleads by concealing the true logical form, but that we tend to misunderstand
the plurality of uses we make of language. Wrong ideas about the manner in which
language becomes meaningful mislead our philosophical thought. Therefore one of
the continuities in Wittgenstein’s thought is his therapeutic vision of philosophy. There
is not something like a philosophical doctrine articulated in propositions, but philoso-

phy is an attempt to clarify problems by reflection on language use. In this process of
clarification, Wittgenstein hopes, the philosophical problems will disappear like an
illness when correctly treated.
Wittgenstein is not a philosopher of language in the sense that language would
be the primary object of his investigations. He is interested in language as a means of
articulation of philosophical thought, being convinced with most analytic philoso-
phers that reflection on language is the key to doing philosophy in the right way. But of
course, in using the analysis of language as a method in philosophy, Wittgenstein has
to develop ideas about what language is and how it functions, and for those ideas lin-
guists have often shown critical interest. It is not so much his idea about the Â�question
why language matters to philosophy, than his ideas about the nature of language which
change considerably during the development of Wittgenstein’s thought.
3.â•… The Tractatus and the picture-theory of meaning
Although not in the focus of Wittgenstein’s interest in the Tractatus, one can find in
this extremely condensed book a theory of (linguistic) representation known as the
picture theory of meaning. The picture theory starts from two basic primitive ideas
which are developed further in a very abstract manner. The first of these is the idea
that names stand in for objects. “In a proposition a name is the representative of
an object” (T 3.22). But what language represents are not just objects but states of
affairs, and to be able to represent states of affairs, referring to objects by means
of names is not enough. For example, naming two soccer teams is not enough to depict
the state of affairs which one won or is the leader in the competition. If we read the
results of the matches, or the arrangement of the names in the table, the spatial form
of the names is an important aspect of depicting. We read from the table that the team
at the top is the leader or we read from the results that Real Madrid is the winner,
because in the ‘picture’ Real Madrid – Arsenal London 2:1, Madrid is placed left, and
the left number is higher than the right one. This illustrates that naming is not enough,
but that the arrangement of names (for example a spatial arrangement as in writing)
plays an important part in the possibility of linguistic or symbolic Â�representation.
Â�Wittgenstein’s picture-theory is an attempt to generalize such a view.
In order to have meaningful representation there must be a connection between
the elements of the picture and elements of reality. Names, in order to be names, must
have bearers. And to the possible combination of the elements in reality (making up
states of affairs) corresponds a structured combination of the elements of the picture.
If we have a language (a special case of a picture) in which the names have bearers
and the possibilties of the elements to participate in state of affairs are mirrored by the
300 Joachim Leilich
possibility of names to combine into propositions, the following ideal situation is met:
everything which can be said to be Â�meaningful represents a Â�possible state of affairs, and
every possible state of affairs can be expressed in language. Therefore Wittgenstein –
anticipating the famous last sentence of his book – can write in his Preface: “The whole
sense of the book might be summed up in the following words: what can be said at all
can be said clearly, and what we cannot talk about we must pass over in silence”. This is
not a demand which an ideal language should fulfill, as Wittgenstein is often misinter-
preted, but if there is something like meaningful language at all, this demand already
has to be fulfilled. “In fact, all the propositions of our everyday language, just as they
stand, are in perfect logical order” (T 5.5563). But the surface grammar of language
disguises thought and its relation to reality: “Language disguises thought. So much
so, that from the outward form of the clothing it is impossible to infer the form of the
thought beneath it, because the outward form of the clothing is not designed to reveal
the form of the body, but for entirely different purposes” (T 4.002).
How is it possible that this correspondence between language and reality exists?
“If a fact is to be a picture, it must have something in common with what it depicts”
(T 2.16). A picture can (but must not) share with reality what it represents. With colour
samples we can represent all colours, for example, or in calling the names of the run-
ners in a temporal order, it is possible to represent the temporal order of the arrival
at the finish. A three-dimensional model can represent all three-dimensional spatial
relations. But of course, it would be strange to request that picture and depicted reality
must share some material qualities. Nevertheless Wittgenstein insists that representa-
tion is only possible because the representation and what is represented have some-
thing in common. What they have in common Wittgenstein calls the ‘logical form’.
But what does ‘logical form’ mean? The Tractatus gives no commentary. Following an
account by Erik Stenius (1960) the picture theory is often interpreted as the view that
language (or pictures in general) and reality are isomorphic, that there is a structural
identity. As an illustration we could use the idea of the possibilities of representing
which places in a parking lot are occupied and which ones are not. In order to do this
we need to name the places, for example by means of numbers. The fact that a place
is occupied or not could be represented in many ways: by making marks with a pencil
and erasing them if the car leaves, by means of an display with numbered small lights
which burn or do not burn, by means of numbered pieces of numbered tickets which
are handed out, etc. But all these ways of proceeding would have something in com-
mon: there is the relation between the numbers and the places, which means there is
some form of naming. And the possibilities of states of affairs – in this case the places
being occupied or not – must be represented by a device which mirrors the combina-
tory possibilities, here the simple case of being occupied or not by a lamp which is
burning or not, a pencil stroke which is made or erased, or a ticket which is handed out
or not. In all these cases – whatever the chosen ‘representational form’ – there would
be an isomorphic relation between the different pictures and the depicted reality. “In
a proposition there must be exactly as many distinguishable parts as in the situation
that it represents. The two must possess the same logical (mathematical) Â�multiplicity”
(T 4.04). Of course the situation can be more complicated: we can take account of dif-
ferent floors, different kinds of cars and so on. But the core idea that representation
becomes possible by naming elements and isomorphic relations between the elements
in reality and in the representation would remain the same.
Nowhere in the Tractatus does Wittgenstein give an example of what he means with
a name or the corresponding category of an object. He cannot mean ordinary names
or ordinary objects, because he attributes to objects rather unusual properties such as
never being composite or being unalterable. Wittgenstein’s objects are postulated as the
results of a complete analysis of ordinary linguistic entities and entities in the world.
Wittgenstein’s ontology, sketched very briefly at the beginning of the Â�Tractatus, mirrors
his linguistic categories: name-object, elementary proposition – state of affairs. Often
Wittgenstein is accused of a primitive form of realism, where ready-made mind- and
language-independent objects are named, but others see in his views a more sophisti-
cated form of a linguistic transformation of Kantian transcendental philosophy, where
from the fact of language insights into the structure of a reality (in so far as it can
be represented) are derived in an a priori style of philosophy. The later titel Tracta-
tus logico-philosophicus, contains such a hint at a non-empirical aprioristic manner of
doing philosophy, by its similarity with the title of a famous treatise by Spinoza.
4.â•… Logical constants and the doctrine of saying and showing
Although normally most attempts at understanding the Tractatus focus on the Â�so-called
picture theory, these views form only a small part of the book, and from Wittgenstein’s
own point of view surely not the most important. Wittgenstein’s Tractatus is foremost
an attempt to win a clear insight into the nature of logic. Logical relations are articulated
through the use of the logical constants (in opposition to the propositional or individual
variables), i.e. signs such as conjunction, disjunction, implication, negation, some, all,
etc., as known from every basic course in logic. We can compose logically complex sen-
tences with the aid of such logical constants. What is the relation between those complex
or compound sentences and reality? Wittgenstein articulates the insight that the logical
constants cannot contribute to what those compound sentences depict. He sees it even
as one of his “fundamental ideas (…) that the ‘logical constants’ are not representatives”
(T 4.0312). This is for Wittgenstein of outmost importance because only in that way is
the pure non-empirical apriori character of logic understandable. Logical truth must
not depend on facts in the world. That logical constants do not have a depicting or
descriptive function can be made easily understandable. If we have two sentences,
302 Joachim Leilich
p and q, we know a priori, without further experience, the truth values of p and q, p or
q, if p then q, etc. if we know the truth values of p and q. If the logical constants would
have a descriptive function, knowing the truth value of the two propositions could not
be sufficient to establish the truth value of the compound sentences. If we identify the
sense of a sentence with its descriptive content, logical constants do not contribute to
the sense of sentences. They say nothing. Wittgenstein’s ontology therefore does not
recognize facts which are complex from a logical point of view. Between facts there
cannot be logical relations: “States of affairs are independent of one another. From the
existence or non-existence of one state of affairs it is impossible to infer the existence
or non-existence of another” (T 2.061–2.062). Most of the Tractarian insights into the
character of logic and first of all the truth-functional character of the logical constants,
are today commonplace and we no longer realize that what seems to us perhaps trivial
today had to be discovered once.
As the most important aspect of his Tractatus Wittgenstein himself considered
his distinction between saying and showing which is intimately connected with the
famous last proposition of the Tractatus: “What we cannot speak about we must pass
over in silence” (T 7). But not everything that cannot be spoken about is considered
by Wittgenstein as nonsense. There are important things we cannot speak about but
which show themselves. The list of what cannot be spoken about but only shown is
rather extensive. It contains among other things the logical form, the meaning of signs
and the sense of propositions, logical relations between propositions and the mystical
(ethics and esthetics). Against Russell this doctrine contains the refutation of the idea
of a meta-language.
5.â•… Wittgenstein’s later philosophy
When Wittgenstein returned to Cambridge in 1929 his thinking was in the beginning

influenced by his discussions with some of the members of Schlick’s Vienna Circle,
but soon he began to articulate thoughts which became characteristic of his later phi-
losophy. Wittgenstein had a lot of difficulties articulating his thought in a systematic
manner and at first sight the Philosophical Investigations could appear as a rather loose
collection of remarks. In the preface to the Investigations Wittgenstein recommends
his readers to read his new insights together with his Tractatus. One of the reasons
surely is that the Philosophical Investigations contain a lot of (implicit) selfcriticism.
The most striking difference is that Wittgenstein no longer believes that there is some-
thing like an essence of human language, some general characteristics language must
have to be able to function as a means of representation. The standpoint that language
becomes meaningful through sharing the logical form with reality is not replaced by a
single new standpoint concerning the functioning of language. Instead, Â�Wittgenstein
articulates the position that there is not just one way in which language becomes
meaningful.
The Philosophical Investigations contain the view that language does not have an
essence and that there is a multitude of language-games, already in the first few para-
graphs. They start with a quote from Saint Augustine in which naming is the cen-
tral function of language. Wittgenstein sums up this Augustinian Picture as follows:
“The individual words in language name objects – sentences are combinations of such
names. – In this picture of language we find the roots of the following idea: Every word
has a meaning. This meaning is correlated with the word. It is the object for which
the word stands” (§1). Wittgenstein confronts this theoretical view with two different
situations of language use. In the first one someone is shopping. When he demands
five red apples the shop assistant is counting up to five and takes with every numeral
an apple. These apples are in a drawer on which the word ‘apples’ is written and the
assistant is comparing the colour of the apples with a red colour sample. Wittgenstein
discusses this situation with the intention to show that it makes no sense to speak
here about the naming of an object. Counting doesn’t function – and surely not in this
example – as naming something, and the colour samples serve as an instrument of
comparison. This primitive language-game is confronted with another one, for which,
as Wittgenstein explicitly concedes, the Augustinian view is correct. Builders refer
with words to objects in order for those objects to be handed over by a helper. May be
one has to take Wittgenstein’s judgement that in the last case the Augustinian view is
correct only as a tactical consideration, because from his point of view naming objects
is never sufficient to give meaning to an expression. Nevertheless, in the builder’s case,
the idea of naming or referring to objects, even if this is only part of the story, fits more
naturally than in the case of numerals or color terms. Wittgenstein’s conclusion from
the comparison of the two language-games is that the Augustinian view is correct
only for some ways of using language – as for example (some aspects of) the builder’s
Â�language – but not for others, e.g. not for the use of numerals.
In so far as speaking of the meaning of a word could suggest that there is only
one way in which a word could become meaningful – by referring to an object –
Â�Wittgenstein proposes that we should rather look at how language is used instead of
asking what the meaning of a word consists in: “The meaning of a word is its use in
the language” (§43).
This view on language is summarized by the famous discussion of games. If some-
one would ask what a game is, one would have difficulties to give a general definition,
because each definition would let out some activity we would call a game. There is no
collection of necessary features such as an adversary or a board, which all games would
have to share in order to be games. Instead of general features which are common to all
games we find family-resemblance, a network of overlapping and criss-crossing simi-
larities. This discussion of games does not so much have the function to be a criticism
304 Joachim Leilich
of definition by means of necessary and sufficient conditions, but rather to throw light
on the nature of language by analogy. Like the different games, the different ways of
using language such as in giving orders, describing an object, asking, swearing, lying,
confessing or testing a hypothesis (see PI §23) do not all function in the same way,
which could be grasped by a definition. Speaking of language-games emphasizes the
impossibility of reducing the different uses of language to one common essence.
Although considered fundamental, the notion of a language-game remains vague
and ambiguous. Wittgenstein uses the term as a unit of a type of language use. Some-
times he uses the term to indicate a more primitive form of language use than ours,
because in such primitive language-games some aspects are clearer (such as really
using labels in order to name objects). Those more primitive forms of language serve
as ‘objects of comparison’ (PI §130) which shed light on certain features of language.
But Wittgenstein is not consistent in this use of ‘language-game’ because he also uses
the term for ordinary linguistic practices, which are not Â�artificially Â�constructed in
order to illustrate a certain point.
The analogy of games bears another important analogy to language which is often
discussed by Wittgenstein: the analogy that speaking a language and playing games
are both rule-guided activities. Some of these analogies are easy to understand. Witt-
genstein illustrates for example the shortcomings of the Augustinian view that naming
plays an important role through the analogy of chess-playing in §31. Someone who
would know the names of the figures would not at all be able to play chess because he
would not thereby know which uses can be made of the different chess-pieces. Giving
names or knowing the names would only be a Â�preparational step in order to be able to
learn, know and use the rules.
In Wittgenstein’s later work there are long and painstakingly difficult discussions of
what it means for a practice (among which linguistic practices) to be guided by rules. If
we say that a rule obliges us to take a certain step, it seems as if this step is already con-
tained in the rule, or in the more arithmetical contexts Wittgenstein often discusses, in
a formula. But certainly in a formula or a rule-formulation as such nothing is contained.
It’s just the formula or formulation and it is a fictitious idea that this rule formulation
would or could contain its future application. We often think and speak as if a rule
formulation could be mentally active and prescribe something. But how can dead signs
on paper prescribe anything at all? If rules prescribe something, it is not so much the
rule which does something with us, but something has the character of a rule because
we make a certain use of it. Something becomes a rule because we establish a certain
social practice and without this practice rules could not exist. One could illustrate this
point for example with rules of orthography. It is evident that the rule does not simply
causally generate its correct application. It’s not the rule itself which spells correctly. The
rule without the practice is no rule. If nobody would follow the rule it would be strange
to say that the rule exists. If a rule is systematically not followed at all, instead of saying
that the rule is no longer followed, we could with the same right say that it no longer
exists. So the notion of following rules must stand in the center of discussion. Following
a rule means making a certain use of a rule formulation. And this practice has typical
aspects, such as teaching the rules to the participants, consulting the rules in cases of
doubt and following it in actual fact. There would not be a rule of orthography if it
would be generally neglected. It is the common practice which establishes the rule and
you can’t have the rule without the practice. It seems to us as if rules guide our practice
but it is the practice which is constitutive for the rule. We can’t separate the rule from a
certain practice. In this sense the practice is a primitive fact which establishes rules, and
it is not rules which establish a certain practice.
In Wittgenstein’s later work these discussions of rules are related to problems of
philosophical psychology. Wittgenstein had articulated the view that following a rule
is necessarily a social practice: “(…) ‘obeying a rule’ is a practice. And to think one is
obeying a rule is not to obey a rule. Hence it is not possible to obey a rule ‘privately’”
(§202). And this view is often understood as the claim that there must be a com-
munity in order to have rules at all. From this it would follow that a person alone
without being embedded in a linguistic community could not follow rules. This gives
rise to the famous (and dark) problem of private language. Would a private language,
a language which would be understandable only by its one and only user be possible?
But this question immediately interferes with another one: must we not concede that
private languages really exist? Because we have private experiences (for example pain)
and are able to name such experiences in a regular way by following the rule of calling
the same private experience ‘pain’ again if we are in the same state later.
The discussion of this private-language problem contains two claims. The more
unproblematic one is that our language-games in which we speak about our sensations
and our experiences are not private languages: they can be taught and understood by
other members of our community. The idea that the language-game of speaking about
subjective sensations must be considered as a private language is according to Witt-
genstein the result of a misunderstanding of the way in which sensation words func-
tion: they cannot be interpreted as names for entities which are hidden for everyone
except the language user. Here the discussion of private languages fuses with the more
general discussion of whether certain ideas about naming are adequate to understand
language. But in refuting the idea that sensation language must be understood as a case
of a private language nothing is said about the more general question of whether a pri-
vate language – even if it does not exist – could be possible at all. Here the discussion
centers around the question of whether the notion of correctness – without any doubt
central in all discussions about following rules – could be applied to an individual
considered in isolation from a linguistic community. Especially Saul Kripke’s contri-
bution in Wittgenstein on Rules and Private Language (1982) stimulated an unprec-
edented interest in the remarks on rule following, but being more the Â�discussion of
306 Joachim Leilich
Â�
“Wittgenstein’s Â�argument as it struck Kripke” (Kripkenstein), the debate is often con-
ducted in disregard of Wittgenstein’s own writings.
Among the texts written later than of the typescript which became Part One of
the Philosophical Investigations, the remarks published as On Certainty are the most
important ones. Even if here Wittgenstein’s thoughts move in new directions, there is
too much continuity to speak of a Third Wittgenstein. In On Certainty he explores scep-
ticism and the limits of meaningful doubt. Universal doubt isn’t meaningful because
all doubt and all justification occurs in a system of thought which cannot be justified
or doubted in its totality. In distinguishing between the movement of thought in the
‘riverbed’ and the change of the riverbed itself, Wittgenstein’s ideas show some similar-
ity with Thomas Kuhn’s concept of a paradigm in scientific thought.
6.â•… Influence
The Tractatus and the later works prompted a huge secondary literature of an often
very detailed exegetical nature. Besides the countless attempts to understand and
to comment on the texts, Wittgenstein’s philosophy inspired other philosophers or
scholars from diverse other fields of study in the development of their own thought.
The Tractatus made a great impression on the members of the neo-positivistic Vienna
Circle, who saw in Wittgenstein’s early philosophy a contribution to their own anti-
metaphysical endeavour to develop a scientific world view. Books like A.J. Ayer’s early
Language, Truth, and Logic (1936) or Carnap’s much more technical The Logical Syn-
tax of Language (1937) are representative examples of a neo-positivistic reading of the
Tractatus. It took a rather long time before one realized, mainly under the influence of
Allan Janik and Stephen Toulmin’s Wittgenstein’s Vienna (1973), that such an under-
standing was rather one-sided and that contributing to a scientific world view or to the
development of our understanding of formal logic was not the only and perhaps not
even the foremost of Wittgenstein’s aims.
The idea of language-games with its stress on the differences in the use of language
inspired a lot of research, which tried to apply the spirit of Wittgenstein’s later phi-
losophy in areas of investigation which Wittgenstein had only touched on occasionally
or not at all. The numerous books having in their titles phrases such as The Language
of …, The Concept of … or The Logic of … are symptomatic of the Linguistic Turn in the
philosophy of the 20th Century which was mainly due to Wittgenstein, who shifted
the focus from thoughts to the expression of thought in language. R.M. Hare, by way
of example, explored The Language of Morals (1972), Peter Winch argued that we need
concepts from Wittgenstein’s later philosophy in order to understand The Idea of a
Social Science (1970), Gilbert Ryle wrote his famous Concept of Mind (1949), religious
belief statements and rituals where analyzed by D.Z. Phillips (1965) or Wittgenstein’s
pupil N. Malcolm (1993). Other authors, for example H. Pitkin (1972), explored the sig-
nificance of Wittgenstein for political science or law. M.R. Bennett and P.M.S. Hacker
(2003) made a great effort to correct from a Wittgensteinian point of view what they
considered fundamental misunderstandings of contemporary neuroscience.
It is beyond any doubt that Wittgenstein’s influence on analytic philosophy is
incomparable, as it played an important role in most of its different strands like logical
analysis, linguistic philosophy, conceptual analysis or ordinary language philosophy.
The name of Wittgenstein is regularly mentioned in linguistic studies and this can-
not be surprising considering his fundamental interest in language. But Wittgenstein’s
interest in language is not directly comparable to the linguist’s and one should not
even consider him as a typical philosopher of language. His aim in philosophy is not
so much to gain insight into language, but to correct philosophical misunderstandings
which are, from his point of view the result of misunderstanding language, for example
as a result of false analogies caused by superficial similarities. It is this therapeutic use
of the analysis of language in discussing philosophical problems which makes it diffi-
cult to find some direct influence from Wittgenstein in linguistic theory. Nevertheless,
his views are often mentioned in this field without being worked out in a theoretically
profound and systematic manner, or they are discussed in a context which is foreign
to Wittgenstein’s aims.
The famous game analogy is a typical case. If Wittgenstein says that the Â�concept
of a game cannot be defined by means of necessary and sufficient criteria for mem-
bership, his intention is not really to contribute to the discussion of a Â�special class
of expressions in the field of linguistic categorization. He wants to refute his earlier
opinion, developed in the Picture Theory of Meaning in the Tractatus, that language
must have an essence. As in the case of games, the different language-games are only
related in the weak sense of family resemblance. Notwithstanding the self-critical
character of Wittgenstein’s remarks on games or family Â�resemblance, his discussion
inspired linguistics, not because linguists wanted to correct a general philosophical
view about language which eventually misled philosophers, but because they saw that
they could use the idea of family resemblance in the field of semantics. From such a
linguistic point of view, Wittgenstein anticipated inadequacies of the classical theory
of categorization, although such inadequacies in theories of categorization were not at
all the problem Wittgenstein was interested in when he discussed games. Nevertheless,
prototype theory can be understood as a further development of Wittgenstein’s insight
that we often can understand categories as ‘family-resemblance concepts’ by adding
the idea that some examples of a category can be considered more central than others.
Another case where linguists make use of concepts developed by Wittgenstein
we can find in A. Duranti’s Linguistic Anthropology (1993). The author thinks that
Â�Austin’s and Searle’s account of speech acts with its stress on individual intentions, is
too poor to be really useful in linguistic anthropolgy or ethnolinguistics. He claims
308 Joachim Leilich
that Â�Wittgenstein’s notion of a language game offers a much richer framework for lin-
guistic analysis, because it allows for (social) activities and forms of life to play an
important role in the analysis of speech.
References
Ayer, A.J. (1936). Language, Truth, and Logic. Penguin.

Bennett M.R. & P.M.S. Hacker (2003). Philosophical Foundations of Neuroscience. Blackwell.
Carnap, R. (1937). The Logical Syntax of Language. Routledge.
Duranti, A. (1993). Linguistic Anthropology. Cambridge University Press.
Hacker, P.M.S. (1996). Wittgenstein’s Place in Twentieth-Century Analytic Philosophy. Blackwell.
Hare, R.M. (1972). The Language of Morals. Oxford University Press.
Janik, A. & S. Toulmin (1973). Wittgenstein’s Vienna. Simon and Schuster.
Kripke, S. (1982) Wittgenstein on Rules and Private Language. Harvard University Press.
Malcolm, N. & P. Winch (1993). Wittgenstein – A Religious Point of View. Routledge.
Phillips, D.Z. (1965). The Concept of Prayer. Routledge.
Pitkin, H.F. (1972). Wittgenstein and Justice. University of California Press.
Ryle, G. (1949). The Concept of Mind. Hutchinson.
Stenius, E. (1960). Wittgenstein’s Tractatus. Blackwell.
Winch, P. (1970). The Idea of a Social Science and its Relation to Philosophy. Routledge.
Wittgenstein, L. (1961 [1922]). Tractatus Logico-Philosophicus/Logisch-Philosophische Abhandlung.
Routledge.
——— (1953). Philosophical Investigations/Philosophische Untersuchungen. Blackwell.
——— (1969). On Certainty/Über Gewissheit. Blackwell.
——— (1993ff). Ludwig Wittgenstein – Wiener Ausgabe, ed. M. Nedo. Springer.
——— (1998ff). Wittgenstein’s Nachlass. The Bergen Electronic Edition. Oxford University Press.
Index
A analytical practice, B
A Behaviorst’s Account of the experimentation,â•‡ 97, 98 Bach, K.,â•‡ 121, 214, 232
Significant Symbol (Journal analytical practice, Baker, G.P.,â•‡ 231
of Philosophy),â•‡ 180 interventional,â•‡ 98 Baker, J.,â•‡ 108, 117–19
A Derrida Dictionary (2004),â•‡ 75 anaphoric interpretation,â•‡ Bakhtin, concrete utterance,â•‡ 44
A Study of the Relations of Signs 140, 145 Bakhtin, linguistic
and Values,â•‡ 183 anaphoric pronouns,â•‡ 137, 139–40, pragmatics,â•‡ 41
absolute value (Kantotle),â•‡ 144 Bakhtin, Mikhail,â•‡ 38–50
119–20 anaphors,â•‡ 141 Bakhtin, M.M.,â•‡ 38–50
abstract system,â•‡ 46 answering comprehension,â•‡ 191 Bakhtin Circle,â•‡ 38, 72
accessibility relations R,â•‡ 211 Apel, K.O.,â•‡ 126, 289–90, Bakhtin industry, The,â•‡ 40–1
accomplishment verbs,â•‡ 161 292–5 Bakhtin’s biography,â•‡ 38–40
acquaintance,â•‡ 257 Apel’s transcendental Bakhtin’s view of language,â•‡
action,â•‡ 34,â•‡ 222 pragmatics,â•‡ 292–5 41–50
action and responsibility,â•‡ 226 appropriateness,â•‡ 22 Bally, C.,â•‡ 128
adjectives,â•‡ 132 appropriation of discourses,â•‡ 97 Bar-Hillel, Y.,â•‡ 21, 123, 232
Adorno, T.W.,â•‡ 289 Archéologie du savoir,â•‡ 86 Barcan formula,â•‡ 166
adverbs,â•‡ 132 archive,â•‡ 91 Barwise, J.,â•‡ 35, 175, 176, 209
agent causation,â•‡ 223 Aristotle,â•‡ 153 Basic Set,â•‡ 54
Aho, T.,â•‡ 175 Aristotle’s modalities,â•‡ 163 Bayesian decision theory,â•‡ 222–3
alethic modalities,â•‡ 163 Aristotle’s theory,â•‡ 163 Beaney, M.,â•‡ 253
Aloni, M.,â•‡ 249,â•‡ 250 Armstrong, D.M.,â•‡ 236 Bedeutung,â•‡ 253–5
Amaral, F.S.,â•‡ 275 Art and answerability,â•‡ 38 Begriffschrift (1879),â•‡ 164
American pragmatic assertion and truth,â•‡ 32 behavioral dispositions,â•‡ 236
movement,â•‡ 193 assurance theory,â•‡ 83 behavioral semiotics,â•‡ 193
analytic philosophy see Attardo, S.,â•‡ 123 behaviorism,â•‡ 182, 235
analytical philosophy attributive,â•‡ 271 behavioristics,â•‡ 182
Analytica Priora 1,â•‡ 163 Augustinian Picture,â•‡ 303 being,â•‡ 75
analytical philosophy, Austin, J.L.,â•‡ 16–17, 26–36, Bennett, M.R.,â•‡ 307
definition,â•‡ 11 27–36 Bernard, J.,â•‡ 250
analytical philosophy, Austin and Wittgenstein,â•‡ 26 Bezuidenhout, A.,â•‡ 64
meta-philosophical Austin’s approach to Bianchi, A.,â•‡ 253
issues,â•‡ 11 philosophy,â•‡ 26 Bianchi, C.,â•‡ 53–68, 253–77
analytical philosophy, state of Austin’s language isolation,â•‡ 28 binarisms,â•‡ 74
art,â•‡ 22 Austin’s philosophical biosemantics,â•‡ 240
analytical philosophy and method,â•‡ 26–8 Black, M.,â•‡ 33, 122
pragmatics,â•‡ 21–3 Austin’s tendencies,â•‡ 28–9 Blackburn, S.,â•‡ 166, 282
analytical philosophy availability principle,â•‡ 62 Blakemore, D.,â•‡ 284, 285, 287
developments,â•‡ 18–21 Avramides, A.,â•‡ 123 Blanchot, M.,â•‡ 76
analytical practice,â•‡ 97–100 axiology,â•‡ 193, 194 Block, N.,â•‡ 236, 237, 238, 240
analytical practice, changes in axiom of extensionality,â•‡ 158 Bloomfield, L.,â•‡ 127, 184
social order of discourse,â•‡ 99 axiom types,â•‡ 281 Blue Book (Wittgenstein,
analytical practice, axiomatic systems of modal 1933–34),â•‡ 26
discourse-knowledge- logic,â•‡ 245 Boeckh, P.,â•‡ 127
power,â•‡ 100 Ayer, A.J.,â•‡ 26, 104, 306 Bolzano, B.,â•‡ 13, 174
310 Index
Boolos, G.,â•‡ 158 classical philosophy of Couturat, L.,â•‡ 245

Borg, E.,â•‡ 54, 57 consciousness,â•‡ 292 creature construction,â•‡ 117–19
bound anaphors,â•‡ 141, 143 classical semiotical theories,â•‡ 292 Cresswell, M.J.,â•‡ 167, 245, 248
box operator,â•‡ 165, 246 cognitive command,â•‡ 260 critical hermeneutics,â•‡ 126
Brandist, C.,â•‡ 38, 42 cognitive linguistics,â•‡ 128–9 critical theory,â•‡ 289
Brandom, R.,â•‡ 20 cognitive pragmatics,â•‡ 240–2 cross-sentential/discourse
Braun, D.,â•‡ 148, 264 cognitive relation,â•‡ 274 anaphora,â•‡ 145, 146
Brentano, F.,â•‡ 217, 218 cognitive sciences,â•‡ 235 CSA, see Context Shifting
Bromberger, S.,â•‡ 231 cognitive semiotics,â•‡ 197 Argument,â•‡ 60
Burge, T.,â•‡ 83, 231 Cohen, L.J.,â•‡ 122 cutting up the semiotic pie,â•‡ 112
coindexation vs. Cuypers, S.E.,â•‡ 235–42
C anaphoricity,â•‡ 141
California: The Ultimate combinatorics vs. D
Guidebook,â•‡ 138 preferences,â•‡ 177 Dancy, J.,â•‡ 79
The Cambridge Introduction to common sense knowledge,â•‡ 176 Darwinian biology,â•‡ 186
Jacques Derrida (2007),â•‡ 75 complex demonstratives,â•‡ 132, Das Apriori der
Cann, R.,â•‡ 202, 207–13 135, 137 Kommunikationsgemein-
canonical language,â•‡ 15 compositionality and schaft und die Grundlagen
Cappelen, H.,â•‡ 54, 55, 57, 60, 63, context-dependence,â•‡ 161 der Ethik (1973b),â•‡ 295
133, 149 conceived value,â•‡ 199 Dascal, M.,â•‡ 240
Carnap, R.,â•‡ 14, 15, 21, 53, 155, Concept of Mind (1949),â•‡ 306 Davidson, A.,â•‡ 76
166, 168, 181, 182, 195, 210, concept words Davidson, D.,â•‡ 20, 34, 80, 156,
229, 245–6, 256 (Begriffswörter),â•‡ 253 175, 224–6, 232, 240, 282–3
Carnap, Rudolf and conceptual role semantics,â•‡ 240 Davidson, principle of
encyclopedia of unified concrete utterance,â•‡ 46 charity,â•‡ 20
science,â•‡ 14–15 connectives,â•‡ 164 Davis, S.,â•‡ 232
carnival,â•‡ 49–50 consciousness,â•‡ 237 Davis, W.A.,â•‡ 123
carnival, categories of,â•‡ 50 constatives,â•‡ 32 The day of art,â•‡ 38
carnivalistic mésalliances,â•‡ 50 constructivist structuralist,â•‡ 95 de dicto vs. de re
carnivalization,â•‡ 49–50 consumerist semantics,â•‡ 265 constructions,â•‡ 166
Carruthers, P.,â•‡ 241 content,â•‡ 54 De interpretatione
Carston, R.,â•‡ 22, 57, 64, 241, context,â•‡ 198 (Aristotle),â•‡ 163
280, 284 context-free grammar,â•‡ 203 de Rose, K.,â•‡ 23
cartesian product,â•‡ 212 context of use,â•‡ 132 de Saussure, F.,â•‡ 41, 47
Casanovas, P.,â•‡ 88 context of utterance,â•‡ 213, 270 deconstruction,â•‡ 71–7
Castañeda, H.-N.,â•‡ 135, 140 context sensitive,â•‡ 63, 65 deconstruction, criticism
causal relations,â•‡ 236 context-sensitive expression,â•‡ against,â•‡ 77
Cavell, S.,â•‡ 29 57, 132 deconstruction, Derrida’s
Chapman, S.,â•‡ 122 context-sensitivity,â•‡ 136 work on,â•‡ 71
character,â•‡ 54, 213 context shifting argument,â•‡ 60 deconstruction, Freud’s
characteristic function,â•‡ 207 contextualism,â•‡ 22 contribution to,â•‡ 71
Chartier, R.,â•‡ 90 contextualism, defined,â•‡ 54 deconstruction, historical
Chierchia, G.,â•‡ 203, 211 contextualism, background,â•‡ 71–3
Chisholm, R.,â•‡ 223, 224 perspectives,â•‡ 53–4 deconstruction, textual
Chomsky, N.,â•‡ 107, 230, 231 conversation analysis,â•‡ 129 unconscious,â•‡ 71
Chouliaraki, L.,â•‡ 99 conversational implicatures,â•‡ 114 deconstruction in literature and
chronotope,â•‡ 49 conversationalist linguistics,â•‡ 76–7
Church, A.,â•‡ 158 hypothesis,â•‡ 112–14 deconstruction tenets,â•‡ 73–6
Churchland, P.M.,â•‡ 237 Cooper, R.,â•‡ 209 deconstruction theory and
Churchland Smith, P.,â•‡ 237 Corazza, E.,â•‡ 131–50 practice,â•‡ 71
circumstances of evaluation,â•‡ 213 correspondence theory,â•‡ 174 definite description,â•‡ 154, 265–77
Clark, H.H.,â•‡ 38, 182 cosmology,â•‡ 194 deflationary,â•‡ 64
Clark, L.,â•‡ 122 counts-as parameter,â•‡ 65–6 deictic,â•‡ 141
Index 311
deictic frame,â•‡ 168 doctrine of saying and event causation,â•‡ 223

deictic temporal reference,â•‡ 168 showing,â•‡ 301–2 Evnine, S.,â•‡ 283
deixis, study of,â•‡ 21 Dokic, J.,â•‡ 64, 131 Excuses (1979),â•‡ 33
deliberation and action,â•‡ 222–3 Donnellan, K.S.,â•‡ 260–5, 271–5 existential generalization,â•‡ 156
demonstrative conventions,â•‡ 32 double-voiced discourse experimentalism,â•‡ 187
demonstrative pronouns,â•‡ 132 (dvugolosoe slovo),â•‡ 43 experimentation,â•‡ 88, 98
demonstratives,â•‡ 54, 134–5 Dowty, D.R.,â•‡ 157, 232, 283 expressions,â•‡ 54
Dennett, D.C.,â•‡ 104, 237–9 Dretske, F.I.,â•‡ 240 extension,â•‡ 210
denotation function,â•‡ 204 Dreyfus, H.L.,â•‡ 85 extensional operator,â•‡ 212
Derrida, J.,â•‡ 71, 73–6 Ducrot, O.,â•‡ 87 extensional vs. intensional
descriptive conventions,â•‡ 32 Dummett, M.,â•‡ 12, 13, 20, 229, 231 grammatical
designatum,â•‡ 197 Duranti, A.,â•‡ 100 constructions,â•‡ 156–7
Destruktion,â•‡ 72 Dynamic Predicate extensionality principle and
Devitt, M.,â•‡ 271, 275–6 Calculus,â•‡ 249 failures,â•‡ 153–4
Dewey, J.,â•‡ 180, 181, 185, 187, 189, dynamic semantics,â•‡ 249
192, 195 dynamical interpretant,â•‡ 189 F
Di Giovanna, J.,â•‡ 28 F-equivalence,â•‡ 155
dialogic conception,â•‡ 42 E F-truth,â•‡ 155
Dialogized heteroglossia,â•‡ 43 e-languages,â•‡ 230 Fairclough, N.,â•‡ 48, 77,
dialogue,â•‡ 42–3 eccentricity,â•‡ 50 97, 99
dialogue, sense of,â•‡ 42 Edelman, G.,â•‡ 129 familiar contact among people,
dialogue and utterances,â•‡ 42 eidetic variation,â•‡ 219 carnival,â•‡ 50
diamond operator,â•‡ 165, 246 Eigennamen,â•‡ 253–5 familiar speech of the street
differance,â•‡ 74 Elements of Symbolic Logic,â•‡ 168 (famil’iarno-ploshchadnaia
Dilthey, W.,â•‡ 126 eliminative materialism,â•‡ 237 rech’),â•‡ 50
direct reference theory,â•‡ 262 embodiment,â•‡ 128 Faulkner, P.,â•‡ 82–4
discourse,â•‡ 89–92 Emerson, C.,â•‡ 40, 41, 43, 45, 50 Festival (1966),â•‡ 195
discourse, experiences,â•‡ 92 Emotions and Will (1859),â•‡ 186 Field, H.H.,â•‡ 238
discourse, forms of The End of History and the Last first philosophy,â•‡ 79, 229, 292
rationality,â•‡ 92 Man (1992),â•‡ 75 Fixation of Belief (1877),â•‡ 186
discourse, knowledge and Engel, P.,â•‡ 12 Florence, M.,â•‡ 87
power,â•‡ 92–5 English as a formal Fodor, J.A.,â•‡ 20, 21, 238, 239,
discourse, reason,â•‡ 92 language,â•‡ 157 240, 284
discourse, regularities,â•‡ 91 English Futilitarianism,â•‡ 107 Fodor, J.D.,â•‡ 286
discourse, relations of force,â•‡ 92 enounced (énoncé),â•‡ 89 folk psychology,â•‡ 237–9
discourse, strategic relations,â•‡ 92 Enrici, A.,â•‡ 85 Forguson, L.W.,â•‡ 30, 35
discourse as practice,â•‡ 89–92 entia realissima,â•‡ 120 The formal method in literary
discourse as process,â•‡ 91 enunciation,â•‡ 96 scholarship,â•‡ 39, 72
discourse as system of rules,â•‡ 91 enunciative system,â•‡ 89 formal semantics,â•‡ 202
Discourse in the novel,â•‡ 39, 45 epistemic modality,â•‡ 164 formal semantics,
discourse representation epistemological breakdown,â•‡ 86 principles,â•‡ 202–3
theory,â•‡ 249 epistemology,â•‡ 28–30, 79–81, Formal’nyi metod v
discourses,â•‡ 248 193, 229 literaturovedenii (1928),â•‡ 39
discursive event,â•‡ 91 epistemology, as part of forms of rationality,â•‡ 92
discursive formation,â•‡ 89 philosophy,â•‡ 79 Forms of time and of the
discursive intervention,â•‡ 98 equal exchange semiotics,â•‡ 195 chronotope in the novel
discursive practice,â•‡ 90 essential indexicals,â•‡ 135 (1981),â•‡ 39
Diskurs und Verantwortung Etchemendy, J.,â•‡ 163, 175–6 Foucault, condition of
(1988),â•‡ 295 Ethica Nicomachea possibility,â•‡ 86–7
Diskursethik (discursive ethics) (Aristotle),â•‡ 222 Foucault, Michel,â•‡ 85–103
(Apel 1988; Habermas ethnomethodology,â•‡ 221 Foucault and discursive
1983),â•‡ 290 evaluational contextualism,â•‡ 65 turn,â•‡ 86–9
dissemination,â•‡ 74 Evans, G.,â•‡ 265, 284 foundationalism,â•‡ 79
312 Index
Foundations of the Theory of Grice, philosophy of Huang, Y.,â•‡ 143–5

Signs (Morris),â•‡ 181, 182, language,â•‡ 104 Hughes, G.E.,â•‡ 167
189, 191 Gricean thesis,â•‡ 62 Husserl, E.,â•‡ 13, 28, 117, 217–21
frame,â•‡ 247 Grice’s cooperative principle,â•‡ 113 hybrid contexts, problem
frame of reference,â•‡ 168 Grice’s theory of meaning,â•‡ 283 of,â•‡ 155–6
free carnival,â•‡ 50 Groenendijk, J.,â•‡ 249 hymen,â•‡ 74
freedom and responsibility,â•‡ 34–5 Grosz, E.,â•‡ 75, 76 hyperintensionality,â•‡ 159–60
Frege, G.,â•‡ 11, 12–13, 26, 132, 154, Grundlagen der Arithmetik,â•‡ 26
253–6, 265 Gärdenfors, P.,â•‡ 209 I
Frege theory of sense and Gurwitsch, A.,â•‡ 220 i-languages,â•‡ 230
reference,â•‡ 154 Guttenplan, S.,â•‡ 241 ICD. See Inter-Contextual
Frege’s influence on analytical Disquotational Indirect
philosophy,â•‡ 12–13 H Report
frequently used symbols,â•‡ 214–15 Habermas, J.,â•‡ 289–96 ICD test,â•‡ 133–4
Friggieri, J.,â•‡ 33, 34 Habermas’ universal The Idea of a Social Science
function of existence,â•‡ 89 pragmatics,â•‡ 290–2 (1970),â•‡ 306
Functional Calculus of habits of behavior,â•‡ 188 idealistic metaphysics,â•‡ 188
First-Order Based on Strict Hacker, P.M.S.,â•‡ 13, 231, 307 identification of a sign,â•‡ 190
Implication,â•‡ 166 Halle, M.,â•‡ 231 identification semiotics,â•‡ 195
functionalism,â•‡ 235, 236 Hamm, F.,â•‡ 158, 159, 161 identity-theory,â•‡ 235, 236
functioning of discourses,â•‡ 91 Hare, R.M.,â•‡ 21, 306 Illustrations of the Logic of
functions-in-intension,â•‡ 158 Harman, G.,â•‡ 122, 240 Science,â•‡ 186
Harnish, R.M.,â•‡ 122, 232 Image (1976),â•‡ 195
G Harris, R.,â•‡ 87, 127 immediate interpretant,â•‡ 189
Gabilondo, A.,â•‡ 85–101 Hasker, W.,â•‡ 235 inappropriateness
Gadamer, H.-G.,â•‡ 126 Haugeland, J.C.,â•‡ 237, 238, 240 argument,â•‡ 60, 61
Gamut, L.T.F.,â•‡ 215, 249 Heidegger, M.,â•‡ 71, 73, 76, indexical expressions,â•‡ 132–4
Garcia-Carpintero, S–M.,â•‡ 175 126, 289 indexical semantics,â•‡ 285
Garfinkel, H.,â•‡ 221 Hempel, C.G.,â•‡ 236 indexicalism,â•‡ 54, 55–7
Gasparov, B.,â•‡ 46, 48 hermeneuein,â•‡ 125 indexicality,â•‡ 131, 285–6
Gauker, C.,â•‡ 22, 56, 65 hermeneutic enterprise,â•‡ 126 indexicals,â•‡ 132
Gazdar, G.,â•‡ 285 hermeneutic thinking, indexicals, direct reference
Gellner, E.,â•‡ 27 evolution of,â•‡ 126 view,â•‡ 136
generalized conversational hermeneutic thinking, indexicals and contexts,â•‡ 148–50
implicature (GCI),â•‡ 114 origins of,â•‡ 125–6 indexicals as singular
generative linguistics,â•‡ 283 hermeneutical circle,â•‡ 126 terms,â•‡ 136–7
Gerbrandy, J.,â•‡ 251 hermeneutical philosophy,â•‡ 289 indirect discourse,â•‡ 154–5
Gibbs, R.,â•‡ 62 hermeneutics,â•‡ 125 informational semantics,â•‡ 240
Gillett, C.,â•‡ 235 hermeneutics of Dasein,â•‡ 126 Inquiry into Meaning and Truth
Glock, H.-J.,â•‡ 2, 12 heteroglossia,â•‡ 43 (Russell),â•‡ 153
Gochet, P.,â•‡ 153, 163, 171, 175, Higginbotham, J.,â•‡ 284 instrumentalism,â•‡ 187
244, 250 Hilpinen, R.,â•‡ 231 intension,â•‡ 210
Goldman, A.I.,â•‡ 34 Hintikka, J.,â•‡ 232, 248, 251 intension determines
Grace, W.,â•‡ 85, 89 Histoire de la folie à l’âge extension,â•‡ 157
Graham, K.,â•‡ 27, 29, 35 classique,â•‡ 88 intension problem,â•‡ 65
grammar subsystem,â•‡ 177 Histoire de la sexualité,â•‡ 88 intension vs. extension,â•‡ 153
Grandy, R.E.,â•‡ 105, 108, 117, Hodge, R.,â•‡ 77, 91 Intensional constructions,â•‡
120, 122 Hodges, W.,â•‡ 249 156–8
Gribomont, P.,â•‡ 250 Holquist, M.,â•‡ 38, 39 intensional contexts,
Grice, H.P.,â•‡ 17–18, 22, 104–20 How to Do Things with Words Frege-Carnap treatment
Grice, life of,â•‡ 105–8 (1975),â•‡ 31–3, 227 of,â•‡ 154–5
Grice, P.,â•‡ 3, 5, 9, 27, 30, 35, How to Make our Ideas Clear Intensional expressions,â•‡ 212
54, 83 (Peirce, 1878),â•‡ 186 intensional operator,â•‡ 212
Index 313
intensional semantics to Journal of Speculative Philosophy language and literature,â•‡ 88

discourse representation (1868),â•‡ 186 Language and reality,â•‡ 118
theory,â•‡ 19 Journal of Symbolic Logic,â•‡ 166 language and thought,â•‡ 229
intensionality,â•‡ 209–12 Jäger, M.,â•‡ 99 language-based conceptual
intention-convention Jäger, S.,â•‡ 99 analysis,â•‡ 22
approach,â•‡ 283 language-games,â•‡ 16, 304, 306
intentional action,â•‡ 224–5 K language-mind connection,â•‡ 283
intentional communication,â•‡ 109 K filosofii postupka (1986),â•‡ 40 language of thought,â•‡ 20, 238
intentional eliminativism,â•‡ 237 Kameyama, M.,â•‡ 174, 176, 177 language-world connection,â•‡ 283
intentional Kamp, H.,â•‡ 19, 169 langue,â•‡ 73, 128, 290
instrumentalism,â•‡ 238 Kaplan, D.,â•‡ 21, 54, 132, 134, 136, Lappin, S.,â•‡ 232
intentional realism,â•‡ 238 148, 150, 213–14, 230, 232, L’Archéologie du Savoir,â•‡ 91
intentional stance theory 245, 261–2, 265 Larson, R.,â•‡ 282
(IST),â•‡ 238 Kaplan, M.,â•‡ 30 Lasnik, H.,â•‡ 139, 145, 146
intentionality,â•‡ 237 Kaplanian model,â•‡ 139 The Late Derrida (2007),â•‡ 76
intentionality, problem Kasher, A.,â•‡ 122, 228–32 Later Derrida (2003),â•‡ 76
of,â•‡ 237–9 Katz, J.,â•‡ 171 Laws of Arithmetic,â•‡ 297
intentionality intuition,â•‡ 261 Kempson, R.,â•‡ 282 Lazar, M.,â•‡ 99
intentionality of propositional Kenny, A.,â•‡ 222 le seuil du classicisme à la
attitudes,â•‡ 238–9 Kim, J.,â•‡ 236, 237 modernité,â•‡ 87
Inter-Contextual Disquotational King, J.,â•‡ 137, 265 L’Ecole française d’analyse du
Indirect Report,â•‡ 133–4 Kölbel, M.,â•‡ 66 discours,â•‡ 86
interaction,â•‡ 289 knowing-who,â•‡ 251 Lectures on philosophy,â•‡ 13–14
International encyclopedia of knowledge and belief,â•‡ 29–30 Lemmon, E.J.,â•‡ 244, 245
unified science,â•‡ 15, 21 kommunikativnye fragmenty Leonardi, P.,â•‡ 277
interpersonal anaphora,â•‡ 146 (communicative Lepore, E.,â•‡ 7, 54, 60, 63, 133,
interpretant,â•‡ 189, 197 fragments),â•‡ 46 149, 240
interpretation semiotics,â•‡ 195 Konkin, S.S.,â•‡ 38, 40 Les mots et les choses,â•‡ 87, 88
interpretationism,â•‡ 240 Konkina, L.S.,â•‡ 38, 40 Levinson, S.C.,â•‡ 109, 112, 114,
interpretative repertories,â•‡ 94 Kratzer, A.,â•‡ 211 129, 286
interpreter,â•‡ 189, 197 Kress, G.,â•‡ 49, 77, 99 Lewis, D.K.,â•‡ 111, 139, 145, 236,
intertextuality,â•‡ 46 Kripke, S.A.,â•‡ 19, 80, 139, 145, 232, 283–6
Introduction to Montague 246–7, 261–5, 271, 273–7, 283 Lewis, I.,â•‡ 164–5, 185, 244
Semantics,â•‡ 157 Kripke’s hypothesis,â•‡ 276 liberty of indifference,â•‡ 225
intuitive truth-conditions,â•‡ 61 Kristeva, J.,â•‡ 40, 46 liberty of spontaneity,â•‡ 225
The Irony of Deconstruction Limited Inc (Derrida 1988),â•‡ 74
and the Example of L linguistic anthropology,â•‡ 128
Marx,â•‡ 76 L-equivalence,â•‡ 155 Linguistic Anthropology
Irving Lewis’ contribution to L-truth,â•‡ 155 (1993),â•‡ 307
modal logic,â•‡ 164–5 la langue,â•‡ 128 linguistic botanizing,â•‡ 27
Iskusstvo i otvetstvennost la parole,â•‡ 128 linguistic code,â•‡ 113
(Den’ iskusstva),â•‡ 38 Lackey, J.,â•‡ 82 Linguistic Contact Without
isomorphism,â•‡ 61 Lakoff, G.,â•‡ 128, 129 Cognitive Contact,â•‡ 265
Ivanov, V.V.,â•‡ 40, 41 Lambalgen, M. van,â•‡ 158, 159, 161 linguistic context,â•‡ 176
lambda abstraction,â•‡ 207–8 linguistic direction,â•‡ 13
J lambda calculus,â•‡ 158 linguistic market,â•‡ 97
Jackendoff, R.,â•‡ 284 lambda conversion,â•‡ 207 linguistic meaning of
Jackson, F.,â•‡ 22 lambda operator,â•‡ 207 indexicals,â•‡ 132–3
Jakobson, R.,â•‡ 5, 220 Langford, C.,â•‡ 244 linguistic memory,â•‡ 46
James, W.,â•‡ 106, 107, 180, 181, The Language of Morals linguistic phenomenology,â•‡
185–7, 192, 195, 220 (1972),â•‡ 306 17, 28
Janik, A.,â•‡ 306 Language, Truth, and Logic linguistic theory,â•‡ 232
Jaspers, K.,â•‡ 219 (1936),â•‡ 306 linguistic turn,â•‡ 12, 87
314 Index
linguistic unification,â•‡ 45 material thing,â•‡ 28, 30 Montague, R.,â•‡ 19, 157, 172,
linguistics and mathematical models,â•‡ 204 202–3, 212, 215, 245, 248,
hermeneutics,â•‡ 126–9 Mathesius, V., 220 282, 283, 285
literalism,â•‡ 53 McConnell-Ginet, S.,â•‡ 203, Montague semantics, see formal
Loewer, B.,â•‡ 235 211, 283 semantics
Logic and conversation McDowell, J.,â•‡ 284 Montague’s intensional
(1975),â•‡ 18, 241 McGinn, C.,â•‡ 286 model,â•‡ 157
logic program,â•‡ 159 Mchoul, A.,â•‡ 85 Montague’s semantics,â•‡ 157
logical atomism,â•‡ 13, 14 McQuillan, M.,â•‡ 76 Montague’s theory,â•‡ 205
logical behaviorism,â•‡ 236 Mead, G.H.,â•‡ 180–3, 185, 192, Moore, G.E.,â•‡ 13, 14, 26, 104, 106
logical-connection 195, 198 Moran, R.,â•‡ 83
argument,â•‡ 223–5 meaning,â•‡ 31, 108–12, 183, Morris, C.L.,â•‡ 1, 6, 15, 21, 180–99,
logical consequence,â•‡ 247 231, 269 228, 232, 292, 293
logical constants,â•‡ 301–2 Meaning and necessity,â•‡ 155 Morris’s behavioristics and
logical constructions,â•‡ 28 meaning and pragmatics,â•‡ 180–5
logical empiricism,â•‡ 15, 18 Â�understanding,â•‡ 19–20 Morris’s pragmatics,â•‡ 185–92
logical final interpretant,â•‡ 189 meaning as use,â•‡ 231 Morris’s semiotics,â•‡ 182
logical form,â•‡ 232, 300 Medvedev, P.N.,â•‡ 38, 39 Morson, G.S.,â•‡ 39, 40, 43, 45
logical modalities,â•‡ 163 mental antecedent and Moschovakis, Y.,â•‡ 158, 159
logical necessity,â•‡ 163 action,â•‡ 223 Mueller-Vollmer, K.,â•‡ 125, 126
logical positivism,â•‡ 15 mental states,â•‡ 236 multiple-description thesis,â•‡ 225
Logical Positivism, Pragmatism Merleau-Ponty, M.,â•‡ 220 Muskens, R.,â•‡ 159, 160, 175
and Scientific Empiricism metalinguistic view vs. linguistic
(1937),â•‡ 180–1 view,â•‡ 44 N
The Logical Syntax of Language metalinguistics,â•‡ 41, 44–7 Nagel, T.,â•‡ 237
(1937),â•‡ 306 metalingvistika Naissance de la clinique,â•‡ 88
Logisch-Philosophische (metalinguistics),â•‡ 44 Naming and Necessity,â•‡ 262
Abhandlung see Tractatus Metaphysical Club,â•‡ 186 natural language,â•‡ 230–1
Logische Untersuchungen,â•‡ 220 metaphysical natural language, aspects of,â•‡ 231
logocentricism,â•‡ 71 transubstantiation,â•‡ 119 natural language, nature
L’ordre du discours and Surveiller metaphysics,â•‡ 117, 229 of,â•‡ 230–1
et punir,â•‡ 88 Method in philosophical natural meaning,â•‡ 108
Lähteenmäki, M.,â•‡ 42, 45 psychology,â•‡ 118 natural realism,â•‡ 30
Luria, A.,â•‡ 129 Miller, D.,â•‡ 71, 159 naturalistic approach,â•‡ 21
Millikan, R.G.,â•‡ 21, 240 Naturalized epistemology,â•‡ 79
M Mind, Self, and Society,â•‡ 182 Neale, S.,â•‡ 122, 123, 275
M. M. Bakhtin: sobranie mind-body problem,â•‡ 20, 235–7 neo-Hegelian absolute
sochinenii,â•‡ 41 mirror of mind,â•‡ 86 Â�idealism,â•‡ 13
M-intentions,â•‡ 117 mirror of nature,â•‡ 86 Neo-Marxist forms of
Maas, U.,â•‡ 99 Mitchell, W.,â•‡ 76 Â�deconstruction,â•‡ 71
MacKay, A.F.,â•‡ 123, 274 modal auxiliaries,â•‡ 164 Neurath, O.,â•‡ 14, 182
Maingueneau, D.,â•‡ 86, 91 modal logic,â•‡ 165–6 new theory of reference,â•‡ 262
Malcolm, N.,â•‡ 223, 307 modal logic development,â•‡ 163–4 The New York Times,â•‡ 82
mandatory semantic process of Modalities and Quantification Nichomachean Ethics,â•‡ 33, 105
saturation,â•‡ 55, 56 (Carnap 1946),â•‡ 166 Niiniluoto, I.,â•‡ 160
Mandelbaum, D.,â•‡ 128 modality of existence,â•‡ 89 non-reductive theory,â•‡ 84
Marková, I.,â•‡ 40 model theory,â•‡ 203–5 non-truth-conditional aspects of
Marksizm i filosofiia iazyka model-theory for contexts,â•‡ semantics,â•‡ 286–7
(1929),â•‡ 39 177–8 nonindexical
Martin, R.M.,â•‡ 219, 232 moderate relativism,â•‡ 35 contextualism,â•‡ 64–6
Martín Rojo, L.,â•‡ 85–101 Moise, J.,â•‡ 62 nonnatural meaning,â•‡ 108–9
Marxism and the philosophy of Monist (1905),â•‡ 187 normalization,â•‡ 94
language (1973),â•‡ 39, 72 monologic,â•‡ 42 Nunberg, G.,â•‡ 213
Index 315
O Peters, S.,â•‡ 157, 232, 283 polyphony,â•‡ 43–4

object value,â•‡ 198 Petitot, J.,â•‡ 220 Ponzio, A.,â•‡ 198
Of grammatology (Derrida Petrilli, S.,â•‡ 180–99 Popular Science Monthly,â•‡ 186
1976),â•‡ 75 pharmakon,â•‡ 74 possessive adjectives,â•‡ 132
On certainty,â•‡ 80 phatic act,â•‡ 33 possible worlds,â•‡ 232
On Interpretation,â•‡ 125 phenomena,â•‡ 217 possible worlds semantics,â•‡
ontic modalities,â•‡ 163 phenomenological 244–51
ontological Marxism,â•‡ 120 movement,â•‡ 217–19 possible worlds semantics,
ontology,â•‡ 204, 229 phenomenology, definition,â•‡ emergence of,â•‡ 245–7
Opacity and the attitudes,â•‡ 160 217, 218 possible worlds semantics to
open sentences,â•‡ 153, 154 phenomenology, linguistics, and pragmatics,â•‡ 248–51
The Open Self (1948),â•‡ 183, 195 social sciences,â•‡ 219–21 postsemantic level context,â•‡
operative value,â•‡ 198 phenomenology of 149–50
order of discourse,â•‡ 95–7 language,â•‡ 220 poststructuralism,â•‡ 73
ordinary language approach,â•‡ 18 Phenomenology of mind,â•‡ 217 Potter, J.,â•‡ 94
ordinary language Phillips, D.Z.,â•‡ 306 Powell, G.,â•‡ 241
philosophy,â•‡ 26, 106 philosophical analysis by Moore practicalism,â•‡ 193
ordinary language philosophy, and Russell,â•‡ 13–14 pragmatic intrusion,â•‡ 104
decline of,â•‡ 18 Philosophical Conceptions and The Pragmatic Movement in
overt semantic structure,â•‡ 255 Practical Results,â•‡ 186 American Philosophy
Oxford ordinary language Philosophical (Morris),â•‡ 181, 192
philosophy,â•‡ 22 Investigations,â•‡ 302–3, 306 pragmatic movement in
Oxford philosophers,â•‡ 16–18 philosophical propositions,â•‡ 14 America,â•‡ 194
philosophical studies of natural pragmatic philosophy in
P language, purposes of,â•‡ 228 US,â•‡ 192–5
Paducheva, E.V.,â•‡ 50 philosophical theory,â•‡ 232 pragmatic principles,â•‡ 250
Palmer, F.R.,â•‡ 164 Philosophische Texte (2009),â•‡ 289 pragmatic theory,â•‡ 160
Palmer, R.,â•‡ 125 Philosophische Untersuchungen/ pragmatical dimension of
Papineau, D.,â•‡ 240 Philosophical semiosis,â•‡ 183
paradigm of philosophy,â•‡ 13 Investigations,â•‡ 298 pragmatics,â•‡ 21, 285
parergon,â•‡ 74 philosophy of action,â•‡ 33–5, pragmatics, definition by
parole,â•‡ 73, 128, 220, 290 222–7, 229 Charles Morris,â•‡ 180
Partee, B.,â•‡ 203, 204, 209 philosophy of art,â•‡ 229 pragmatics defined by
Passmore, J.,â•‡ 16 philosophy of history,â•‡ 229 Â�Morris,â•‡ 195
Pateman, T.,â•‡ 231 philosophy of language,â•‡ 30, pragmatics subsystem,â•‡ 177
Paths of Life (1942),â•‡ 183, 195 31–3, 228–32, 239 pragmatisch,â•‡ 193
patterns unification,â•‡ 159 philosophy of meaning,â•‡ 239 Pragmatism (1907),â•‡ 187
Pegasus,â•‡ 167 philosophy of mind,â•‡ 20–1, pre-rational structure,â•‡ 118
Peirce, C.S.,â•‡ 180–3, 185–97, 289, 229–30 Predelli, S.,â•‡ 64, 66, 137, 139, 145
292, 294 philosophy of mind predicate logic,â•‡ 203–5
Peirce’s pragmaticism,â•‡ 185–92 naturalized,â•‡ 235 presemantic level context,â•‡ 148
Peirce’s semiotics,â•‡ 187 philosophy of religion,â•‡ 229 presuppositions,â•‡ 142
Peirce’s studies on signs,â•‡ 188 philosophy of science,â•‡ 228 primary speech genres,â•‡ 48
perception,â•‡ 30 phonetic act,â•‡ 33 Principia mathematica
performative/constative physicalism,â•‡ 235 (1910–13), 164, 297
dichotomy hypothesis,â•‡ 33 picture theory of meaning,â•‡ 299 principle of acquaintance,â•‡ 257
performative hypothesis,â•‡ 286 Pitcher, G.,â•‡ 32 principle of charity,â•‡ 20
performative Pitkin, H.F.,â•‡ 307 principle of
self-contradiction,â•‡ 294 Place, U.T.,â•‡ 236 compositionality,â•‡ 161,
performative utterances,â•‡ 31–2 The Play Group,â•‡ 106 203, 240
Perry, J.,â•‡ 22, 35, 58, 67, 131, 132, polarization,â•‡ 96 principle of identifying
135, 136 The Politics of Deconstruction descriptions,â•‡ 260
personal pronouns,â•‡ 54, 132 (2007),â•‡ 76 principle of intensionality,â•‡ 155
316 Index
principle of verifiability,â•‡ 15 Quine, W.V.O.,â•‡ 6, 18–19, 79, 107, rule of necessitation,â•‡ 245
The problem of speech genres,â•‡ 160, 239, 250 Russell, B.,â•‡ 13–14, 53, 154, 213,
39, 47 Quinean approach to 256–8, 265–8, 297
The problem of the text epistemology,â•‡ 79 Russell’s theory of
(1959–61),â•‡ 44 description,â•‡ 268
Problema rechevykh zhanrov R Ryle, G.,â•‡ 16, 104, 106, 223,
(1952–3),â•‡ 39 Rabelais and his world 236, 306
problematization,â•‡ 88–9, 98 (1968),â•‡ 39
problematization procedure,â•‡ 99 Rabinow, P.,â•‡ 85 S
Problems of Dostoevsky’s creative radical contextualism,â•‡ 53, 57–64 Saarinen, E.,â•‡ 160
art,â•‡ 39 radical contextualism, salience,â•‡ 177
Problems of Dostoevsky’s poetics motivations for,â•‡ 60–2 Salmon, N.,â•‡ 275
(1973),â•‡ 39 radical contextualism, Sandu, G.,â•‡ 175
Problemy poètiki Dostoevskogo objections to,â•‡ 62–4 Sapir-Whorf hypothesis,â•‡ 289
(1963),â•‡ 39 radical pragmatic,â•‡ 105 Sarangi, S.K.,â•‡ 100, 123
Problemy tvorchestva Railton, P.,â•‡ 9 Sbisà, M.,â•‡ 1, 5, 11, 26, 33, 119, 131
Dostoevskogo (1929),â•‡ 39 range of sentence,â•‡ 246 scepticism,â•‡ 81
process of objectivization,â•‡ 93 rationality,â•‡ 114–17 Schiffer, S.R.,â•‡ 107, 109, 122
profanation,â•‡ 50 rationality, evaluative Schleiermacher, F.,â•‡ 126
programmatic theories,â•‡ 231–2 principles,â•‡ 115 scholasticism,â•‡ 108
Prolegomena to an Apology Raz, J.,â•‡ 223 Schuetz, A.,â•‡ 220
for Pragmaticism re-problematization,â•‡ 98 science of ideology,â•‡ 86
(1906),â•‡ 187–8 reason,â•‡ 92 Scientific Empiricism
Promoting extensionality,â•‡ 160 reconstructive science,â•‡ 294 (Morris),â•‡ 180, 181
prononcer les idées,â•‡ 86 recursive definition of Scott, D.,â•‡ 154, 165, 259, 267, 268
proper name as rigid truth,â•‡ 247 Searle, John,â•‡ 258–60
Â�designator,â•‡ 263 reductionism,â•‡ 183 Searle, J.R.,â•‡ 5, 20, 33, 57, 58–60,
proper names,â•‡ 253–65, 257, reductive theory,â•‡ 84 106, 111, 121, 122, 195, 227,
258–9 reference,â•‡ 210, 257, 260 232, 235, 237, 241, 258–60
The proper Treatment of reference pointer,â•‡ 169 second-level constructs,â•‡ 221
Quantification in reference shift,â•‡ 156 secondary speech genres,â•‡ 48
ordinary English referential,â•‡ 271 self-controlled conduct,â•‡ 193
(Montague 1974),â•‡ 157 Reichenbach, H.,â•‡ 15, 21, 132, 168 semantic ambiguity,â•‡ 156
propositional attitude relations of force,â•‡ 92 semantic argument,â•‡ 263
Â�psychology,â•‡ 238–9 relativism,â•‡ 66 semantic atomism,â•‡ 240
propositional attitudes,â•‡ 237 relevance theory,â•‡ 22,â•‡ 241 semantic competence,â•‡ 246
propositional attitudes and representation of a semantic content,â•‡ 63
pragmatics,â•‡ 160–1 representation semantic holism,â•‡ 239
propositional calculus,â•‡ 165 (thought),â•‡ 86 semantic level context,â•‡ 148–9
propositional functions,â•‡ representational form,â•‡ 300 semantic minimalism,â•‡ 53, 54–5
154, 266 representational theory of the semantic problems,â•‡ 15
proto-perceptions,â•‡ 135 mind (RTM),â•‡ 238 semantic reference,â•‡ 275
psychosemantics,â•‡ 239–40 Reyle, U.,â•‡ 19 semantic representation,â•‡ 284
pure indexicals,â•‡ 134–5 rhetic act,â•‡ 33 semantic significance,â•‡ 248
Putnam, H.,â•‡ 64, 66, 131, 137, 138, Rijke, M. de,â•‡ 166 semantic theory,â•‡ 15, 155
139, 146, 147 Roberts, C.,â•‡ 100 semantic types,â•‡ 205–7
Pêcheux, M.,â•‡ 99 Rorty, R.,â•‡ 12, 76, 85 semantic
Roulet, E.,â•‡ 44 underdetermination,â•‡ 53
Q Recanati, F.,â•‡ 22, 35, 53, 57, 58, semantical dimension of
quantified modal logic,â•‡ 166–7 62–6, 109, 136, 160, 161, semiosis,â•‡ 183
quantifiers,â•‡ 175–6, 208–9 232, 241 semantics,â•‡ 21, 197
Questions Concerning Certain rule-by-rule hypothesis,â•‡ 203 semantics, approaches to,â•‡
Faculties Claimed for Man rule of functional 171–2
(1868),â•‡ 188 application,â•‡ 206 semantics, origins of term,â•‡ 171
Index 317
semantics to pragmatics, shift of Soams, S.,â•‡ 232 syntactical elements,â•‡ 54, 56

attention,â•‡ 19 social order of discourse,â•‡ 97 Szabò, Z.G.,â•‡ 55, 56, 62, 63, 67,
semantics vs. pragmatics,â•‡ 274–7 Some Consequences of Four 241, 242
semioethics,â•‡ 197 Incapacities (1868),â•‡ 188, 189
semiosical theory of mind,â•‡ 193 Sosa, E.,â•‡ 79, 89 T
semiosis,â•‡ 184 The space of literature T-sentences,â•‡ 280
semiotic sign-function.,â•‡ 292 (Blanchot 1989),â•‡ 76 tacit initiator,â•‡ 141–2
semiotically interpreted spatial indexicals,â•‡ 138 Tannen, D.,â•‡ 129
logic,â•‡ 193 speaker’s reference,â•‡ 275 Tarski, A.,â•‡ 3, 15, 53, 156, 157, 171,
semiotics,â•‡ 197 Specters of Marx (1994),â•‡ 75, 76 172, 174, 282, 283
sense-datum,â•‡ 28, 30 spectrality,â•‡ 75 Tarskian formula,â•‡ 282
sentence nonliterality,â•‡ 63 speech act,â•‡ 32–3, 110 Tarskian semantics,â•‡ 156–7,
sentences (Sätze),â•‡ 253 speech act, facets,â•‡ 33 172–4
sentential functions,â•‡ 154 speech act content,â•‡ 63 Tarski’s semantics, scope and
sentential utterance,â•‡ 111–12 speech act pluralism,â•‡ 55 limits of,â•‡ 174–5
set-theory,â•‡ 158 speech act theory,â•‡ 33 tautology,â•‡ 244–5
Shepherd, D.,â•‡ 38 speech acts,â•‡ 232, 290 Taylor, K.,â•‡ 63,â•‡ 76, 241
Shoemaker, S.,â•‡ 237 Speech acts,â•‡ 227, 291 teleological semantics,â•‡ 240
Sidelle, A.,â•‡ 137 speech genres,â•‡ 47–8 tense logic,â•‡ 167–8
sign,â•‡ 188 Sperber, D.,â•‡ 22, 57, 104, 122, 241, tense logic to pragmatics,â•‡
sign activity,â•‡ 191 284, 286, 287 168–9
sign-processes,â•‡ 187 Spiegelberg, H.,â•‡ 219 Ter Meulen, A.,â•‡ 203, 204
sign/sign vehicle,â•‡ 197 Stalnaker, R.C.,â•‡ 22, 142, 232 testimonial knowledge,â•‡ 83
signification,â•‡ 197–8 standard semantics, Thayse, A.,â•‡ 157, 248
Signification and Significance inadequacies of,â•‡ 158–9 Theorie des kommunikativen
(1964),â•‡ 183, 194, 195, 196 Stanley, J.,â•‡ 55, 56, 62, 149 Handelns (1981),â•‡ 289
significatum,â•‡ 189, 197 state-description,â•‡ 246 theory of action,â•‡ 35
significs,â•‡ 197 Stenius, E.,â•‡ 300 theory of interpretation,â•‡ 239
signifiers,â•‡ 72 Sterelny, K.,â•‡ 282 theory of meaning,â•‡ 19, 171, 220
Signs, Language and Stevenson, C.L.,â•‡ 21 theory of mental
Behavior,â•‡ 182, 187, 189, 191, Stokhof, M.,â•‡ 249 representations,â•‡ 238
193, 194 strategic relations,â•‡ 92 theory of reference,â•‡ 171
Signs about Signs about Strawson, Peter,â•‡ 269–71 theory of signs,â•‡ 193, 194
Signs,â•‡ 189 Strawson, P.F.,â•‡ 17, 22, 32, 54, 106, theory of truth-conditions,â•‡ 280
signs and values,â•‡ 195–9 213, 232, 269–71, 283 Thomas, W.,â•‡ 127, 192
single-voiced discourse strict conditional,â•‡ 165 Thomason, R.,â•‡ 159
(odnogolosoe slovo),â•‡ 43 strict implication,â•‡ 165, 244 thought (Gedanke),â•‡ 254
singular propositions,â•‡ 136 strong moderate relativism,â•‡ 65 Tichy, P.,â•‡ 156
situational context,â•‡ 176 structuralism,â•‡ 128 Tihanov, G.,â•‡ 38
situationalism,â•‡ 64 structure as metanarrative,â•‡ 73 Toward a philosophy of the act
Six Theories of Mind subjection of discourses,â•‡ 94 (1993),â•‡ 40
(Morris),â•‡ 180, 186 subjection of individuals,â•‡ 94 trace,â•‡ 73
Slavic and east European subjective meaning-intentions Tractatus
journal (1986),â•‡ 39 (Bedeutungsinten- logico-philosophicus,â•‡ 14,
Slembrouck, S.,â•‡ 123 tionen),â•‡ 220 297, 298, 299–301, 301
slovo,â•‡ 43 subjectivist semantics,â•‡ 261 transcendental
Slovo v romane (1934–5),â•‡ 39 subjectivization,â•‡ 87 consciousness,â•‡ 219
Smart, J.J.C.,â•‡ 236 supplement,â•‡ 75 transcendental-
Smith, N.V.,â•‡ 111 syllogism,â•‡ 163 phenomenological
Smith, P.K.,â•‡ 237, 241 Symbolism and Reality (Charles reduction,â•‡ 218
Smith, Q.,â•‡ 137–9 Morris),â•‡ 180 transcendental-pragmatic
Smullyan, A.F.,â•‡ 268 syntactic analysis,â•‡ 203 argumentation
Sneddon, A.,â•‡ 34 syntactical dimension of strategy,â•‡ 294
Soames, S.,â•‡ 12, 30, 271, 275 semiosis,â•‡ 183 transcendental pragmatics,â•‡ 289
318 Index
transcendental utterance-type meaning,â•‡ 107 William James Today

re-interpretation,â•‡ 293 utterer’s meaning,â•‡ 107 (1942a),â•‡ 181
transcendental-semiotical,â•‡ 293 Williamson, T.,â•‡ 6, 22
transcendental signifier,â•‡ 74, 75 V Wilson, D.,â•‡ 57, 241, 284, 286, 287
Transformation,â•‡ 295 validity-claims,â•‡ 290 Wilson, R.A.,â•‡ 122
Transformation der Philosophie Vallée, R.,â•‡ 136 Winch, P.,â•‡ 306
(Apel),â•‡ 290, 293, 295 value,â•‡ 198 Wisdom, J.,â•‡ 16
translation rules,â•‡ 204 value dimensions,â•‡ 199 Wittgenstein, Austin, Searle,
Travis, C.,â•‡ 22, 32, 35, 57–62 value situation,â•‡ 198 and Travis view on
Treatise on human Van Dijk, T.,â•‡ 96 contextualism,â•‡ 58–60
understanding,â•‡ 225 van Fraassen, B.C.,â•‡ 228 Wittgenstein, Ludwig
triadic sign-relation,â•‡ 292 Vanderveken, D.,â•‡ 232 (biography),â•‡ 297–8
true indexicals,â•‡ 54 variable assignment Wittgenstein on Rules and
truth-conditional function,â•‡ 208 Private Language
approach,â•‡ 283, 284 Varieties of Human Value (1982),â•‡ 305
truth-conditional semantics, (1956),â•‡ 183, 196 Wittgenstein’s influence and
definition,â•‡ 280 Veltman, F.,â•‡ 249 ordinary language
truth-conditional semantics and Verständigung,â•‡ 290 philosophy,â•‡ 16
pragmatics,â•‡ 284–7 Verstehen,â•‡ 219 Wittgenstein’s later
truth-theoretic approach,â•‡ 282 Vienna Circle,â•‡ 14 philosophy,â•‡ 302–6
Turner, K.,â•‡ 67, 104 Vološinov, V.N.,â•‡ 38, 39, 45, 46, Wittgenstein’s philosophy,â•‡
Turner, R.,â•‡ 67, 104 71, 72, 74, 196 298–9
Tvorchestvo Fransua Rable Wittgenstein’s thought,
i narodnaia kul’tura W evolution of,â•‡ 16
srednevekov’ia i Renessansa Waismann, F.,â•‡ 54, 59 Wittgenstein’s Tractatus,â•‡ 14,
(1965),â•‡ 39 Wall, R.E.,â•‡ 157, 203, 204, 232, 283 161, 301
two poles of the text,â•‡ 46 Warner, R.,â•‡ 108, 110, 117, 118, 120 Wittgenstein’s Vienna
type raising,â•‡ 209 Warnock, G.J.,â•‡ 26, 30, 32, 33 (1973),â•‡ 306
typed logical language,â•‡ 205 Was heißt Universalpragmatik Wodak, R.,â•‡ 48, 99
(Habermas 1976),â•‡ 289 Wolper, P.,â•‡ 166
U Wetherell, M.,â•‡ 94, 95 words-world relationship,â•‡ 28
Über Sinn und Bedeutung,â•‡ 154 Wettstein, H.K.,â•‡ 232, 261, 265, writing,â•‡ 74
unbound anaphors,â•‡ 143–5 271, 275
unitary language,â•‡ 45 Whitehead, A.N.,â•‡ 13, 164, Y
universal grammar,â•‡ 232 165, 297 Yeager, D.,â•‡ 34
universal pragmatics,â•‡ 289 Whittaker, R.,â•‡ 96
universe narrowers,â•‡ 135 why?-questions,â•‡ 228–9 Z
Update Semantics,â•‡ 249 Wiener Ausgabe,â•‡ 298 Žižek, S.,â•‡ 77
Urmson, J.O.,â•‡ 13, 32 Wierzbicka, A.,â•‡ 48 Znaniecki, F.,â•‡ 127
utterance and context,â•‡ 22 Will to Believe (1897),â•‡ 187 Zolotova, G.A.,â•‡ 49
In the series Handbook of Pragmatics Highlights the following titles have been published thus far or
are scheduled for publication:
10 SBISÀ, Marina, Jan-Ola ÖSTMAN and Jef VERSCHUEREN (eds.): Philosophical Perspectives for Pragmatics.
2011. xv, 318 pp.
9 ÖSTMAN, Jan-Ola and Jef VERSCHUEREN (eds.): Pragmatics in Practice. ca. 250 pp. Expected Forthcoming
8 ZIENKOWSKI, Jan, Jan-Ola ÖSTMAN and Jef VERSCHUEREN (eds.): Discursive Pragmatics.
xv, 296 pp. + index. Expected July 2011
7 JASPERS, Jürgen, Jan-Ola ÖSTMAN and Jef VERSCHUEREN (eds.): Society and Language Use. 2010. xiii, 324 pp.
6 FRIED, Mirjam, Jan-Ola ÖSTMAN and Jef VERSCHUEREN (eds.): Variation and Change. Pragmatic perspectives.
2010. x, 275 pp.
5 BRISARD, Frank, Jan-Ola ÖSTMAN and Jef VERSCHUEREN (eds.): Grammar, Meaning and Pragmatics. 2009.
xiii, 308 pp.
4 D’HONDT, Sigurd, Jan-Ola ÖSTMAN and Jef VERSCHUEREN (eds.): The Pragmatics of Interaction. 2009.
xiii, 262 pp.
3 SANDRA, Dominiek, Jan-Ola ÖSTMAN and Jef VERSCHUEREN (eds.): Cognition and Pragmatics. 2009.
xvii, 399 pp.
2 SENFT, Gunter, Jan-Ola ÖSTMAN and Jef VERSCHUEREN (eds.): Culture and Language Use. 2009. xiii, 280 pp.
1 VERSCHUEREN, Jef and Jan-Ola ÖSTMAN (eds.): Key Notions for Pragmatics. 2009. xiii, 253 pp.

Philosophical Perspectives For Pragmatics

Uploaded by

Copyright:

Available Formats

Philosophical Perspectives For Pragmatics

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Philosophical Perspectives For Pragmatics

Uploaded by

Copyright:

Available Formats

What is the focus of each volume of the Handbook of Pragmatics Highlights?

What is the focus of each volume of the Handbook of Pragmatics Highlights?

What information do the Handbook of Pragmatics and Handbook of Pragmatics Online provide?

What information do the Handbook of Pragmatics and Handbook of Pragmatics Online provide?

Philosophical Perspectives for Pragmatics

Handbook of Pragmatics Highlights (HoPH)

The ten volumes of Handbook of Pragmatics Highlights focus on the

John Benjamins Publishing Company

Library of Congress Cataloging-in-Publication Data

© 2011 – John Benjamins B.V.

Preface to the series xiii

Analytical philosophy – Ordinary language philosophy 11

4.3 Meaning and understandingâ•… 19

H.P. Grice 104

3.  alue and the new metaphysicsâ•… 117

Indexicals and demonstratives 131

Intensional logic 153

8. Propositional attitudes and pragmaticsâ•… 160

Modal logic 163

Model-theoretic semantics 171

Charles Morris 180

Notation in formal semantics 202

Philosophy of action 222

Philosophy of language 228

Philosophy of mind 235

Possible worlds semantics 244

Reference and descriptions 253

1.4 Taking stockâ•… 260

Truth-conditional semantics 280

Universal and transcendental pragmatics 289

Ludwig Wittgenstein 297

In 1995, the first installments of the Handbook of Pragmatics (HoP) were

– Key notions for pragmatics

– The pragmatics of interaction

 Jan-Ola Östman (University of Helsinki) &

 Jan-Ola Östman (University of Helsinki) &

1.â•… Pragmatics and philosophy

2.â•… Conceptions of meaning

3.â•… Speech as action

while it is correct to trace meaning back to use, it is necessary to make distinctions

4.â•… Mind and self

5.â•… Doing pragmatics, doing philosophy

Taking Wittgenstein’s way of doing philosophy a bit too literally (philosophy as an

Glock, H.-J. (2004). What is analytic philosophy? Cambridge University Press.

1.â•… Philosophy as analysis

An overview of the main authors and problems of analytical philosophy in the

1.1â•… The ‘linguistic turn’

1.2â•… The influence of Frege

1.3â•… Analysis in G.E. Moore and B. Russell

2.â•… Analysis and the ideal of scientific language

2.1â•… Wittgenstein’s Tractatus

2.2â•… Rudolf Carnap and the encyclopedia of unified science

3.â•… Analysis and ordinary language

3.1â•… The evolution of Wittgenstein’s thought

3.2â•… Wittgenstein’s influence and ordinary language philosophy

3.3â•… Some Oxford philosophers

3.3.1â•… J.L. Austin

labeled ‘linguistic phenomenology’ (but others labeled also ‘linguistic botanizing’:

3.3.2â•… P.F. Strawson

Preface to the series xiii

Analytical philosophy – Ordinary language philosophy 11

H.P. Grice 104

3. alue and the new metaphysicsâ•… 117

Indexicals and demonstratives 131

Intensional logic 153

Modal logic 163

Model-theoretic semantics 171

Charles Morris 180

Notation in formal semantics 202

Philosophy of action 222

Philosophy of language 228

Philosophy of mind 235

Possible worlds semantics 244

Reference and descriptions 253

Truth-conditional semantics 280

Universal and transcendental pragmatics 289

Ludwig Wittgenstein 297

Jan-Ola Östman (University of Helsinki) &

Jan-Ola Östman (University of Helsinki) &

1.â•… Pragmatics and philosophy

2.â•… Conceptions of meaning

3.â•… Speech as action

4.â•… Mind and self

5.â•… Doing pragmatics, doing philosophy

1.â•… Philosophy as analysis

1.1â•… The ‘linguistic turn’

1.2â•… The influence of Frege

1.3â•… Analysis in G.E. Moore and B. Russell

2.â•… Analysis and the ideal of scientific language

2.1â•… Wittgenstein’s Tractatus

2.2â•… Rudolf Carnap and the encyclopedia of unified science

3.â•… Analysis and ordinary language

3.1â•… The evolution of Wittgenstein’s thought

3.2â•… Wittgenstein’s influence and ordinary language philosophy

3.3â•… Some Oxford philosophers

4.â•… Further developments of analytical philosophy

4.1â•… W.V.O. Quine: From analysis to naturalization

4.2â•… From intensional semantics to discourse representation theory

4.3â•… Meaning and understanding

4.4â•… Philosophy of mind

5.â•… Analytical philosophy and pragmatics