Philosophical Perspectives for Pragmatics (Handbook of Pragmatics Highlights)

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. Each€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 University of Antwerp

Jan-Ola Östman 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

8

TM

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

Library of Congress Cataloging-in-Publication Data Philosophical perspectives for pragmatics / edited by Marina Sbisà, Jan-Ola Östman and Jef Verschueren. 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 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 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

1

11

VI

Philosophical Perspectives for Pragmatics

5.

4.3 Meaning and understandingâ•… 19 4.4 Philosophy of mindâ•… 20 Analytical philosophy and pragmaticsâ•… 21

John L. Austin 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

26

Mikhail Bakhtin 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

38

Contextualism Claudia Bianchi 1. Two perspectivesâ•… 53 2. Semantic minimalismâ•… 54

53



3. 4.

5. 6.

Table of contents

Indexicalismâ•… 55 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 Nonindexical contextualismâ•… 64 Conclusionâ•… 67

Deconstruction Tony Schirato 1. Introductionâ•… 71 2. Historical backgroundâ•… 71 3. Basic tenetsâ•… 73 4. Deconstruction in literature and linguisticsâ•… 76 5. Against deconstructionâ•… 77

71

Epistemology Filip Buekens

79

Epistemology of testimony Paul Faulkner

82

Michel Foucault Luisa Martín Rojo & Angel Gabilondo Pujol 1. Introductionâ•… 85 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

85

H.P. Grice Frank Brisard 1. Lifeâ•… 105 2. Languageâ•… 108 2.1 Meaningâ•… 108 2.2 The conversationalist hypothesisâ•… 112 2.3 Rationalityâ•… 114

104

VII

VIII Philosophical Perspectives for Pragmatics

3.

4. 5.

 alue and the new metaphysicsâ•… 117 V 3.1 Creature constructionâ•… 117 3.2 Absolute value (Kantotle)â•… 119 Concluding remarksâ•… 120 Further readingâ•… 122

Hermeneutics Piet Van de Craen 1. Introductionâ•… 125 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

125

Indexicals and demonstratives Eros Corazza 1. Introductionâ•… 131 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 7. Conclusionâ•… 150

131

Intensional logic 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

153



8. 9.

Table of contents

Propositional attitudes and pragmaticsâ•… 160 Intension, compositionality and context-dependenceâ•… 161

Modal logic 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

163

Model-theoretic semantics 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

171

Charles Morris 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

180

Notation in formal semantics 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

202

IX

X

Philosophical Perspectives for Pragmatics

Phenomenology 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

217

Philosophy of action Filip Buekens

222

Philosophy of language Asa Kasher 1. Introductionâ•… 228 2. Natural languageâ•… 230 3. Certain aspects of natural languageâ•… 231 4. Programmatic theoriesâ•… 231

228

Philosophy of mind 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

235

Possible worlds semantics 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

244

Reference and descriptions Andrea Bianchi 1. Do proper names describe?â•… 253 1.1 Gottlob Fregeâ•… 253 1.2 Bertrand Russellâ•… 256 1.3 John Searleâ•… 258

253



2.

3.

Table of contents

1.4 Taking stockâ•… 260 1.5 Saul Kripkeâ•… 261 Do definite descriptions refer?â•… 265 2.1 Russell againâ•… 265 2.2 Peter Strawsonâ•… 269 2.3 Keith Donnellanâ•… 271 One further issue: Semantics versus pragmaticsâ•… 274

Truth-conditional semantics 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

280

Universal and transcendental pragmatics Joachim Leilich 1. Originsâ•… 289 2. Habermas’ universal pragmaticsâ•… 290 3. Apel’s transcendental pragmaticsâ•… 292

289

Ludwig Wittgenstein 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

297

Index

309

XI

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 pragmatic 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 methodologies, 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 intelligibility 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 information 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 overview 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, cultural, variational, interactional, or discursive, can always have the most relevant encyclopedic articles at their fingertips. Affordability, topical organization and selectivity also turn these books into practical teaching tools which can be used as reading materials for a wide range of pragmatics-related linguistics courses. With this endeavor, we hope to make a further contribution to the goals underlying 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, �without the help of dozens, even hundreds of scholars. First of all, there are the authors themselves, who sometimes had to work under extreme conditions of time �pressure. Further, most members of the IPrA Consultation Board have occasionally, and some repeatedly, been called upon to review contribu�tions. Innumerable additional �scholars were thanked in the initial versions of handbook entries. All this makes the Handbook of Pragmatics a truly joint endeavor by the pragmatics community world-wide. We are greatly indebted to you all. We do want to specifically mention the important contributions over the years of three scholars: the co-editors of the Manual and the first eight annual installments, 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 Versluys. 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 communication, 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 problems in pragmatics are often not only a matter of research within one accepted paradigm, 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 possible routes of reflection at the interface between pragmatics (as a interdisciplinary field) and philosophy. The Â�collection provides information about some of the philosophical 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, Reference 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 semantics). If the collection does not comprise articles dealing with such classic topics of philosophical pragmatics as speech acts, presupposition, implicature, and the semantics 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 various 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 philosophical 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 constantly 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 studying such an activity requires taking for granted that the texts by means of which it is



Introduction

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 definitions. 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 sentence 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 conceptions (speaker meaning and meaning as truth conditions) has triggered the development 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

3

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 enunciation, 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. illocutionary and perlocutionary acts. Austin maintained that illocutionary acts can be successfully performed only if the audience’s uptake is secured, and that their performance involves conventional effects. Perlocutionary acts correspond to the speaker’s responsibility for further effects or consequences on the audience. In Austin’s view,



Introduction

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 subsequent 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 relationships. 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 questioning 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 relation to Foucault and Austin: illocution viewed as producing changes in what participants can or have to do (see Sbisà 1984, 2006: 157–159; 2009: 43–46) basically involves matters of power.

5

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, entitlements, or to say it in one word, of pouvoir – the Â�nominalized modal verb usually translated 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 utterances 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 “private” 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 century. 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 pragmatics. Those trends of pragmatic research that investigate verbal interaction, relationship 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

Other trends of research, particularly so-called cognitive pragmatics, are mainly concerned 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 philosophical presuppositions, which may require overt philosophical discussion. What philosophy can say about utterance processing, moreover, hardly counts as a description 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 naturalism), tends to assume that speaker and hearer (as individuals, endowed with individual 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 interactional 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 postphenomenological philosophy and is not represented in this volume: Stanley Cavell’s analysis of interpersonal acknowledgement, showing how it is up to us to acknowledge 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

7

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 intermediate 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 philosophers prefer to assume some kind of mediation and therefore indirectness in cognitive 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 discussed 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 interaction, 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 ambiguity between research in pragmatics and philosophical reflection. When the pragmatic investigation of a text, or of a given interactional event, succeeds in clarifying the intersubjective 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 philosophy 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 clarification (of social and interpersonal relations, of interaction, of implicit Â�meaning…),



Introduction

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 common 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 justified (cf. Grice 1991:  79–83). Moreover, pragmatics’ highlighting interpersonal relations (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 philosophy 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 investigations. 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.

9

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 University of Trieste

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 precise 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 propositions 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 origins 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 rational 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 distinguishes 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 propositional attitudes.



Analytical philosophy – Ordinary language philosophy

Nowadays he is generally held to have laid the foundations for analytical philosophy. At the basis of the analytical Linguistic Turn lies a fundamental conception that Frege shared with other philosophical writers who used the German language (Bolzano, 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 propositions. 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 propositions, 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 propositions of common sense are true with certainty, but he thought also that the analysis 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

13

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 philosophy, 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, however, 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 tautologies. 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 mathematicians, 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



Analytical philosophy – Ordinary language philosophy

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 empiricists 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), proposing 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. Correspondingly, 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.

15

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. Nevertheless, 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, immediately influenced his Cambridge students as well as other philosophers. In Cambridge, John Â�Wisdom focused attention on a restatement of the nature of philosophical activity, emphasizing its similarity to an illness and its misleading, although illuminating, character (Wisdom 1953). In Oxford, Gilbert Ryle searched for the origin of metaphysical 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 language 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 investigation 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 interested 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



Analytical philosophy – Ordinary language philosophy

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 philosopher 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 conceptual 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 distinctions 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 linguistics 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 elaborated 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 language 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 philosophers already in the 1940s, but his most relevant contributions to philosophy were to

17

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 attention 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 natural 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 framework in which they were collocated, by challenging the dichotomy between analytic and Â�synthetic propositions (Quine 1953). According to him, synonymity as an intensional notion, which was involved in the received definition of analyticity, was too vague and uncontrollable. He outlined, therefore, a view of language both behavioristic and holistic: meaning assignment (beyond the identification of the perceptual stimulus 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.



Analytical philosophy – Ordinary language philosophy

In this framework, philosophical analysis is no longer possible either as elucidation of concepts or as translation of sentences into synonymous ones. Also, the separation 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 systematization to the language of science, purifying it from intensional language, and considered 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 independently 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 pragmatics 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 controllable 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

19

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 contributions 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). Functionalism, 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



Analytical philosophy – Ordinary language philosophy

intentionality, it remains open whether it is a fact (intentional realism) or a delusion (eliminativism) or the result of our interpretive strategies (interpretationism, see� 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 internalists 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 guaranteeing the connection between external states of affairs and their mental representations 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), semantics (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 encyclopedia 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 indexicals 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

21

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 communicative, 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 felicity conditions for speech acts (Austin 1962: 12–44) and the notion of presupposition (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 pragmatic 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 pragmatics are going to be. One salient fact is that philosophy has been influenced by cognitive pragmatics, particularly Relevance Theory (Sperber & Wilson 1986; Carston 2002). Another salient fact is that several philosophers have recently focused attention 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 sentences often contain elements not spelled out in words but necessary to the truthevaluability 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 conclusions 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



Analytical philosophy – Ordinary language philosophy

criteria for the attribution of knowledge change with context have argued for contextualism 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 language 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.

23

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.



Analytical philosophy – Ordinary language philosophy

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. Oxford University Press. 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.

25

John L. Austin Marina Sbisà University of Trieste

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 remarkable 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 activities 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 neopositivism, 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 philosophical investigations not to the construction of artificial languages but to the consideration of the workings of ordinary language, and did not restrict his considerations 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

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 useful 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 conceptual distinctions simply invented by philosophers are often arbitrary and misleading. 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 levelled (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 language, 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 language, 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 knowledge 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 manifests 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

27

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 methodological 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 isolating 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 complexity of reality by resorting to unifying abstraction, by operating reductions, or introducing symmetrical, supposedly exhaustive dichotomies, Austin is suspicious of “logical constructions” (among which the notions of truth and of statement), 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 philosophical 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

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 sharpened 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 challenge 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.;

29

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 consisting 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 perceiving 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 attention 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 principle indistinguishable from cases of veridical perception, as is alleged, since nothing is perceived in isolation, and the context of an experience usually makes discrimination 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 silvery speck is a huge star (1962: 98–99). Austin also disputes the idea that sense-data statements are incorrigible and therefore the foundations of knowledge. He argues that it is possible to misdescribe sensations 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

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 meanings 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 different 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 Aristotle, 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 common 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 different 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 promise” 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 syntactic forms besides the original performatives can be said to perform an action, provided 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

31

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 correspondence 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 similarity between language and the world, but requires a match between “descriptive conventions” linking the sentence to a type of situation and “demonstrative conventions” 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 “truthmakers” 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 contrast 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 constatives, and concludes that the dichotomy has to be abandoned. He then proceeds to formulate a new framework for the study of language, based on the notion of speech act.



John L. Austin

A few critics have misunderstood Austin’s intents and have taken him as actually seeking 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 analysis 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.

33

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 something (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 identification 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 corresponding 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 different 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

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 making, 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 recognizedly “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 occurrence 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 influenced 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 philosophy 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

35

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 perceptual 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

——— (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.

37

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, however, 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 figure, 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

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 Rabelais, 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 controversy, 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 linguistics, 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

39

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 western 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 Bakhtinian 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). Moreover, 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 considerably differ from Russian versions; or versions by literary scholars may differ from versions 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 distinction 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, however, 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

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 concrete whole utterances used in given contexts dates back to 1924, when he wrote an article entitled ‘Problema soderzhaniia, materiala, i formy v slovesnom khudozhestvennom 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 linguistics 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)

.╅ De Saussure is, however, not mentioned by name in the 1924 article or the 1929 Dostoevsky book.

41

42

Martina Björklund

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, dialogue, speech, treatise, novel, and so on” (Bakhtin 1990: 293). Bakhtin further maintained 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 contrary, 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

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, presupposing 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 utterances 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 singlevoiced 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 socioideological ‘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

43

44

Martina Björklund

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 character 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, however, 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 dialogic 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

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 similar 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 considers 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 selfidentical 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 “theoretical 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 uninterrupted 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

45

46

Martina Björklund

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 everything that conforms to the language system, viz. everything “that is repeated and reproduced, everything repeatable and reproducible, everything that can be given outside 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

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 normative 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

47

48

Martina Björklund

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 command to the multivolume novel (pp. 60–61). Further, Bakhtin (pp. 61–62) draws attention 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 genericstylistic 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

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 specifically, 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 Rabelais 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

49

50

Martina Björklund

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 (combination 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 specifically, 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’iarnoploshchadnaia 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 contemporary 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

——— (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 literaturnoe 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 University 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 introduction 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”.

51

52

Martina Björklund 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”. Zhanry Rechi: Sbornik Nauchnykh Statei 1999(2). GosUNTS “Kolledzh”. Zhanry Rechi: Sbornik Nauchnykh Statei 2002(3). 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 distinguish 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 phenomenon 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 “radical contextualism”2 – no sentence of a natural language expresses a complete proposition, 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 communication, 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 perspective 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 expressions such as true indexicals and demonstratives) – namely only when the conventional 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”), personal 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 content (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

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 context 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.

55

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 context 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

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 “obviously” indexical – namely predicates like “know”, quantified phrases, adjectives like “tall” or “green”; moreover, moderate contextualism is doomed to collapse into radical contextualism. 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 contextsensitive 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 context 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 constituent, 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 scholars: 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];

57

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 propositions 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 logical form of (2) a covert indexical element for the relevant place. Similarly, the quantificational 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 (“unarticulated constituents”): “No proposition could be expressed without some unarticulated 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

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 impossible 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 �travelling in interstellar space, the russet leaves of a Japanese maple painted green, Bill �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.

59

60

Claudia Bianchi

show that every sentence has a literal meaning only against a background of contextual 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 sentence, if taken independently of any context whatsoever, underdetermines its truthconditions. 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. According to Travis, in C1 and C2 (17) has the same conventional meaning, but a crucially different interpretation:



Contextualism

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 �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 different 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 contextualists claim that the pragmatic enrichments10 are necessary in order to account for the intuitions speakers have about the truth-conditions of their utterances. The pragmatic processes of completion and enrichment are pervasive, and generally unconscious; 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, according 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. Departure 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 correspond 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.

61

62

Claudia Bianchi

intuitions of the participants in the conversational exchange. Understanding an utterance 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 reasonable 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, completion, quantifier domain restriction, free enrichment. To avoid this kind of objection, contextualists propose a range of tests (Binding, Optionality, Availability, Embedding 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 proposition”; 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 conversional 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

in determining what is said by an utterance, instead of, for example, what is communicated by a speaker. Moreover, intuitions on the propositional content of a sentence tend to be sensitive to extralinguistic information and are likely to reflect interpretations 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 immediate 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, radical 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)) minimalists 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 propositionality 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.

63

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 indexicalists 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 explanatory and predictive power – and must account for the stronger role of the contextual 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 context 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 pragmatics, 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 contextualism 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 systematically 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

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 different things (express different propositions) in different contexts; in a minimalist perspective, on the contrary, there is an invariant semantic content: every utterance 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 proposition 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 proposition 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 proposition 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 evaluation. 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 different speech acts having different contents. 22.â•… MacFarlane’s example is “Chiara is tall”.

65

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 discussing 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 traditional view, indicating a form of contextuality intervening “at the post-compositional 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 parameter. 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 MacFarlane’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 parameter (“counts-as”) is relativistic in the moderate sense of Kölbel and Récanati, but not in MacFarlane’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 initialized by the context of use, relativism lets it be initialized by the context of assessment”.



Contextualism

6.â•… Conclusion Several issues are still open. For one thing, some clarification on the notion of semantic intuitions would surely be welcome, along with more extensive reflection on the role that intuitions play – or must play – in semantic theories. Moreover, contextualism (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 considered 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 syntactic 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 minimalist perspective, it is possible to assign truth-conditions to sentences of natural Â�languages, relying on their conventional meaning and contextually fixing the reference of Â�indexicals and demonstratives. At the other end lies radical contextualism, which claims that in order to obtain a complete proposition (i.e. complete truth-conditions), it is always necessary to resort to pragmatic processes. Indexicalism 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 indexical Â�contextualist solution succeeds in explaining away certain apparent contradictions in the linguistic practices relative to epistemic modals, knowledge ascriptions,

67

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 perspective 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 meaning 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.) ContextSensitivity 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

——— (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: 565–614. Oxford University Press. 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: 240–250. Oxford University Press. ——— (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.

69

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, contradictions, 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 writing 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 phenomenological concept of intentionality; rather, for Saussure, meaning is produced within a system of signifiers, more specifically in terms of the difference between signifiers. 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 structuralist 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 relationship 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 culturally 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 technical 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 scholarship) 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 twentieth 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 structuralism, and extends them to produce what have been called ‘poststructuralist’ theories 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 constantly deferred. The reasoning behind this is straightforward enough: if, as we suggested earlier, the signifier woman can only be understood in terms of its ‘other’ (man or



Deconstruction

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 meaning, 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 textuality occupy a particularly central, yet ambiguous, place in deconstruction: deconstruction 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 difficulty 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 deconstruction) 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

73

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 produced 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 ‘dissemination’. Dissemination is used by Derrida and deconstructionists to suggest an antiteleological 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 ‘portable’; 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 deconstruction demonstrates that such a controlling centre, what has been termed a ‘transcendental signifier’, is always somehow implicated in the deferment of meaning. A



Deconstruction

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 structural 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, temporality 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 potential, 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 supplementarity. 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 Marxism; 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 accessible (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

75

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, narratives, ideologies and philosophy (democracy, globalization, the idea of Europe, liberalism, 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 influence 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 movement, the demonstration and the practical deconstruction of the representation of what was done with literature” (Derrida 1982: 69). Derrida’s exploration and (partial) privileging of ‘literary’ texts is an example of the influence exerted over his work by the writings 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), marxism (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 deconstructive approaches are employed by some discourse analysts, Hallidayians and



Deconstruction

Â� stylisticians such as Norman Fairclough, Bob Hodge, Gunther Kress and Terry Threadgold. Although most prominent discourse analysts are more likely to be influenced by, and to work with, Foucaldian notions of power, knowledge and discourse, deconstruction 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 meaningless, 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.

77

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 associated 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 inaugurated 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. Naturalized 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 epistemological 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 theories 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 verification 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 investigations, 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

theory of meaning is not shared conventions or a communal set of rules one is supposed 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.

81

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 content 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

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 recognise 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 something 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 conventional 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. Testimony 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 knowing that p that explains an audience’s knowing that p on the basis of the speaker’s testimony 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

83

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 speaker’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 Â�University Press. Matilal, B.K., and A. Chakrabarti, eds. (1994). Knowing From Words- Western and Indian Philosophical 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

1.â•… Introduction 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 concepts we consider to have shaped a new understanding of discourse, in a process 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 consequences 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 associated 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 controversies, 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 perspective 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 discourse 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 presented, appeared simultaneously. This school looked for an ‘epistemological breakdown’, 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 classical 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 objective 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

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 viewing 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 objective 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

87

88

Luisa Martín Rojo & Angel Gabilondo Pujol

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 dramatic 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 contemplation 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.

.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

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 signification. 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 enunciative 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 formation’ 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 various 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).

89

90

Luisa Martín Rojo & Angel Gabilondo Pujol

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 nondeterministic 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

.╅ Discursive relations are not, as we can see, internal to discourse: they do not connect �concepts or words with one another; they do not establish a deductive or rhetorical structure between �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 �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

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, concepts, 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 extrinsic 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 characterized 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 embracing 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 ‘regularities’; 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

91

92

Luisa Martín Rojo & Angel Gabilondo Pujol

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 embodied 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, reinforces it, but also exposes it, makes it fragile and permits its localization and detention (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

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 conceptualization 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 eighteenth century. These transformations had different consequences – among them, the higher juridical and moral value placed on property relations, and a correlative extension 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

93

94

Luisa Martín Rojo & Angel Gabilondo Pujol

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 question 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 discursive 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 conditions 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 ‘normalization’, 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

.╅ Power is exercised by exercising destructive forces (executions, occupations) or by uninterrupted constraints imposed in practices of discipline and training. The exercise of power on �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

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 contradictory 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 perspective 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 distributed, within which context, which features of the discourse can authorize them,

95

96

Luisa Martín Rojo & Angel Gabilondo Pujol

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 controls 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 relationships.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 discursive 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 representations 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 reproduction of discourses: while dominant/legitimated/authorized discourses are constantly evoked, the discourses of minorities are ignored. Discourses which are

.╅ 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

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 representations, values, and ideologies are constantly reproduced, while others are silenced. C. Control of the conditions of production of discourses, by establishing restrictions 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 production, circulation and reception of discourses are not random but ordered. The influence 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 interdiscursive 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 application 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.

97

98

Luisa Martín Rojo & Angel Gabilondo Pujol

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 Foucault’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 domination) has been evaluated in different ways. However, Foucault’s approach to discourse 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 consecrations. 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 sociolinguistics, pragmatics, and discourse analysis. They appear in regulations of linguistic practices, 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 inextricably 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

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 introduce 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 production. 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 theoretical 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 positioning, 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 incorporation 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

99

100 Luisa Martín Rojo & Angel Gabilondo Pujol

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 speakers’ spatial configurations are limited by power relationships; but also at a macrolevel, 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, Foucault’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 individual 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, teachers, 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 organization of conversation, of the cooperative building up of meaning (conversational inferences) 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 influence 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 problematization or experimentation. His understandings of and approaches to discourse



Michel Foucault

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 thinking 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.].

101

102 Luisa Martín Rojo & Angel Gabilondo Pujol ——— (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

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.

103

H.P. Grice Frank Brisard University of Antwerp

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 unfamiliarity, 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) edition 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

.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

inferential (argumentative) work on the part of the hearer, as guided by the conversational hypothesis. As such, Grice is one of the prime instigators of a ‘Â�radically pragmatic’ take on meaning that suggests the viability of maintaining a rigorous Â�theoretical distinction between semantics and pragmatics, while ensuring a paradigmatic continuity 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 sufficient) 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 properties 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 follows, 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 necessary 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 development 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

105

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 leadership 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 sympathetic 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 overall picture left room for much diversity. The exact relationship between linguistic Â�phenomena and specific philosophical theses would remain a heated topic of debate, .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

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 trying 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 ‘utterance-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 fortune 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 philosophical 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 minimal 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

107

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 pragmatics (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 instrument 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) publications 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

.â•… In a truly pragmatic spirit, Grice distinguishes between two types of use of the English expression 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’ meaning 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 communicative 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

.â•… For a critique of the nesting or iteration of (representational within) self-referential communicative 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 reflexivity, 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. .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 interpretation 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

.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). .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 propositional 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 propositional 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

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 necessarily coincide with, a fourth conception of meaning, that of the ‘utterer’s occasionmeaning’, 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 project 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 utterancetype 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. .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.

111

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 particular 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 meanings. 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 conditional sentences, and in particular of the relation between the material conditional

.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

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 (informativeness), 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 utterance, on the assumption that the speaker is in fact being cooperative. Under this condition, 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 observation 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).

.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).

113

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 assumption 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 implicature 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 truthconditional). Conventional implicatures may look like GCIs, because they, too, are supposed 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 language” (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

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; emphases 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, expectations 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 conceived 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 conceived 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

.â•… Here, Grice offers a variation on Moore’s defense of ‘common sense’.

115

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 interests 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 person 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 truthconditional 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 linguistic 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 .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

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. survive, 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 achieving 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. .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.

117

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 constructivist 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 developing 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 interaction 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 redesign 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

.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). .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

one. By this transformation (a process which Grice called ‘Metaphysical Transubstantiation’; 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 interaction 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 evaluative 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 accidental. 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 critical in the projection involved in creature construction, because absolute value, and

119

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, grounding 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 scientific. 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 intellectual 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

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 tempting 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 century. Yet we have noticed that quite a few premises generally adopted in the cognitive 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 distinction 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 ‘information’ 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

121

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 systematic theory of communication within the broader framework of social interaction, adapting Grice’s notion of (reflexive) M-intentions and thereby challenging conceptions 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 linguistically 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 ‘experimental 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 pragmatic 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 reissue 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

MacKay (1972). The volume edited by Bar-Hillel (1971) proposes an early appreciation of general philosophical problems conjured up by Grice’s (and others’) logical 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.

123

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

1.â•… Introduction 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 convey 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 interpretation 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 hermeneutic thinking: theology, philology, jurisprudence and philosophy. In the 16th century a number of protestant theologians, such as Flavius, Luther and Melanchton, considered the Scriptures as insufficiently understood. A linguistic and hermeneutic 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 Justinian, ultimately lies at the basis of legal hermeneutics, which developed together with philology. Finally, following Aristotle’s analysis of logic in his essay On Interpretation, the Enlightenment philosophers of the 18th century tried to develop hermeneutics as a general method for all fields of knowledge based on interpretation. 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. MuellerVollmer 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 considerable influence on philology, to such an extent that those systems of interpretation 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 context 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 hermeneutics’ was emancipation. This means, first, to make transparent the underlying individual and social processes of social actions and, second, to develop one’s consciousness accordingly (cf. Apel 1976). Habermas’s interpretation of hermeneutics 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

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 dialectical 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, objective 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 hermeneutical 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 methodological 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 hermeneutics. However, though the link is rarely made explicitly, traces of Â�hermeneutic

127

128 Piet Van de Craen

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 manage 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 communication. Language, Sapir says “can discover meanings for its speakers which are not simply 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 cognitive 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

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.

129

130 Piet Van de Craen ——— (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

1.â•… Introduction 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 something because we are in a given context. We are, we could say, Â�context-bound creatures. 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 ‘yesterday’” (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 indexicals 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 accompanying 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



Indexicals and demonstratives

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 referent/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 InterContextual 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 utterance 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)

133

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 foreigner, 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 accompanying 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.



Indexicals and demonstratives

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 demonstratives (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 bothering 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 perception 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).

135

136 Eros Corazza

into the following: essential indexicals have a cognitive impact, for they trigger selfcentered 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), utterances containing indexicals express singular propositions, i.e. propositions whose constituents 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 indexicals 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 context-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’,



Indexicals and demonstratives

‘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 presence 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 previously fixed reference. ‘That boy’ in “That boy we encountered yesterday was in trouble 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.

137

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 �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).



Indexicals and demonstratives

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 Kaplanian model. Yet they wouldn’t refer as pure indexicals do. They would rather be considered as demonstrative. Hence, this analysis jeopardizes the traditional distinction between demonstratives and pure indexicals. According to this interpretation, ‘here’ and ‘now’ are ambiguous between being used as essential indexicals or demonstratives.11 Furthermore, one could claim, pace Quentin Smith, that the relevant contextual parameter to which the character of ‘now’ and ‘here’ is sensitive is not the location 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 parameters 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/demonstratives 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.

139

140 Eros Corazza

assume that in examples like these, ‘now’ and ‘here’ can be interpreted as anaphoric 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 antecedent, ‘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 dependent 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-�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 quasiindexical ‘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)



Indexicals and demonstratives

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, anaphoricity is not represented by indices (subscripts), but by dependence and independence markers (superscripts). In other words, the superscripts, not the subscripts, represent 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 occurrences 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 proposed (AI), function as an anaphoric term; but then, where is the antecedent? According 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)

141

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 difficult 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 propositions 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 proposition 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 proposition 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).



Indexicals and demonstratives

(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 indexicals 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. d.

∀x (x won the lottery) & y is now rich ∀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.

143

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 anaphoric 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 referentially 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.

(ii) a. b.

Mozart1i adored his1/2d music b. He1i adored Mozart’s2/*1d music The bus1i came trundling round the bend. The vehicle1d almost flattened a pedestrian The vehicle1i came trundling round the bend. The bus2/*1d almost flattened a pedestrian



Indexicals and demonstratives

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 coreferential with another NP not in virtue of inheriting its reference but simply because the latter 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 interpretation: (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.

145

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 economical 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,



Indexicals and demonstratives

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 interchangeable without any loss or gain in a communication. My New Shorter Oxford Dictionary 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. (21) b.

We are at this point of the argument/proof/…0i. Now/here0d I am going to prove lemma S 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) Bound

Unbound

Deictic (independent) Pure Indexicals Essential

non-Essential

Demonstratives Simple

Complex

24.╅ 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 spatiotemporal 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 �reasons. These phenomena are linked to our cognitive architecture; they are not �intrinsic to the semantics of our language. Thanks to Predelli for this suggestion.

147

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 content 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 impossible, 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.

Indexicals and demonstratives 149



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 standing 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 different 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 utterance’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. b. c.

An utterance u of “It’s raining” is true iff it’s raining An utterance u of “It’s raining” is true iff it’s raining in l 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 unarticulated constituent of the proposition expressed, viz. the location enters the proposition 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 entering 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, insofar 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 intrusion 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 7.â•… Conclusion In dealing with the problem of context sensitivity, in general, and indexicality, in particular, we saw that a few distinctions need to be made. We first saw how the traditional 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 distinction 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 �University Press. Carston, R. (2002). Thoughts and utterances. Blackwell.

28.â•… For a defense of this position see Corazza (2007) and Corazza & Dokic (2007). A similar position 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).



Indexicals and demonstratives

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. Oxford University Press. Kaplan, D. (1989). Afterthoughts. In J. Almog et  al. (eds.) (1989). Themes from Kaplan, 565–614. Oxford University Press. 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, 233–249. Oxford University Press. 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, 240–250. Oxford University Press. 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. Oxford University Press. 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.

151

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, 41–62. Oxford University Press. 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, 47–62. Oxford University Press. 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 tradition. 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 abovementioned 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 possibility 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 (Russell 1940: 168) substituted for ϕ.

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 functions); introduction of truth-functional connectives such as ‘not’, ‘and’, ‘or’, ‘if … then …’; and quantification. It fails for constructions involving modalities such as ‘It is necessary 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 constructions such as indirect discourse. A verb of propositional attitude (‘A believes that …’)



Intensional logic

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 someone 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 conjectured 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 semantic 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-equivalence (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 designators (singular terms, general terms, or sentences, respectively) are L-equivalents, their intensions (individual concepts, properties or relations-in-intension, propositions, respectively) are identical. If two designators are F-equivalents, their extensions (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

155

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 creates 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 occurrences 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 opposition 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 proposition 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 differences 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



Intensional logic

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 semantics 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 giving 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 ordinary 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 languages. 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 meaning of a sentence before assessing its truth-value. Montague’s semantics formally captures 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 characteristic 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 arguments, 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

157

158 Paul Gochet

the division of linguistic labor (Putnam 1975: 246). I shall however refrain for belaboring this point here. Montague’s account of intensions as functions from possible worlds to extensions 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 extensionality, stipulates that two sets are identical if and only if they share the same members. 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 calculus 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.



Intensional logic

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, actually 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 constrained 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 finding 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 undecidable and the unification algorithm wich is the core of logic programming has been proved to be undecidable beyond first-order. This objection against Muskens’ enterprise 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

159

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 mainstream 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 propositional 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 contribute 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 occurring 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 complicated 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 wellknown example: from ‘Philip believes that Cicero denounced Catilina’ and ‘Cicero = Tully’, ‘Philip believes that Tully denounced Catilina’ does not follow. Quine initially suggested 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 attitude 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



Intensional logic

‘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 proposition without having its sense explained to me” (Wittgenstein 1960 [1921]: 41). Language 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 simply 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.

161

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 philosophical 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. Montague, R. (1974). Formal philosophy. Yale University Press. 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, Philosophical 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 University of Liège

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 statement 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 belonging, necessarily belonging and being able to belong. Modalities qualify the onto-logical relation of belonging that obtains between attribute and substance. Aristotle’s modalities 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 necessarily 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 modalities 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 structure contains both modal terms and quantifiers. Can we from ‘It is possible that everybody 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 interdiction. 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 mathematica (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 connective ∨ (‘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 understood ‘if…then…’ in the modal sense as ‘It is impossible that… and not…’.



Modal logic



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 distinguished 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 calculus 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 meanings 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 possible 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

165

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 relational 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 formalisms. Usually modal logic has good computational properties (decidability, expressivity) 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: ‘Functional 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 foreshadowed 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

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

167

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) conception 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 University of Liège

1.â•… The meeting of two different approaches to semantics Quine once argued that semantics had split into two provinces “so fundamentally distinct 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 analyticity, 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 interpret 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 proposition 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 considerably 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 symbols (a countable set of individual variables x, x’, x’, …, the connectives ¬, ∨, →, ↔⇔ and &, the quantifiers ∃ and ∀, the identity symbol = and parentheses) and (2) nonlogical symbols (a finite set of n-place predicates), and with a syntax allowing three



Model-theoretic semantics

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 connectives. 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 nonempty 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 twoplace 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 interpretation I. There is a clause for each atomic predicate constant, a clause for each connective 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.

173

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 interpretation 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 standard 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 deductive 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 satisfied 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



Model-theoretic semantics

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 intension 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 overcome 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 connectives 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 = ∅ }

175

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).



Model-theoretic semantics

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’; according 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 bakery. 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 infelicitous, to use a word borrowed from Speech Act Theory. We have described the interplay of various pragmatic constraints which determine 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 processing architecture which accounts for the incremental disambiguation of underspecified 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 entities and propositions (ordered by salience); L represents linguistic knowledge; and K represents knowledge about the world.

177

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 knowledge 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 Conference 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 properties. 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. Pin´on & H. de Swart (eds.) Quantifiers, Deduction and Context: 111–138. CSLI. Kanazawa, M. (1996). Indefeasible Semantics and Defeasible Pragmatics. In M. Kanazawa, Ch. Pin´on & 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.



Model-theoretic semantics

Montague, R. (1974). Formal philosophy. Yale University Press. 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.

179

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 professional 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 Reality, of 1925 (but only published in 1993 preceded by a German translation in 1981), includes a chapter entitled ‘Some Psychological and Biological Considerations’. THerefore 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

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 Morris’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 contextual 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., contextualized 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 behaviorism, 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 characteristic of the physicalist project. His biological perspective was also important in

181

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 construct 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 relationship 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 ‘behavioristics’, 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 Behavior, 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 Theory 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, Language 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

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 different 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 consolidated the connection between semiotics and axiology. The English term ‘meaning’ has a dual accceptation referring not only to the semantic dimension of signs (signification) 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 confirms 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 pragmatical dimension of semiosis from the syntactical and semantical dimensions, as early as 1938. That meaning is generally associated with the semantical dimension of semiosis 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

183

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.

DESIGNATION DENOTATUM

SIGN VEHICLE

Pragmatical Dimension of semiosis

Syntactical Dimension of semiosis

OTHER SIGN VEHICLES

Semantical Dimension of semiosis

SEMIOSIS

INTERPRETANT INTERPRETER

SEMIOTIC

SYNTACTICS

SEMANTICS

PRAGMATICS

The top part of the diagram represents semiosis, the process by which something functions 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 pragmatics. 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 semantical 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 dimension 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 distinguish between the relation of the sign-vehicle to the designatum (and eventually the



Charles Morris

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 throughout 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 dimension 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 pragmatist, 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.

185

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 natural 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 proposal 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 relations as in classical empiricism, but a function or activity exercised by verbal and nonverbal 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 assertion, arising from our conception of the sensible effects of things. For a clear understanding 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

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 pragmatism, which he denominated ‘experimentalism’ or ‘instrumentalism’. In Italy, pragmatism 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 distanced 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 pragmatism, 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 relation 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 mediation 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

187

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 intervenes, 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 without 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 studies 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 interpreter. Indeed, given that the interpreter does not subsist as such if not as a modification 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 interpretant may be understood, despite Morris’s attempt to reduce ambiguity by introducing the term significatum (designatum in Foundations) alongside it. Dewey employed the term ‘interpretant’ with the same meaning Morris attributed to significatum, without realizing that for the latter ‘interpretant’ indicated the effect of a sign on an interpreter. 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 certain 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 goalseeking 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 precedes 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, therefore, 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

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 behavior 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 responding 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 ‘dynamical 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;

191

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 ‘biological 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: adenosinemonophosphate, 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 interesting 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, focusing 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 interpretive 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 problems 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

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 diversity 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 differences in personality, in characteristic problems treated, and even in proposed solutions, 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; semiotically 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 cognition’, 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 practical, 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 concerned then with the relation between “deliberate conduct” and “the intellectual purport of symbols” (cf. CP 8.322, 5.442). Therefore, intelligent, purposive behavior, 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

193

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, comprising 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 precise 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 semiotics, 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

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, pathological 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 nonlinguistic signs much later (1955), Morris’s conception of pragmatics concerns both verbal 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 mentioned 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 meaning commonly used to cover (at least) intention, signification, and value, and maintained 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 distinguish 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 theory. The connection between linguistics and economics, or semiotics and economics, promised a fuller understanding of the disciplines in question as well as of their interrelations (cf. Rossi-Landi 1975b). Saussure referred specifically to economic value as

195

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 materialism 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, RossiLandi stated that after Foundations, Morris’s research developed in two different directions. 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 Signification 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 problem 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 axiological issues. In truth, Peirce’s writings (cf. 1923) did not fail to deal with the problem 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 problem of knowledge necessarily involved orientations and issues of an evaluative order,



Charles Morris

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 ‘teleosemiotics’, 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 contributes 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 theory 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 addition 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 semioticians 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 stimulus. 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 functions as the stimulus or sign vehicle. Instead, the part not fully perceived functions

197

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 perceived 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 fundamental 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 conditions 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 knowability, 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 classifies 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’ signifies 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 necessarily 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 perceptual, 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 identify 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

——— (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.

201

Notation in formal semantics Walter De Mulder University of Antwerp

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.

Notation in formal semantics 203



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 �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 �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 �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. �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 analysis 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 �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 assignment 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.



Notation in formal semantics 205

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 {} – where {a,b} symbolises a set composed of the elements a and b and 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 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 correspondence 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 ; 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>.

206 Walter De Mulder

It remains to convert the syntactic categories of the grammar into the �appropriate logical types (an operation that is much simplified if one already uses a categorial �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 �constructed in accordance with the rule of functional application: (8)

Rule of functional application: If f is an expression of type , and a is an expression of type a, then f(a) is an expression of type b, or, slightly more informally, an combines with an a to form a b.

This rule allows us to find the type corresponding to an intransitive verb, i.e. an �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 language 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 structure 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 �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 �theory. We already said that a transitive verb denotes a set of ordered pairs. An intransitive 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

Notation in formal semantics 207



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 �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 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 entities 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 semantic 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 (Cann 1993: 116) e, then λ x[Φ] is a well-formed expression of type <e,t>.

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’)

208 Walter De Mulder

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



Notation in formal semantics 209

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 quantified 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 possibility 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 wellknown 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.

210 Walter De Mulder

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 distinguished 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 expressions, 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 circumstances 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 formula 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 ◊.



Notation in formal semantics

(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 suggests 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) (20bis)

If φ is a formula, then [☐φ]M,g,wn = 1 iff [φ]M,g,wm is 1 for all wm ∈ W where wnRwm. 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.

211

212

Walter De Mulder

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 operator 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 expression. 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 notational 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 natural 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



Notation in formal semantics

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 referent 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: Context Circumstances

Character Intension

Proposition (intension) 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 linguistic 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.

213

214 Walter De Mulder

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



Notation in formal semantics

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) 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 expressions 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 assignment 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, 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)

215

216 Walter De Mulder

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: 481–563. Oxford University Press. 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. Montague, R. (1974). Formal philosophy. Yale University 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 University of Antwerp

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 Philosophy 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 phenomenology 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 separate 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 prepared 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 consciousness and their objects. It studies the way in which specific kinds of objects correlate 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 prescribes 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 different 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 mathematical 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. Phenomenology 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

can only be studied on the basis of a second important methodological constraint, which Husserl calls eidetic variation. Starting from the particular, individual and subjective experiences, one has to modify them in order to identify their inalterable essential 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 temporal characteristics. In that sense, phenomenological method led Husserl to a fullfledged 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 consciousness and returned to basic ontological questions. Jean-Paul Sartre’s existentialist philosophy equally has its roots in the phenomenological method. Maurice MerleauPonty’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 distance from his idealist philosophy. Being a real philosophical movement, where participants 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 products 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

219

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 consciousness, which came close to the Fregean approach of sense and reference. Subjective meaning-intentions (Bedeutungsintentionen) animate language and found its meaningfulness. 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 influenced the functionalism of the Prague school (see, e.g. Mathesius 1964). Moreover, a ‘phenomenology of language’ was programmatically anticipated by most of the philosophers 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 phenomenological 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

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 pragmatics, there is also a direct line from phenomenology to pragmatics via the ethnomethodological 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 phenomena 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 organization 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 University 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.

221

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 philosophical 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 century 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 relation between one’s beliefs and desires and the action is a justificatory one: we understand 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



Philosophy of action

(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 deliberation leading to the conclusion that a particular type of action is desirable is, furthermore, 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 antecedent 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 connection 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 conceptual relation between a person’s wanting (willing, intending) to A and her A-ing, the latter (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 contentful 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

223

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 physiological 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 pressure 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, intentional 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



Philosophy of action

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 understanding, 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. explanations in terms of the physical properties of those causes and effects? After all, when we

225

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 statement 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 explaining 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 intentional are not equivalent to physical descriptions. Application of intentional descriptions 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 constitutive 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 responsibility 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 description 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.



Philosophy of action

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, illocutionary 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 semiautonomous 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. Searle, J.R. (1969). Speech acts. Cambridge University Press. Wittgenstein, L. (1953). Philosophical investigations. Blackwell.

227

Philosophy of language Asa Kasher Tel Aviv University

1.â•… Introduction 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 alternatives, 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 understanding of a pragmatic facet of science, namely the activity of explaining, rests on



Philosophy of language 229

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 ingredient 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 various 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 essential role. On this view, since it is impossible to understand belief and desire, knowledge 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 philosophy 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, logical 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 analysis of language” (Carnap 1932). Analytical philosophy of the second half of the twentieth 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 sometimes 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 philosophical 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 consideration? Discussions of this problem have had two foci. One problem is that of the identification 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 similar 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



Philosophy of language

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 produce 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 communication, 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 similar 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 consideration. 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 philosophical interest, and secondly, developments within mathematical logic have provided philosophers with powerful tools for presenting, discussing, and often solving philosophical problems of certain types, first and foremost, problems related to meaning 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 family of facts is under consideration on both occasions. The distinction between those

231

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 philosophical 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 conditions (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 obviates 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 without 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 frameworks 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 pragmatics (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 philosophical work has also been done on particular speech acts, such as asserting, promising, 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).



Philosophy of language

References Almog, J., J. Perry & H. Wettstein (eds.) (1989). Themes from Kaplan. Oxford University Press. 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. Wettstein (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.

233

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. Searle, J.R. (1969). Speech acts. Cambridge University Press. 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. Wittgenstein, L. (1953). Philosophical investigations. Blackwell.

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 cognitive sciences. Adopting a general scientistic outlook and, in particular, the naturalistic 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 ontological 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 nothing 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 philosophical 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 physical dispositions such as the brittleness of glas and are analysed in terms of hypotheticals 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 identical to inner brainstates, mental causality is unproblematically a form of physical causality. 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 inadequately 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 macroscopic 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 implemented 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



Philosophy of mind

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, functionalism 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 eliminativism, 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

237

238 Stefaan E. Cuypers

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 memory 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 representational theory of the mind (RTM) (Fodor 1981; compare also Field 1978). This view involves a basically positive attitude towards folk psychology; this intentional psychology 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 propositional 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 computationally 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 intentional 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 psychology 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 speaking, 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.



Philosophy of mind 239

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 instrumentalistic fashion the (neo-)behavioristic theory of the mind must, according to Dennett, be enlarged with a theory of interpretation. Belief-desire profiles are attributed to an intentional system from the intentional stance in order to rationally interpret its behaviors. Folk psychology is a ‘dramatic’ interpretation game (Quine 1960) played according to rational and normative principles by players who adopt the intentional 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 intentionality or semanticity. The psychological theory of the intentional content of propositional attitudes runs parallel to the linguistic theory of the semantical content of languageexpressions, 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 conflicting 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

240 Stefaan E. Cuypers

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 meaning primarily depends upon the internal relation between (mental) symbols. The doctrine 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 difficulties, 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 isolation, even if cut off from everything else. As a consequence, the constitution of meaning 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 difficulties, 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 contributions 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 considered as secondary to and derived from the psychological theory of intentionality



Philosophy of mind

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 psychosemantics 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 principle 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). Conversational 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 maximization 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 interpretation, 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 semantics/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 truthconditional content of utterances, there is also only a single inferential Â�mechanism of mutual adjustment that processes (explicit) linguistically encoded meanings and

241

242 Stefaan E. Cuypers

(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.



Philosophy of mind

———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 identity theory: 52–66. Macmillan. Sperber, D. & D. Wilson (1995). Relevance: communication and cognition, 2nd ed. Blackwell. Taylor, K. (1998). Truth and Meaning. Blackwell. Wittgenstein, L. (1953). Philosophical investigations. Blackwell.

243

Possible worlds semantics Paul Gochet University of Liège

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 relation 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 necessary 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 conditional 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 systems can be designed to capture the strict conditional, it is no longer true that the latter 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 



Possible worlds semantics

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 substitute 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 interpretation 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 propositions 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 necessary 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).

245

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 necessary 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 semanticists: “[…] 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 ‘absolute’ 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 quantifier 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



Possible worlds semantics



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 statement, it is necessary and sufficient to require that R be transitive. Thanks to possible worlds semantics, different interpretations have been associated with the box operator 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 conditionals ( ) 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 fundamental 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)

247

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 expressions 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 (Cresswell 1978: 12) omniscient, we will not know whether that world is actual.

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 formalization of that kind fails, however, to cater for the possibility of appending new items to the sequence (1), such as

Possible worlds semantics 249





(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 whistling. 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 Predicate 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 reconsider 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 truthconditional 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 information state has been updated with the information that it is raining, it is no longer consistent 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 logicians 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, however 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). (Intermediate steps are omitted here.) Sentence (iii), however, is counter-Â�intuitive. Ralph is illinformed 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 selection 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.



Possible worlds semantics

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, knowing-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.

251

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 perspectives 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 Supplementary 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 context, 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

.â•… In this article, I shall focus on singular definite descriptions and say nothing about plural definite 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 descriptions bear to their denotatum, without taking any stand on what other kinds of descriptions do. .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 “reference” 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 truthvalues. “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 interesting 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 mention 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, .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).



Reference and descriptions

“[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 linguistic 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 syntactic 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 speakers). 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

255

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 intension 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 something 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. unstructured) 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 cannot 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

.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).



Reference and descriptions

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 supposition 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 Principle 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”

.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. .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 supplement 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. .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).

257

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 logical 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 concerning 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 describing, 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 having 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, .â•… See Kaplan 2005: 973–1000 for a careful discussion of the entanglement of semantic and epistemological issues in Russell’s philosophy, and of his “allowing the demands of his epistemology to override his semantic convictions” (990).



Reference and descriptions 259

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

.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 properties 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 (descriptive) 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 description. 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, concerning (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 command consists in knowing a condition that is uniquely satisfied by the entity referred to, Russell that it consists in a direct awareness of it.

.╅ 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.



Reference and descriptions

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,

261

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 criticisms 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 description 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 reference 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

.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.



Reference and descriptions 263

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 designator, 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). Actually, 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

.╅ 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 troubling 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.

Reference and descriptions 265



to develop, reference, at least when proper names are concerned, is basically a historical 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 concept, 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

.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). .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 constituted 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 condition 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.

.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).

Reference and descriptions 267



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 existential 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 constituted by an open formula, which individuates a (quite complex) condition or propositional 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 uniqueness 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

.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 (488) denotation of “the author of Waverley”

Note, however, that the relation of denotation is of no semantic import. In fact, according 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 individual (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 philosophy. 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 quantified) rather than about any particular individual that may happen to be denoted by the descriptions in them.

.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. .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. “Babylonians believed that Hesperus was not Phosphorous”). Unfortunately, as Kripke has shown (see Section 1.5), proper names are not definite descriptions in disguise.

Reference and descriptions 269



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 philosophers 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 particular, “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

.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 different 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 (16) of utterance, shows what unique reference is being made

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



Reference and descriptions

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. According 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).

.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.

271

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 something? 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



Reference and descriptions

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 suppose 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 something 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 person who murdered Smith (in fact, Adams) would not be insane (situation a). According 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 individuated 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.

273

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 ascribing 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 straining 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 relation, 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 reference 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 fundamental 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

.╅ Donnellan (1968) tries to defend himself from the accusation, which was actually brought against him by Alfred MacKay (1968).



Reference and descriptions

Â� distinction is not a semantic one. It is, rather, a pragmatic one, and as such is fully compatible with Russell’s account of the semantics of definite descriptions.22 To show this, Kripke distinguishes, “following Grice” (Kripke 1977: 262), two different 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)

.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 description in referential use to refer to something which does not fit it. Indeed, they tend to assimilate 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). .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).

275

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 worried, 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 .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.



Reference and descriptions

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. Almog, J., J. Perry & H. Wettstein (eds.) (1989). Themes from Kaplan. Oxford University Press. 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.

.╅ 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.

277

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 epistemology 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 Â�University Press. 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.



Reference and descriptions 279

Soames, S. (2005). Why incomplete definite descriptions do not defeat Russell’s theory of descriptions. 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 misunderstanding 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 uninterpreted 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



Truth-conditional semantics

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. b. c.

Reference axioms for names: e.g. ‘Sigmund Freud’ refers to Sigmund Freud. Satisfaction axioms for predicates: e.g. ‘Intelligent’ is satisfied by intelligent things. 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 Zegarese 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.

281

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 truthconditional 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



Truth-conditional semantics 283

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 propositional 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]; �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 intentions 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 meaning 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 linguistic competence: phonological, syntactic, semantic. This knowledge is characterised as

284 Robyn Carston

systems of mental representations and rules for transforming one sort of representation 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 semantic 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) anomaly (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 translational 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 interpreted, 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 semantics 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



Truth-conditional semantics 285

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 truthconditional 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, indicated 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 ‘Moonlight 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 pragmatics; 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 version 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 truthcondition. McGinn (1977), for instance, treats imperatives as having fulfilment conditions 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.

Truth-conditional semantics 287



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 utterance. Expressions like ‘but’, ‘therefore’ and ‘moreover’, that Grice characterized as communicating 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. b. c. d.

Frankly, Ian is a fool. Fortunately, it didn’t rain. You’re going, eh? I have to go, alas.

In each of these, the split between the truth-conditional content and the non-truthconditional 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 truthconditional 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 vagueness, 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 linguistically 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 truthconditional 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. Montague, R. (1974). Formal philosophy. Yale University 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 representations: 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 University of Antwerp

1.â•… Origins The work of Jürgen Habermas is normally regarded as a continuation of the movement 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 sciences (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 discussions 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 ‘hermeneutical 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 SapirWhorf 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 kinship, 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 illocutionary 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 sincerity. 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 understandable, 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 legitimacy 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



Universal and transcendental pragmatics

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 understood 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 presuppositions 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 representative function of speech and in Habermas to the validity claim of truth. Bühler’s appelative 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 communication 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 coordination 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.

291

292 Joachim Leilich

From the four validity claims only two – truth and correctness – can be thematized 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 question 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 universalibility 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. Morris. 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 philosophy. They can be characterized through the question of which of the three dimensions 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 allowing the possibility for the knowledge of objects to be mediated by signs or the knowing 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 reflection. This first paradigm abstracts from two of the three dimensions of the semiotic sign-function. With the second paradigm – classical philosophy of consciousness – which dominated philosophy from Descartes to Husserl, the mediation of objective knowledge through consciousness comes into play. But for this paradigm the signs are not relevant, 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 consciousness but not signs are a theme of philosophical reflection.



Universal and transcendental pragmatics 293

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 Transformation 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 empiricalbehavioristic (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 knowledge. 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 transcendental pragmatics within the frame of a transcendental semiotics. Insofar as semantical 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 possibility 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 justification 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 fallibilism. But on the level of a reflection on the ultimate presuppositions of intersubjective 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 presuppositions 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 foundations. 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 differences 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.



Universal and transcendental pragmatics 295

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 argumentation we can discover ethical presuppositions. One of the results of his transcendentalpragmatic reflections on the conditions of the possibility of communication was the insight that in all rational argumentation we are obliged to presuppose the accountability 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 wellknown 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 convinced 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 followed 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 Maynard 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 philosophy 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, prepared 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 philosophy. 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



Ludwig Wittgenstein 299

is not something like a philosophical doctrine articulated in propositions, but philosophy 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 philosophers 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 linguists 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 misinterpreted, 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 runners 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 representation is only possible because the representation and what is represented have something 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 common: 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 combinatory 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



Ludwig Wittgenstein

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 different 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 sophisticated 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 Tractatus 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 sentences 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,

301

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 philosophy. 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 something 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



Ludwig Wittgenstein 303

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 paragraphs. They start with a quote from Saint Augustine in which naming is the central 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 someone 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 similarities. 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. Sometimes 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. Wittgenstein 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 contained 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



Ludwig Wittgenstein 305

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 community 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 Wittgenstein the result of a misunderstanding of the way in which sensation words function: 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 private 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 contribution in Wittgenstein on Rules and Private Language (1982) stimulated an unprecedented 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 conducted 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 scepticism 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 similarity 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 antimetaphysical 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 Syntax 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 understanding 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 philosophy 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



Ludwig Wittgenstein 307

pupil N. Malcolm (1993). Other authors, for example H. Pitkin (1972), explored the significance 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 cannot 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 difficult 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 membership, 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 linguistic 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 A Behaviorst’s Account of the Significant Symbol (Journal of Philosophy),╇ 180 A Derrida Dictionary (2004),╇ 75 A Study of the Relations of Signs and Values,╇ 183 absolute value (Kantotle),╇ 119–20 abstract system,╇ 46 accessibility relations R,╇ 211 accomplishment verbs,╇ 161 acquaintance,╇ 257 action,╇ 34,╇ 222 action and responsibility,╇ 226 adjectives,╇ 132 Adorno, T.W.,╇ 289 adverbs,╇ 132 agent causation,╇ 223 Aho, T.,╇ 175 alethic modalities,╇ 163 Aloni, M.,╇ 249,╇ 250 Amaral, F.S.,╇ 275 American pragmatic movement,╇ 193 analytic philosophy see analytical philosophy Analytica Priora 1,╇ 163 analytical philosophy, definition,╇ 11 analytical philosophy, meta-philosophical issues,╇ 11 analytical philosophy, state of art,╇ 22 analytical philosophy and pragmatics,╇ 21–3 analytical philosophy developments,╇ 18–21 analytical practice,╇ 97–100 analytical practice, changes in social order of discourse,╇ 99 analytical practice, discourse-knowledgepower,╇ 100

analytical practice, experimentation,╇ 97, 98 analytical practice, interventional,╇ 98 anaphoric interpretation,╇ 140, 145 anaphoric pronouns,╇ 137, 139–40, 144 anaphors,╇ 141 answering comprehension,╇ 191 Apel, K.O.,╇ 126, 289–90, 292–5 Apel’s transcendental pragmatics,╇ 292–5 appropriateness,╇ 22 appropriation of discourses,╇ 97 Archéologie du savoir,╇ 86 archive,╇ 91 Aristotle,╇ 153 Aristotle’s modalities,╇ 163 Aristotle’s theory,╇ 163 Armstrong, D.M.,╇ 236 Art and answerability,╇ 38 assertion and truth,╇ 32 assurance theory,╇ 83 Attardo, S.,╇ 123 attributive,╇ 271 Augustinian Picture,╇ 303 Austin, J.L.,╇ 16–17, 26–36, 27–36 Austin and Wittgenstein,╇ 26 Austin’s approach to philosophy,╇ 26 Austin’s language isolation,╇ 28 Austin’s philosophical method,╇ 26–8 Austin’s tendencies,╇ 28–9 availability principle,╇ 62 Avramides, A.,╇ 123 axiology,╇ 193, 194 axiom of extensionality,╇ 158 axiom types,╇ 281 axiomatic systems of modal logic,╇ 245 Ayer, A.J.,╇ 26, 104, 306

B Bach, K.,╇ 121, 214, 232 Baker, G.P.,╇ 231 Baker, J.,╇ 108, 117–19 Bakhtin, concrete utterance,╇ 44 Bakhtin, linguistic pragmatics,╇ 41 Bakhtin, Mikhail,╇ 38–50 Bakhtin, M.M.,╇ 38–50 Bakhtin Circle,╇ 38, 72 Bakhtin industry, The,╇ 40–1 Bakhtin’s biography,╇ 38–40 Bakhtin’s view of language,╇ 41–50 Bally, C.,╇ 128 Bar-Hillel, Y.,╇ 21, 123, 232 Barcan formula,╇ 166 Barwise, J.,╇ 35, 175, 176, 209 Basic Set,╇ 54 Bayesian decision theory,╇ 222–3 Beaney, M.,╇ 253 Bedeutung,╇ 253–5 Begriffschrift (1879),╇ 164 behavioral dispositions,╇ 236 behavioral semiotics,╇ 193 behaviorism,╇ 182, 235 behavioristics,╇ 182 being,╇ 75 Bennett, M.R.,╇ 307 Bernard, J.,╇ 250 Bezuidenhout, A.,╇ 64 Bianchi, A.,╇ 253 Bianchi, C.,╇ 53–68, 253–77 binarisms,╇ 74 biosemantics,╇ 240 Black, M.,╇ 33, 122 Blackburn, S.,╇ 166, 282 Blakemore, D.,╇ 284, 285, 287 Blanchot, M.,╇ 76 Block, N.,╇ 236, 237, 238, 240 Bloomfield, L.,╇ 127, 184 Blue Book (Wittgenstein, 1933–34),╇ 26 Boeckh, P.,╇ 127 Bolzano, B.,╇ 13, 174

310 Index Boolos, G.,╇ 158 Borg, E.,╇ 54, 57 bound anaphors,╇ 141, 143 box operator,╇ 165, 246 Brandist, C.,╇ 38, 42 Brandom, R.,╇ 20 Braun, D.,╇ 148, 264 Brentano, F.,╇ 217, 218 Bromberger, S.,╇ 231 Burge, T.,╇ 83, 231 C California: The Ultimate Guidebook,╇ 138 The Cambridge Introduction to Jacques Derrida (2007),╇ 75 Cann, R.,╇ 202, 207–13 canonical language,╇ 15 Cappelen, H.,╇ 54, 55, 57, 60, 63, 133, 149 Carnap, R.,╇ 14, 15, 21, 53, 155, 166, 168, 181, 182, 195, 210, 229, 245–6, 256 Carnap, Rudolf and encyclopedia of unified science,╇ 14–15 carnival,╇ 49–50 carnival, categories of,╇ 50 carnivalistic mésalliances,╇ 50 carnivalization,╇ 49–50 Carruthers, P.,╇ 241 Carston, R.,╇ 22, 57, 64, 241, 280, 284 cartesian product,╇ 212 Casanovas, P.,╇ 88 Castañeda, H.-N.,╇ 135, 140 causal relations,╇ 236 Cavell, S.,╇ 29 Chapman, S.,╇ 122 character,╇ 54, 213 characteristic function,╇ 207 Chartier, R.,╇ 90 Chierchia, G.,╇ 203, 211 Chisholm, R.,╇ 223, 224 Chomsky, N.,╇ 107, 230, 231 Chouliaraki, L.,╇ 99 chronotope,╇ 49 Church, A.,╇ 158 Churchland, P.M.,╇ 237 Churchland Smith, P.,╇ 237 circumstances of evaluation,╇ 213 Clark, H.H.,╇ 38, 182 Clark, L.,╇ 122

classical philosophy of consciousness,╇ 292 classical semiotical theories,╇ 292 cognitive command,╇ 260 cognitive linguistics,╇ 128–9 cognitive pragmatics,╇ 240–2 cognitive relation,╇ 274 cognitive sciences,╇ 235 cognitive semiotics,╇ 197 Cohen, L.J.,╇ 122 coindexation vs. anaphoricity,╇ 141 combinatorics vs. preferences,╇ 177 common sense knowledge,╇ 176 complex demonstratives,╇ 132, 135, 137 compositionality and context-dependence,╇ 161 conceived value,╇ 199 Concept of Mind (1949),╇ 306 concept words (Begriffswörter),╇ 253 conceptual role semantics,╇ 240 concrete utterance,╇ 46 connectives,╇ 164 consciousness,╇ 237 constatives,╇ 32 constructivist structuralist,╇ 95 consumerist semantics,╇ 265 content,╇ 54 context,╇ 198 context-free grammar,╇ 203 context of use,╇ 132 context of utterance,╇ 213, 270 context sensitive,╇ 63, 65 context-sensitive expression,╇ 57, 132 context-sensitivity,╇ 136 context shifting argument,╇ 60 contextualism,╇ 22 contextualism, defined,╇ 54 contextualism, perspectives,╇ 53–4 conversation analysis,╇ 129 conversational implicatures,╇ 114 conversationalist hypothesis,╇ 112–14 Cooper, R.,╇ 209 Corazza, E.,╇ 131–50 correspondence theory,╇ 174 cosmology,╇ 194 counts-as parameter,╇ 65–6

Couturat, L.,╇ 245 creature construction,╇ 117–19 Cresswell, M.J.,╇ 167, 245, 248 critical hermeneutics,╇ 126 critical theory,╇ 289 cross-sentential/discourse anaphora,╇ 145, 146 CSA, see Context Shifting Argument,╇ 60 cutting up the semiotic pie,╇ 112 Cuypers, S.E.,╇ 235–42 D Dancy, J.,╇ 79 Darwinian biology,╇ 186 Das Apriori der Kommunikationsgemeinschaft und die Grundlagen der Ethik (1973b),╇ 295 Dascal, M.,╇ 240 Davidson, A.,╇ 76 Davidson, D.,╇ 20, 34, 80, 156, 175, 224–6, 232, 240, 282–3 Davidson, principle of charity,╇ 20 Davis, S.,╇ 232 Davis, W.A.,╇ 123 The day of art,╇ 38 de dicto vs. de re constructions,╇ 166 De interpretatione (Aristotle),╇ 163 de Rose, K.,╇ 23 de Saussure, F.,╇ 41, 47 deconstruction,╇ 71–7 deconstruction, criticism against,╇ 77 deconstruction, Derrida’s work on,╇ 71 deconstruction, Freud’s contribution to,╇ 71 deconstruction, historical background,╇ 71–3 deconstruction, textual unconscious,╇ 71 deconstruction in literature and linguistics,╇ 76–7 deconstruction tenets,╇ 73–6 deconstruction theory and practice,╇ 71 definite description,╇ 154, 265–77 deflationary,╇ 64 deictic,╇ 141

deictic frame,╇ 168 deictic temporal reference,╇ 168 deixis, study of,╇ 21 deliberation and action,╇ 222–3 demonstrative conventions,╇ 32 demonstrative pronouns,╇ 132 demonstratives,╇ 54, 134–5 Dennett, D.C.,╇ 104, 237–9 denotation function,╇ 204 Derrida, J.,╇ 71, 73–6 descriptive conventions,╇ 32 designatum,╇ 197 Destruktion,╇ 72 Devitt, M.,╇ 271, 275–6 Dewey, J.,╇ 180, 181, 185, 187, 189, 192, 195 Di Giovanna, J.,╇ 28 dialogic conception,╇ 42 Dialogized heteroglossia,╇ 43 dialogue,╇ 42–3 dialogue, sense of,╇ 42 dialogue and utterances,╇ 42 diamond operator,╇ 165, 246 differance,╇ 74 Dilthey, W.,╇ 126 direct reference theory,╇ 262 discourse,╇ 89–92 discourse, experiences,╇ 92 discourse, forms of rationality,╇ 92 discourse, knowledge and power,╇ 92–5 discourse, reason,╇ 92 discourse, regularities,╇ 91 discourse, relations of force,╇ 92 discourse, strategic relations,╇ 92 discourse as practice,╇ 89–92 discourse as process,╇ 91 discourse as system of rules,╇ 91 Discourse in the novel,╇ 39, 45 discourse representation theory,╇ 249 discourses,╇ 248 discursive event,╇ 91 discursive formation,╇ 89 discursive intervention,╇ 98 discursive practice,╇ 90 Diskurs und Verantwortung (1988),╇ 295 Diskursethik (discursive ethics) (Apel 1988; Habermas 1983),╇ 290 dissemination,╇ 74

Index doctrine of saying and showing,╇ 301–2 Dokic, J.,╇ 64, 131 Donnellan, K.S.,╇ 260–5, 271–5 double-voiced discourse (dvugolosoe slovo),╇ 43 Dowty, D.R.,╇ 157, 232, 283 Dretske, F.I.,╇ 240 Dreyfus, H.L.,╇ 85 Ducrot, O.,╇ 87 Dummett, M.,╇ 12, 13, 20, 229, 231 Duranti, A.,╇ 100 Dynamic Predicate Calculus,╇ 249 dynamic semantics,╇ 249 dynamical interpretant,╇ 189 E e-languages,╇ 230 eccentricity,╇ 50 Edelman, G.,╇ 129 eidetic variation,╇ 219 Eigennamen,╇ 253–5 Elements of Symbolic Logic,╇ 168 eliminative materialism,╇ 237 embodiment,╇ 128 Emerson, C.,╇ 40, 41, 43, 45, 50 Emotions and Will (1859),╇ 186 The End of History and the Last Man (1992),╇ 75 Engel, P.,╇ 12 English as a formal language,╇ 157 English Futilitarianism,╇ 107 enounced (énoncé),╇ 89 Enrici, A.,╇ 85 entia realissima,╇ 120 enunciation,╇ 96 enunciative system,╇ 89 epistemic modality,╇ 164 epistemological breakdown,╇ 86 epistemology,╇ 28–30, 79–81, 193, 229 epistemology, as part of philosophy,╇ 79 equal exchange semiotics,╇ 195 essential indexicals,╇ 135 Etchemendy, J.,╇ 163, 175–6 Ethica Nicomachea (Aristotle),╇ 222 ethnomethodology,╇ 221 evaluational contextualism,╇ 65 Evans, G.,╇ 265, 284

event causation,╇ 223 Evnine, S.,╇ 283 Excuses (1979),╇ 33 existential generalization,╇ 156 experimentalism,╇ 187 experimentation,╇ 88, 98 expressions,╇ 54 extension,╇ 210 extensional operator,╇ 212 extensional vs. intensional grammatical constructions,╇ 156–7 extensionality principle and failures,╇ 153–4 F F-equivalence,╇ 155 F-truth,╇ 155 Fairclough, N.,╇ 48, 77, 97, 99 familiar contact among people, carnival,╇ 50 familiar speech of the street (famil’iarno-ploshchadnaia rech’),╇ 50 Faulkner, P.,╇ 82–4 Festival (1966),╇ 195 Field, H.H.,╇ 238 first philosophy,╇ 79, 229, 292 Fixation of Belief (1877),╇ 186 Florence, M.,╇ 87 Fodor, J.A.,╇ 20, 21, 238, 239, 240, 284 Fodor, J.D.,╇ 286 folk psychology,╇ 237–9 Forguson, L.W.,╇ 30, 35 The formal method in literary scholarship,╇ 39, 72 formal semantics,╇ 202 formal semantics, principles,╇ 202–3 Formal’nyi metod v literaturovedenii (1928),╇ 39 forms of rationality,╇ 92 Forms of time and of the chronotope in the novel (1981),╇ 39 Foucault, condition of possibility,╇ 86–7 Foucault, Michel,╇ 85–103 Foucault and discursive turn,╇ 86–9 foundationalism,╇ 79

311

312

Index Foundations of the Theory of Signs (Morris),╇ 181, 182, 189, 191 frame,╇ 247 frame of reference,╇ 168 free carnival,╇ 50 freedom and responsibility,╇ 34–5 Frege, G.,╇ 11, 12–13, 26, 132, 154, 253–6, 265 Frege theory of sense and reference,╇ 154 Frege’s influence on analytical philosophy,╇ 12–13 frequently used symbols,╇ 214–15 Friggieri, J.,╇ 33, 34 function of existence,╇ 89 Functional Calculus of First-Order Based on Strict Implication,╇ 166 functionalism,╇ 235, 236 functioning of discourses,╇ 91 functions-in-intension,╇ 158 G Gabilondo, A.,╇ 85–101 Gadamer, H.-G.,╇ 126 Gamut, L.T.F.,╇ 215, 249 Garcia-Carpintero, S–M.,╇ 175 Garfinkel, H.,╇ 221 Gasparov, B.,╇ 46, 48 Gauker, C.,╇ 22, 56, 65 Gazdar, G.,╇ 285 Gellner, E.,╇ 27 generalized conversational implicature (GCI),╇ 114 generative linguistics,╇ 283 Gerbrandy, J.,╇ 251 Gibbs, R.,╇ 62 Gillett, C.,╇ 235 Glock, H.-J.,╇ 2, 12 Gochet, P.,╇ 153, 163, 171, 175, 244, 250 Goldman, A.I.,╇ 34 Grace, W.,╇ 85, 89 Graham, K.,╇ 27, 29, 35 grammar subsystem,╇ 177 Grandy, R.E.,╇ 105, 108, 117, 120, 122 Gribomont, P.,╇ 250 Grice, H.P.,╇ 17–18, 22, 104–20 Grice, life of,╇ 105–8 Grice, P.,╇ 3, 5, 9, 27, 30, 35, 54, 83

Grice, philosophy of language,╇ 104 Gricean thesis,╇ 62 Grice’s cooperative principle,╇ 113 Grice’s theory of meaning,╇ 283 Groenendijk, J.,╇ 249 Grosz, E.,╇ 75, 76 Grundlagen der Arithmetik,╇ 26 Gärdenfors, P.,╇ 209 Gurwitsch, A.,╇ 220 Guttenplan, S.,╇ 241 H Habermas, J.,╇ 289–96 Habermas’ universal pragmatics,╇ 290–2 habits of behavior,╇ 188 Hacker, P.M.S.,╇ 13, 231, 307 Halle, M.,╇ 231 Hamm, F.,╇ 158, 159, 161 Hare, R.M.,╇ 21, 306 Harman, G.,╇ 122, 240 Harnish, R.M.,╇ 122, 232 Harris, R.,╇ 87, 127 Hasker, W.,╇ 235 Haugeland, J.C.,╇ 237, 238, 240 Heidegger, M.,╇ 71, 73, 76, 126, 289 Hempel, C.G.,╇ 236 hermeneuein,╇ 125 hermeneutic enterprise,╇ 126 hermeneutic thinking, evolution of,╇ 126 hermeneutic thinking, origins of,╇ 125–6 hermeneutical circle,╇ 126 hermeneutical philosophy,╇ 289 hermeneutics,╇ 125 hermeneutics of Dasein,╇ 126 heteroglossia,╇ 43 Higginbotham, J.,╇ 284 Hilpinen, R.,╇ 231 Hintikka, J.,╇ 232, 248, 251 Histoire de la folie à l’âge classique,╇ 88 Histoire de la sexualité,╇ 88 Hodge, R.,╇ 77, 91 Hodges, W.,╇ 249 Holquist, M.,╇ 38, 39 How to Do Things with Words (1975),╇ 31–3, 227 How to Make our Ideas Clear (Peirce, 1878),╇ 186

Huang, Y.,╇ 143–5 Hughes, G.E.,╇ 167 Husserl, E.,╇ 13, 28, 117, 217–21 hybrid contexts, problem of,╇ 155–6 hymen,╇ 74 hyperintensionality,╇ 159–60 I i-languages,╇ 230 ICD. See Inter-Contextual Disquotational Indirect Report ICD test,╇ 133–4 The Idea of a Social Science (1970),╇ 306 idealistic metaphysics,╇ 188 identification of a sign,╇ 190 identification semiotics,╇ 195 identity-theory,╇ 235, 236 Illustrations of the Logic of Science,╇ 186 Image (1976),╇ 195 immediate interpretant,╇ 189 inappropriateness argument,╇ 60, 61 indexical expressions,╇ 132–4 indexical semantics,╇ 285 indexicalism,╇ 54, 55–7 indexicality,╇ 131, 285–6 indexicals,╇ 132 indexicals, direct reference view,╇ 136 indexicals and contexts,╇ 148–50 indexicals as singular terms,╇ 136–7 indirect discourse,╇ 154–5 informational semantics,╇ 240 Inquiry into Meaning and Truth (Russell),╇ 153 instrumentalism,╇ 187 intension,╇ 210 intension determines extension,╇ 157 intension problem,╇ 65 intension vs. extension,╇ 153 Intensional constructions,╇ 156–8 intensional contexts, Frege-Carnap treatment of,╇ 154–5 Intensional expressions,╇ 212 intensional operator,╇ 212

intensional semantics to discourse representation theory,╇ 19 intensionality,╇ 209–12 intention-convention approach,╇ 283 intentional action,╇ 224–5 intentional communication,╇ 109 intentional eliminativism,╇ 237 intentional instrumentalism,╇ 238 intentional realism,╇ 238 intentional stance theory (IST),╇ 238 intentionality,╇ 237 intentionality, problem of,╇ 237–9 intentionality intuition,╇ 261 intentionality of propositional attitudes,╇ 238–9 Inter-Contextual Disquotational Indirect Report,╇ 133–4 interaction,╇ 289 International encyclopedia of unified science,╇ 15, 21 interpersonal anaphora,╇ 146 interpretant,╇ 189, 197 interpretation semiotics,╇ 195 interpretationism,╇ 240 interpretative repertories,╇ 94 interpreter,╇ 189, 197 intertextuality,╇ 46 Introduction to Montague Semantics,╇ 157 intuitive truth-conditions,╇ 61 The Irony of Deconstruction and the Example of Marx,╇ 76 Irving Lewis’ contribution to modal logic,╇ 164–5 Iskusstvo i otvetstvennost (Den’ iskusstva),╇ 38 isomorphism,╇ 61 Ivanov, V.V.,╇ 40, 41 J Jackendoff, R.,╇ 284 Jackson, F.,╇ 22 Jakobson, R.,╇ 5, 220 James, W.,╇ 106, 107, 180, 181, 185–7, 192, 195, 220 Janik, A.,╇ 306 Jaspers, K.,╇ 219

Index Journal of Speculative Philosophy (1868),╇ 186 Journal of Symbolic Logic,╇ 166 Jäger, M.,╇ 99 Jäger, S.,╇ 99 K K filosofii postupka (1986),╇ 40 Kameyama, M.,╇ 174, 176, 177 Kamp, H.,╇ 19, 169 Kaplan, D.,╇ 21, 54, 132, 134, 136, 148, 150, 213–14, 230, 232, 245, 261–2, 265 Kaplan, M.,╇ 30 Kaplanian model,╇ 139 Kasher, A.,╇ 122, 228–32 Katz, J.,╇ 171 Kempson, R.,╇ 282 Kenny, A.,╇ 222 Kim, J.,╇ 236, 237 King, J.,╇ 137, 265 Kölbel, M.,╇ 66 knowing-who,╇ 251 knowledge and belief,╇ 29–30 kommunikativnye fragmenty (communicative fragments),╇ 46 Konkin, S.S.,╇ 38, 40 Konkina, L.S.,╇ 38, 40 Kratzer, A.,╇ 211 Kress, G.,╇ 49, 77, 99 Kripke, S.A.,╇ 19, 80, 139, 145, 232, 246–7, 261–5, 271, 273–7, 283 Kripke’s hypothesis,╇ 276 Kristeva, J.,╇ 40, 46 L L-equivalence,╇ 155 L-truth,╇ 155 la langue,╇ 128 la parole,╇ 128 Lackey, J.,╇ 82 Lakoff, G.,╇ 128, 129 Lambalgen, M. van,╇ 158, 159, 161 lambda abstraction,╇ 207–8 lambda calculus,╇ 158 lambda conversion,╇ 207 lambda operator,╇ 207 Langford, C.,╇ 244 The Language of Morals (1972),╇ 306 Language, Truth, and Logic (1936),╇ 306

language and literature,╇ 88 Language and reality,╇ 118 language and thought,╇ 229 language-based conceptual analysis,╇ 22 language-games,╇ 16, 304, 306 language-mind connection,╇ 283 language of thought,╇ 20, 238 language-world connection,╇ 283 langue,╇ 73, 128, 290 Lappin, S.,╇ 232 L’Archéologie du Savoir,╇ 91 Larson, R.,╇ 282 Lasnik, H.,╇ 139, 145, 146 The Late Derrida (2007),╇ 76 Later Derrida (2003),╇ 76 Laws of Arithmetic,╇ 297 Lazar, M.,╇ 99 le seuil du classicisme à la modernité,╇ 87 L’Ecole française d’analyse du discours,╇ 86 Lectures on philosophy,╇ 13–14 Lemmon, E.J.,╇ 244, 245 Leonardi, P.,╇ 277 Lepore, E.,╇ 7, 54, 60, 63, 133, 149, 240 Les mots et les choses,╇ 87, 88 Levinson, S.C.,╇ 109, 112, 114, 129, 286 Lewis, D.K.,╇ 111, 139, 145, 236, 283–6 Lewis, I.,╇ 164–5, 185, 244 liberty of indifference,╇ 225 liberty of spontaneity,╇ 225 Limited Inc (Derrida 1988),╇ 74 linguistic anthropology,╇ 128 Linguistic Anthropology (1993),╇ 307 linguistic botanizing,╇ 27 linguistic code,╇ 113 Linguistic Contact Without Cognitive Contact,╇ 265 linguistic context,╇ 176 linguistic direction,╇ 13 linguistic market,╇ 97 linguistic meaning of indexicals,╇ 132–3 linguistic memory,╇ 46 linguistic phenomenology,╇ 17, 28 linguistic theory,╇ 232 linguistic turn,╇ 12, 87

313

314 Index linguistic unification,╇ 45 linguistics and hermeneutics,╇ 126–9 literalism,╇ 53 Loewer, B.,╇ 235 Logic and conversation (1975),╇ 18, 241 logic program,╇ 159 logical atomism,╇ 13, 14 logical behaviorism,╇ 236 logical-connection argument,╇ 223–5 logical consequence,╇ 247 logical constants,╇ 301–2 logical constructions,╇ 28 logical empiricism,╇ 15, 18 logical final interpretant,╇ 189 logical form,╇ 232, 300 logical modalities,╇ 163 logical necessity,╇ 163 logical positivism,╇ 15 Logical Positivism, Pragmatism and Scientific Empiricism (1937),╇ 180–1 The Logical Syntax of Language (1937),╇ 306 Logisch-Philosophische Abhandlung see Tractatus Logische Untersuchungen,╇ 220 logocentricism,╇ 71 L’ordre du discours and Surveiller et punir,╇ 88 Lähteenmäki, M.,╇ 42, 45 Luria, A.,╇ 129 M M. M. Bakhtin: sobranie sochinenii,╇ 41 M-intentions,╇ 117 Maas, U.,╇ 99 MacKay, A.F.,╇ 123, 274 Maingueneau, D.,╇ 86, 91 Malcolm, N.,╇ 223, 307 mandatory semantic process of saturation,╇ 55, 56 Mandelbaum, D.,╇ 128 Marková, I.,╇ 40 Marksizm i filosofiia iazyka (1929),╇ 39 Martin, R.M.,╇ 219, 232 Martín Rojo, L.,╇ 85–101 Marxism and the philosophy of language (1973),╇ 39, 72

material thing,╇ 28, 30 mathematical models,╇ 204 Mathesius, V., 220 McConnell-Ginet, S.,╇ 203, 211, 283 McDowell, J.,╇ 284 McGinn, C.,╇ 286 Mchoul, A.,╇ 85 McQuillan, M.,╇ 76 Mead, G.H.,╇ 180–3, 185, 192, 195, 198 meaning,╇ 31, 108–12, 183, 231, 269 Meaning and necessity,╇ 155 meaning and Â�understanding,╇ 19–20 meaning as use,╇ 231 Medvedev, P.N.,╇ 38, 39 mental antecedent and action,╇ 223 mental states,╇ 236 Merleau-Ponty, M.,╇ 220 metalinguistic view vs. linguistic view,╇ 44 metalinguistics,╇ 41, 44–7 metalingvistika (metalinguistics),╇ 44 Metaphysical Club,╇ 186 metaphysical transubstantiation,╇ 119 metaphysics,╇ 117, 229 Method in philosophical psychology,╇ 118 Miller, D.,╇ 71, 159 Millikan, R.G.,╇ 21, 240 Mind, Self, and Society,╇ 182 mind-body problem,╇ 20, 235–7 mirror of mind,╇ 86 mirror of nature,╇ 86 Mitchell, W.,╇ 76 modal auxiliaries,╇ 164 modal logic,╇ 165–6 modal logic development,╇ 163–4 Modalities and Quantification (Carnap 1946),╇ 166 modality of existence,╇ 89 model theory,╇ 203–5 model-theory for contexts,╇ 177–8 moderate relativism,╇ 35 Moise, J.,╇ 62 Monist (1905),╇ 187 monologic,╇ 42

Montague, R.,╇ 19, 157, 172, 202–3, 212, 215, 245, 248, 282, 283, 285 Montague semantics, see formal semantics Montague’s intensional model,╇ 157 Montague’s semantics,╇ 157 Montague’s theory,╇ 205 Moore, G.E.,╇ 13, 14, 26, 104, 106 Moran, R.,╇ 83 Morris, C.L.,╇ 1, 6, 15, 21, 180–99, 228, 232, 292, 293 Morris’s behavioristics and pragmatics,╇ 180–5 Morris’s pragmatics,╇ 185–92 Morris’s semiotics,╇ 182 Morson, G.S.,╇ 39, 40, 43, 45 Moschovakis, Y.,╇ 158, 159 Mueller-Vollmer, K.,╇ 125, 126 multiple-description thesis,╇ 225 Muskens, R.,╇ 159, 160, 175 N Nagel, T.,╇ 237 Naissance de la clinique,╇ 88 Naming and Necessity,╇ 262 natural language,╇ 230–1 natural language, aspects of,╇ 231 natural language, nature of,╇ 230–1 natural meaning,╇ 108 natural realism,╇ 30 naturalistic approach,╇ 21 Naturalized epistemology,╇ 79 Neale, S.,╇ 122, 123, 275 neo-Hegelian absolute Â�idealism,╇ 13 Neo-Marxist forms of Â�deconstruction,╇ 71 Neurath, O.,╇ 14, 182 new theory of reference,╇ 262 The New York Times,╇ 82 Nichomachean Ethics,╇ 33, 105 Niiniluoto, I.,╇ 160 non-reductive theory,╇ 84 non-truth-conditional aspects of semantics,╇ 286–7 nonindexical contextualism,╇ 64–6 nonnatural meaning,╇ 108–9 normalization,╇ 94 Nunberg, G.,╇ 213

O object value,╇ 198 Of grammatology (Derrida 1976),╇ 75 On certainty,╇ 80 On Interpretation,╇ 125 ontic modalities,╇ 163 ontological Marxism,╇ 120 ontology,╇ 204, 229 Opacity and the attitudes,╇ 160 open sentences,╇ 153, 154 The Open Self (1948),╇ 183, 195 operative value,╇ 198 order of discourse,╇ 95–7 ordinary language approach,╇ 18 ordinary language philosophy,╇ 26, 106 ordinary language philosophy, decline of,╇ 18 overt semantic structure,╇ 255 Oxford ordinary language philosophy,╇ 22 Oxford philosophers,╇ 16–18 P Paducheva, E.V.,╇ 50 Palmer, F.R.,╇ 164 Palmer, R.,╇ 125 Papineau, D.,╇ 240 paradigm of philosophy,╇ 13 parergon,╇ 74 parole,╇ 73, 128, 220, 290 Partee, B.,╇ 203, 204, 209 Passmore, J.,╇ 16 Pateman, T.,╇ 231 Paths of Life (1942),╇ 183, 195 patterns unification,╇ 159 Pegasus,╇ 167 Peirce, C.S.,╇ 180–3, 185–97, 289, 292, 294 Peirce’s pragmaticism,╇ 185–92 Peirce’s semiotics,╇ 187 Peirce’s studies on signs,╇ 188 perception,╇ 30 performative/constative dichotomy hypothesis,╇ 33 performative hypothesis,╇ 286 performative self-contradiction,╇ 294 performative utterances,╇ 31–2 Perry, J.,╇ 22, 35, 58, 67, 131, 132, 135, 136 personal pronouns,╇ 54, 132

Index Peters, S.,╇ 157, 232, 283 Petitot, J.,╇ 220 Petrilli, S.,╇ 180–99 pharmakon,╇ 74 phatic act,╇ 33 phenomena,╇ 217 phenomenological movement,╇ 217–19 phenomenology, definition,╇ 217, 218 phenomenology, linguistics, and social sciences,╇ 219–21 phenomenology of language,╇ 220 Phenomenology of mind,╇ 217 Phillips, D.Z.,╇ 306 philosophical analysis by Moore and Russell,╇ 13–14 Philosophical Conceptions and Practical Results,╇ 186 Philosophical Investigations,╇ 302–3, 306 philosophical propositions,╇ 14 philosophical studies of natural language, purposes of,╇ 228 philosophical theory,╇ 232 Philosophische Texte (2009),╇ 289 Philosophische Untersuchungen/ Philosophical Investigations,╇ 298 philosophy of action,╇ 33–5, 222–7, 229 philosophy of art,╇ 229 philosophy of history,╇ 229 philosophy of language,╇ 30, 31–3, 228–32, 239 philosophy of meaning,╇ 239 philosophy of mind,╇ 20–1, 229–30 philosophy of mind naturalized,╇ 235 philosophy of religion,╇ 229 philosophy of science,╇ 228 phonetic act,╇ 33 physicalism,╇ 235 picture theory of meaning,╇ 299 Pitcher, G.,╇ 32 Pitkin, H.F.,╇ 307 Place, U.T.,╇ 236 The Play Group,╇ 106 polarization,╇ 96 The Politics of Deconstruction (2007),╇ 76

polyphony,╇ 43–4 Ponzio, A.,╇ 198 Popular Science Monthly,╇ 186 possessive adjectives,╇ 132 possible worlds,╇ 232 possible worlds semantics,╇ 244–51 possible worlds semantics, emergence of,╇ 245–7 possible worlds semantics to pragmatics,╇ 248–51 postsemantic level context,╇ 149–50 poststructuralism,╇ 73 Potter, J.,╇ 94 Powell, G.,╇ 241 practicalism,╇ 193 pragmatic intrusion,╇ 104 The Pragmatic Movement in American Philosophy (Morris),╇ 181, 192 pragmatic movement in America,╇ 194 pragmatic philosophy in US,╇ 192–5 pragmatic principles,╇ 250 pragmatic theory,╇ 160 pragmatical dimension of semiosis,╇ 183 pragmatics,╇ 21, 285 pragmatics, definition by Charles Morris,╇ 180 pragmatics defined by Â�Morris,╇ 195 pragmatics subsystem,╇ 177 pragmatisch,╇ 193 Pragmatism (1907),╇ 187 pre-rational structure,╇ 118 Predelli, S.,╇ 64, 66, 137, 139, 145 predicate logic,╇ 203–5 presemantic level context,╇ 148 presuppositions,╇ 142 primary speech genres,╇ 48 Principia mathematica (1910–13), 164, 297 principle of acquaintance,╇ 257 principle of charity,╇ 20 principle of compositionality,╇ 161, 203, 240 principle of identifying descriptions,╇ 260 principle of intensionality,╇ 155

315

316 Index principle of verifiability,╇ 15 The problem of speech genres,╇ 39, 47 The problem of the text (1959–61),╇ 44 Problema rechevykh zhanrov (1952–3),╇ 39 problematization,╇ 88–9, 98 problematization procedure,╇ 99 Problems of Dostoevsky’s creative art,╇ 39 Problems of Dostoevsky’s poetics (1973),╇ 39 Problemy poètiki Dostoevskogo (1963),╇ 39 Problemy tvorchestva Dostoevskogo (1929),╇ 39 process of objectivization,╇ 93 profanation,╇ 50 programmatic theories,╇ 231–2 Prolegomena to an Apology for Pragmaticism (1906),╇ 187–8 Promoting extensionality,╇ 160 prononcer les idées,╇ 86 proper name as rigid Â�designator,╇ 263 proper names,╇ 253–65, 257, 258–9 The proper Treatment of Quantification in ordinary English (Montague 1974),╇ 157 propositional attitude Â�psychology,╇ 238–9 propositional attitudes,╇ 237 propositional attitudes and pragmatics,╇ 160–1 propositional calculus,╇ 165 propositional functions,╇ 154, 266 proto-perceptions,╇ 135 psychosemantics,╇ 239–40 pure indexicals,╇ 134–5 Putnam, H.,╇ 64, 66, 131, 137, 138, 139, 146, 147 Pêcheux, M.,╇ 99 Q quantified modal logic,╇ 166–7 quantifiers,╇ 175–6, 208–9 Questions Concerning Certain Faculties Claimed for Man (1868),╇ 188

Quine, W.V.O.,╇ 6, 18–19, 79, 107, 160, 239, 250 Quinean approach to epistemology,╇ 79 R Rabelais and his world (1968),╇ 39 Rabinow, P.,╇ 85 radical contextualism,╇ 53, 57–64 radical contextualism, motivations for,╇ 60–2 radical contextualism, objections to,╇ 62–4 radical pragmatic,╇ 105 Railton, P.,╇ 9 range of sentence,╇ 246 rationality,╇ 114–17 rationality, evaluative principles,╇ 115 Raz, J.,╇ 223 re-problematization,╇ 98 reason,╇ 92 reconstructive science,╇ 294 recursive definition of truth,╇ 247 reductionism,╇ 183 reductive theory,╇ 84 reference,╇ 210, 257, 260 reference pointer,╇ 169 reference shift,╇ 156 referential,╇ 271 Reichenbach, H.,╇ 15, 21, 132, 168 relations of force,╇ 92 relativism,╇ 66 relevance theory,╇ 22,╇ 241 representation of a representation (thought),╇ 86 representational form,╇ 300 representational theory of the mind (RTM),╇ 238 Reyle, U.,╇ 19 rhetic act,╇ 33 Rijke, M. de,╇ 166 Roberts, C.,╇ 100 Rorty, R.,╇ 12, 76, 85 Roulet, E.,╇ 44 Recanati, F.,╇ 22, 35, 53, 57, 58, 62–6, 109, 136, 160, 161, 232, 241 rule-by-rule hypothesis,╇ 203 rule of functional application,╇ 206

rule of necessitation,╇ 245 Russell, B.,╇ 13–14, 53, 154, 213, 256–8, 265–8, 297 Russell’s theory of description,╇ 268 Ryle, G.,╇ 16, 104, 106, 223, 236, 306 S Saarinen, E.,╇ 160 salience,╇ 177 Salmon, N.,╇ 275 Sandu, G.,╇ 175 Sapir-Whorf hypothesis,╇ 289 Sarangi, S.K.,╇ 100, 123 Sbisà, M.,╇ 1, 5, 11, 26, 33, 119, 131 scepticism,╇ 81 Schiffer, S.R.,╇ 107, 109, 122 Schleiermacher, F.,╇ 126 scholasticism,╇ 108 Schuetz, A.,╇ 220 science of ideology,╇ 86 Scientific Empiricism (Morris),╇ 180, 181 Scott, D.,╇ 154, 165, 259, 267, 268 Searle, John,╇ 258–60 Searle, J.R.,╇ 5, 20, 33, 57, 58–60, 106, 111, 121, 122, 195, 227, 232, 235, 237, 241, 258–60 second-level constructs,╇ 221 secondary speech genres,╇ 48 self-controlled conduct,╇ 193 semantic ambiguity,╇ 156 semantic argument,╇ 263 semantic atomism,╇ 240 semantic competence,╇ 246 semantic content,╇ 63 semantic holism,╇ 239 semantic level context,╇ 148–9 semantic minimalism,╇ 53, 54–5 semantic problems,╇ 15 semantic reference,╇ 275 semantic representation,╇ 284 semantic significance,╇ 248 semantic theory,╇ 15, 155 semantic types,╇ 205–7 semantic underdetermination,╇ 53 semantical dimension of semiosis,╇ 183 semantics,╇ 21, 197 semantics, approaches to,╇ 171–2 semantics, origins of term,╇ 171

semantics to pragmatics, shift of attention,╇ 19 semantics vs. pragmatics,╇ 274–7 semioethics,╇ 197 semiosical theory of mind,╇ 193 semiosis,╇ 184 semiotic sign-function.,╇ 292 semiotically interpreted logic,╇ 193 semiotics,╇ 197 sense-datum,╇ 28, 30 sentence nonliterality,╇ 63 sentences (Sätze),╇ 253 sentential functions,╇ 154 sentential utterance,╇ 111–12 set-theory,╇ 158 Shepherd, D.,╇ 38 Shoemaker, S.,╇ 237 Sidelle, A.,╇ 137 sign,╇ 188 sign activity,╇ 191 sign-processes,╇ 187 sign/sign vehicle,╇ 197 signification,╇ 197–8 Signification and Significance (1964),╇ 183, 194, 195, 196 significatum,╇ 189, 197 significs,╇ 197 signifiers,╇ 72 Signs, Language and Behavior,╇ 182, 187, 189, 191, 193, 194 Signs about Signs about Signs,╇ 189 signs and values,╇ 195–9 single-voiced discourse (odnogolosoe slovo),╇ 43 singular propositions,╇ 136 situational context,╇ 176 situationalism,╇ 64 Six Theories of Mind (Morris),╇ 180, 186 Slavic and east European journal (1986),╇ 39 Slembrouck, S.,╇ 123 slovo,╇ 43 Slovo v romane (1934–5),╇ 39 Smart, J.J.C.,╇ 236 Smith, N.V.,╇ 111 Smith, P.K.,╇ 237, 241 Smith, Q.,╇ 137–9 Smullyan, A.F.,╇ 268 Sneddon, A.,╇ 34 Soames, S.,╇ 12, 30, 271, 275

Index Soams, S.,╇ 232 social order of discourse,╇ 97 Some Consequences of Four Incapacities (1868),╇ 188, 189 Sosa, E.,╇ 79, 89 The space of literature (Blanchot 1989),╇ 76 spatial indexicals,╇ 138 speaker’s reference,╇ 275 Specters of Marx (1994),╇ 75, 76 spectrality,╇ 75 speech act,╇ 32–3, 110 speech act, facets,╇ 33 speech act content,╇ 63 speech act pluralism,╇ 55 speech act theory,╇ 33 speech acts,╇ 232, 290 Speech acts,╇ 227, 291 speech genres,╇ 47–8 Sperber, D.,╇ 22, 57, 104, 122, 241, 284, 286, 287 Spiegelberg, H.,╇ 219 Stalnaker, R.C.,╇ 22, 142, 232 standard semantics, inadequacies of,╇ 158–9 Stanley, J.,╇ 55, 56, 62, 149 state-description,╇ 246 Stenius, E.,╇ 300 Sterelny, K.,╇ 282 Stevenson, C.L.,╇ 21 Stokhof, M.,╇ 249 strategic relations,╇ 92 Strawson, Peter,╇ 269–71 Strawson, P.F.,╇ 17, 22, 32, 54, 106, 213, 232, 269–71, 283 strict conditional,╇ 165 strict implication,╇ 165, 244 strong moderate relativism,╇ 65 structuralism,╇ 128 structure as metanarrative,╇ 73 subjection of discourses,╇ 94 subjection of individuals,╇ 94 subjective meaning-intentions (Bedeutungsintentionen),╇ 220 subjectivist semantics,╇ 261 subjectivization,╇ 87 supplement,╇ 75 syllogism,╇ 163 Symbolism and Reality (Charles Morris),╇ 180 syntactic analysis,╇ 203 syntactical dimension of semiosis,╇ 183

syntactical elements,╇ 54, 56 Szabò, Z.G.,╇ 55, 56, 62, 63, 67, 241, 242 T T-sentences,╇ 280 tacit initiator,╇ 141–2 Tannen, D.,╇ 129 Tarski, A.,╇ 3, 15, 53, 156, 157, 171, 172, 174, 282, 283 Tarskian formula,╇ 282 Tarskian semantics,╇ 156–7, 172–4 Tarski’s semantics, scope and limits of,╇ 174–5 tautology,╇ 244–5 Taylor, K.,╇ 63,╇ 76, 241 teleological semantics,╇ 240 tense logic,╇ 167–8 tense logic to pragmatics,╇ 168–9 Ter Meulen, A.,╇ 203, 204 testimonial knowledge,╇ 83 Thayse, A.,╇ 157, 248 Theorie des kommunikativen Handelns (1981),╇ 289 theory of action,╇ 35 theory of interpretation,╇ 239 theory of meaning,╇ 19, 171, 220 theory of mental representations,╇ 238 theory of reference,╇ 171 theory of signs,╇ 193, 194 theory of truth-conditions,╇ 280 Thomas, W.,╇ 127, 192 Thomason, R.,╇ 159 thought (Gedanke),╇ 254 Tichy, P.,╇ 156 Tihanov, G.,╇ 38 Toward a philosophy of the act (1993),╇ 40 trace,╇ 73 Tractatus logico-philosophicus,╇ 14, 297, 298, 299–301, 301 transcendental consciousness,╇ 219 transcendentalphenomenological reduction,╇ 218 transcendental-pragmatic argumentation strategy,╇ 294 transcendental pragmatics,╇ 289

317

318 Index transcendental re-interpretation,╇ 293 transcendental-semiotical,╇ 293 transcendental signifier,╇ 74, 75 Transformation,╇ 295 Transformation der Philosophie (Apel),╇ 290, 293, 295 translation rules,╇ 204 Travis, C.,╇ 22, 32, 35, 57–62 Treatise on human understanding,╇ 225 triadic sign-relation,╇ 292 true indexicals,╇ 54 truth-conditional approach,╇ 283, 284 truth-conditional semantics, definition,╇ 280 truth-conditional semantics and pragmatics,╇ 284–7 truth-theoretic approach,╇ 282 Turner, K.,╇ 67, 104 Turner, R.,╇ 67, 104 Tvorchestvo Fransua Rable i narodnaia kul’tura srednevekov’ia i Renessansa (1965),╇ 39 two poles of the text,╇ 46 type raising,╇ 209 typed logical language,╇ 205 U Über Sinn und Bedeutung,╇ 154 unbound anaphors,╇ 143–5 unitary language,╇ 45 universal grammar,╇ 232 universal pragmatics,╇ 289 universe narrowers,╇ 135 Update Semantics,╇ 249 Urmson, J.O.,╇ 13, 32 utterance and context,╇ 22

utterance-type meaning,╇ 107 utterer’s meaning,╇ 107 V validity-claims,╇ 290 Vallée, R.,╇ 136 value,╇ 198 value dimensions,╇ 199 value situation,╇ 198 Van Dijk, T.,╇ 96 van Fraassen, B.C.,╇ 228 Vanderveken, D.,╇ 232 variable assignment function,╇ 208 Varieties of Human Value (1956),╇ 183, 196 Veltman, F.,╇ 249 Verständigung,╇ 290 Verstehen,╇ 219 Vienna Circle,╇ 14 Vološinov, V.N.,╇ 38, 39, 45, 46, 71, 72, 74, 196 W Waismann, F.,╇ 54, 59 Wall, R.E.,╇ 157, 203, 204, 232, 283 Warner, R.,╇ 108, 110, 117, 118, 120 Warnock, G.J.,╇ 26, 30, 32, 33 Was heißt Universalpragmatik (Habermas 1976),╇ 289 Wetherell, M.,╇ 94, 95 Wettstein, H.K.,╇ 232, 261, 265, 271, 275 Whitehead, A.N.,╇ 13, 164, 165, 297 Whittaker, R.,╇ 96 why?-questions,╇ 228–9 Wiener Ausgabe,╇ 298 Wierzbicka, A.,╇ 48 Will to Believe (1897),╇ 187

William James Today (1942a),╇ 181 Williamson, T.,╇ 6, 22 Wilson, D.,╇ 57, 241, 284, 286, 287 Wilson, R.A.,╇ 122 Winch, P.,╇ 306 Wisdom, J.,╇ 16 Wittgenstein, Austin, Searle, and Travis view on contextualism,╇ 58–60 Wittgenstein, Ludwig (biography),╇ 297–8 Wittgenstein on Rules and Private Language (1982),╇ 305 Wittgenstein’s influence and ordinary language philosophy,╇ 16 Wittgenstein’s later philosophy,╇ 302–6 Wittgenstein’s philosophy,╇ 298–9 Wittgenstein’s thought, evolution of,╇ 16 Wittgenstein’s Tractatus,╇ 14, 161, 301 Wittgenstein’s Vienna (1973),╇ 306 Wodak, R.,╇ 48, 99 Wolper, P.,╇ 166 words-world relationship,╇ 28 writing,╇ 74 Y Yeager, D.,╇ 34 Z Žižek, S.,╇ 77 Znaniecki, F.,╇ 127 Zolotova, G.A.,╇ 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.