Essays on Speech Acts and Other Topics in Pragmatics 9780192844125

Table of contents :
Cover
Essays on Speech Acts and Other Topics in Pragmatics
Copyright
Contents
Preface
Introduction
1. The Approach
2. The Essays
3. Some Background
4. Where All This Is Heading To
1: On Illocutionary Types
1. Introduction
2. Austin’s Classification of Illocutionary Acts
2.1 How to Describe Illocutionary Effects
2.2 Who Is Affected by Illocutionary Effects?
2.3 Illocutionary Effects and Felicity Conditions
3. Four Illocutionary Types
3.1 Exercitives
3.2 Commissives
3.3 Verdictives
3.4 Behabitives
4. Some Speech-Act Theoretical Games
4.1 Type-Grouping Game
4.2 Type-Shifting Game
4.3 Group-Switching Game
2: Speech Acts, Effects, and Responses
1. Introduction
2. Uptake and Response
3. Uptake and Perlocution
4. Uptake and Allegedly “Conditional” Illocutionary Acts
5. Two Kinds of Inappropriate Response
6. From Inappropriateness to Shifts in Illocutionary Force
7. What Happens in an Adjacency Pair
8. Speech Acts in Conversation
3: Ideology and the Persuasive Use of Presupposition
1. Introduction
2. Informative and Persuasive Uses of Presupposition in the Italian Daily Press
3. Some Doubts about the Received Account of Pragmatic Presupposition
4. Toward an Account of the Persuasive Use of Presupposition
5. Presupposition and Ideology
4: Intentions from the Other Side
1. Introduction
2. The Analysis of Non-natural Meaning
2.1 Some Evidence for a Reading of Grice’s Analysis of Meaningnn Which Focuses on the Audience
2.2 How Grice’s Meaningnn Looks if Seen from the Other Side
2.3 How Objections to Grice’s Original Analysis Look if Seen from the Other Side
3. The Cooperative Principle
3.1 Some Evidence for a Reading of Grice’s CP Which Focuses on the Audience
3.2 Some Evidence against Conversational Implicatures Being Speaker’s Intentions
3.3 How the CP Works if Seen from the Other Side
4. The Notion of a Person
4.1 Value, Persons, and Rationality
4.2 A Retrospective Glance
5: Presupposition, Implicature, and Context in Text Understanding
1. Two Perspectives on Text Understanding
2. On Text and Context
3. The Dynamic Relation between Text and Context
4. Assertion
5. Implicature
6. Presupposition
7. The Distinction between Presupposition and Implicature
7.1 Interactions with Asserted Contents
7.2 Implicatures and Background Assumptions
7.3 Negative Sentences and Presupposition Cancellation
8. Concluding Remarks
6: Illocutionary Force and Degrees of Strength in Language Use
1. Introduction
2. Why a Unified Account of Mitigation/Reinforcement and Illocutionary Force Is Desirable
3. Conditions for a Unified Account of Illocutionary Force and Mitigation/Reinforcement Phenomena
3.1 The Standard Conception of Illocutionary Force
3.2 A Reformulated Conception of Illocutionary Force
4. Cases of Mitigation and Reinforcement
4.1 Speaker’s Entitlement
4.2 Modal Values Assigned to the Addressee
4.3 The Speaker’s Commitment
4.4 Degrees of Strength in Expressed Attitudes and Inner States
4.5 Perlocutionary Goals of the Speech Act
5. Concluding Remarks
7: Speech Acts in Context
1. Context in the Received Speech Act Theory
2. What Should the Context of a Speech Act Be Like?
2.1 Given vs Constructed Context
2.2 Limited vs Unlimited Context
2.3 Objective vs Cognitive Context
3. Some Alleged Incompatibilities
4. Speech Acts and Context Change
5. Concluding Remarks
8: Cognition and Narrativity in Speech Act Sequences
1. Introduction
2. Speech Act Sequencing and the Production of Effects
2.1 What Effects Do Illocutionary Acts Bring About?
2.2 The Role of Sequencing in the Production of Effects
3. The Dynamics of Speech Act Sequences
4. The Role of Narrativity in Speech Act Sequencing
5. The Narrative Schema and the Understanding of Insisting
6. Some Conclusions
9: Two Conceptions of Rationality in Grice’s Theory of Implicature
1. Implicature
2. Is Implicature Rational?
3. Grice on the Rationality of Conversational Implicature
4. The Nature of the Calculability Requirement
5. Two Ideas of Rationality
6. Concluding Remarks
10: How to Read Austin
1. Four Assumptions to Be Disputed
2. In Pursuit of Performatives?
3. Illocutionary Acts and Their Effects
4. What Is a Perlocutionary Act?
5. Did Austin Exclude “Non-Serious” Cases?
6. Some Implications of Austin’s Claims
11: Uptake and Conventionality in Illocution
1. Introduction
2. The Role of Uptake in Illocution
3. The Divide between Conventional and Intention-Based Illocutionary Acts
4. The Conventionality of Illocutionary Effects
4.1 The “Taking Effect” of the Illocutionary Act
4.2 From Conventional Procedures to Conventional Effects
4.3 From Conventionality to the Securing of Uptake
5. How to Achieve a (Conventional) Illocutionary Effect
12: Illocution and Silencing
1. Speech Acts and Power Relationships
2. Pornography and Speech Acts: A Debate
3. The Case of the Ineffective Refusal
4. What Is Wrong in Silencing?
5. Conclusions
13: The Austinian Conception of Illocution and Its Implications for Value Judgments and Social Ontology
1. Introduction
2. Austinian Illocutionary Effects
2.1 When Is an Effect “Conventional”?
2.2 How to Describe Illocutionary Effects
3. Philosophical Implications for the Value–Fact Distinction
4. Philosophical Implications for Social Ontology
5. A Puzzle and Its Proposed Solution
14: Varieties of Speech Act Norms
1. Introduction
2. Constitutive Rules
3. Maxims
4. Objective Requirements
5. Three Examples
6. What about Assertion?
7. Are There “Rules of Accommodation”?
8. Are Speech Act Norms Conventional?
9. Concluding Remarks
15: Ways to Be Concerned with Gender in Philosophy
1. Introduction
2. Critique of Gender through Discourse Analysis
3. Critique of Gender Essentialism
4. How Do Gender Concepts Work?
5. Gender Issues and Intersubjective Recognition
16: Assertion among the Speech Acts
1. Assertion as Illocution
2. Assertion within the Assertive Family
2.1 Members of the Assertive Family and Their Differences
2.2 The Assertive Family: Limits and Structure
3. Assertion within Illocutionary Act Classifications
3.1 Definitions of Assertion and Illocutionary Classes
3.2 The Place of Assertion in Austin’s Illocutionary Classes
3.3 Do Assertions Have Conventional Effects?
4. Is Assertion “Special”?
4.1 Assertion as Central to Inferential Semantics
4.2 Assertion and the Truth/Falsity Assessment
4.3 Assertion and Knowledge
17: Illocution and Power Imbalance
1. Introduction
2. Exercitives
3. Not Only Exercitives
4. Everyday-Life Authorities
5. Established, Accommodated, or Contested Authorities
6. A Back-Door Exercitive and How to Block It
7. Concluding Remarks
References
Index

Citation preview

Essays on Speech Acts and Other Topics in Pragmatics

Essays on Speech Acts and Other Topics in Pragmatics MARINA SBISÀ

Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © Marina Sbisà 2023 The moral rights of the author have been asserted All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2022951821 ISBN 978–0–19–284412–5 DOI: 10.1093/oso/9780192844125.001.0001 Printed and bound in the UK by Clays Ltd, Elcograf S.p.A. Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

Contents Preface

Introduction

vii

1

1. On Illocutionary Types

23

2. Speech Acts, Effects, and Responses

43

3. Ideology and the Persuasive Use of Presupposition

53

4. Intentions from the Other Side

72

5. Presupposition, Implicature, and Context in Text Understanding

90

6. Illocutionary Force and Degrees of Strength in Language Use

105

7. Speech Acts in Context

129

8. Cognition and Narrativity in Speech Act Sequences

144

9. Two Conceptions of Rationality in Grice’s Theory of Implicature

167

10. How to Read Austin

181

11. Uptake and Conventionality in Illocution

195

12. Illocution and Silencing

212

13. The Austinian Conception of Illocution and Its Implications for Value Judgments and Social Ontology

219

14. Varieties of Speech Act Norms

231

15. Ways to Be Concerned with Gender in Philosophy

254

16. Assertion among the Speech Acts

267

17. Illocution and Power Imbalance

288

References Index

307 323

Preface This volume collects seventeen papers of mine dealing with speech acts and related issues of a pragmatic character, covering a range of thirty-six years from 1984 to 2020. The idea and the decision to submit the collection to Oxford University Press were inspired by Rae Langton, to whom I am extremely grateful. I am also grateful for advice and support on this project to Claudia Bianchi, Laura Caponetto, Bianca Cepollaro, and Paolo Labinaz. It was no easy matter to review my life through its academic products but, in the end, the main lines, motivations, and outcomes of my approach to speech as action have now emerged with much greater clarity than before from the numerous separate papers scattered across the years and around various sites of publication. The papers are ordered by date of composition. I have deliberately excluded from the selection papers of a merely introductory nature, review articles, papers addressed to an Italian readership analysing samples from texts or conversations in Italian, and works written in collaboration with other authors. Sixteen of the papers selected, already published in English, were subject to some editing (especially concerning language, style, and footnotes), while one paper was translated into English from Italian. Any new footnotes or additions to the original ones are in square brackets. All references have been gathered in one list at the end of the volume. I am grateful to an enormous number of people whom I met in the course of my career for exchanges of views, discussion, collaboration, encouragement, or help with the English of my papers. They include: Bruno Ambroise, Anita Avramides, Elvio Baccarini, Thomas Ballmer, Lynn Baker, Carla Bazzanella, Claudia Bianchi, Paolo Bouquet, Claudia Caffi, Laura Caponetto, Bianca Cepollaro, Elena Collavin, François Cooren, Alice Crary, Paolo Fabbri, Giolo Fele, Anita Fetzer, Michel Fournel, Bruce Fraser, Christopher Gauker, Hans-Johann Glock, Mitch Green, Sanford Goldberg, William Hanks, Mike Harnish, David Holdcroft, Jennifer Hornsby, Andreas Kemmerling, Paolo Labinaz, Eric Landowski, Rae Langton, Paolo Leonardi, Barry Loewer, Diego Marconi, Jacob Mey, Judy Moss, Gabriella Nuciforo Paoletti, Etsuko Oishi, Herman Parret, Carlo Penco, Snježana PrijćSamaržjia, Paola Rodari, François Recanati, Constantine Sandis, Jennifer Saul, Ken Turner, Achille Varzi, Jef Verschueren, Tim Williamson, Iwona WitczakPlisiecka, Maciej Witek, Tomoyuki Yamada, and Igor Žagar (the list should also include members of my family, but they prefer not to appear). Among the people who contributed to my formation in early years, I would like to remember Amedeo G. Conte, Maria-Elisabeth Conte, Carmela Di Lallo Metelli,

viii



Gaetano Kanizsa, Anthony Kenny, Eugenio Lecaldano, Brian McGuinness, Guido Morpurgo-Tagliabue, Renzo Piovesan, and James O. Urmson. I am also extremely grateful to George Wilson and Paul Smolensky and the Departments of Philosophy, and respectively Cognitive Science of Johns Hopkins University for jointly hosting me as a Visiting Scholar in Spring 1999, to Brian McLaughlin and the Department of Philosophy of Rutgers University for hosting me as a Visiting Scholar in Fall 2002, to Magdalen College, Oxford, for hosting me as a Visiting Fellow in Hilary Term 2006, to Sandra Laugier and Bruno Ambroise for inviting me as a Visiting Professor at the University of Amiens and at CURAPP (CNRS) (France) in May 2010, to the Philosophy Faculty of Oxford University for hosting me as an Academic Visitor on various occasions between 2008 and 2017, and to New College, Oxford, for hosting me as a Visiting Fellow on the sponsorship of Timothy Williamson in Trinity Term 2019.

Introduction The essays collected in this volume examine the categories of speech act theory and apply and develop them in the context of natural discourse and conversation, with the purpose of providing an accurate analysis of how speech can be action. Particular attention is paid to normative aspects of language and language use, with an eye to understanding the social and political dimensions of linguistic activity. Issues concerning implicitness are also considered in the same light. In Section 1 of this Introduction, I shall briefly introduce my approach to speech acts and to implicitness. In Section 2, I introduce the contents of the essays collected here and the contributions they make to speech act theory or neighboring pragmatic issues. Since the collection ranges over different stages in the development of my views, Section 3 will be devoted to providing some narrative background: how my thoughts were formed, what voices influenced them, and how each essay fits in the general context of my research activities. Section 4 provides some details about my current views as they have evolved as a result of the research activity that this collection of essays attests to.

1. The Approach My approach to speech acts is inspired by the philosophy of language of John L. Austin. The central core of my work is the reflection on the nature of illocution, accompanied by the investigation of how illocution works in naturally produced speech. Austin’s conception of illocution goes beyond the received idea that language serves various functions by identifying one kind of thing that is done in speaking, illocution, and distinguishing it from speech itself on the one hand, and the goals to which linguistic activity is directed on the other. It is also innovative because it does not perpetuate the usual opposition between speech acts performed in uttering declarative sentences and other speech acts, but instead aims to throw light on the fact that utterances of any kind may embody an illocution. I follow Austin in taking illocutionary acts to be “conventional.” By this, I mean above all that the effects they bring about are conventional: they are not the output of a chain of causes, which, once they are the case, cannot be “undone” (or made never to have been the case), but rather states of affairs that obtain depending on social or interpersonal agreement about the speaker’s performance and the norms

Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0001

2



governing it, and are annulled if the speaker’s performance is found to be defective in certain ways. This idea of the conventionality of illocution fits with Austin’s claim that speakers, in order to make their illocutionary acts “take effect,” must “secure” the audience’s uptake (1975, 116–17). I describe illocutionary effects in terms of attributions or cancellations of deontic properties to and from the participants in an interactional event. By “deontic” properties I mean such properties as having a right, being entitled or having the authority to do something, having to do something (because obliged or legitimately expected to do so, or committed to doing so), and being recognized as capable of doing something or knowledgeable about something. These properties belong to agents in virtue of social or interpersonal agreement. This kind of description of illocutionary effects represents participants in a linguistic exchange as having at any given time a certain set of deontic properties on the basis of which they perform speech acts and which is altered by the speech acts they perform. This is similar, on the one hand, to the “modal competence” of narrative semiotics (Greimas and Courtés 1979, 54) which each actant must be assigned at each step of a narrative and which changes as the narrative proceeds and, on the other, to David Lewis’s “scorekeeping” associated with each participant in a language game, which varies depending on the moves they make (Lewis 1979). My idea differs from both “modal competence” and “scorekeeping” because, while they take into consideration a mixture of deontic modalities and various other attitudes, I focus on the sub-set of deontic properties which are assigned to agents or withdrawn from them in a conventional and defeasible way (changes in attitudes such as desire or belief, instead, are matters of fact pertaining to individual or social psychology). The participants in an event of verbal interaction are usually assumed to be speaker and hearer (or audience), but according to my analysis, further distinctions must be introduced in order to account for the different roles that speakers and audience, or members of the audience, play in the dynamics of illocution. Speakers are responsible for the illocutionary effects of their utterances, but the illocutionary effect itself does not merely affect one target participant: it affects a relationship, since to any change in the deontic properties of one of the participants, there corresponds another change in the deontic properties of another (including the participant who is also responsible for the illocutionary act). Representing participants in an interactional event as endowed with deontic properties, and their sets of deontic properties as changing over time as an effect of the linguistic exchange, enables us to realize how illocution contributes to forming and modifying interpersonal and social relations, including power relations. I agree with the tradition (not only Austin but also Grice and Strawson) that the performance of an illocutionary act must be recognizable as such. But since I also take it that illocutionary effects are brought about thanks to social or interpersonal



3

agreement, I assign an important role to how an utterance is received and responded to. In ways and to an extent depending on the nature and structure of the interactional event in which an utterance is framed, the uptake of the relevant participants establishes, or contributes to establishing, what illocutionary act was performed. About infelicities, I maintain that attempts at performing illocutionary acts may fail if the circumstances are inappropriate or the execution is gravely flawed, while utterances achieve their illocutionary effect if the infelicity consists merely in the absence of the appropriate psychological states of the speaker. However, I do not take felicity to be a matter of complying with necessary and sufficient conditions: when it is clear enough what an illocutionary act an utterance is designed to perform, it is assumed by default, that is, unless there is reason to suspect that some circumstance is inappropriate or there is some flaw in the execution or that the speaker lacks the appropriate psychological states. Illocutionary acts, so conceived, should be classified according to the effects they produce or more precisely, according to the whole configuration of deontic properties in which their preconditions and their effects consist. Four out of five of Austin’s classes of illocutionary acts are suitable for being characterized in this way. So, in my theoretical considerations about illocution as well as in my use of illocution in the analysis of discourse or conversation, I highlight four main illocutionary types, corresponding to my re-definitions of the Austinian classes of Verdictives, Exercitives, Commissives, and Behabitives. In Austin’s class of Expositives, various configurations of deontic properties can be found, while the contents of those deontic properties concern matters related to the structure and development of discourse or conversation. So, it is possible to identify a group of Expositives within each of the four main classes of Verdictives, Exercitives, Commissives, and Behabitives. Speech acts are usually considered as a special, self-contained topic in pragmatics and not as a theme whose in-depth treatment can shed light on other issues. In fact, how speech acts are conceived is often influenced by the ways in which issues such as presupposition and common ground or, respectively, implicature and relevance-driven inferences, are dealt with. But my way with speech acts has implications for how to deal with issues concerning implicitness. If implicitness is viewed as a phenomenon concerning the speech act (at some level or other), it becomes natural to inquire about the functions it serves with respect to what the speaker is doing with their words. In particular, suggestions about when and why speech acts convey certain kinds of implicit content may come from the investigation of the dynamics of illocution. If not only illocutionary felicity but also the appropriateness of utterances and the participants’ cooperativity are assumed by default (that is, unless there are reasons not to do so), this may be used to account for both so-called presupposition accommodation and the working out of conversational implicature.

4



My main idea about presupposition is that it is misleading to give two distinct accounts of speaker presupposition and presupposition accommodation. As activated by a speech act (or also, by a text comprising various speech acts), presupposition should be recognized as one and the same phenomenon whether the presupposed content is already shared or not. This is made possible by considering presuppositions not as shared assumptions, but as assumptions which ought to be shared. The speech act requires the assumption that things are in a certain way to be true for its own appropriateness or felicity, irrespective of whether we accommodate that assumption or already know that things are so. This view of presupposition accounts for its informative and persuasive functions as well as for the cases in which the content that is linguistically marked as presupposed is already part of the common ground. Presuppositions encode content that may be new for at least some members of the audience in a syntactically compact and therefore economic way and help the speaker avoid the (face-threatening) implication that no one in the audience already possesses that information. Presuppositions make the addressee of a speech act feel they are a member of the group as soon as they share, or accept to accommodate, the presupposed assumptions. The fact that challenging presuppositions is costly, both cognitively and interactionally, explains why they are suitable for conveying content that the speaker does not want to be challenged and may be exploited as a powerful means of persuasion. As to implicature, I consider it from the point of view of the audience. I therefore focus on why it is legitimate to understand the speaker’s utterance as having a certain implicature and how the implicated content can be retrieved. Implicature makes additional or redressive information available to the audience. It backgrounds the speaker’s commitment to the content made available and involves the audience in the process of working it out, thus fostering a sort of complicity between speaker and audience. I distinguish presupposition from implicature by two main criteria inspired by the ways in which their contents are made explicit and by the functions they have with respect to the overall significance of a text. In the explicitation of an implicature, we infer a content that is not expressed on the text’s surface; in the explicitation of a presupposition, we reformulate and foreground content which is already present in the text in a backgrounded form. Likewise, a content cannot be (merely) made available by implicature when it is explicitly affirmed, while content linguistically marked as presupposed can be at the same time spelled out explicitly while still conserving the status of presupposed information.

2. The Essays In Essay 1, “On Illocutionary Types” (published in Journal of Pragmatics, 1984), I examine the classification of illocutionary acts outlined by John L. Austin and



5

elaborate upon it. I reconsider Austin’s four classes of Exercitives, Commissives, Verdictives, and Behabitives and characterize them as involving distinct types of conventional effect on the interactional relation. These types of effect are described in terms of assignment or cancellation of deontic predicates to or from the participants and to the purpose of this description, the participants are considered in terms of the “actants” they express, borrowing this notion from narrative semiotics. Illocution is thus presented not as a matter of a speaker achieving an effect on the audience, but of a Destinator assigning or withdrawing deontic properties to or from each one of two Destinees, thus affecting their relation. Shortly (and approximately) said, in Exercitives, Commissives, and Verdictives, the Destinator has a deontic property of the “can” kind (amounting to some kind of authority, recognized capacity, or recognized competence), while in Behabitives the Destinator fulfills the task of reacting to some action, event, or situation and has therefore a deontic property of the “ought” kind. The relationship between Destinee 1 and Destinee 2 is changed by Exercitives by creating or eliminating obligations and by Commissives by assigning a right or legitimate expectation to Destinee 1 and an obligation to Destinee 2. Both Verdictives and Behabitives entitle Destinee 1 to possess knowledge, but in the case of Verdictives, it is the knowledge formulated in the speech act and Destinee 2 is assigned the obligation to guarantee for it, while in the case of Behabitives, it is the Destinator’s reaction itself which becomes an object of knowledge for Destinee 1, while Destinee 2 is freed from previous obligations as regards Destinee 1, so that their relation can have a new start. Austin’s illocutionary classes overlap with one another and have fuzzy borders. I claim that this is not a weakness, but a strong point of his classification (or typology), since it enables us to tackle various complexities in the analysis of both illocutionary verbs and the illocutionary force of utterance tokens. Likewise, distinguishing illocutionary types in the way I propose is not pidgeon-holing existing practices, but outlining a flexible descriptive mechanism, capable to account for marginal and hybrid cases as well as prototypical ones. I conclude Essay 1 with an invitation to use the typology it presents as a point of departure for further analyses, cast in the form of the proposal of some speech-act theoretical “games” one might want to play with it: grouping types according to similarities, shifting a kind of illocutionary act or the illocutionary force of a token utterance from a type to another, reframing the illocutionary act performed in issuing a token utterance within a group of illocutionary types to which it would not be expected to belong. Essay 2, “Speech Acts, Effects, and Responses,” written as a reaction to John Searle’s paper “Conversation” and published in a collection discussing it (Searle et al. 1992, 7–29), examines the relationship between individual speech acts and conversational sequences. In it, I claim that instead of inquiring whether one can

6



extend speech act theory (with the received apparatus of felicity conditions) to the study of conversation, one should consider that the study of speech acts already presupposes some reference to conversational sequences. Reading Austin as claiming that the hearer’s uptake is necessary for an illocutionary act to take effect, I suggest that the response to a speech act, which manifests how the interlocutor understood it, contributes to letting that speech act take effect according to a certain illocutionary force. While admitting that there are cases in which a response that is inappropriate to the apparent force of a speech act can be dismissed as due to misunderstanding, I argue that in the course of informal conversations, the response that a speech act actually receives often reveals how it was taken and what illocutionary effect it brought about. So, in order to tell what an illocutionary act is performed in issuing an utterance, we should not consider that utterance in isolation, but frame it within the conversational sequence to which it belongs. In Essay 3, “Ideology and the Persuasive Use of Presupposition,” first presented at the 6th International Pragmatics Conference (1998), whose special topic was “Language and Ideology,” I show how presupposition is used to convey new information and even to achieve persuasion (especially in fields pertaining to ideology), by discussing examples drawn from political news in Italian daily newspapers. I challenge the received account of pragmatic presupposition as unable to yield a plausible explanation of these phenomena, and argue that presupposition and presupposition accommodation should not be given two separate accounts: instead, presupposition should be recognized as one and the same phenomenon, irrespective of whether the presupposed content is already shared or new. I propose to consider presuppositions not as shared assumptions, but as assumptions which ought to be shared. This normative feature helps account for their informative and persuasive uses. Against the risks of uncritical persuasion, I also claim, the first defense lies in the ability to make presuppositions explicit, turning them into assertions or assessments that, being openly performed, do not escape conscious control and are liable to criticism. Essay 4, “Intentions from the Other Side,” read at the Conference “Paul Grice’s Heritage” (S. Marino, International Center for Semiotic and Cognitive Studies, 1998), revisits Paul Grice’s philosophy of language. It examines three Gricean ideas: non-natural meaning, the Cooperative Principle, and the notion of a person that Grice puts forward in the context of his theory of value. My approach to those ideas does not focus on the speaker’s mind and its internal states, but on the audience and how they should take the speaker’s utterances. Viewing non-natural meaning “from the other side” (as the title of Essay 4 reads) amounts to reformulating its Gricean definition as spelling out what intentions the audience should attribute to a speaker when considering that speaker as non-naturally meaning something. Likewise, the Cooperative Principle and its maxims are no rules that must be actually followed by speakers, but assumptions in the light of which the



7

audience can best understand speakers and which speakers may exploit to make themselves understood. So, some traditional objections to Grice’s definition of non-natural meaning turn out to be harmless and the use of the Cooperative Principle in the analysis of discourse is fully justified. As to Grice’s notion of a person as the essentially rational being (1991), such an essence of personhood is an object of self-attribution by human beings and its borders are fuzzy: as one would expect from an essence that is (so to speak) essentially attributed, there will always be borderline cases, in which it is unclear whether the attribution is due or not. I take this as confirming my intuition that there is an ethical burden implicit in the recognition of others as subjects. In Essay 5, “Presupposition, Implicature, and Context in Text Understanding,” read at a Conference of the international interdisciplinary network “Context” held in Trento in 1999, I sketch an overview of the field of implicitness which has its roots in my Austin-inspired speech-act theoretical perspective. I examine the roles that presupposition and implicature play in the process of text understanding. Context is part of the picture in two ways: as the objective and mind-transcendent context to which linguistically indicated presuppositions have to conform (and are by default taken as conforming to) and as the representation of context whose construction and update are part of the process of text understanding. I describe presupposition and implicature as two different ways in which changes in the representation of the context are induced, both different from the way in which assertion changes it. On the basis of this description, I claim that there are good reasons to consider presupposition and implicature as distinct linguisticpragmatic phenomena. Indeed, a linguistically activated presupposition remains such even if its content has already been asserted, while the previous assertion of the same context makes an implicature useless and therefore prevents it from arising. Essay 6, “Illocutionary Force and Degrees of Strength in Language Use” was also written in 1999, with the intent of drawing my own conclusions—after conducting further research on my own into speech acts in conversation—from my previous collaboration with Claudia Caffi and Carla Bazzanella on mitigation and reinforcement (see Bazzanella, Caffi, and Sbisà 1991). I felt uneasy about some of the results of that joint research, especially as to its taking “propositions” as one of the aspects of the speech act whose modifications contribute to variations in what we called the “degree of strength” of illocutionary force (following Searle’s terminology: Searle 1979, 5; Searle and Vanderveken 1985). The main idea in Essay 6 is that those variations affect aspects of the illocutionary effect and of its preconditions, both of which consist in deontic properties of the participants. Analysts do not face two phenomena, illocutionary force (or illocutionary point) on the one hand, and mitigation/reinforcement on the other, but have only the (multi-faceted) task of describing the dynamics of illocution in its shades (both qualitative and quantitative) and its relations to the locutionary and

8



perlocutionary aspects of the speech act. In support of the claim that mitigation and reinforcement are adjustments of the illocutionary effects of utterances, I analyze examples from recorded conversations in Italian. The paper was submitted to the Journal of Pragmatics and was accepted after lengthy and instructive discussions. Essay 7, “Speech Acts in Context,” written in 2000, examines the notion of context and its uses in speech act theory. I claim that the requirements for the felicity of performative utterances discussed by Austin have to be satisfied by how things are objectively, while Searle’s felicity conditions are mostly requirements on the speaker’s intentions and other attitudes. I then argue in favor of a view of the context of a speech act as constructed (therefore, potentially dynamic) as opposed to merely given, limited as opposed to indefinitely extendable, and (insofar as the speech act has to be assessed against it) objective as opposed to cognitive. I also reaffirm the context-changing role of speech acts and the difference between the illocutionary and perlocutionary dimensions of the changes they bring about. In Essay 8, “Cognition and Narrativity in Speech Act Sequences,” read at the 7th International Pragmatics Conference (2000), I examine the interactional dynamics of speech acts, making an extensive use of examples from natural conversations and paying particular attention to speech act sequences. I develop some of the insights about the hearer’s uptake that were already present in Essays 1 and 2, arguing that what is constitutive of the audience’s uptake (and therefore contributes to making the speaker’s utterance illocutionarily effective) are the audience’s verbal and behavioral responses as opposed to their cognitive representation of what the speaker meant. I then show how the analysis of action sequences put forward by narrative semiotics, which envisages a three-place structure (the narrative schema) consisting of a manipulating move, an action, and a sanctioning move, can be used to obtain a better understanding of what is going on in a speech act sequence, shedding light on the sequential relations between speech acts. I also demonstrate that the narrative schema can be applied to the same conversational sequences in different positions, revealing a variety of possible sequence-depending aspects for each conversational contribution. In Essay 9, “Two Conceptions of Rationality in Grice’s Theory of Implicature,” read at the international conference “Rationality in Belief and Action” held in Rijeka in 2004, I take another look at Grice’s philosophy of language. I explore why conversational implicature is considered by Grice, as well as by post-Gricean and neo-Gricean scholars, as something “rational” and show that two distinct conceptions of rationality, “instrumental” and “argumentative,” co-exist in Grice’s thought. The “instrumental” rationality of conversational implicature is based on the efficacy of the Cooperative Principle in maximizing understanding: conversational implicature is rational in this sense because conversation is more efficient if the Cooperative Principle is assumed to hold. “Argumentative” rationality, on the other hand, is a matter of having reasons for what one says or does: if we focus on



9

it, conversational implicature is rational because it is calculable, that is, retrievable by reasoning in accordance with a certain argumentative path. While post- and neo-Gricean theories (Sperber and Wilson 1995; Levinson 2000) focus on instrumental rationality, Grice’s posthumous writings on value and reason (1991, 2001) pay attention to argumentative rationality. I argue that conversational implicature is best understood in the light of the argumentative conception of rationality. Essay 10, “How to Read Austin,” originally read as a plenary lecture at the 9th International Pragmatics Conference (2005), special topic “Pragmatics and Philosophy,” reassesses the contributions to pragmatics made by John L. Austin’s How to Do Things with Words. I discuss some of the assumptions made by the received readings of that volume, concerning its aim and structure, the role of the performative/constative dichotomy, the conceptions of illocution and perlocution, and the alleged exclusion of non-seriousness. I argue that the volume is structured as a proof by contradiction of the claim that all speech should be considered as action, that illocution has conventional effects, that perlocution presupposes a conception of action as responsibility, and that Austin had reasons to avoid dealing with non-seriousness on a par with illocution and perlocution. As to the conventionality of illocution in particular, I claim that illocutionary effects are conventional insofar as they can be annulled, making the illocutionary act “undone” (which is something that cannot happen to acts that bring about their effects by natural causation). Essay 11, “Uptake and Conventionality in Illocution,” written in 2008 and published in a speech-act theoretical issue of Lodz Papers in Pragmatics, tackles the issue of the conventionality of illocutionary acts in its relation to the “securing of uptake.” While the indispensability of the latter is widely accepted as a hallmark of illocution, it has also been taken as evidence of the intention-based nature of illocutionary acts as opposed to their conventionality. I question the reading of the securing of uptake offered by Strawson (1964), who considers the audience’s uptake as the only effect proper to all illocutionary acts. Strawson, I claim, is right in seeing a connection between the alleged conventionality of performatives (and illocution) and the indispensability of the securing of uptake, but incorrectly identifies the kind of connection. For most (informal) kinds of illocutionary act, he explains conventionality away as the expression, on the part of the speaker, of Gricean meaning intentions that must be made recognizable to the hearer. In my reading of Austin, the securing of uptake is a matter of enabling the hearer to recognize the procedure that the speaker is executing, and its indispensability is the hallmark of the conventionality of illocution understood as the illocutionary act’s “taking effect” (Austin 1975, 117) thanks to the participants’ agreement on what is being done. In Essay 12, “Illocution and Silencing,” written in 2008 as a contribution to a Festschrift for Jacob Mey (the founder of the Journal of Pragmatics), I consider the problem of how to account for a person’s being “silenced” as it emerges from the

10



debate on pornography and speech acts initiated by MacKinnon (1987) and Langton (1993). I consider the case of the ineffective refusal of sexual advances, distinguishing various ways of explaining what goes wrong (perlocutionary failure, illocutionary disablement, misfire, lack of uptake, epistemic injustice) and arguing that none of them provides the complete picture of what is going on. I propose to analyze sexual refusal as an exercitive illocutionary act and its silencing as a denial of the woman’s authority over her own body and her entitlement to autonomous decision. Essay 13, “The Austinian Conception of Illocution and its Implications for Value Judgments and Social Ontology,” started out as a contribution to a joint seminar of the University of Trieste and the University of Bamberg, Germany, on “Moral Realism and Political Decisions,” held in December 2013. My main aim in this paper is to explain how to deal with value judgments and social ontology in my Austin-inspired perspective. After providing a brief account of Austin’s speech act theory as I read it and some examples of conventional illocutionary effects, I discuss the implications of Austin’s classification of illocutionary acts for value judgments, arguing that Austin challenges the fact/value dichotomy by assimilating statements of fact to value judgments as both members of the class of Verdictives. Moreover, I argue that illocution gives us an easy path into social ontology: indeed, many social realities can be described in terms of states of affairs consisting of the distribution of deontic properties among the social agents involved and such states of affairs share the same nature as the effects of illocutionary acts. When participants have certain deontic properties, for example a right or an obligation, that right or that obligation, albeit not by itself institutions, can be dealt with as socially constituted objects. The deontic properties of social agents are the material out of which institutions are made, and they come into being thanks to illocutionary acts. So, in the last resort, it is illocution that serves as the basic source of social ontology and institutions. In Essay 14, “Varieties of Speech Act Norms,” written in 2016 as a contribution to a volume on speech acts and their normativity edited by Maciej Witek and Iwona Witzak-Plisiecka, I explore various kinds of norms governing speech acts and their roles in the dynamics of illocution. I distinguish constitutive rules (upon which the performance of illocutionary acts depends), maxims (based on rational motivations and encoding regulative advice for optimal speech act performance in the perspective of the participants), and objective requirements for the overall correctness of the accomplished speech act. In so doing, I represent the performance of an illocutionary act as the execution of a procedure, including an initial situation, the verbal or non-verbal behavior which is expected on the part of the performer, and the designed outcome: speech act norms specify parts of a procedure or other features of the situation associated with it. The proposal to consider rules of accommodation as a further kind of speech act norm is discussed and rejected: procedures are structured wholes, and any part of a procedure that



11

suffices to make the whole recognizable invites the by-default assumption that the other parts are instantiated as well. I also argue that, among speech act norms, only the rules I here call “constitutive” are conventional. All speech acts are, therefore, conventional for certain aspects and non-conventional for others. Essay 15 was prepared as an invited contribution to the Conference “Methods of Philosophy,” organized by the University Vita–Salute S. Raffaele (Milan) in 2017. The paper illustrates my understanding of some philosophical ways of dealing with gender, harking back to my own experience of feminism. I argue for the philosophical relevance of critiques of gender roles made by means of the analysis of language and speech. I reflect on the oscillations from the emancipatory perspective to the positive assessment of gender difference and then back (due to worries with essentialism) to the (mere?) quest for equal rights and opportunities. I consider the role of generic statements about the characteristics of genders in fostering essentialism on the one hand, and producing an overrating of the normativity of gender concepts on the other. I argue that people’s subjectivity should be recognized by other people (including point of view, attitudes, and rights) in an unconditional way; that is, with regard to gender, whether they fit a given gender model or not. By the way, I do not think that agreeing to be called “he” or “she” should be equated to agreeing to fit the leading model (or any fixed model, for that matter) of the masculine or feminine gender. Unconditional recognition of the other’s subjectivity has the power to deconstruct the normativity of gender concepts without falling back into homologation (that is, denial of gender- or sex-related differences), which would also be a constraint on identity. Essay 16, “Assertion among the Speech Acts,” is a contribution to the Oxford Handbook of Assertion (Goldberg ed. 2020). In it, I discuss how assertion is collocated among the other speech acts, starting from the assumption that, in speech-act theoretical terms, assertion is an illocutionary act. I examine how assertion relates to other illocutionary acts involving the utterance of plain declarative sentences and how it should be collocated within the whole gamut of illocutionary acts. While the former exploration relies upon an intuitive grasp of the family of assertive illocutionary acts, the latter requires a more complete characterization of assertion in the framework of illocutionary act classification. Using Austin’s terms, I describe assertion as an expositive Verdictive: an act involving judgment and allowing for the transfer of knowledge, which affects discursive and conversational relations. I also argue that from an Austin-inspired perspective there is no good reason to grant assertion a special place among illocutionary acts. The challenge lies, rather, in putting assertion on a par with other kinds of illocutionary act, making the analysis of these throw new light upon assertion and the other way around. The contents of Essay 17, “Illocution and Power Imbalance,” came together gradually and took various forms across a number of lectures and conference papers read in Italian and English between 2015 and 2019, until the paper was

12



published in Italian in 2020. My commitment to admit of Exercitives as one of the main types of illocutionary act dates back to the beginnings of my research on speech acts (see Essay 1) and characterizes my use of illocution in discourse analysis (see, e.g., Essay 12 and 2006c).¹ In Essay 17, I describe Exercitives as consisting of the exercise of authority or prestige and affecting the rights and obligations of the participants. I focus attention on examples of Exercitives from informal speech situations, showing how the exercitive force of an utterance is indicated and what effects it may have on the degree of power of the participants with respect to each other. Indeed, studying Exercitives involves studying kinds and degrees of that power which is granted to agents by social or interpersonal agreement (especially authority, formal and informal), and the ways in which agents manage or fail to get it, including “back-door” strategies (Langton 2018a) that involve the accommodation of some kind of authority as the initial condition of a tacitly accepted exercitive procedure. In this connection, I also discuss a case in which the characteristic effects of Exercitives are achieved in a back-door fashion and the audience’s responses attempt to block them. I conclude by arguing that the problem with Exercitives is not so much that they introduce power imbalance, but that they can be exploited to make it stable and mono-directional.

3. Some Background When I first read Austin, it was 1969 and I was reading philosophy at university in my home town, Trieste. I was looking for recent philosophical work which might help me understand the human condition at the time—an evergreen aim for a philosopher, I dare say. I turned to Austin after considering pragmatism, the Frankfurt school, some phenomenology, and various authors in the analytic and neo-positivist traditions from Wittgenstein to Ayer and Hare. Wittgenstein’s Tractatus (1922), translated into Italian in 1964 by Amedeo G. Conte, struck me with its Silence. I found neo-positivism irritating, its verificationism unconvincing, and its meta-ethical emotivism upsetting. I found no real way out in metaethical prescriptivism. There was interest for Wittgenstein in Italy in those years, but without being aware of any particular reason for this choice, I did not feel like concentrating on his approach (or approaches: it was not clear how many Wittgensteins there had to be); I was to return to Wittgenstein for about a decade after 1973. In the Italian environment, there could also be heard echoes of Ordinary Language Philosophy, particularly thanks to Ferruccio Rossi-Landi (1961) (who later turned to a peculiar form of Marxist semiotics) and Renzo

¹ References specified only by the year of publication indicate works of mine. Full references to them can be found under my name in the final list of References.



13

Piovesan (a highly perceptive thinker and teacher, based in Padova, who published too little but gave the most in personal discussion). Austin offered me a way of being philosophically concerned with language that did not seem to presuppose any metaphysics (thus making me feel I was complying with Wittgenstein’s Silence) and was promising in at least two directions: finding a way out of the fact/value dichotomy, and working out an approach to language and its uses with good chances to prove revealing about human relations and the cultural and ideological frameworks surrounding them. In the former direction, I started reflecting on the classification of illocutionary acts, which according to Austin himself, should be a key to deconstructing the fact/value dichotomy. But soon, the issue of illocutionary act classification became an object of research for me in its own right (1972). In my subsequent work, while never forgetting about the fact/value problem, I did not return to it in a systematic way (my regret about this is expressed in 2012b; but see also Essay 13). In the latter direction, I soon realized that the idea of speech as action as outlined by Austin offered some interesting potential for social and cultural critique. It was summer 1975 and I had just obtained my first temporary research position at the University of Trieste, when I was invited to contribute to a local left-wing Festival by giving a talk on the language of women’s magazines. I began to examine a sample of women’s magazines through the lens of speech act theory, supplemented with the notion of presupposition. The results of that research took the form of a paper (1976), which was my first experience of speech act-oriented discourse analysis. On that occasion, I also started to realize that in the study of speech acts and more generally in pragmatics (which was then a new and rapidly expanding research field), there has to be complementarity between theoretical reflection and empirical exploration of natural speech in its various forms (oral, written, or whatever they might be). Since then, I have expressed this conviction on many occasions. I thought of the philosophy of language not so much as a set of theoretical claims, but as a methodological framework for the analysis of actual speech (or “text” in the sense of Hjelmslev 1961) (see 1989a, 10). Those were the beginnings of two trends in my research that are present and intertwined also in this volume: theoretical interest in speech act theory and fascination with (or commitment to) the task of analyzing discourse with the tools of philosophical pragmatics. It should be pointed out, however, that I also had other interests and activities in those early years, all of which contributed to the formation of my views. I spent the autumn of 1972 in Oxford, collaborating with J. O. Urmson on the second revised edition of J. L. Austin’s How to Do Things with Words, then published as Austin (1975). While in Oxford, I received precious advice from Brian McGuinness and Anthony Kenny about the short introduction to Wittgenstein I was planning to write for an Italian publisher (1975). As regards Wittgenstein, I continued to pay attention to certain themes of his philosophy

14



until the mid-1980s, forming the conviction that the rejection of metalinguistic explanations as never-ending and therefore ultimately useless played an important role in his philosophical development and in the ways in which he attempted to practice philosophy in a non-metaphysical style (1980, 1981, 1983b, 1984d). Between 1974 and 1978, I collaborated with Paolo Leonardi on presupposition (Leonardi and Sbisà 1977, 1978). After studying with Piovesan in Padova, Leonardi got a grant that enabled him to visit Berkeley and was therefore well aware of the new trends at the interface between philosophy and linguistics. Exchanges of ideas with Leonardi continued after that collaboration, even across the differences between our respective paths in philosophy. From 1976 onwards, my curiosity about the world of Italian and French semiotics (Umberto Eco and Paolo Fabbri in Bologna; in Paris, Algirdas Greimas and his research group, in which Fabbri participated) led me to explore whether semiotics could offer some way of dealing with meaning other than the analytic orthodoxy of compositional truth-conditions, of which I was wary. To be honest, I did not find anything convincing enough to satisfy my quest. Nevertheless, I found a lot of interesting notions and methods for the analysis of texts and (given the focus of Greimas’s work on narrativity) of actions too. I collaborated with Fabbri on a comparison between Searle’s and Austin’s ways of dealing with speech acts on the one hand, and models of social interaction of a structuralist and, respectively, interactionist kind on the other (Sbisà and Fabbri 1980). That collaboration influenced me under various respects. Not only did Fabbri make me appreciate various insights into language, action, and human interaction coming from semiotics and from the sociology of interaction revisited with a “semiotic eye” (see Fabbri 1973), but in my discussions with him, I had the opportunity to make up my mind about the tendency to detach the notion of an agent (or subject or self) from that of an individual human being and its pros and cons. Last but not least, the late 1970s were the heyday of the feminist movement. I participated in feminist discussion groups from 1974 onwards. Childbirth, which I experienced in 1973 and again in 1974, appeared to me as having the potential to unmask a number of mystifications about women: at least, that was its effect on me. To substantiate my intuitions, of course, I read feminist thinkers, with particular sympathy for Irigaray (1974). I then started a line of research on discourse on childbirth and the maternal stereotype, which led to the publication of a booklet centered on the analysis of implicit communication about gender and motherhood in the vulgarization of medical advice on pregnancy and birth (1984c), followed by other studies and publications on neighboring subjects later on (1988, 1992d, 1996, 1999d, 2017b). In the 1980s, my main job was reordering and completing my views on speech acts. I wrote Essay 1 and various other papers in Italian, English, or French (1983a, 1984b, 1986a, 1987a, 1987b) and then the volume Linguaggio,



15

ragione, interazione. Per una teoria pragmatica degli atti linguistici (“Language, Reason, Interaction. Toward a Pragmatic Speech Act Theory”) (1989a). That volume argues for a speech-act theoretical framework in which illocutionary acts produce conventional effects (thus developing the outline already sketched out in Essay 1), taking up ideas from the different fields, authors, and themes I had experience of, including: Austin, Wittgenstein, speech act theory, context and context change, narrative semiotics, and Goffman’s sociology of interaction. I remember I proofread it in hospital, where I was striving to delay the birth of my fourth child as much as possible, who was about to be born prematurely. Soon after 1989, I considered and then gave up the project of translating my book into English. I am grateful to Anthony Kenny for pointing out to me that some international publishers might be willing to publish its translation. But I felt unsure to what extent my remarks and arguments would have survived the filter of translation, I knew I would have to check the translation word by word to avoid any misunderstandings, and above all I was unwilling to spend any more time on that text: I really wanted to move on. Moreover, I thought that my views needed to be tested against analyses of naturally produced speech, particularly conversation. Only seeing them put to work in that way would make me feel confident enough about their soundness. I therefore expended my efforts in that direction, and it was not until later, from around 2000 onwards, that I felt like resuming, reformulating, discussing more thoroughly, or further developing some of the ideas contained in my volume on speech act theory (1989a) in papers written in English (see Essays 7, 8, 10, and 11). In 1990–1, I collaborated with two Italian linguists working on pragmatic issues, Claudia Caffi and Carla Bazzanella, on the “mitigation” and “reinforcement” of speech acts. The results of the collaboration were published as Bazzanella, Caffi, and Sbisà (1991). Writing a co-authored paper required negotiation and compromise, but at any rate I continued to reflect upon mitigation and reinforcement and in 1999 this eventually led me to write Essay 6. The collaboration with Caffi and Bazzanella gave great support to my project to test my speech-act theoretical framework against analyses of samples of natural conversation. In various papers, often making reference to examples, I investigated the deontic aspect of illocution, the relationship of illocution to affect, the role of the hearers’ uptake and of the utterance’s sequential position, and the characteristics of particular speech act kinds (Essays 2 and 8; 1987b, 1987c, 1989b, 1990, 1992b, 1992c, 1994, 1995b). In the late 1980s, I also set out to resume a line of enquiry into implicitness, whose beginnings go back to the observation (made while studying presupposition in the 1970s) that presuppositions in natural speech are too often new information for them to be plausibly defined as shared assumptions (1979). I made this decision when one of my kids, aged 12, asked me to explain a sentence in his history textbook which he could not understand: in complying with his

16



request, I realized I was making explicit a great deal of content linguistically marked as presupposed. Hence the idea that an awareness of implicit content and how to make it explicit should be part of teachers’ toolkit (first expressed in Sbisà and Rodari 1989 and later developed in 1991, 1999c, 2007b, 2007c; Sbisà and Regina 2003). Most of the products of this line of research were published in Italian, in order to be accessible to Italian teachers and communicators. But I was also interested in theoretical accounts of the persuasive use of presupposition (that is, why presuppositions tend to be accommodated, often with long-lasting persuasive effects), the difference I felt to exist between presupposition and implicature, and the conditions for attributing an utterance with an implicature (Essays 3, 4, 5, 9; see also 2007c, 2021a). As to implicature, it should be pointed out that while well aware of the importance of Grice’s philosophy for philosophical pragmatics (I was the first to bring a xerox copy of his 1967 lectures on “Logic and Conversation” to Italy in 1972), I found his influence on speech act theory highly problematic, especially for its repercussions on the conception of illocution. It is not by chance that my re-reading of Grice and my use of some of his categories are audience-centered rather than speaker-centered (Essay 4). I found myself sympathetic with Jennifer Saul’s (2002a) idea that implicature is meaning made available by the speaker to the audience, which I endorse in Essay 9. My research on implicitness was accompanied by the critical examination of the notion of context, in the course of which I benefited from e-mail correspondence and personal discussion with Christopher Gauker and Carlo Penco. I found Gauker’s work on presupposition and context (1997, 1998) particularly illuminating. I knew that in the work I was doing on the subject, there was something basically foreign to mainstream views of presupposition. But I was not well aware of what was at stake. Gauker’s clear-cut distinction between context as a set of assumptions that the participants in a conversation share and assume to be shared, and context as the set of sentences which, if asserted, would objectively help the participants achieve the goals of their conversation, made me realize that my account of the informative and persuasive potential of presupposition had to be connected to the objective, rather than cognitive, nature of the relevant context. The normative character of presupposition depends, indeed, on the fact that linguistic form and force indicators pose requirements on the appropriateness of utterances, which have to be satisfied objectively (so that taking an utterance to be appropriate commits the audience to grant that its context actually comprises certain features). The connection between the felicitous performance of speech acts and the objective character of the context of utterance against which they are assessed is also emphasized in Essay 7. That paper was read by R. M. Harnish (2009) as giving an “externalist” picture of speech acts as opposed to the “internalist” picture provided by the theory he endorsed (Bach and Harnish 1979). He argued that an internalist framework is richer and more comprehensive than an externalist one. I had an opportunity to reply to Harnish in Essay 11, but in fact



17

did so only indirectly, since I felt it a priority to explain why and how my speech-act theoretical framework differs from the theories that accept the Gricean reading of Austin offered by Strawson (1964). It was in 2002 or shortly afterwards that I started realizing I should go back to the philosophical inputs at the basis of my elaborations and analyses, especially J. L. Austin’s philosophy of language. The role of keynote speaker at the 9th International Pragmatics Conference (2005) gave me the opportunity to make a public statement of what I thought to be the central contributions offered by How to Do Things with Words (Austin 1975) to pragmatics and philosophy (Essay 10). In the following years, I explored again and again what Austin said or implied about speech acts and related topics, such as action, context, truth, meaning, or “propositions” (Essays 11, 13, 16; see also 2006a, 2011, 2012a, 2012b, 2014c, 2015). In 2006, I prepared the project of a volume on Austin’s philosophy extending far beyond the theme of speech acts, but the real work on the volume (now still in preparation) only began in 2017. At any rate, in the two decades following 2000, I also kept on discussing and updating my speech-act theoretical framework and my views on implicitness (Essays 11, 12, 14, 16, 17; see also 2007c, 2021a, 2021b) and using them in the analysis of discourse (2003, 2006c, 2014a, 2017a; Sbisà and Vascotto 2007). I was also engaged in a collaboration on the analysis of so-called “knowledge dissemination” through the web and especially social media (Labinaz and Sbisà 2020, 2021a, 2021b): here, too, what I write about assertion as a speech act from the theoretical point of view (e.g., Essay 16) finds its counterpart in the analysis of naturally produced discourse.

4. Where All This Is Heading To My overall project can be described as the construction of a non-metaphysical theoretical framework for the study of speech as action. By “non-metaphysical” I do not mean naturalistic (naturalism itself may be metaphysics in its own way). I mean that my theoretical framework admits of the context-bound and intersubjectively based character of our use of language, tries to work on speech as action from within these limits, and aims to provide tools for a better understanding of what the texts we produce mean and do and why they mean and do just that. I went some way in that direction: how far, it is up to the reader to say.² For example, I realize that my treatment of classic issues such as the conventionality of language and of illocution is far from exhaustive. At any rate, the development of my views in the years through the essays republished in this collection and other research work has led to some clarifications and refinements with respect to my approach as introduced in Section 1. Concluding this introduction, I mention briefly some areas to which these clarifications or refinements are relevant. ² Precious feedback to my work is provided by Caponetto and Labinaz eds. (forthcoming).

18

 (i) Speech-act theoretical terminology. In various essays republished here, “speech act” is used as a synonym for “illocutionary act,” notwithstanding my conviction that these expressions should be kept distinct. Each (total) speech act (meaning by this what is done by a speaker who utters certain words on a certain occasion, considered in its entirety) comprises an illocutionary act, that is, can be described as an illocutionary act (in consideration of its being designed for, or succeeding in, achieving a certain illocutionary effect); rather obviously, however, it comprises also a locutionary act, or else it would not be an act of speech. The fact that acts of these two kinds at least (not to consider perlocutionary ones) regularly co-exist does not justify using “speech act” to mean “illocutionary act” or vice versa. Instead, that fact gives sense to the use of “in” made by Austin when he calls the illocutionary act an act performed “in saying something”: it is not an act of speech, but an act that inhabits speech. Moreover, “speech act” may have loose uses. For example, we may refer to an utterance of a certain speaker on a certain occasion as their “speech act,” even before examining whether it was really grammatical and meaningful, or really felicitous and illocutionarily effective, or “seriously” issued as opposed to being reframed as part of some “aetiolated” performance. We may say that lying is a kind of speech act without implying that it is a kind of illocutionary act, and call shaming a speech act, meaning by that a kind of speech event in which the utterer is responsible for shaming effects (to be further analyzed at the locutionary, illocutionary, and perlocutionary levels). Writing on my views of speech acts anew, I would be more rigorous in the use of those phrases. A second point of which I became aware gradually is the misleading character of the expression “counts as” when applied to utterances in the making of which the speaker performs an illocutionary act. “Counts as” has a reductionist flavor, tracing a “conventional” action back to a description of the allegedly basic action. This implies that the conventional action does not amount to a real action distinct from the basic one. Instead, in my view, both the locution and the illocution are real actions: the speaker is legitimately held responsible for both. So, recently I have given up the use of “counts as” in this context. A third point is that Austin’s notion of “procedure” is rarely used in speech act theory, but, when I realized it has a remarkable explanatory potential, I introduced the term “procedure” into my lexicon (see Essays 14, 16, and 17). By “procedure” I mean a complex script, which may have a higher or lower number of options at each step, such that whoever executes its steps completely enough in appropriate circumstances should be ceteris paribus recognized as performing the corresponding illocutionary act. The execution of a given procedure may follow its steps more or less closely, and a performance may be hybrid or marginal with respect to



19

the socially accepted repertoire of procedures. The advantages of introducing such a notion of procedure are that it is by far more flexible than the received apparatus of felicity conditions, it may be appealed to not only when illocutionary indicators give clear indication of the procedure invoked but also in cases of vagueness, ambiguity, hybridity, marginality, and interactional negotiation, and it may be used (as I claim in Essay 14) to account for accommodation phenomena. (ii) The securing of uptake. What the “securing of uptake” should consist in and which is its function in the dynamics of illocution are still hot topics in speech act theory. As to the former question, I think that it is primarily the speaker who has to indicate what procedure they are invoking clearly enough for the audience to recognize it. But for a speaker to invoke a procedure may amount to behaving in the way envisaged by that procedure: indeed, it is on the ground of the speaker’s verbal and non-verbal behavior that the audience legitimately takes the speaker’s utterance as the performance of a certain illocutionary act. Moreover, how the audience understands a speech act is often not a matter of interpretation or explicit assessment, but a matter of responding to the speech act in a way appropriate to speech acts of a certain illocutionary force. When, especially in informal interaction, conversational contributions give merely vague or ambiguous indications of their force and leave room for the interlocutor to decide about how to proceed, uptake can be deemed to be “secured” insofar as it actually occurs, that is, insofar as the interlocutor’s response indicates that the speaker’s utterance was taken as having an illocutionary effect reasonably compatible with the speaker’s behavior. As to the latter question, I think that the audience’s uptake, explicit or tacit, by default or on reasons, is part of that interpersonal agreement (explicit or tacit, by default or negotiated) which makes utterances illocutionarily effective. In the case of disagreement (recently discussed, e.g., by McDonald 2020), I do not think there is one solution only. We should take into consideration not merely the uptake made manifest by the interlocutor’s response, but also the speaker’s reaction to that response, by which they might (after all) show acceptance of the uptake obtained or attempt to correct it by reformulating their speech act. Moreover, there are many more possible shades and combinations of circumstances which are worth examining. For example, there are situations in which one of the participants (addressee, speaker, or observer) is granted greater conversational power, so that their response determines what the speaker did. There are situations, extended in time or suitably reframed, in which participants who are observers or bystanders with respect to a certain speech act may reject and correct the way in which its original audience took it. At any rate, in no case would I locate the “securing of uptake”

20

 among the felicity conditions of an illocutionary act, since felicity conditions, like procedures, are specific for each kind of illocutionary act, while the securing of uptake is a general requirement and uptake itself is a phase in illocutionary dynamics. (iii) Varieties of implicitness. I still distinguish presupposition from implicature by means of criteria inspired by the ways in which their contents are made explicit and the functions they serve: in the explicitation of an implicature, we infer a content that is not expressed on the text’s surface; in the explicitation of a presupposition, we reformulate and foreground content which is already present in the text in a linguistically backgrounded form. Likewise, a content cannot be both explicitly affirmed and made available by implicature, while content linguistically marked as presupposed can conserve that status when explicitly formulated. Presupposition (when not already shared) calls for accommodation, may transmit new content as taken for granted or give taken-for-granted status to old content, and fosters in-group participation. Implicature makes its content available to the audience without the speaker being explicitly committed to its truth, creates shades of meaning, and fosters the co-responsibility of audience and speaker. However, I am no longer willing to consider Gricean conventional implicatures as proper cases of implicature (as I did in Essay 5): a more promising approach to them would be (as I propose in 2021a, 183–4) to redefine them as preconditions of speech acts with Expositive force (what Grice calls “non-central speech acts”: Grice 1989, 121–2; see Austin 1975, 152, 161–3). In this case, the field of implicitness would comprise, on the one hand, conversational implicature (with its varieties ranging from generalized to context-specific implicature and from implicature preserving the observance of the maxims, to flouting implicature) and, on the other hand, presupposition, which in turn should be divided at least into force-related presupposition, appropriateness-related presupposition, and content that, while expressed on the text’s surface, is syntactically marked as not-at-issue (2021a, 182–3). How to best organize the field of implicitness is, at any rate, not only a matter of linguistic or logical theory but also a matter of refining our awareness and mastery of explicitation practices. (iv) My “externalism.” Harnish (2009) dubs my approach to speech acts “externalist.” In fact, I maintain that, while it may be useful to consider the speaker’s (expected) internal states when characterizing a kind of illocutionary act, the attribution of an internal state to the speaker on a particular occasion cannot determine what the speaker is doing, since it is itself made on the basis of “external” facts such as what words the speaker utters in what circumstances. While Harnish maintains that an internalist framework is richer and more comprehensive than an externalist one,



21

I think that focusing on internal states and processes might even hinder the researcher from paying attention to intersubjective dynamics and its potential for changing states of affairs in the world. In my approach to speech as action, attributions of intentions to speakers are part of the phenomena to be accounted for and cannot be invoked as grounds for attributing locutionary and illocutionary agency or implicit meaning: in order to use them in this way, we would need access to speakers’ intentions regardless of what words they utter and how they behave. The firstperson perspective is no way-out, because there is no viable path from subjectivity to intersubjectivity (as Avramides 2001 has argued). Instead, both speech and action are located in the intersubjectively accessible world, and in it, also acknowledgments of each other’s subjectivity among human agents take place. Likewise, I do not resort to the participants’ actual beliefs to help determine what is implicit in a speech act performed on a certain occasion, unless the participants make their beliefs manifest in some relevant way. As regards implicitness, too, the starting point must be what participants say and do in what circumstances. My approach thus strives to keep theory firmly rooted in the intersubjective world. (v) The underlying philosophy of action. My view of speech acts calls for completion as to the notion of action to be used in it. Since the early 1980s, I have been aware that the conception of illocution and of its relations with locution, perlocution, the (total) speech act, and aetiolation, depends, in part at least, on how action is understood and defined. But, after some explorations (1987a, 1989a) inspired on the one hand by narrative semiotics and, on the other, by the philosophy of action of G. H. von Wright, I did not tackle the task of specifying the conception of action that matches my views on speech acts until recently, when I set out to examine the connections between Austin’s outline of speech act theory and his philosophy of action (Essay 10, 2013a, 2014c; see also in preparation). Austin’s view of action, and mine too, are very far from the standard view inspired by Davidson (1980), according to which action is behavior appropriately caused by an intention (for criticism of this kind of view, see Hornsby 2004, 2010). I associate action with (legitimate) ascription of responsibility, rather than (actual) presence of intention. I admit that action is connected to causation, but do not see the action as something caused, whether by the agent or by any of their states or attitudes, including intention. Whenever a state of affairs is identified as caused or co-determined by the behavior of an agent, the agent is by default ascribed the responsibility for bringing it about: so, causation is in the picture simply because the action itself is the causing of its own outcome. This view of action, whose details should be further developed

22

 and discussed, appears to be suitably coherent with my view of speech acts, because they both focus on outcomes or effects as the grounds for by-default ascriptions of responsibility (and therefore agency), and moreover, leave room for adjustments, infelicities, or extenuation in non-default conditions.

1 On Illocutionary Types 1. Introduction The aim of this paper is to propose a reconsideration of the old (and perhaps old-fashioned) problem of the classification of illocutionary acts. This does not mean that we are going to discover anything new about illocutionary acts. Rather, we shall try to reformulate what we already know, according to different, and in some respects more sophisticated criteria. Our inquiry shall attempt to avoid both the traditional search for clear-cut theoretical classifications, and the recent orientation toward an empirical attitude that refrains from all classification. The main trouble with the former approach is its tendency toward fixism: the available types of illocutionary acts seem to be established once and for all and can hardly account for the theoretician’s own examples. The chief drawback of the latter approach is that it risks a loss of relevance: when everything is empirical, and all that is empirical is relevant, everything is relevant, and the researcher is left without any criterion for selecting his/her object and categories. Nevertheless, both rival approaches have their positive features, and it would certainly be worth our while to try and salvage these. After all, the analysis of empirical data needs theoretical distinctions and relevance criteria, if it is to give rise to interesting hypotheses; and theoretical distinctions could themselves be formulated in a more dynamic way, incorporating “cross-breeding” phenomena, shifts of relevance, in short, all those non-standard cases that are so frequent in our everyday experience and that cannot stop occurring in our analysis. In order to proceed, however, we shall for a moment go back to the Austinian notion of an illocutionary act. Since a direct reference to John L. Austin has become rather unusual in speech act theory, I shall briefly indicate which aspects of his thought I am using as to a source of methodological suggestions. (i) In introducing his classes of illocutionary acts, Austin admits that his “classification” includes marginal and peculiar cases as well as overlaps between the classes. Furthermore, he dedicates most of his exposition to the discussion of such overlaps. This suggests that his “classification” should rather be considered as a descriptive mechanism which is not Originally published in Journal of Pragmatics 8 (1), special issue on “Speech Acts after Speech Act Theory,” ed. Paolo Leonardi and Marina Sbisà: 93–112. Amsterdam: Elsevier, 1984.

Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0002

24

   without a certain hybrid-generating potential. Our reformulation will attempt to exploit such a feature, which could be of use in the task of describing empirically produced illocutionary acts. (ii) Austin shows a tendency toward foregrounding the effect of the act rather than the means for performing it. In this perspective, an illocutionary act should be defined not by the fact that certain means are used, but by the fact that using such means amounts to invoking a socially acceptable way of achieving a socially recognized kind of effect. We also shall foreground effects. Moreover, we shall maintain that the Austinian notion of an act is compatible with the notion, inspired by narrative semiotics, of an act as the change from one state into another, the responsibility for which is assigned to an agent (cf. Greimas and Courtés 1979, 5).¹ (iii) From the very beginning of his inquiry, Austin hints at the bilaterality of performatives: without the interlocutor’s agreement, the relevant effect is not realized (1975, 9, 22). Later on, he reframes such a bilaterality as the fact that unless a certain effect is achieved (namely, the hearer’s uptake of the meaning and of the force of the utterance), the illocutionary act will not have been successfully performed (1975, 116–17). Since the hearer’s uptake is in general manifested by his/her response or reaction, one might even conclude that Austin’s theory does not allow definite illocutionary forces to be assigned to isolated speech acts. Here, we shall take it for granted that a speech act receives a definite illocutionary force, and brings about a definite illocutionary effect, in conformity with the hearer’s uptake; moreover, we shall admit of bilaterality not only in the performance of illocutionary acts but also in their effects, and suggest that sequential factors are involved not merely in the interpretation of empirically produced speech acts, but also in our reformulated definitions of the Austinian classes.

2. Austin’s Classification of Illocutionary Acts As is well known, Austin’s classification of illocutionary acts includes: Verdictives, that “consist in the delivering of a finding, official or unofficial, upon evidence or reasons” (1975, 153); Exercitives, that are the exercising of powers, rights, or influence in the giving of decisions or in the advocacy of decisions in favor of or against a certain course of action (151, 155); Commissives, that commit the speaker to a certain course of action (157); Behabitives, that include various kinds of reactions to various kinds of actions and events (150); and Expositives, that are acts used in the organization of discourse and of conversation (152, 161).

¹ I have discussed this notion in Sbisà 1983a and 1987a [referred to as “forthcoming” in the original publication of this Essay].

  

25

Though Austin does not make any explicit reference to criteria of classification, we shall assume that his classes are not unmotivated. Our reconsideration will, however, take into account only the four classes of Verdictives, Exercitives, Commissives, and Behabitives, since the reasons for admitting of a class of Expositives seem to be related to a different dimension of classification (Sbisà 1972; Searle 1979, 10). The four Austinian classes with which we are concerned differ from each other in some intuitive ways, among which the most apparent one is the different hierarchical organization of their sets of felicity conditions. With regard to Austin’s outline of felicity conditions (including a condition requiring, among other things, the appropriateness of persons and circumstances, which seems to be a counterpart of Searle’s preparatory conditions (1969); a sincerity condition; and the requirement of the appropriateness of subsequent behavior, which we shall call a consistency condition: Austin 1975, 14–15), it can be pointed out that Exercitives are crucially related to the appropriateness of persons and circumstances, Behabitives to sincerity, and Commissives to consistency, while Verdictives seem to be liable to all kinds of infelicity in equal measure (Sbisà 1972, 6–7). But the fact that one of these felicity conditions should prevail (or that all of them should have the same importance), though it confirms our feeling that Austin’s distinctions are not unmotivated, is not enough to define the four illocutionary classes: it is too vague, it admits of degrees, and, finally, it does not affect the roles of the felicity conditions in determining the successfulness and appropriateness of the illocutionary act. For acts of all the four classes, it remains true that the inappropriateness of persons and circumstances can make the act void or voidable, while insincerity and inconsistency—whichever their degree of importance—make it rather an abuse. Therefore, the prevailing importance of one (or no) felicity condition(s) either is not relevant to our purposes or should be traced back to some more central feature of the illocutionary act. Indeed, there seems to be a correspondence between the prevailing importance of a felicity condition and the kind of effect which is typical of each class of illocutionary acts: for Behabitives, the importance of the sincerity condition is clearly related to the fact that their effect is the manifestation of reactions which also involve attitudes, feelings, and the like; for Commissives, the importance of the consistency condition lends itself as a basis for the undertaking of commitments; and in the cases of Exercitives or of Verdictives, there is a correlation between the speaker’s power (authority, influence . . . ) and the effect achieved by exercising it, respectively between the general reliability of the speaker (his/her competence, sincerity, and consistency) and the utterance’s claim to truth.² Thus, the intuitive consideration

² Of course, not all Verdictives are literally true or false. They do, however, put forward a claim to some positive value such as fairness or correctness, belonging to a dimension of assessment that Austin (1975, 140–5) considers as expressing an “assessment of the accomplished utterance” logically subsequent to the assessment of the utterance itself as a felicitous one, and related with correspondence with facts.

26

  

of the four Austinian classes matches the theoretical requirement that the definition of illocutionary acts be given by reference to their effects; one could be tempted to define the four classes by taking into account their most relevant felicity condition, along with the typical effect of each act. This is what we shall try to do in our next section, where, however, we shall no longer talk of classes of illocutionary acts, but of illocutionary types, because we want to make it clear that we are focusing on prototypical cases and do not commit ourselves to such (impossible?) undertakings as finding a unique classificatory slot for every illocutionary force—let alone the illocutionary force of every empirically produced utterance.³ For the moment, however, we shall comment on some further difficulties which our project meets on its way.

2.1 How to Describe Illocutionary Effects A first difficulty is that of isolating a genuinely illocutionary effect: something that can be represented as an effect of the speech act (a change in the state of something within the context of the interaction, occurring under the responsibility of the speaker-agent), but which is not yet a perlocutionary effect. Searle’s own classification of illocutionary acts, too, is affected by this difficulty: among his illocutionary points there is at least one, viz. that of Directives (attempting to get the hearer to do something) which, if it describes an illocutionary effect at all, describes it in perlocutionary terms, i.e., as a mere attempt to achieve a perlocutionary effect. Besides, the illocutionary points of Assertives, viz. to commit the speaker to the truth of the expressed proposition (1979, 12), and of Expressives, viz. to express the psychological state specified in the sincerity condition (1979, 15), are not effects in our sense, since they do not seem to be describable as changes of something within the context of the interaction: rather, they have to do with attitudes and standpoints of the speaker quite apart from the interactional relation, and with such philosophically puzzling matters as “truth” and “expression.” Our way out of this difficulty will be to exploit an unusual sense of the old, and often criticized label of “conventionality,” as attached by Austin to illocutionary acts, by contrast to perlocutionary ones. Indeed, illocutionary effects are achieved on the basis of an explicit or implicit agreement between the participants (think of the role of the hearer’s uptake) and, at least in certain cases (such as the discovery of infelicities of certain kinds), can be “undone” through further agreements. This means that the changes that are brought about by the illocutionary act are not, or not only, material, irreversible changes; they do not affect things or individuals, but representations, in so far as these are agreed upon. We shall consider illocutionary ³ Hereby I also hope to escape those kinds of criticism of the classifications of illocutionary acts which show them not to be “real” classifications (Ballmer 1979).

  

27

effects as “conventional” in this sense and as affecting not merely the addressee and/or the speaker, but the speaker–addressee relation or, more specifically, such aspects of that relation as exist by virtue of an interactionally accepted definition. Such changes in the speaker–addressee relation, or in its definition, are easily represented by the assignment, subsequent to the speech act, of some modal or propositional-attitude predicate such as “can,” “ought,” “know,” “believe,” “want” to each participant.⁴ A first and provisional example is offered by the schematic description of prototypical Exercitives outlined in Figure 1. Our schematic descriptions of illocutionary types will employ a restricted number of modal predicates. We shall not employ “want,” “believe,” or propositional attitude verbs in general, since the assignment of such predicates is related to the expression of the corresponding attitudes by some individual, and the attitudes themselves cannot be produced or cancelled, by mere agreement, in the individual. We shall, however, employ “can” (in the deontic sense of “being allowed” or “entitled,” “being in one’s right,” and the like) and “ought” (taken as the equivalent of the German “sollen”). Deontic modalities seem to fit our case, since their assignment relies on social understanding and acceptance, and it is possible to cancel them. We shall also employ the epistemic modal predicate “know,” which can be assigned and eliminated in a way parallel to that of deontic modalities in so far as it involves the idea of being socially recognized as capable of issuing reliable assertions on a certain matter.⁵

2.2 Who Is Affected by Illocutionary Effects? A second difficulty arises with respect to the terms of the relation that we want to describe. This difficulty becomes apparent, for example, if we try to extend the scheme in Figure 1 to Commissives, as is done in Figure 1a. In Figure 1a, the obligation, which is the most apparent illocutionary effect of Commissives, is assigned by the speaker to him/herself while the addressee,

Exercitive illocutionary type: Speaker

Addressee

can

ought Figure 1

⁴ A somewhat analogous use of modal and propositional-attitude predicates is to be found in Wunderlich’s interpretation of illocutionary types at the level of institutional pragmatics (1976, 89–99). Here, however, I rely mainly upon a semiotic notion of “modal competence,” that is, the set of modal predicates belonging to a subject at a given stage of a narrative sequence (Greimas 1976, Greimas and Courtés 1979, 53). ⁵ Such a notion of knowledge is inspired by some Wittgensteinian suggestions (see, e.g., Wittgenstein 1953, § 150), that I have discussed elsewhere (Sbisà 1980).

28

  

Commissive illocutionary type: Speaker

Addressee

can

?

ought Figure 1a

strangely enough, plays the role of an observer. However, it cannot be denied that promises are always made to someone; if not to anybody else, at least to oneself: someone must understand and accept the speaker’s commitment for it to be valid. Thus, the addressee neither can be eliminated from our scheme nor is assigned a precise role within it. This difficulty (along with another, partly analogous one, singled out by Searle with respect to the relation between Commissives and Directives in his classification: 1979, 14–15) is linked to the fact that our schemes take into consideration the speaker and the addressee qua actors of the linguistic exchange. As a way out, I propose a shift to the more abstract level of the actants underlying the interactional event, namely, the source of the illocutionary effect and the target affected by it. The distinction between actants and actors, between the interactional functions and the persons who perform them, is helpful here (as in various other cultural contexts, more or less remote from speech act theory) for throwing some light on the different behavior of notions such as that of individual or person, respectively that of subject or “self,” and on the fact that the analysis of social interaction often requires the self expressed by one social actor to be fragmented into a number of different aspects, not only in marginal and deviant cases but in the very routine of everyday life. In particular, the distinction between actors and actants proves useful in all cases in which the same actantial function is performed by more than one actor, or in which the same actor expresses more than one actant.⁶ The commissive illocutionary type seems to stage a situation of the latter kind: Destinator and Destinee⁷ (source and target) of the illocutionary effect are in general expressed by the same actor, namely, the speaker. If, however, a scheme such as that in Figure 2 is, again, not sufficient for characterizing Commissives

⁶ For the semiotic definitions of “actant” and “actor,” see Greimas and Courtés (1979, 3–5, 7–8). It should be pointed out that my use of these notions is only one of their possible applications. Moreover, the use of “actor” to indicate the speaker is an oversimplification of the fact that a description of the speaker as a person or individual would require, in comparison with the description of the underlying actants, a great deal of thematic investments. ⁷ I use “Destinator–Destinee” for the French technical terms “destinateur–destinataire.” I have not adopted the terms suggested by Greimas and Courtés (1979), viz. “addresser–addressee,” since these also translate the French “énonciateur–énonciataire” and are perhaps more suited to this role than “enunciator–enunciatee,” suggested by Greimas and Courtés (1979). I have not employed the terms adopted by the English translation of Greimas and Courtés (1979), viz. “sender” and “receiver,” because of their ambiguity between the material and the semiotic level (individuals as opposed to actants).

  

29

Commissive illocutionary type: Destinator

Destinee

can

ought Figure 2

Commissive illocutionary type: Destinator

Destinee 1

can

Destinee 2

can ought Figure 2a

(since it blurs the intuitive distinction between Exercitives and Commissives), we can enrich it by redoubling the Destinee, as shown in Figure 2a. Thus, we represent the illocutionary effect of Commissives as an effect on a couple of Destinees, one of which⁸ is assigned a new obligation, while the other is assigned the right to expect the fulfillment of such an obligation. The part of Destinee, here indicated as Destinee 2, is most often performed by the same actor who expresses the Destinator, or by an actor who is more or less directly related to him/her; however, the actant–actor distinction allows also for non-standard cases. We shall extend the structure displayed in Figure 2a (one Destinator, a couple of Destinees) to the other illocutionary types, since the reduplication of the Destinee seems apt to account for the fact that the illocutionary effect is an effect on a relation.

2.3 Illocutionary Effects and Felicity Conditions A parallel difficulty arises from some aspects of the relationship between felicity conditions and illocutionary effects. Felicity conditions, in an Austinian framework, are intersubjective requirements on which hearers rely in their understanding of illocutionary acts: in particular, preparatory conditions require that persons and circumstances be appropriate, i.e., accepted as appropriate by the participants as opposed to merely intended as appropriate by the speaker. Therefore, such conditions should be taken to refer to the state of the context previous to the occurrence of the illocutionary act. Accordingly, in all our schemes from Figure 1 to Figure 2a, the modal predicate assigned to the speaker (or to the Destinator of the illocutionary act) has been represented as the point of departure of an arrow ⁸ Since gender roles are thematic roles, the actant is not yet concerned with them. Therefore, it does not make sense to refer to the actant as to he/she: I take it to be neuter (and shall write “It,” “Its” with a capital I in order to facilitate anaphora).

30

  

directed toward the illocutionary effect. Yet, speakers can give their hearers new information by way of implicitly communicating the preparatory conditions of their illocutionary act, so that, in many cases, the hearers do not seem to be aware of the contextual features required by the preparatory conditions until they understand the speaker’s illocutionary act and accept it as a felicitous one. Therefore, the representation of preparatory conditions is in some respect subsequent to the illocutionary act; and doubts may arise as to whether or not we should represent the assignment of the modal predicate specified by the preparatory condition as a backward effect of the act on the speaker, as shown in Figure 3. The scheme in Figure 3 describes those cases in which the speaker’s power or authority is increased, or even established by the accepted successfulness of the illocutionary act (think of the first Exercitives issued by a rebel leader) and can perhaps be extended to all Exercitives. But it effaces the distinction between what is held to have been the case before the illocutionary act, and what is held to be the case “now” (from the illocutionary act on). This becomes apparent in the tentative extension of that scheme to Commissives, which is represented in Figure 3a. The scheme in Figure 3a may appear simpler than that in Figure 2a and therefore preferable; however, there is little doubt that the predicate “ought,” which it assigns to the speaker, takes effect from the illocutionary act on, whereas the predicate “can,” that the scheme in Figure 3 assigns to the speaker, represents the modal state of the speaker not only as it is after the act but also as it must have been beforehand. The distinction between actors and actants again offers a way out. The notion of a Destinator involves the idea of some priority with respect to the effect on the Destinee(s), so that the arrow from the modal predicate assigned to the Destinator to that assigned to the Destinee(s) is no longer necessary. At the same time, an actant Destinator cannot exist apart from Its act, nor apart from the latter’s Destinee(s), so that the modal predicate that fits Its description cannot be selected until Its act has achieved its effects. This feature can account for the

Exercitive illocutionary type: Speaker

Addressee ought

can Figure 3

Commissive illocutionary type: Speaker

Addressee can

ought Figure 3a

  

31

backward effects of the illocutionary act without any need to represent these by means of a backward oriented arrow. Therefore, our arrows can simply be dispensed with. Moreover, the doubling of the Destinee establishes a firm distinction between the actantial description of the state of the speaker–addressee relation subsequent to the speech act, and the description of the speaker qua Destinator. Concluding then, our schematic descriptions will express: (i) the way in which the previous modal state of the actor, who performs the Destinator of the illocutionary act, is to be described qua the Destinator’s state, if the illocutionary act is considered as successfully performed; (ii) the modal states, subsequent to the illocutionary act and correlated to each other, of the actors who express Destinee 1 and Destinee 2, qua Destinees. This will be done by means of the modal predicates “can,” “ought,” and “know.” The positive assignment of one of these predicates will be indicated by the symbol “+,” its elimination by the symbol “ ”.

3. Four Illocutionary Types As a partial reformulation of the Austinian classes of illocutionary acts, I propose the four schematic descriptions of illocutionary types, which are collected in Table 1. These schematic descriptions do not claim to be full descriptions of empirically produced tokens of illocutionary acts; their aim is to offer a model for such descriptions, which in fact should be richer and more complicated. Our schematic descriptions, however, are oversimplified even as models and could be expanded in several ways: (i) by admitting of more than one modal predicate for each slot; (ii) by using more specific labels instead of, or together with, the highly general labels “can,” “ought,” “know”; (iii) by indicating which kind of propositional content is to be embedded by each occurrence of each modal predicate;

Table 1 Illocutionary type

Destinator

Destinee 1

Destinee 2

Exercitives Commissives Verdictives Behabitives

+ can + can + can + ought

± ought + can + knows + knows

± ought + ought + ought ought

32

   (iv) by including those modal predicates (such as “want” and “believe”) that are assigned to the Destinator by the sincerity condition of the act; (v) by including the modal predicates of the sincerity condition kind,⁹ in order to represent the possible alignment of the Destinee(s) with the perlocutionary object of the act.

In this paper, we shall not have space for these further problems, which anyway are not essential to our aim of preparing a first, workable guide to the effectcentered description of illocutionary acts.

3.1 Exercitives According to Searle (1979, 10), the notion of an Exercitive relies on features such as status, or relationship to institutions, which are not relevant to the dimensions of classification upon which Searle focuses his attention. Therefore, he does not admit of an exercitive illocutionary point, and splits up the Austinian Exercitives into two groups, the one included among Directives, the other among Declarations. Exercitives are split up in a parallel way also by Vendler (1972, 6–26), who distinguishes between Exercitives proper and a class of Operatives. The distinction that is singled out by both authors has its own point and can be accepted by us, provided it is not a distinction between kinds of illocutionary effects. For us, all Exercitives can be said to presuppose a power (their Destinator is assigned “+can”) and to bring about a creation or elimination of obligations. But the linguistic performance of, as well as the report on Exercitives assigning obligations relative to extralinguistic institutions does not need to specify the contents of such obligations (these have been established, or are even listed, elsewhere), whereas Exercitives that operate with reference to particular, concrete occasions of interaction, without any interference with the possible institutional features of their context, must specify the particular contents of at least one of the assigned, or eliminated, obligations. This determines the syntactic differences pointed out by both Vendler (1972, 20–3) and Searle (1979, 22, 26–7). Four further problems that are worth mentioning are: (i) Destinee 2 does not seem to play an essential role in the description of Exercitives. However, it should be stressed that the assignment or elimination of a predicate of the “ought” kind is often, or even in general, accompanied by the assignment or elimination of a complementary obligation to, or from, a partner of the addressee (who could also be a

⁹ [What I had in mind were predicates attributing attitudes as opposed to deontic modalities.]

  

33

collective actor). An order establishes for Destinee 2 (usually performed by the speaker) that he/she ought not to carry out the action him/herself, and, moreover, that he/she ought to evaluate the subsequent performances of the actor expressing Destinee 1 by reference to the obligation assigned to him/her by the order itself. Christian baptism assigns to Destinee 1 the obligations shared by all members of Christianity and to Destinee 2 (Christianity, not the baptizer: the latter, in particular cases, might not him/herself be a Christian) the obligation to accept Destinee 1 as one of its members.¹⁰ (ii) There are Exercitives which could be described as assigning to Destinee 1 a modal predicate “± can” (for example, permission, prohibition, and certain kinds of appointments). Such acts can be related to our exercitive illocutionary type insofar as the assignment or elimination of the predicate “can” can also be formulated as the elimination or the assignment of a predicate of the “ought” kind, embedding a negative proposition; or else, insofar as they can be described as belonging to a sub-type which joins to the assignment of “±ought” the assignment of “±can.” However, it should be pointed out that nominations and appointments are to be considered as Exercitives (as opposed to Commissives) as far as the charge or title they assign is considered primarily as a task and, therefore, as a set of obligations. (iii) A speaker may issue commands on the basis not of a recognized authority, but of the brute fact that he/she possesses a gun (Searle 1979, 7). Such a fact is, of course, not an institutional fact, nor even a “conventional” fact in our sense. Thus, it seems that, at least for some Exercitives, one of the assigned modal predicates does not need to be “conventional.” But a second look at this example will easily show how we should deal with it. The utterance of an imperative sentence by a speaker possessing a gun produces the effect represented as “+ought,” namely a deontic obligation, only if the speaker is recognized by the addressee as entitled to produce such an effect (or if his/her power is recognized as a power to engender deontic obligation). Such brute facts as the possession of a gun (perhaps, together with further assumptions about the legitimacy of the possession) may count as grounds for this recognition. But if the recognition does not occur, the possession of a gun can merely create a state of necessity (as opposed to one of deontic obligation), and the speaker’s utterance does

¹⁰ [In this example, I was considering baptism qua illocutionary act, not qua sacrament. When assimilating the ritual formulas of sacraments to performative utterances, and therefore sacraments to illocutions, it should be considered that what is believed to bring about their specifically sacramental effects is not the minister’s words, but God’s grace.]

34

   not count as an Exercitive in our sense, in spite of its apparent, or even successful, directive goal. (iv) Directivity is obviously connected with the use of the imperative mood; the latter is also one of the most widespread devices for performing certain varieties of exercitive illocutionary acts. Moreover, the most obvious means of getting the addressee to do something is to achieve agreement about a definition of the interactional relation, such that the addressee is assigned the obligation to do something. This can explain why so many Directives are Exercitives as well, but does not yet clarify the connection between exercitivity and directivity. Here, we cannot enter into a detailed discussion of this problem, which would also give rise to a wider debate on the relationship between illocutionary forces and perlocutionary objects. We can merely point out that, in our framework, the directive, perlocutionary object of getting someone to do something can be pursued not only by exercitive illocutionary acts but also by Verdictives (such as evaluations), Behabitives (such as challenges), and Commissives (such as threats).

3.2 Commissives Our schematic description for the commissive illocutionary type does not apply only to prototypical promises. It can account for the cases in which Destinator and Destinee 2 are expressed by distinct actors, as in many cases of guaranteeing, or in those promises which involve the future action of a subordinate. It can also be used in describing “espousals” (consenting, siding with, favoring, and the like: cf. Austin 1975, 152, 158). In acts of espousal, the proposition embedded by the predicate assigned to Destinee 2 (“+ought”) does not specify a particular action but a more general orientation of conduct, or some constraints on it, which, however, correspond to the expectations that the addressee (who expresses Destinee 1) is thereby entitled to have. Some problems are raised by the two different occurrences of one and the same predicate “+can.” If assigned to Destinee 1, “+can” is to be understood as representing the right to expect or even exact the fulfillment of the obligation which is assigned to Destinee 2. If assigned to the Destinator, “+can” is more puzzling. It should not be understood as representing the Destinator’s capacity to do something (i.e., to fulfill the obligation assigned to Destinee 2), since this would conflate Destinator and Destinee 2. Nor is it enough to say that such a “+can” represents a capacity to cause something to be done by Destinee 2, since this is not necessarily a “conventional” feature and, moreover, would link Commissives to one among their possible perlocutionary goals. I would rather say that “+can” represents here the acknowledgment that the Destinator has a right to authorize Destinee 1 to

  

35

hold certain expectations with respect to Destinee 2. The recognition of a capacity is, however, to be regarded as a ground for assigning such a right to the Destinator.

3.3 Verdictives In our schematic description of the verdictive illocutionary type, Verdictives appear essentially as formulations of knowledge which rely on the Destinator’s modal predicate “+can.” Such a predicate expresses, in this case, the competence of the Destinator with regard to the subject matter of the Verdictive. Such a competence could also be represented by a predicate “+know,” since competence is often identifiable with the knowledge of the criteria according to which the Destinator formulates the relevant kind of knowledge, or (when these criteria are held to be commonly shared, as in the case of everyday empirical statements) with the Destinator’s acquaintance with the relevant “facts.” However, the choice of the predicate “+can” stresses that the Destinator is competent in so far as It is recognized as competent and, therefore, entitled to formulate a certain piece of knowledge. Moreover, “+can” could be extended to include social and cultural constraints on the performance of Verdictives (not everybody is equally likely to be authorized to formulate knowledge about everything: think of such social actors as children or women as opposed to male adults, or lay people as opposed to priests, and reporters as opposed to juries). As to the Destinees, it could be observed that many Verdictives do not encode the reference to an addressee. But for us this cannot mean that they have no Destinee: Verdictives take effect on the interactional relation as much as all other illocutionary acts; furthermore, we are considering them just insofar as they have such effects (by the same token, we shall leave aside the question as to whether merely mental formulations of knowledge should be represented as involving one or more actants). In the case of many Verdictives, Destinee 1 (which, after the speech act, can be said to possess the piece of knowledge formulated in the speech act) is performed by any hearer, and even by hearers to whom the act is apparently not addressed (think of spies). As to Destinee 2, Its presence in the schematic description stresses that the formulation of knowledge is to be guaranteed by someone, who is thereby assigned a predicate of the “+ought” kind. Such a Destinee 2 is usually performed by the actor who also expresses the Destinator (as in Commissives), but there are cases in which the two are performed by distinct actors, for example those Verdictives in which the speaker does not commit him/herself to give evidence, but sends Destinee 1 back to some external “authority” (ranging all the way from Aristotle to the computer), that is held to possess the relevant evidence. It should be pointed out that our Verdictives do not correspond to Searle’s Assertives. The latter include only those speech acts that can be said to be true or false, while our Verdictives, like Austin’s, include also acts which should rather be

36

  

evaluated as fair/unfair, sound/unsound, correct/incorrect, and even good/bad. A further problem lies in the relationship of Assertives, respectively Verdictives, to a possible dimension of speech act classification that considers the linguistic or extralinguistic character of illocutionary effects. All Assertives seem to belong to Expositives as well: the cases in which their effects involve matters external to discourse or conversation are classed by Searle separately as “Assertive Declarations” (1979, 19–20). However, here we shall not attempt to draw such a distinction, which is a difficult one in its own right, to be discussed with reference to all of our four illocutionary types.

3.4 Behabitives The idea underlying our schematic description of the behabitive illocutionary type is that of a reaction of the Destinator, offered by It to Destinee 1 as an object of knowledge. This enables us to represent the Behabitives as operating on, and within, an interactional relation. It is mainly in this respect that our Behabitives differ from Searle’s Expressives. However, Searle’s definition of Expressives accounts for what we have called the prevailing importance of the sincerity condition by linking the illocutionary point to such a condition. In order to account for this feature of Behabitives, we shall have to specify that what Destinee 1 comes to be acquainted with includes the state which is assigned to the Destinator by the sincerity condition of the act. It should be pointed out that this does not commit us to take into consideration the actual occurrence of a certain psychological state in the speaker’s mind. Behabitives, like other illocutionary acts, can be valid, though insincere; insincerity makes them abuses, not misfires (cf. Austin 1975, 40). Moreover, our use of the predicate “+know” is relative to the interactional situation in which its assignment occurs, and does not commit us to accept the embedded proposition as a true one. Our notion of knowledge, here as well as in the case of Verdictives, is a notion of what counts, or is recognized, as knowledge in a certain situation, but could still turn out to be false or cease to be recognized as genuine knowledge, as opposed to a notion of “absolute” knowledge which is supposed to have found “the facts” once and for all. Offering one’s own act as an object of knowledge to someone else does not state any requirement of competence, rights, or authority on the part of the Destinator. Rather, it seems appropriate to say that a Behabitive is successfully performed insofar as it is, in some sense, a due reaction: Behabitives involve the manifestation of a “self,” which is represented as being called upon. Thus, if the Destinator of a Behabitive is to be assigned any of our modal predicates, the most obvious candidate is “+ought.” The deontic character of such a predicate is apparent in many cases (apologies, thanks, wishes, challenges); however, also for those cases in which the Behabitive seems to react to “brute” facts (e.g., needs or wishes, as

  

37

opposed to obligations), it can be said that the Behabitive is an appropriate one in so far as the need or wish are accepted by the participants as good reasons for a certain reaction. The predicate “ ought,” when assigned to Destinee 2, expresses the elimination of an obligation: as soon as Destinee 1 acquires the relevant piece of knowledge, the relation between It and Destinee 2 can, under certain aspects, be said to start anew. It should be pointed out that the “miscellaneous” character of Behabitives (cf. Austin 1975, 152, 158) might demand the specification of further modal changes of Destinee 2, which would vary from sub-type to sub-type: our “ ought” is therefore a minimal modal change (albeit a characteristic one).

4. Some Speech-Act Theoretical Games Our four illocutionary types lend themselves to a number of speech-act theoretical games. Here, we shall not act out any of these completely, but only give some examples and, so to speak, some instructions for use. I should like to emphasize that the point of our games is to suggest some ways in which our descriptive, typological work can enable us to distinguish “family resemblances” and combinatory possibilities, and even to make a dynamic use of illocutionary types in the analysis of discourse and conversation.

4.1 Type-Grouping Game This consists in the attempt to find ways for assembling two or more illocutionary types into one group. Its point is to show various kinds of similarities and oppositions between illocutionary types. (i) The three types whose Destinators are assigned “+can” can be grouped together and contrasted with Behabitives. We thus point to a difference between those types of acts that cannot be performed unless the speaker has some recognized authority, capacity, or competence, and those types of acts that can be performed by whoever finds him/herself in an appropriate interactional situation. This grouping suggests that speakers who, in a given situation, have no recognized power or less power than anybody else, are likely to perform Behabitives instead of Exercitives, Commissives, or Verdictives. Perhaps they will themselves indicate the behabitive features of their own speech acts by using such mitigating devices as, for example, hedges (think of so-called “women’s language”). (ii) Exercitives and Commissives can be gathered into one group, to contrast with the group of Verdictives and Behabitives. An apparent ground

38

   for such a grouping is that our descriptions employ, for Exercitives and Commissives, the sole predicates “can” and “ought,” while Verdictives and Behabitives need also the predicate “know.” Exercitives and Commissives, grouped together, can be labeled “acts of manipulation.” Such a label emphasizes that the Destinator assigns or cancels rights and obligations to or from Its Destinees in view of some future action on the part of at least one of the latter. Verdictives and Behabitives can be labeled “acts of sanction,” since they share an orientation toward what has already been done or what is already the case, and react to, or comment upon it. These labels, borrowed once more from narrative semiotics (Greimas and Courtés 1979), indicate for us an initiating, respectively closing position of an act within a narrative sequence. The idea behind this grouping is to point out the similarities between Exercitives and Commissives, respectively Verdictives and Behabitives, as well as to hint at a sequential orientation of our illocutionary types, by the aid of a comparison with narrative-semiotic categories.¹¹

However, it should be kept in mind that our use of the categories of “manipulation” and of “sanction” is not parallel to their use in narrative semiotics. For example, we do not want to restrict our acts of sanction to the acts that comment on the performance of a task that has been previously assigned to a Subject by the Destinator Itself, as is the case in narrative sequences of the Manipulation– Action–Sanction form. Interactions such as “Tom, give me a spoon”–[Tom performs the action]–“Thank you,” or sequences such as the Elicitation–Answer– Evaluation sequences which are so frequent in classroom discourse (cf. Sinclair and Coulthard 1975) display an apparent Manipulation–Action–Sanction form, but life is not always like this. A certain amount of generalization both for the category of sanction and for that of manipulation is needed, in order to make them more widely applicable. Moreover, we have not introduced nor discussed the notion of an actant Subject, not because of theoretical reasons, but owing to differences of focus between the relationship Destinator–Subject and the relationship Destinator–Destinee(s) which is of use here. (iii) Exercitives and Verdictives can be opposed to Commissives and Behabitives, since, in describing the effects of the former, we naturally focus attention on Destinee 1, while the description of the effects of the latter assigns the most prominent role to Destinee 2. However, no change in Destinee 2 of Commissives and Behabitives, respectively in Destinee 1 of Exercitives and Verdictives, can be achieved unless there is a corresponding change in the other Destinee, playing the role of a support or of ¹¹ As far as our illocutionary types are comparable to those distinguished by Wunderlich, their characterizations as initiative or reactive, too, match each other (cf. Wunderlich 1976, 77–8).

  

39

an indispensable witness. The point of such a grouping is to emphasize some of the differences that a dichotomy between acts of manipulation and acts of sanction would blur, and to stimulate a more attentive consideration of the complementarity between the Destinees.

4.2 Type-Shifting Game This game presupposes the former and consists of finding out how either illocutionary sub-types and varieties¹² or empirically produced illocutionary acts may shift from type to type within a given group. Its point is to foreground the varieties that are intermediate between the types or that can be described as belonging to more than one type, and to suggest hybrid interpretations for ambiguous, actual utterances. Within the group of acts of manipulation (Exercitives and Commissives), a variety such as permission can be described either as eliminating an obligation for Destinee 1 and limiting the rights of Destinee 2 (an exercitive illocutionary effect) or as conceding a right to Destinee 1 and assigning an obligation to Destinee 2 (a commissive illocutionary effect). There are also acts that seem to have, at the same time, effects of both types: the partners in a bet, or even in a marriage, assign obligations to each other (an exercitive feature) as soon as they accept their own obligations with respect to each other (a commissive feature). Within the group formed by Exercitives and Verdictives, there are such intermediate varieties as interpreting and predicting, or even advising and warning. Interpreting and predicting become Exercitives as soon as the competence of the Destinator is considered as an authority: correspondingly, the knowledge assigned to Destinee 1 turns into an obligation to believe. Advising and warning come near to being Verdictives when it is stressed that the obligation assigned to Destinee 1 includes the recognition of a certain piece of knowledge, so that the authority of the Destinator can, correspondingly, be redefined as competence. The group of acts of sanction does not admit of intermediate varieties: the fact that the knowledge produced by a Verdictive is formulated in the speech act, while the knowledge produced by a Behabitive is a knowledge of the speech act, does not admit of degrees. But the actual production of utterances which are ambiguous between the two types seems to make fun of these theoretical difficulties. Thus, expressions of opinion may still count as statements, and therefore as Verdictives, but they may also foreground the good faith of the speaker rather than his/her competence and claim to truth, and therefore count as Behabitives manifesting the speaker’s beliefs. Moreover, when the statements about one’s own mental states, ¹² By “illocutionary varieties,” I mean lexically distinguished sub-types (or sub-sub-types) of illocutionary acts. They are, however, types as opposed to tokens.

40

  

feelings, sense data, and the like do not involve any difficulty that has to be overcome by means of the competence of the Destinator, there is no reason for assigning to the Destinator a predicate “+can,” and the utterances are to be interpreted as mere reactions to some emotional and/or cognitive experience. Of course, in all the cases in which it is possible to describe a shift from type to type at the level of illocutionary varieties, one can reasonably expect to find further and more detailed cases of shift at the level of actual discourse and conversation. I should also like to add that there are diachronic cases of type-shifting: thus, such apparently clear-cut varieties of the verdictive type as assertion and evaluation have shifted from type to type in their philosophical definitions and perhaps, at least to some extent, also in the underlying practice of language use. The prescriptivist and emotivist theories of ethics are examples of the shift of evaluation from Verdictives to Exercitives and to Behabitives. Assertion itself, which has counted and still may count, in certain contexts, as an Exercitive relying on the authority of the Destinator, is also moving toward the commissive and even the behabitive illocutionary type. The shift of assertion toward Behabitives is indicated by the quite widespread inclination to describe it as communicating belief. As for its shift toward Commissives (cf., e.g., Wunderlich 1977), such a change of type is not foreseen by our examples of type-grouping, but could be represented as occurring within a further group formed by Verdictives and Commissives, both of which assign to Destinee 2 a predicate “+ought,” or as involving our next game.

4.3 Group-Switching Game This game presupposes again groups of types. Given an utterance (actually produced or imaginary) that displays illocutionary indicators related to a type which belongs to a certain group of types, we attempt to detect co-textual and/or contextual features that would suggest locating it within the opposite group of types. This is not a matter of gradual shift, but of radically changing or drastically selecting the illocutionary effects of the speech act. The connection of group-switching with co-textual and contextual features does not commit us to explore the entire co-text and context of the speech act: we need to focus attention merely on those features which are relevant to the kind of switching under consideration. Here are some examples of group-switching, each of which relies on one of our examples of type grouping. (i) The relevant features for a group-switching from Exercitives–Commissives– Verdictives to Behabitives are the assignment to the actor who expresses the Destinator, previously to the speech act under consideration, of some modal predicate of the “+can” type and, respectively, the existence of constraints

  

41

which hinder such an assignment. Thus, Exercitives can turn into expressions of will, Commissives into expressions of intention or of sympathy, and Verdictives into expressions of belief. Moreover, Behabitives issued by speakers who are unlikely to express a powerless Destinator can turn out to achieve exercitive, commissive, or verdictive illocutionary effects. In this connection, one could also wonder whether Searle’s classification, which keeps matters of power apart from all illocutionary points except that of Declarations, tends to foreground the expressive features of the illocutionary acts that are assigned to its first four classes. In particular, it could be pointed out that the most typical Directive, that is, request, is an illocutionary variety which was not included among Austin’s Exercitives and which cannot be said to assign obligations but rather, like our Behabitives, to react to a given situation by manifesting the Destinator’s will to a Destinee. In this framework, we can also explain why Austin lists challenges as Behabitives, whereas Searle counts them as Directives. They are, in fact, both, since they can be described as reactions to a supposed threat to the face of a self, and also have an apparent directive goal (which, however, does not rely on recognized rights or authority). (ii) The type-grouping which contrasts acts of manipulation with acts of sanction suggests, as has already been noticed, that illocutionary types have certain roughly sequential properties. Therefore, to play groupswitching with this in mind amounts to looking for interferences between illocutionary type and sequential position. Such interferences occur whenever a speech act displaying exercitive or commissive features occurs in a position which is not, or can be interpreted as not being, an initiating one; similarly, whenever a speech act displaying verdictive or behabitive features occurs in a position which is not, or can be interpreted as not being, a closing one. (Note, in passing, that such interferences occur almost everywhere, and their discovery and description might turn out to be a main device for the understanding of the dynamics of illocutionary effects in actual interaction). However, this group-switching game is far from being unproblematic. To start with, not being in an initiating position does not mean being in a closing one, and vice versa. Indeed, the narrative sequence which has inspired our groups has a central slot, Action, which has not been taken into account here, since it does not characterize any group of illocutionary types in particular (we cannot consider acts of some illocutionary types as being more of an action than others!). One might then introduce a further distinction between a closing position (which is also a reactive one), and a merely reactive position (which is not a closing one). But there remains the further problem that not all sequences are triplets, or analyzable into triplets. In order to cope with these difficulties

42

   and to enable ourselves to play a game which is sufficiently interesting, we shall rely on a number of assumptions, which can be made explicit as follows. First, we shall assume that whenever a speech act is located (and interpreted as located) in the central slot of a narrative sequence Manipulation–Action–Sanction, there is neither a positive nor a negative interference between its illocutionary indicators and such a sequential position. Secondly, we shall assume that almost every speech act may be considered as initiating a sequence (in so far as we can focus attention on its connection with what follows), or as being a closing reaction (in so far as we can focus attention on its connection with what precedes, and abstract from what follows). Thirdly, we shall admit of both verbal and non-verbal steps in the relevant action sequences, as well as (following Goffman 1974) of more than one “track,” and therefore more than one sequence, within each interactional event. These assumptions enable us to play group-switching between acts of manipulation and acts of sanction according to two strategies: either (a) we first delimit a sequence (on whatever criterion), and then consider whether the position of a certain speech act within the sequence interferes with its own illocutionary indicators (e.g., enforces or contrasts them); or (b) we first give an illocutionary interpretation to a speech act and look for the next and/or the preceding step in an appropriate sequence, and then attempt to single out other sequences (through a different “punctuation” of the interaction or through a change of “track”), within which the speech act is likely to find a different collocation. It should be pointed out that both strategies commit the player to further inquiries on the selection, delimitation, and “punctuation” of conversational sequences, of complex action sequences, and of interactional “tracks.” (iii) In the group-switching between Exercitives/Verdictives and Commissives/ Behabitives, the relevant features are the contextual and co-textual relevance of the effects on Destinee 1, respectively Destinee 2. A simple test would be to consider whether the perlocutionary effects which the utterance is aimed at are better represented as consequences of the modal change in Destinee 1 or of that in Destinee 2. However, a perlocutionary effect can also be selected as relevant by its very occurrence, whether expected or unexpected, in the following interaction, or by the occurrence of insisting, or of some negative evaluation of the non-occurrence of the intended effect. Therefore, this kind of group-switching game would commit the player to further inquiries on perlocution.

2 Speech Acts, Effects, and Responses 1. Introduction In his paper “Conversation” (1992), John Searle begins by questioning the traditional limitation of speech act theory to the study of individual speech acts; but ends up by pointing out a feature of conversation (that of having a background) which is again a feature of individual speech acts.¹ His speech act theory and its derivates do not seem to be able (nor perhaps willing) to leave the traditional sentential perspective for a textual (here, conversational) one. I shall now try to put forward some comments on certain aspects of the theses argued by Searle, from a point of view which is internal to speech act theory (and thus different from, say, the point of view of Conversation Analysis), but is not committed to an “individual speech act” kind of sentential perspective. Our considerations ought to be accompanied by a careful discussion of their consequences on various methodological issues in speech act theory and pragmatics, but here there will not be sufficient space to do so.

2. Uptake and Response J. L. Austin maintained that “an effect must be achieved on the audience if the illocutionary act is to be carried out” and called such an effect “uptake,” meaning by this “the understanding of the meaning and the force of the locution” (Austin 1975, 117). This notion has sometimes received exaggerated attention (as in Bach and Harnish 1979, for example, where uptake appears to be all the illocutionary act is up to), but its consequences have rarely been discussed (Gazdar 1981 is an isolated example). I would like now to draw attention to some of these consequences. If the hearer’s uptake is necessary for the carrying out of an illocutionary act, in order to know whether a certain illocutionary act has been carried out we should Originally published in J. R. Searle et al., (On) Searle on Conversation (Pragmatics & Beyond New Series 21), compiled and introduced by Herman Parret and Jef Verschueren, 101–11. Amsterdam: John Benjamins, 1992. ¹ [It follows from Searle’s claim that “All interpretation, understanding, and meaning . . . functions against a background of mental capacities . . . that are not themselves interpretations, meanings, understandings” (1992, 29; my italics), that isolated utterances too are understood against such a background.]

Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0003

44

 , ,  

first know whether an uptake has been achieved. And this we can know from a consideration of the response (verbal or non-verbal) which follows the illocutionary act under examination, since each response makes manifest how the hearer has taken the speaker’s illocutionary act. It seems therefore that, when we want to assign a definite illocutionary force to a certain speech act, we should take the hearer’s response into account. But if this is so, we have to admit that our consideration of isolated speech acts leads only to provisional results, until we know not only in which context a speech act has been issued but also which response it has received. Thus the consideration of sequential phenomena seems to be relevant to the consideration of the very illocutionary force of an utterance. When asking ourselves how the study of conversation is to be related to the study of speech acts, we should not try simply to extend speech act theory to conversation, since the study of speech acts already presupposes some reference to conversational sequences. It should be noted that if a response can confirm or disconfirm the supposed illocutionary force of the utterance to which it reacts, it is reasonable to expect that, if the utterance displays some illocutionary ambiguity, the response it is given can be used for disambiguating its force and, therefore, for letting it count as a definite illocutionary act. It is less clear what we should say when a speech act, which at first sight appears to be unambiguous, receives a response which would be appropriate to a different illocutionary force: sometimes we would speak of a “misunderstanding,” but in other cases the deviant response may have the effect of re-defining the situation and thus the illocutionary force of the previous speech act.

3. Uptake and Perlocution Let us now consider some possible objections to the perspective on speech acts that we are proposing. First of all, in claiming that the consideration of sequential phenomena is relevant to the very illocutionary force of an utterance, are we not confounding illocutionary and perlocutionary acts? A response, or any other verbal or non-verbal act of the hearer, which is performed because of the understanding by the hearer of the meaning and/or the force of the speaker’s utterance, counts as a perlocutionary effect of the speech act, and the act by the speaker of bringing it about counts as a perlocutionary act of the speaker (cf. Austin 1975, 101; Davis 1981). Thus the two facts (i) that the perlocutionary act does not coincide with the illocutionary act and (ii) that the response given by the hearer, or generally the perlocutionary effect on her/ him, presuppose a certain understanding of the meaning and/or the force of the speech act, both follow from the definition of perlocution and do not conflict with each other. The case of the response which confirms, defines, or redefines the

 , ,  

45

illocutionary force of a speech act is a case of perlocution: in particular, it is the case of a perlocutionary effect consisting not merely in a psychological reaction, but in an act of the hearer, when such an effect presupposes an understanding of the force of the speech act (and not, as may also happen, merely of its meaning). However, the response produces its own backward effect on the illocutionary force of the previous speech act not by virtue of its being a perlocutionary effect, but because it presupposes and therefore indicates how the speech act has been taken. It should also be noted that such a response need not be a response of compliance with respect to the perlocutionary goals of the speaker (we can disobey a command, while our response to it shows we recognize it as a command; disbelieve— and contradict—an assertion, while recognizing it as an assertion; refuse to give the information we are asked for, while recognizing we are asked for it . . . ).

4. Uptake and Allegedly “Conditional” Illocutionary Acts Secondly, up to this point we have been considering uptake as being necessary to all kinds of illocutionary acts; but is that really the case? According to various authors (Searle 1992; Katz 1977; Bach and Harnish 1979) the phenomena we have been considering as general ones are in fact limited to “conditional” speech acts such as bets and offers. Let us now see whether bets and offers are so different from the other illocutionary acts (and so similar to each other). A bet, if it is accepted, commits the speaker to doing something (specified by the utterance which counts as a bet) only if something else (also specified by the utterance) happens or turns out to be the case. An offer, if it is accepted, commits the speaker to do what s/he offered to do. But does “is accepted” mean just the same in these two cases? As for the bet, it is relevant that it is accepted as a bet, namely, that the addressee agrees that the speaker is betting and that her/his bet is felicitous. In this case the bet “takes effect” (I recall here the distinction drawn by Austin between three kinds of effects of the speech act: uptake, “taking effect,” and the inviting of perlocutionary responses or sequels; cf. Austin 1975, 116–18): namely, if the specified event happens or turns out to be the case, the speaker will have to perform the specified action (say, pay the hearer a certain amount of money). As for the offer, if it is to create a straightforward obligation for the speaker to do something, it has to be accepted in a stronger sense, that is, the hearer has not only to accept that it is an offer (that the speaker is offering and that her/his offer is felicitous) but also to accept the offer (by saying “Yes, please” rather than “No, thanks”). As to those speech acts, which are held not to be conditional, it is easy to see that their being accepted as illocutionary acts of certain kinds is relevant to their “taking effect” in the corresponding ways. If your interlocutor does think that, in spite of your good will, you will never be able to do what you are promising to do,

46

 , ,  

s/he will not accept your promise as a promise, and you will not be bound by it. Similarly, I can reject your justifications or your excuses or your compliments by refusing to recognize them as such, which is a more radical kind of rejection than maintaining that your excuse or compliments were insincere or even finding out you gave a false justification. And if an assertion is not accepted as an assertion, the speaker will not count as being committed to the truth of the proposition expressed: the problem of the truth or falsity of what s/he said will never seriously arise; the hearer will not ask the speaker for evidence, reasons, and the like, nor perhaps expect her/him to avoid self-contradiction. Think of what happened to moral judgments when philosophers decided they were not real assertions. Should we therefore conclude that all speech acts are “conditional,” in the sense that they all “take effect” only if they are accepted as performances of some illocutionary act? This is, in other words, what we have already claimed by arguing that the hearer’s uptake, together with the response which makes it manifest, plays a central role in the assignment of an illocutionary force to an utterance. What is then the difference between so called “conditional” speech acts and other speech acts? Bets seem to “take effect” only when they are accepted as bets: they are like “non-conditional” speech acts, but for the fact that the obligation produced by their “taking effect” is itself conditional (an obligation to do something if certain circumstances are the case). Offers, on the contrary, have been described above as “conditional” in a stronger sense, since they produce an obligation only if the addressee “accepts” them not only by taking them to be (felicitous) offers but also by manifesting a positive attitude of her/his will toward the offered action. But is it really the case that a bet “takes effect” if it is simply accepted as a bet, by producing an obligation which is itself conditional, while an offer needs to be accepted in a stronger sense, in order for it to “take effect” by producing an obligation which is not itself conditional? If we re-describe offers as producing conditional obligations, that is obligations to do something if something else is the case (in particular, if the hearer wants the speaker to do so), offers appear to be similar to bets and thus to “non-conditional” speech acts: in order for them to “take effect” by producing a conditional obligation, it is enough that they are accepted by the hearer as offers. Thus, it would be convenient to consider offers as producing conditional obligations, and as taking effect if they are simply accepted as offers, independently of the positive or negative orientation of the speaker toward the offered action, which is expressed by the response they (as offers) receive. Under this new description, offers appear to have all the three kinds of effects distinguished by Austin: they need the hearer’s uptake (in order to be offers), they “take effect” (by producing a conditional obligation), they produce further consequences (acceptance—in the stronger sense—or refusal; beyond these and possibly according to these, the performance or non-performance by the speaker of the offered action; and so on). The existence of a “conditional obligation” may seem strange, but is no stranger than the existence of a

 , ,  

47

conditional illocutionary act. From a logical point of view, this is simply a matter of the scope we want to assign to the deontic operator. Moreover, it is already a consequence of our distinction between illocution and perlocution that a speaker may be assigned a certain obligation, without this obligation giving rise to an actual course of action. Our conclusion is, then, that all illocutionary acts (bets and offers included) may be held to be “conditional” in the sense that they have to be accepted as illocutionary acts of certain kinds, if they are to “take effect” in the corresponding ways, while only some illocutionary acts (among which bets and offers, but not promises or commands) produce conditional obligations.

5. Two Kinds of Inappropriate Response We can now come back to the suggested interference between the conversational sequence and the illocutionary force of speech acts. According to Searle, the illocutionary point of an assertion is achieved whether it receives an appropriate response or an inappropriate one, while the illocutionary point of a bet is achieved only if it receives an appropriate answer. I would like to argue that such a distinction refers to different kinds of “appropriateness.” That in such a distinction, different kinds of “appropriateness” are at issue. If we react to a would-be bet as if it were not a bet, we are giving an inappropriate response in an illocutionary sense (our response is inappropriate to the supposed illocutionary force of the utterance). On the contrary, if we react to an utterance such as Searle’s example: (1)

I think the Republicans will win the next election.

by saying (2)

I think the Brazilian government has devaluated the Cruzeiro again (Searle 1992, 10)

we are giving a response whose inappropriateness is due to locutionary (or propositional) features both of the response and of the previous speech act: it is the (asserted) content of our response which is not appropriate (relevant, etc.) to the (asserted) content of the previous utterance.² Apart from this, to issue an assertion as a response to an assertion, that is, to add a piece of information to a piece of information (either in a cooperative or in a competitive way, which may

² [Here I take Searle’s examples to be cases of assertion, following his own reading of them.]

48

 , ,  

involve change of topic) is a way of responding which is wholly compatible with the implicit uptake of the former speech act as an assertion. However, it is not the case that every response to an assertion takes it as an assertion. Imagine responses to (1) in a more or less Alice-in-Wonderland style, such as: (3)

You should not think.

or (4)

But do you hope they will?

These are responses that do not take the assertion as an assertion and thus are inappropriate in the illocutionary sense. When the response to a would-be assertion in a certain speech situation approximates itself to these examples, it would be incorrect to claim that the assertion has been successfully performed: the utterance does not “take effect” as an assertion, namely, it does not commit the speaker to the truth of the expressed proposition, so that the speaker is not obliged to give evidence or reasons for it, and the hearer her/himself is not made competent to issue further assertions on the same topic (that is, s/he would not tell someone else that the Republicans will win) on the sole ground of her/his having listened to that utterance.

6. From Inappropriateness to Shifts in Illocutionary Force If a would-be assertion does not “take effect” as an assertion, however, what does it “do” (in the illocutionary sense)? Does it have any illocutionary effect, and which one? While in formal situations infelicitous speech acts do not seem to get any other illocutionary force apart from the infelicitous (and thus ineffective) one, so that infelicitous bets remain mere attempts at betting, infelicitous christenings remain mere attempts at christening, and so on, in informal talk an utterance which could have a certain illocutionary force but does not get a response appropriate to it, may “take effect” in a different way, according to the uptake which is made manifest by the actual response. Thus, in a dialogue such as: (1a)

A: I think the Republicans will win the next election. B: Why do you think so?

the speaker is asked for reasons for her/his assertion, and this makes it apparent that, according to the hearer, the speech act is an assertion and “takes effect” in the way assertions do. On the contrary, in a dialogue such as:

 , ,   (1b)

49

A: I think the Republicans will win the next election. B: But do you hope they will?

what is at issue is not a judgment to the effect that the Republicans will (probably) win, but the expression of subjective attitudes (opinion, hope . . . ) with respect to the same possible fact: this makes the utterance count as an expressive speech act rather than as an assertion. It should be noted that assertions introduced by prefixes such as “I think” are hedged judgments, and thus have a tendency to oscillate between the force of real assertions and that of Expressives. Thus, it is perhaps true that for such utterances every response may count as an appropriate response. Nevertheless, some responses are more appropriate with respect to an assertive illocutionary force, while others emphasize the expressive features of the speech act. Moreover, the phenomenon described here is not confined to such hedged speech acts. Consider: (5)

It is raining.

together with each of the following responses: (6)

Is it windy too?

(7)

Don’t go out.

(8)

All right, I’ll take my umbrella.

Response (6), asking the speaker for further information, takes it for granted that the preceding speech act has given the hearer a piece of information (namely, that it was an assertion). Response (7), advising the speaker not to go out, indicates that the preceding comment about the weather has been taken by the hearer as a reaction to the situation and in particular as an expression of perplexity about what to do in that situation: thus, (5) is viewed as belonging to a kind of speech act roughly corresponding to the Austinian class of Behabitives, and related to that of Expressives. Response (8) shows that the hearer aligns her/himself with what here counts as a warning (an exercitive illocutionary act in Austin’s terminology and perhaps a Directive as well, since it tries to and succeeds in getting the hearer to modify her/his course of action).

7. What Happens in an Adjacency Pair In this perspective it is not relevant to discuss whether, in order to identify in actual conversation such “adjacency pairs” as question–answer, greeting–greeting, and proposal–acceptation or refusal, we need an independent definition of the

50

 , ,  

speech acts which constitute their first pair-parts, namely question, greeting, and proposal. We have a question or a proposal, and the like, when a speech act is issued and accepted, by speaker and hearer, respectively, as being a question or a proposal; but we can understand whether the hearer accepts it as such only if we take into consideration her/his response. Conversely, if we understand that a certain utterance counts as a (preferred or dispreferred) second pair-part of an adjacency pair of a certain kind, we have also to understand that it has a connection with the corresponding first pair-part and what this connection is. That is, we have to understand which is the effect, produced by the first pair-part, that affects the situation in which the second pair-part is uttered and which this second pair-part reacts to. Thus, in a sense, it is correct to say that each speech act creates a space of possibilities of appropriate response speech acts: but this should be taken to mean that each speech act, by achieving an uptake, produces a corresponding effect which counts as the point of departure, or the initial situation, of the response act, while the response act makes manifest how the speech act has been taken and thus in which way the speech act “takes effect.” Only response acts that count as appropriate responses (in the illocutionary sense) to a certain illocutionary act can make it the case that the preceding speech act counts as an illocutionary act of just that kind. It should be noted that these “appropriate” responses may be preferred or dispreferred ones (in terms proper to speech act theory, they may satisfy or not satisfy the perlocutionary goals of the speaker); what is relevant is that they presuppose (or at least are compatible with) some illocutionary effect, and thus validate its achievement.

8. Speech Acts in Conversation What should then be the relationship between speech act theory and the study of conversation? It seems that the two have to be interrelated, though, of course, Searle is right in pointing out that some methods proper to speech act theory do not fit the study of conversation. For example, a speech act theorist should resist the temptation to extend her/his own idea of the “felicity conditions” for illocutionary acts to aspects of language different from illocution. And conversation is obviously not itself a kind of illocution. If this is what is meant by saying that we cannot get a theory of conversation that matches our theory of speech acts, this conclusion has to be accepted as obvious. However, a theory of conversation could “match” a theory of speech acts in more than one sense. We might agree, for example, that conversation is made up of speech acts producing, among other things, illocutionary effects, and we might investigate, whether different genres of discourse are characterized by different kinds of illocutionary effect (in such an investigation the distinctions drawn by Austin, who devotes more attention than Searle to the deontic features of illocutionary effects, are of easier application:

 , ,  

51

cf. Austin 1975, 151–63; Sbisà 1984a [Essay 1], 1984b, 1987a). This work is very different from the work of conversation analysts, but is not incompatible with it. Moreover, conversation analysts have often been concerned with the deontic aspects of conversational activity, namely, with the rights and the obligations of the participants toward each other, with respect to conversational turn-taking, keeping the floor, interrupting and overlapping of turns, and the like. Maybe these deontic aspects could be compared with the deontic aspects of the illocutionary effects of speech acts: such a comparison might help us to understand the so-called “expositive” illocutionary acts, which seem to be similar to other illocutionary acts but also different from them all (cf. Austin 1975, 152; Searle 1979, 6), and conversation could turn out to be a kind of speech activity which is characterized by the bringing about of certain illocutionary effects of the expositive kind. The analogy to be exploited is not between turn-taking rules (or other conversational rules) and the constitutive rules of speech acts, but between the obligations and the rights assigned to the participants by the fact that one of them is speaking (or is arriving at a “transition point,” or is ceasing to speak altogether . . . ) and the obligations and rights assigned to speaker and hearer by the intersubjectively validated performance of illocutionary acts. This analogy is still to be explored.³ It could be added that, though constitutive rules are an important tool for theoretical work in the field of language use, they are not the only tool to use for every kind of investigation. Thus, we should not feel obliged to either adopt constitutive rules as our unique tool or reject them altogether. Moreover, constitutive rules are not self-explanatory. If we want to build up a dynamic image of how verbal interaction works, we shall have to investigate how it is that rules (both constitutive and regulative, or belonging to other possible kinds) can be accepted, shared, recognized, violated without rejection, or rejected, and here such assumptions as cooperativity and relevance come into play. We shall not tackle the methodological problem of the way in which these assumptions should be used in theory and analysis, but perhaps it is relevant to the aims of this paper to suggest that notions such as cooperativity and relevance should also allow for degrees of intensity and for intersubjective negotiation. It is not the case that such assumptions are always made by the participants in actual interaction. Nevertheless, the fact that they are made (or possibly that some conflicting assumptions, also related to the attitudes of the participants toward each other, are made instead) plays a role in the construction of the intersubjective relationship, and perhaps in the processes leading to the identification and the acceptance of shared rules or even to the identification of each other’s background. The description of this role is another task on which the study of conversation and the theory of speech acts could usefully cooperate; the emphasis posed by Searle ³ [Interesting explorations in this direction have been carried out by Mary Kate McGowan in her work on Conversational Exercitives (2003, 2004).]

52

 , ,  

on the background (which affects both the understanding of individual speech acts and of conversational sequences) may be taken as a suggestion in this direction, though one would hope that his analysis of the connections of the background with the purposes of the participants and with conversational relevance, which leads to denying the possibility of intercultural communication, would not be considered as final. However, apart from these hints at the possible ways in which speech act theory could contribute to or cooperate with the study of conversation, my main suggestion remains that the right direction to take in order to bring together the study of speech acts and the study of conversation is to go from conversation to individual speech acts, and not the other way round: it is the investigation of conversational sequences that could throw some light on the investigation of individual speech acts, and in particular of the various kinds of effects they bring about.

3 Ideology and the Persuasive Use of Presupposition 1. Introduction Pragmatic presuppositions have been defined as assumptions shared by the interlocutors, which form the background of their ongoing discourse (Stalnaker 1973, 1974). Some if not all of these shared background assumptions have linguistic markers: thus Stalnaker (1973) (followed, among others, by Soames 1989) has spoken of presupposition requirements, while others have spoken of presupposition inducers or triggers (Levinson 1983; van der Sandt 1988), referring to those lexical items, morphological devices, or syntactic constructions the use of which may require or activate a pragmatic presupposition.¹ Inasmuch as this paper is an exercise in the pragmatic analysis of discourse and an attempt to reflect more deeply upon the theoretical concepts which make this possible, it will take into consideration only linguistically marked presuppositions. Therefore, I shall speak of presuppositions of utterances (Soames 1982; van der Sandt 1988) rather than of presuppositions of speakers. However, according to the received view of pragmatic presupposition, whenever an utterance has a presupposition, its speaker may be said to make that presupposition. It has long since been observed that presuppositions may have informative uses (Karttunnen 1974; Stalnaker 1974). In other words, when an utterance contains a presupposition inducer, the required presupposition still may not be included among the beliefs shared by the interlocutors and, in particular, it may be new information for the listener, who will “accommodate” the presupposition by adding it to the shared background beliefs (Lewis 1979). Such informative uses of presuppositions are frequent in the press, at least in Italy. A discussion of some examples drawn from the Italian daily papers will show how presuppositions are used to convey information. It will also become apparent how the received Originally published in Language and Ideology. Selected Papers from the 6th International Pragmatics Conference, ed. Jef Verschueren, 492–509. Antwerp: International Pragmatics Association (Copyright: IPrA and the Authors), 1999. ¹ It should be noted that the investigation of presuppositional phenomena had its starting point in reflections on some of these linguistic devices (Fillmore 1971; Kiparsky and Kiparsky 1971; Strawson 1950a), thus highlighting, initially at least, certain semantic aspects of presuppositional phenomena rather than pragmatic ones.

Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0004

54

      

distinction between presupposition proper and informative presupposition does not reflect the reality of social communication processes, in which more subtle distinctions are involved. This paper will focus on the fact that when what is presupposed has to do with values, social norms or ideals, or with perspectives on facts which are proper to a specific social agent, then informative presuppositions seem to serve persuasive aims. It will tackle the question of whether the received notions of pragmatic presupposition and of presupposition accommodation can account for the persuasive use of informative presupposition, and will argue that they cannot; a revised account of presuppositional phenomena will be put forward which can explain their persuasive function. On the basis of the proposed description, it will be argued that presupposition is suitable for transmitting a kind of content which might be called ideological: assumptions, not necessarily conscious but liable to be brought to consciousness, about how our human world is and how it should be. By the same token, criticism of ideology may avail itself of a knowledge of linguistic presupposition inducers and of their communicative dynamics.

2. Informative and Persuasive Uses of Presupposition in the Italian Daily Press The headlines in Italian newspapers² often contain linguistic presupposition inducers. It may be the case that the triggered presupposition is satisfied by the assumptions shared by the headline’s author and the reader, but in many cases, this is not at all certain. The reader may never have read a certain previous issue of the newspaper in which the information, now presupposed, was explicitly stated. Or he or she may have forgotten about that piece of information. In principle, it is also possible that the information concerned was never actually stated explicitly in any issue of that newspaper. It is therefore sensible to consider all presuppositions which convey a piece of information not explicitly stated in the same newspaper issue as at least partly informative. Consider the following headline: (1)

Anche Scalfaro dà l’ultimatum a Bossi. “Scalfaro too delivers an ultimatum to Bossi.” (La Stampa 09/08/1997, 5)

² The examples quoted in this paper are taken from La Stampa, Il Corriere della Sera and Il Giornale. The issues of September 1–15, 1997 and December 15–31, 1997 have been consulted thoroughly. Material from other issues or from other daily newspapers (e.g., Il Piccolo) was also examined, but apart from one or two exceptions, has not been quoted.

      

55

The presupposition inducer “anche” (“too”) triggers the following presupposition: (1a)

Someone other than Scalfaro has delivered an ultimatum to Bossi.

Scalfaro is the President of the Italian Republic and Bossi is the leader of the Lega Nord (“Northern League”), a party proposing federalism or even the independence of Northern Italy. If the reader knows that some days earlier, the Italian premier Romano Prodi had threatened to take Bossi to court if he insists in his claim that the electoral consultation his party is organizing in Northern Italy represents the election of a real Northern Italian autonomous Parliament, then the presupposition does belong to the set of assumptions shared by the author and the reader. But nothing is explicitly stated elsewhere in the same issue of the paper so as to ensure that this information is readily available to any reader before he or she reads the headline in (1). So, the presupposition is likely to function as an informative one and, in the context of that issue, can be counted as such. It is interesting to notice that the generic information conveyed by the headline’s presupposition can be integrated by information conveyed by the article that follows it (which, however, is usually read after and not before the headline). In fact, in the article we find: (2)

Scalfaro plaude a Prodi e al governo per il richiamo sulle elezioni padane. “Scalfaro praises Prodi and the government for the reprimand about the Po Valley elections.”

Here again, however, the information that Prodi and the government have reprimanded someone about the Po Valley elections is not explicitly stated but encoded in a definite description (“il richiamo sulle elezioni padane,” “the reprimand about the Po Valley elections”), the most classic of all presupposition inducers. Our analysis of this example amounts to saying that a reader who is not quite sure who has given an ultimatum to whom about the Po Valley elections, is informed by (1) and (2)—thanks to the presuppositions they trigger—that the premier Prodi, prior to Scalfaro’s ultimatum, has reprimanded Bossi about the Po Valley elections in a manner which can be considered equivalent to an ultimatum. Informative presuppositions play a prominent role in articles containing scientific information or involving reference to scientific knowledge. Consider: (3)

L’Unione Europea era arrivata a Kyoto con la richiesta più avanzata: una riduzione dei gas nocivi che producono l’“effetto serra” (principalmente diossido di carbonio) del 15% rispetto ai livelli del 1990.

56

       “The European Union had come to Kyoto with the most radical request: a reduction in the harmful gases which produce the ‘greenhouse effect’ (mainly carbon dioxide) by 15% with respect to the levels of 1990.” (La Stampa 12/10/1997, 8: “Un giorno per salvare la Terra” “A day to save the Earth”)

Here, whether or not the reader has already been informed and remembers that the so-called “greenhouse effect” is produced by certain gases which are therefore harmful, he or she can gather this information from the presupposition triggered by the definite description “i gas nocivi che producono l’‘effetto serra’ ” (“the harmful gases which produce the ‘greenhouse effect’ ”).³ Sometimes, articles of political commentary use presupposition to activate assumptions they do not argue for: (4)

Tutti i partiti ex cattolici sembrano assolutamente incapaci di dare una risposta a Giovanni Paolo II . . . sulle “emergenze” nazionali che egli ha individuato . . . : politica a sostegno della famiglia, scuola libera e creazione di lavoro vero e stabile. Neanche si accorgono che D’Alema, su questi temi, tenta di scavalcarli . . . “All the ex-Catholic parties seem quite incapable of giving an answer to John Paul II about the national ‘emergencies’ he has indicated: policies in support of the family, parity between state and private schools, and creation of jobs which are real and stable. They do not even realize that D’Alema, on these issues, is trying to get there before they do . . .” (Il Giornale 09/09/1997, 1: “La mosca cocchiera di D’Alema” “The fly that pretends to drive D’Alema’s coach” by Antonio Socci)

Here the information that D’Alema, leader of the Democratic Party of the Left, is trying to supplant the traditional role of the Catholic parties with respect to some political issues singled out by the Pope, is conveyed as presupposed, since it is embedded by the factive verb “accorgersi” (“realize”). Often, however, what is presupposed by a newspaper headline or article has to do with values, social norms or ideals, or with perspectives on facts which are proper to a specific social agent. In September 1997, the daily press used different definite descriptions to refer to the electoral consultation organized by the Northern League for October 26: we find, for example, (5)

le annunciate “elezioni” padane “the announced Po Valley ‘elections’ ” (Il Corriere della Sera 09/04/1997, 1)

³ [In my pragmatic approach to discourse analysis, I count Italian plural noun phrases with the definite article as presupposition inducers similar to definite descriptions. See on this Sbisà 2021a, 181.]

       (6)

le elezioni del 26 ottobre “the elections of 26 October” (La Stampa 09/04/1997, 5)

(7)

il voto padano “the Po Valley vote” (Il Corriere della Sera 09/08/1997, 1)

(8)

le elezioni-sondaggio di ottobre “the opinion poll-elections of October” (Il Giornale 09/12/1997, 7)

57

The presuppositions triggered by these definite descriptions are somewhat different from each other and reflect the different attitudes of the newspapers or of the authors of the articles toward the political role of the alleged elections: (5a)

Something has been announced that someone is calling the Po Valley “elections.”

(6a)

There will be elections on October 26.

(7a)

There is or will be a vote in the Po Valley.

(8a)

There will be opinion poll-elections in October.

(5a) and (8a) do not entail there being elections at all. (6a) and (7a) simply say that there will be elections, and although (6) and (7) are embedded in contexts which under one reading at least would filter their presuppositions, their most direct impact on the reader tends to activate them. In these examples, it should be noted that a suitably chosen definite description is used in place of an assertion or even an argument to the effect that the Po Valley elections are, or are not, real elections. Since different political evaluations are associated with the two opposite assumptions, it is manifest that the use of a given definite description is not merely informative, but has persuasive aims: in particular, there is the aim to get the reader to believe that the Northern League really is going (or is not really going) to make the people of Northern Italy elect their own autonomous Parliament. Another interesting case in which informative presuppositions shift toward being persuasive is the following: (9)

La terapia anticancro ideata dal professore modenese Luigi Di Bella segna un altro clamoroso punto a favore.

58

       “The anti-cancer treatment invented by Luigi Di Bella, the professor from Modena, scores another amazing goal in its own favor.” (Il Giornale 12/17/1997, 1: “Il pretore ordina: usate l’anticancro Di Bella” “Order of the judge: use the Di Bella anti-cancer treatment”)

An inspection of the issue containing this article and of the preceding issues of the same newspaper authorizes us to consider the presuppositions triggered by the definite description “la terapia anticancro ideata dal professore modenese Luigi Di Bella” (“the anti-cancer treatment invented by Luigi Di Bella, the professor from Modena”) and by the iterative “un altro” (“another”) as informative ones. The Di Bella case had been absent from the press for months, and had in any case never received a great deal of attention. So, readers were very likely to know nothing about it. Newspapers use various strategies to cope with this problem, ranging from inserting a brief summary of the previous stages in the story into the middle of the article (an explicitly assertive strategy), to drawing largely on informative presuppositions and on that feeling of familiarity with the presupposed news which they are able to induce. A strategy of the first kind is used by La Stampa, where, however, the explicit report of the previous steps is introduced by a presuppositional expression—note the iterative “riaprire” (“reawaken”): (10)

La . . . decisione è destinata ora a riaprire le polemiche sulla utilità del metodo del professor Di Bella. “The decision is now destined to reawaken the controversy surrounding the usefulness of Professor Di Bella’s method.” (La Stampa 12/17/1997, 13)

A strategy of the second kind is the one used by Il Giornale in the article from which example (9) was drawn. There, the following presuppositions are triggered: (9a)

An anti-cancer treatment exists which was invented by Professor Luigi Di Bella from Modena.

(9b)

The Di Bella anti-cancer treatment has already scored some amazing goals in its own favor.

which provide the reader with pieces of information he or she is unlikely to have and create the impression that there is an ongoing story which began some time ago, and which also try to get the reader to take it for granted that Di Bella has invented an anti-cancer treatment and that the goals scored in its favor are amazing. In fact, Il Giornale was about to launch a campaign in favor of the Di

      

59

Bella therapy, and the persuasive use of informative presuppositions in this article was just a first step. Another use of presuppositions which I take to be aimed at persuasion consists of conveying accusations or criticism without spelling them out explicitly. This use is especially frequent in Il Giornale, a right-wing newspaper clearly opposed to the government. Thus, in the same issue from which example (9) is drawn, we find: (11)

La terapia del professor Di Bella, boicottata e osteggiata dal ministero della Sanità, . . . finisce in Parlamento. “Professor Di Bella’s treatment, which has been boycotted and opposed by the Ministry for Health . . . , ends up in Parliament.” (Il Giornale 12/17/1997, 10: “Polo e Ulivo al parlamento: basta ostracismi al farmaco” “Polo and Ulivo to the Parliament: stop ostracizing the drug”)

Here an accusation levelled at the government (that it is boycotting Di Bella’s treatment) is included in a presuppositional construction. Another example of the same strategy, drawing on the use of a factive verb, is the following: (12)

Eugenio Scalfari . . . vi ha avvisati: lasciate perdere le soluzioni forti. . . . Ma so che non lo ascolterete e non ascolterete neppure me, perché un anno e mezzo al potere vi ha drogati. “Eugenio Scalfari has warned you: forget about strong solutions. But I know you won’t listen to him or to me either, because a year and a half in power has intoxicated you.” (Il Giornale 9/4/1997, 1: “Gli ercolini gonfiano i muscoli” “The little Hercules show off their muscles”)

In consideration of the fact that the author addresses himself to the politicians presently in power, the triggered presuppositions can be expressed as follows: (12a)

The politicians presently in power will not listen to Eugenio Scalfari or to the author of the article.

(12b)

The politicians in power are intoxicated by a year and a half in power.

(12c)

The politicians in power have been in power for one and a half years.

(12d)

The fact in (12b) is a cause of the fact in (12a).

The author does not level an explicit accusation at the politicians in power, but takes the opportunity to presuppose that they are intoxicated by power by

60

      

embedding his explanation of their behavior as a complement clause of the factive verb “sapere” (“know”). Of course, (12b–12d) count as presupposed if the causal subordinate clause “perché (‘because’) . . .” is considered as subordinated to “non lo ascolterete e non ascolterete neppure me” (“you won’t listen to him or to me either”) and not directly to “so” (“I know”). But that is the most natural reading: the causal clause seems to explain why the politicians will not listen to the two journalists, not why it is that the author knows that they will not listen. The author clearly feels hostility for the government coalition. It is likely but not at all certain that his readers already share his feelings. But, in a way, those readers who happen not to share them are the main target of the author’s persuasive aims. Moreover, even in the case of readers who do share them, their hostile feelings for the government are reinforced by being provided with new input. Finally, the persuasive use of presuppositions is manifest in political discourse, when a politician addresses the citizens directly in order to gain their approval, and can often be found in the quotations from such discourses reported by the daily press. There are clear examples in some statements attributed to Umberto Bossi (leader of the Northern League): (13)

Sono momenti in cui è in atto una riorganizzazione, una contrapposizione tra chi vuole mantenere Roma come baricentro dello sfruttamento della Padania e i patrioti padani. “These are times when there is a reorganization going on, a contraposition between those who want to keep Rome as the center of gravity for the exploitation of Padania and Padania’s patriots.” (Il Corriere della Sera 09/04/1997, 11)

Some definite descriptions (as well as the relative clause “chi ‘those who’ . . .”), the proper name Padania,⁴ and the use of the verb “mantenere” (“keep”) trigger the following presuppositions: (13a)

Some people want to keep Rome as the center of gravity for the exploitation of Padania.

(13b)

Rome is the center of gravity for the exploitation of Padania.

(13c)

Padania is being exploited.

(13d)

Padania exists.

⁴ “Padania” is a proper name (geographical areas take the definite article in Italian). Therefore, its function is to point to an object which is taken to exist.

       (13e)

Padania’s patriots exist.

(13f)

Padania is a fatherland.

61

Belief in these assumptions (first of all, belief in the existence of Padania, presently not even a clearly defined region of Italy, and in its being a fatherland which has its patriots) is precisely what Bossi needs to elicit in order to gain wider approval for his policies. Of course, reported statements are usually formulated so as to present the quoted politician’s political line in such a way as to be functional to the political line of the newspaper. For example, Il Corriere della Sera seems to believe that Bossi is dangerous and that some of his attitudes are illegal and/or irresponsible, and the above quotation may be used to support this evaluation. Both presuppositions and, as we shall see, the omission of presuppositions are prominent moves in the persuasive strategies of the press. I shall close this overview on informative and persuasive presuppositions in the Italian press with the following example of a declaration attributed to premier Prodi: (14)

L’unità morale e civile della nazione è radicata e resa salda anche nella vita religiosa e nell’appartenenza cattolica. “The moral and civil unity of the nation is also rooted in, and held fast by, religious life and belonging to the Catholic faith.” (La Stampa 9/09/1997, 1, “Prodi: la fede fa l’Italia unita” “Prodi: the faith makes Italy united”)

In this declaration, the definite description “l’unità morale e civile della nazione” (“the moral and civil unity of the nation”) and the adverb “anche” (“also”) induce the following presuppositions: (14a)

There is a moral and civil unity of the nation.

(14b)

The moral and civil unity of the nation is rooted in and held fast by something else besides religious life and belonging to the Catholic faith.

In the context of the debates taking place at the time about the secessionist aims of the Northern League, presupposition (14a)—although not exactly new information—has clearly persuasive functions: (14) takes one of the theses which is being debated for granted, by presupposing it. Presupposition (14b) should be shared knowledge for all Italians who have studied some history of their country; however, opinions differ about what exactly can be said to make for the unity of the Italian nation; so that (14b) is at least in part informative. Moreover, it is

62

      

relevant that it is presupposed by the premier, Prodi, a Catholic himself, but leading a left-wing coalition. Thus, one informative contribution conveyed by this presupposition is that Prodi believes that there are forces other than Catholicism which hold fast the unity of the nation. That is, by presupposing (14b) Prodi has implicitly recognized the role of lay society with respect to the unity of the nation. But it should also be noted that the speech concerned was made in a Catholic context, where its presupposition (14b) was likely to have the persuasive effect of getting Catholics to recognize—and moreover, recognize as obvious—that there are lay moral values in Italian society. The newspaper omits “anche” (“also”) from the headline “Prodi: la fede fa l’Italia unita” (“Faith makes Italy united”) and by so doing modifies the overall meaning conveyed by the statement quoted. This enables a political commentator to write on the same page: (15)

Siamo uniti, ha detto sostanzialmente Romano Prodi a Loreto, perché siamo cattolici. “Romano Prodi basically said in Loreto that we are united because we are Catholics.” (La Stampa 9/09/1997, 1: Sergio Romano, “Se l’Ulivo diventa guelfo” (“If the Ulivo becomes a clerical party”)

where the focusing adverb “sostanzialmente” (“basically”) is hardly sufficient to make up for the omission of “anche” (“also”), with its associated presupposition (14b), and where the elimination of this presupposition paves the way for an attack on the Ulivo coalition, at risk of being excessively conditioned by Catholicism.

3. Some Doubts about the Received Account of Pragmatic Presupposition Do the received notions of pragmatic presupposition and of presupposition accommodation allow a satisfactory account of the uses of presupposition illustrated above? Firstly, I would argue that presupposition proper and presupposition accommodation should not be given two separate accounts, the one correcting and integrating the other. There is some continuity between the two phenomena, and when we are faced with actual samples of discourse containing presupposition inducers, it is hard to tell whether the triggered presuppositions should be counted as satisfied (that is, as part of a set of shared assumptions) or not (and therefore, as informative presuppositions to be accommodated). In fact, the task of a receiver when faced with a text which induces a presupposition does not differ from one

      

63

situation to the other: he or she is required to identify the presupposition and take it for granted, whether he or she was already acquainted with it or not. As regards the supposedly shared status of presuppositions, we should distinguish at least between (a) cases in which the hearer holds the presupposed assumption and the speaker knows this to be so; (b) cases in which the hearer does not hold the presupposed assumption, but the speaker thinks he or she does; (c) cases in which the speaker does not know whether the hearer holds the presupposed assumption or not; (d) cases in which the speaker thinks that the hearer does not hold the presupposed assumption, while in fact he or she does; (e) cases in which the speaker thinks that the hearer does not hold the presupposed assumption, and he or she in fact does not. In the cases in (d) and (e) the speaker intends to make an informative use of the presupposition, but only in case (e) does he or she succeed. In the cases in (b), the presupposition turns out to be informative beyond the speaker’s intentions. In the cases in (c), it is hard to tell whether the use the speaker makes of the presupposition should be counted as informative or not, since it may be either. This case is undoubtedly very frequent among our examples from the press: a journalist can never be sure which of his or her assumptions are already shared by the reader, and in addition has often to cope with an audience which is likely to be split, partly sharing and partly not sharing his or her assumptions. Therefore, it is apparent that we should not give separate accounts of presupposition proper and of informative or persuasive presupposition. A good account of presupposition should be extendable without modification to informative presupposition, as well as to the intermediate cases. Secondly, I would like to recall some of the arguments against the received notions of pragmatic presupposition and of presupposition accommodation put forward by Gauker (1997, 1998). Gauker is concerned with a problem different from the one discussed here, namely with defining the “context” of the utterance, that is, that set of propositional elements on which the evaluation of the utterance of a sentence depends. He concludes that the context cannot be defined in terms of shared assumptions and that one should adopt two distinct concepts: that of objective context and that of the participants’ take on the context. It is the objective context which must be taken into consideration in order to evaluate the appropriateness of utterances, while it is the participants’ take on the context which is modified by what participants tell each other. On his way to establishing his thesis, he discusses informative presupposition and claims that no satisfactory account of it has been given so far (Gauker 1997, 204–5; 1998). The “accommodation rule” proposed by Lewis (1979), according to which a presupposition comes into existence whenever that is required for the acceptability of what is said, is just the formulation of the problem, not a solution of it: one would like to understand how it happens in this way. Stalnaker (1973, 1974) has claimed that in informative presuppositions the speaker is pretending to presuppose something.

      

64

Against him, Gauker objects that the point of an utterance like the well-known example by Karttunen (1974): (16)

We regret that children cannot accompany their parents to the commencement exercises.

is just to inform that children cannot accompany their parents; if the speaker were acting as if everyone already knew that, he or she would not utter the sentence at all. Soames (1982), in his definition of utterance presupposition, has used the notion of “taking something as uncontroversial” and, in order to cope with informative presupposition, has analyzed this notion as equivalent either to thinking that the presupposition is a shared assumption or to thinking that the audience is prepared to add the presupposition without objection to the context against which the utterance is evaluated. To him, Gauker objects that this would make informative presupposition indistinguishable from unobjectionable information, while, in fact, an utterance conveying an informative presupposition such as (16) may well raise objections; it is therefore at least unclear in which other sense the audience should be prepared to add the presupposition to the context without objection. Besides, it should be noted that the notion of presupposition accommodation as used in Heim (1983) and even more in the anaphoric theory of presupposition (Beaver 1997; van der Sandt 1992) shifts from Lewis’s original pragmatic notion to being a device for making sure that the presuppositions triggered by an uttered sentence are located in their due place in the representation of discourse. Thus, accommodation becomes a general tool for the logico-linguistic description of discourses containing presupposition inducers. There are no longer two separate explanations for presupposition proper and informative presupposition. But the pragmatic aspects of presupposition accommodation fade away, and the question of why discourse should function like this is not raised at all. By now, we may conclude that the received notions of pragmatic presupposition and of presupposition accommodation do not account for the phenomenon of informative presupposition satisfactorily, and therefore cannot account for the persuasive uses of presupposition either.

4. Toward an Account of the Persuasive Use of Presupposition I would now like to make a fresh start by focusing on the persuasive use of presupposition. The existence of such a use raises some questions which may emerge as being central to an account of presupposition: why should presuppositions ever be persuasive? Why should the audience not only identify the presuppositions triggered by an utterance but also agree to take them for granted? Why is

      

65

there a default tendency to do so (unless there are special reasons for rejecting the presuppositions)? I think that one answer to these questions may be couched in the following terms. I propose considering presuppositions not as shared assumptions, but as assumptions which ought to be shared. Thus we allow them a normative feature. Such an admission is consistent with early insights about presupposition such as Strawson’s (1950a). There, presupposition (particularly, the existential presupposition associated with a definite description used as the subject of an assertion) is required in order for the assertion which presupposes it to have a truth-value. One might say that when the context fails to satisfy the presupposition, the assertion violates one of its most fundamental norms: the one according to which it must have a truth-value. Actually, Austin seems to have read Strawson in this way, since he includes presupposition among the felicity conditions of assertive illocutionary acts (1975, 137). Ducrot too (1972) highlighted the normative or “deontic” factors in presupposition, although drawing the disputable conclusion (criticized in Sbisà 1979), that presupposition itself is an illocutionary act. In the recent contributions by Gauker mentioned above, normative factors emerge again. According to Gauker (1998), conversations are governed by objective contexts (that is, by sets of propositional elements, factually constrained and connected with the goals of the ongoing interaction, against which utterances are evaluated), since certain norms of discourse that interlocutors ought to adhere to are formulated in terms of these contexts. It seems to follow from this that an utterance should be regarded as in some way inappropriate or out of order if the speaker, in uttering it, does not take the objective context into due consideration. In this vein, I would like to claim that whenever a hearer takes it that an utterance by a certain speaker presupposes something (because it contains a certain presupposition inducer), if the hearer takes the objective context not to contain the presupposed propositional element, he or she will be bound to consider the speaker not only as being wrong about the facts (as occurs when somebody says something factually false) but also as violating some norm of discourse. Violating norms of discourse may in turn be deemed a kind of uncooperative behavior: it is in fact a kind of behavior which makes it difficult to continue conversational cooperation. How can we reply to a conversational contribution whose presuppositions are not satisfied by the context? To use Strawson’s words (1950a, 12), the question doesn’t arise; we may find it difficult to continue the exchange at all. One solution which might preserve the communicative relationship is to initiate a discussion on those features of the objective context about which the participants disagree. This solution is laborious, because it involves a change of topic from what was explicitly at issue to what was merely presupposed, as well as being risky, because it amounts to openly challenging the entitlement of the speaker to issue the utterances he or she has issued, which may once again lead to a breakdown in the communicative relationship (why should

66

      

anyone talk to someone who says things he or she is not entitled to say?). Since this breakdown in the communicative relationship may even damage the participant who has not violated any norm by leaving him or her deprived of a source of information (though dubious at times), and of a certain amount of cooperative resources (though not wholly reliable), it is obvious that, whenever possible, the hearer will avoid treating the speaker as someone violating norms of discourse. In all the cases in which there is some possibility left that the objective context does contain the presupposition, the default tendency will be to assume that it is so, namely, that the presupposition is in fact satisfied. It should be pointed out that default rules or defeasible reasoning strategies have often been included in accounts of presupposition. A default tendency to accommodation has been noted by Lewis (1979); in the context of the “projection problem,” that is, the problem of calculating the presuppositions of compound sentences, Gazdar (1979) has held that presuppositions are projected by default, unless there are reasons for cancelling them, while Soames (1982) has spoken of “defeasible presumptions” which are created by speakers who utter compound sentences one of whose constituents contains a presupposition inducer; van der Sandt (1988, 1989) has dealt with presuppositions as default inferences; in more recent logico-linguistic theories of presupposition (Beaver 1997; van der Sandt 1992; but see already Heim 1983) there is the default preference for the so-called “global accommodation” of the presupposition, which collocates the presupposition in the global context of the discourse, over its “local accommodation,” which guarantees the acceptability of the utterance without requiring the global context of the discourse to contain the presupposed propositional element. Gauker (1998) has claimed that there is a default tendency to assume that other participants presuppose whatever really does belong to the objective context; his claim is different from the one put forward here (which is, rather, that there is a default tendency to assume that what is presupposed by the utterances of other participants really belongs to the objective context), but in the last resort the two claims are compatible, since both are rooted in the normative character of the relationship between utterance and context. From all of this, we may conclude that the persuasive use of presupposition depends on certain normative or deontic features that presuppositional phenomena have. The utterance of a given sentence is really appropriate only if the objective context really has a certain content. It is among the speaker’s responsibilities to make it so, that his or her utterance is appropriate. It is among the speaker’s responsibilities to issue an utterance only if it is appropriate when the objective context is as it happens to be. Moreover, it is among the speaker’s responsibilities to issue an utterance containing certain presupposition inducers only if the objective context really contains the presuppositions they trigger. Thus, we are describing presuppositions as assumptions that the speaker ought to make, or, however, assumptions for which he or she is responsible. But in order to give

      

67

an appropriate evaluation of the speaker’s utterance, the hearer too ought to make the same assumptions, not because the speaker makes them, but because they are the assumptions which the speaker ought to make. Thus, we are describing presuppositions as assumptions which ought to be shared. Although this description could be extended to those presuppositions of the speaker which are not linguistically marked, but can be inferred by the hearer from the objective context on the basis of the assumption that the speaker has (as he ought to) a correct take on the objective context (Gauker 1998), in our present investigation we are applying it primarily to linguistically marked presupposition: those presuppositions of an utterance which are triggered by linguistic presupposition inducers. When faced with an utterance containing a presupposition inducer, and knowing that the speaker ought to issue it only if the context contains the triggered presupposition (provided he or she wants to keep on communicating with the speaker, and unless there are specific reasons not to do so), the hearer will hold that the presupposition of the speaker’s utterance is contained in the objective context. This can be taken as a generally valid description of the hearer’s position with regard to the (linguistically marked) presuppositions of the speaker’s utterance. Thus, the phenomenon giving rise to the persuasive use of presupposition turns out to be quite general. We have perhaps reached a description of the dynamics of presuppositional phenomena which holds good irrespective of whether the presupposed propositional elements are old or new information for the hearer, and more generally, irrespective of whether they initially belong to the speaker’s take on the context, to the hearer’s, to both, or to neither.

5. Presupposition and Ideology The view of presupposition which emerges from the preceding discussion is that of a communicative device for constructing the participants’ takes on the context, the functioning of which is guided by the underlying normative character of the objective context. Presupposition, so intended, is clearly suitable for transmitting a kind of contents which may be called ideological: assumptions, not necessarily conscious but liable to be brought to consciousness, about how our human world is and how it should be. For various reasons, such assumptions are often difficult to be certain about, even though they are not in general unverifiable. Since asserting and arguing commit the speaker to giving evidence or reasons for what is asserted or argued for, it may be difficult, or uncomfortable, to assert or discuss this kind of assumptions explicitly. However, they manage to play a prominent role in political or ethical choices or in the practical adoption of a way of life, so that speakers may find themselves needing both to convey them and to win their audience’s approval

68

      

of them. From a consideration of some examples of the persuasive use of presuppositions, we can see that the presuppositions they convey display features of this kind. The reader’s attitude toward the Northern League and specifically his or her emotional attitude toward the likelihood of secession will change considerably depending on whether he or she assumes: (7a)

There is or will be a vote in the Po Valley.

which is presupposed by (7), “il voto padano” (“the Po Valley vote”), or: (8a)

There will be opinion poll-elections in October.

which is presupposed by (8), le elezioni-sondaggio di ottobre (“the opinion pollelection of October”). It has to be noted that the contents of (7a) and (8a) are fully manifest in (7) and (8), so that their identification does not demand much processing. But the status of (7a) and (8a) is not that of explicit assertions; the speaker is required to take them for granted, unless he or she wants to challenge the author’s reliability and thus the communicative relationship. The evaluation of the behavior of prof. Di Bella and of the Ministry of Health, and, in particular cases, even the reader’s personal behavioral choices with respect to cancer, can certainly be influenced by: (17a)

There is a Di Bella anti-cancer treatment.

which is presupposed by the headline: (17)

Il pretore ordina: usate l’anticancro Di Bella “Order from the judge: use the Di Bella anti-cancer treatment”

and by the other presuppositions (9a–b) triggered by example (9). It is not clear at all whether one and only one thing which can be described as the Di Bella anticancer treatment actually exists and whether Di Bella’s prescriptions to patients suffering from cancer do anything at all to cure cancer. But presupposing (17a) cuts short of these doubts. It is interesting to note that opposite evaluations of the policy of the Ministry of Health in the Di Bella controversy are often conveyed by constructions presupposing opposing conceptions of the rights of the patient and of the obligations of the State or of the doctors, which hardly ever come to an open confrontation. I would like here to add the following quotation to our sample taken from an article commenting upon the decision of an administrative court in favor of the free distribution to terminally ill patients of one of the drugs used by Di Bella.

       (18)

69

Eppure, mi pare una cosa giusta. Perché tien conto che non si tratta di una sperimentazione che distribuisce incertezza, ma che distribuisce speranza. “Still, I think it’s the right thing. Because it takes into account the fact that this is not one of those experiments that distribute uncertainty, but one that distributes hope.” (Il Piccolo, 03/12/1998: Fernando Camon, “Cittadini uguali solo davanti al Fisco” “Citizens equal only before the tax-man”)

The distribution of hope is here presupposed to be brought about by the experimental adoption of the Di Bella treatment (note the factive “tiene conto che,” “takes into account the fact that”) and taken as a criterion of positive evaluation. Finally, consider presupposition: (13b)

Rome is the center of gravity for the exploitation of Padania.

triggered by “mantenere” (“keep”) in the declaration by Bossi: (13)

Sono momenti in cui è in atto una riorganizzazione, una contrapposizione tra chi vuole mantenere Roma come baricentro dello sfruttamento della Padania e i patrioti padani. “These are times when there is a reorganization going on, a contraposition between those who want to keep Rome as the center of gravity for the exploitation of Padania and Padania’s patriots.”

Here, the content of (13b) is not fully manifest in (13). Speaking of someone who wants something to stay in a certain state, one is not spelling out that it has been in that state up to now, but one is nonetheless taking for granted and inviting the reader to take for granted that it has been in that state, and from linguistic material in (13) it is certainly not difficult to gather which. Moreover, the presupposition quite obviously suggests or fosters resentment against Rome. As to the other presuppositions of (13), (13 a–f) quoted above, it is clear that a follower of Bossi has to find room in his or her universe for items such as Padania (a fatherland, which has its patriots) and Padania’s patriots, and that such a universe is different from the one, say, evoked by Prodi in example (14), where what is presupposed to exist is the moral and civil unity of the Italian nation (and therefore, at the same time, the Italian nation as well). Thus, the presuppositions of (13) and those of (14) require the readers to have very different takes on the context. I shall not discuss under which conditions propositional elements such as (13a–f ) or (14a–b) could be said to belong to an objective context. What is of interest here is that in the communicative situation in which (13) and (14) are produced, there is uncertainty and disagreement about the existence of things such as Padania,

70

      

Padania’s patriots, and the Italian nation with its moral and civil unity: both (13) and (14) try to get beyond this state by their presuppositions. What is implicitly conveyed in these examples has to do with ideology in one of the many senses which the word “ideology” may have. In the framework of the historical-terminological analysis proposed by Rossi-Landi (1978, 15–34), the sense in which we are using the word “ideology” here—assumptions about how our human world is and should be, not necessarily conscious but liable to be brought to consciousness—can be collocated between ideology as common sense and ideology as false consciousness (whether in the form of non-deliberate lie or in the form of conscious deception). On the one hand, common sense assumptions are regularly taken for granted by speakers and may act as criteria for value judgments or choices, but they do not need to be encoded by linguistic presupposition inducers. When this does occur, the discourse which triggers the presuppositions turns out to have hidden didactic aims. For example, in romantic fiction commonplace assumptions about male and female, the rich and the poor, love, sex, jealousy, marriage, and the like have often become the content of a welldesigned implicit sentimental education (Pozzato 1982; Sbisà 1986b). On the other hand, what Marxism has called “false consciousness” is characterized by being just false, although not always consciously recognized as such, and by concealing something non-ideological, thus serving the aims of some socioeconomic or power group. Although its contents too can be conveyed by way of presuppositions, they are often explicitly theorized and even argued for. The assumptions conveyed by our persuasive presuppositions are half-way between these two poles. They are not so clearly commonplace as not to be in need of some linguistic manifestation. In some cases (e.g., with iteratives or verbs of state change) the triggered presupposition can be reconstructed from the linguistic material present in the uttered sentence and the presupposition inducer itself. In other cases, everything that has to be presupposed is in fact linguistically manifest, as in definite descriptions (“il voto Padano,” “the Po Valley vote”) or nominalized clauses introduced by the definite article (“lo sfruttamento della Padania” “the exploitation of Padania”) and in complement clauses of a factive verb (which formulate the presupposition which has to be triggered by the factive). But presuppositions are not fully explicit either: they do not present themselves as explicit claims, as assertions, evaluations, or arguments. These would be verdictive illocutionary acts, therefore liable to be submitted to a truth/falsity, correctness/incorrectness judgment. Presuppositions avoid exposure to such a challenge by requiring the audience to take them for granted in order to keep on communicating. Thus, they may act as a powerful tool for the organization of approval in the mass media: they combine common sense, elements of “false consciousness” and other heterogeneous socio-cultural or evaluative assumptions into a mixture which, conveyed with the complicity of norms of discourse, seems to evade critical analysis.

      

71

However, it should be remembered at this point that presuppositions can be made explicit. All linguistic work on presupposition inducers has in fact used explicitation procedures. The classical discussions about presupposition projection, informative presuppositions, presupposition accommodation, and perhaps even more the recent works which parallel anaphora and presupposition (Beaver 1997; van der Sandt 1989, 1992), all remind us—quite beyond their differences and shortcomings—that presupposition explicitation is a feasible procedure. Moreover, research on presupposition inducers and presupposition projection enables us not only to tell what presuppositions an utterance conveys (this could be done even in absence of such technical tools!) but also to do so on the basis of some specifiable reason or argument. Explicitation of presuppositions and challenge to them are available moves in case of disagreement or misunderstanding, although usually dispreferred ones. If there is an interest, for example, in achieving conscious control over implicitly conveyed assumptions, explicitation may become a deliberate activity and the listener or reader practicing it can consciously profit of the received knowledge about linguistic presupposition inducers. Exercises in presupposition explicitation with adults and even with children⁵ have convinced me that some acquaintance with the kinds of sentence components and constructions which are recognized as triggering presuppositions can be of help in overcoming resistance to focusing explicitly on a presupposition. Once it is made explicit, of course, a presupposition becomes just an assertion and, like all assertions, whilst the speaker is committed to defending it, the listener or reader is allowed to challenge it and even to ask the speaker for his or her grounds for making that assertion. Thus, by facilitating both presupposition explicitation and the monitoring of it, research on presuppositions can help improve reading and comprehension skills in general, and in particular promote and foster criticism of implicitly transmitted ideological contents.⁶

⁵ Cf. “La pragmatica nell’educazione linguistica e nella didattica. Parafrasi e acquisizione di informazioni,” paper read at the International Workshop “Insegnare italiano nella scuola del 2000” (Trieste, November 7–9, 1996), to be published in the Proceedings [Sbisà 1999c]. ⁶ I would like to thank Paola Rodari for the many discussions we have had about the use of presuppositions in written texts since we wrote Sbisà and Rodari (1989) and Christopher Gauker for his insightful e-mail correspondence regarding context and assertibility.

4 Intentions from the Other Side 1. Introduction In this paper I shall discuss some of Grice’s ideas which I find valuable: his first analysis of non-natural meaning (or “meaningnn,” to be contrasted with natural meaning, or “meaningn”; 1957), his notion of a Cooperative Principle underlying human interaction (1975), and his notion of a person (1991). I have learnt to appreciate one of these ideas, the Cooperative Principle, in the context of the analysis of discourse. Analyzing discourse involves giving interpretations to actual cases of language use. Unfortunately, the basis of these interpretations is often merely intuitive and problems are given ad hoc solutions. It may even seem that any interpretation is good, providing it is couched in a rhetorically convincing form. That is why there have been discussions about the limits of interpretation (Collini ed. 1992; Eco 1990). I take it to be a part of Grice’s heritage that the assignment of meaning intentions has to be motivated and, whenever possible, argued for. This is especially clear from the way in which he deals with the distinction between conventional and conversational implicatures (Grice 1975). His preference for an explication of implicated meaning in terms of conversational rather than conventional implicature is to be associated, not only (as he did explicitly) with the need not to multiply conventional senses beyond necessity but also with the appreciation of argumentative activity (which is involved in working out conversational implicature, but not conventional implicature) as responding to a demand for rationality. Thus one thing which Grice teaches to those who are concerned with the analysis of discourse is that they have to give justifications for their interpretations and couch them, whenever possible, in argumentative form. I shall not limit my discussion to the Cooperative Principle, because I think that the significance of this notion and the plausibility of my interpretation of it can best be seen within a wider consideration of Grice’s philosophy. I shall defend the three Gricean ideas I mentioned above at the price of turning some features of his philosophy upside down or, at any rate, the way it has been received. In so doing I shall follow a “red thread” throughout his work by taking as central, not the point of view of the utterer or of a neutral observer, but the point of view of the audience. Originally published in Paul Grice’s Heritage (Semiotic and Cognitive Studies 9), ed. Giovanna Cosenza, 185–206. Turnhout: Brepols, 2001.

Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0005

    

73

This “red thread” is more mine than Grice’s; mine is the concern for the audience’s point of view, in the framework of a view of meaning as interactionally constructed. But, as I shall try to show, there are a number of hints in Grice’s work which justify my interpretive choice. According to the received interpretation, the speaker’s intentions determine whether there is non-natural meaning at all, what is non-naturally meant by an utterance, whether the utterer observes the Cooperative Principle, which implicatures his or her utterance has. Here, I shall reformulate Grice’s claims in terms of ascriptions of intentions to the utterer by the audience. In my opinion, this can throw new light on some traditional criticism of the definition of non-natural meaning, as well as justify the use of the Cooperative Principle in the analysis of discourse, and enable the internal coherence of Grice’s philosophy to be better appreciated.

2. The Analysis of Non-natural Meaning Let us start by asking ourselves what is the point of the analysis of non-natural meaning (henceforth: meaningnn) proposed by Grice (1957, 1969). Is he proposing that we should check the speaker’s intentions in order to know whether his or her behavior has meaningnn? The use of formulas such as “ ‘U meantnn something by uttering x’ is true iff . . .” (1969, 92) can make us think that Grice is spelling out the conditions at which it can be correctly said that an utterer meantnn something. But does it make sense to check whether someone really has a certain intention in order to decide whether what he or she says or does has meaningnn? In our everyday experience of human interaction, we ascribe intentions to our interlocutors just on the basis of the meaning we understand them to convey. So, if Grice’s definition is an analysis of the concept of meaning in terms of the truthconditions of sentences containing the verb “to mean,” it is an analysis which does not help us in telling when we can apply that concept and when we cannot, and it seems to me to have little point. In my opinion, if Grice’s definition of meaningnn has a point at all, it is to spell out what we mean when we say that someone meansnn something, that is, to specify the intentions we are bound to ascribe to an utterer whenever we take his or her behavior as having meaningnn. If I say that U meansnn something by s, I am bound to ascribe certain intentions to him or her. This at least tells us something about what ascribing meaningnn to someone amounts to. It should be noted that this reading of Grice’s definition of meaning (1957) relocates it within the historical context of ordinary language philosophy: what is primarily at stake is not the “real story” of how meaningnn generates in an uttering subject, but the elucidation of what we mean, or perhaps of what we do, when we use the problematic word “mean.” I am aware that such a reading is in some sense

74

    

weaker than the received one. But I am not saying that Grice did not intend to make any stronger claim. Certainly, for example, he does use the “. . . is true iff . . .” formula; and his own reconsideration of meaningnn in “Meaning revisited” (1982) can be taken to show that he really was interested in making claims about what there has to be in the utterer’s mind in order for him or her to meannn something.¹ What I am saying is that Grice (1957) raises at least the weak claim identified by my reading, and that such a claim has a point. Besides, the concern for the meaning of what we say, whether ordinary or stipulated, that is, the concern for what we mean when we use certain problematic expressions, continued to play a role in Grice’s later works (1982, 302; 1986, 58–9; 1989, 339–85). So, it is not unreasonable to trace back his definition of meaningnn to a model of analysis focusing on the use of words.

2.1 Some Evidence for a Reading of Grice’s Analysis of Meaningnn Which Focuses on the Audience The reading I am proposing for the analysis of meaningnn is supported by some hints to be found in Grice’s texts. (i) In “Meaning” (1957) Grice admits that intentions are not conscious plans of an utterer, or at least not necessarily, but are ascribed to him or her by the audience or even by the utterer himself or herself on the basis of his or her behavior and of criteria taking into consideration general usage and “what is normally conveyed” by the utterance. Similarly, in Essay 7 of the “Logic and Conversation” series, he makes it clear that his theory does not affirm any epistemic priority of the utterer’s intentions over the conveyed meaning (Grice 1989, 138–9). Rather, he admits that we ascribe intentions corresponding to the meaning we understand to the utterer and that sometimes at least, for example in cases of linguistic communication, we understand that meaning on the basis of the timeless meaning of the sentence used. (ii) In Essay 5 of the “Logic and Conversation” series (1969, 98–9) Grice shows concern for “the calculations” required of the audience by the utterer. He affirms that “one cannot have intentions to achieve results which one sees no chance of achieving”: if the audience cannot recognize an intention, the utterer cannot have the intention to get the audience to recognize it. This may be taken to mean that the utterer cannot form that ¹ I shall not discuss here the later Gricean “constructive” method, which I take to be a kind of analysis by simulation (instead of distinguishing the elements in a process, Grice starts by positing the elements and adds them one to the other in order to obtain a simulation of that process).

    

75

intention; but it certainly means also that he or she cannot be ascribed that intention. Restrictions on the intentions the utterer can have seem to coincide here with restrictions on the ascription of intentions. (iii) In discussing thought, Grice affirms that psychological attitudes are not “natural causes and effects” of what he takes to be “the word flows constituting thought,” but “they are due or proper antecedents or consequences of the word flows in question and as such are legitimately deemed to be present in those roles” (1989, 142). If this remark can be extended outside the context in which Grice made it (concerning linguistic thought and self-assignment of interpretations by the thinker), one might conclude that psychological attitudes are for him legitimately ascribed to an utterer we take as meaningnn something, without being the natural causes of his or her utterance. Avramides (1989) has convincingly argued that Grice’s analysis of meaning is not reductionist but reciprocal and that he should not be taken as claiming that semantic facts are to be reduced to psychological facts. Moreover, Grice (1991) dissociates himself from reductionism in general. If he cannot be attributed any ontological claim regarding the reduction of meaning to utterer’s intentions, then we can take the hints mentioned above seriously without accusing him of any contradiction. And these hints indicate that it is a mistake to view Grice’s analysis of meaning as an attempt to explain meaning production on the basis of the utterer’s intentions.

2.2 How Grice’s Meaningnn Looks if Seen from the Other Side Here I shall consider Grice’s first definition of meaningnn (1957), according to the received interpretation of which an utterer U, in order to meannn something by s, must have: (i) the intention to achieve effect E on an audience A (ii) the intention to get A to recognize that U intends to achieve effect E on A (iii) the intention that effect E on A is to be achieved on the basis of A’s recognition of U’s intention to achieve effect E on A. I shall refer to these three intentions as the original Gricean intentions. Since I want to discuss their role, I shall not discuss their content, although much could be said, in particular about the third (which has been criticized in the literature: see below, Section 2.3) and about the first, which states the nature of the intended effect but, in my opinion, cannot be easily made to accommodate for the whole variety of speech acts.

76

    

Viewing meaningnn from the other side involves foregrounding the ascription of intentions to the utterer by the audience. The received definition can thus be reformulated as follows. Whenever an audience A takes an utterer U to meannn something by an utterance s: (i) A should ascribe to U the intention to achieve effect E on A (ii) A should ascribe to U the intention to get A to recognize that U intends to achieve effect E on A (iii) A should ascribe to U the intention that effect E on A is to be achieved on the basis of A’s recognition of U’s intention to achieve effect E on A. This reformulation, although it is no longer a definition in the classical sense, helps elucidate how we use the problematic word “mean” in ascriptions of meaningnn by spelling out the ascriptions of intentions to which we are committed in such cases. It interferes in interesting ways with the objections to Grice’s analysis, based on counterexamples, which are meant to show that the three original Gricean intentions are not necessary and sufficient conditions for meaningnn. These objections arise just because it is assumed that the analysis should enable us to check whether in any given case there is or is not meaningnn. But if our aim is not to verify this, that is, if (as suggested here) we take meaningnn as our starting point in the ascription of intentions, and focus on the connection between ascription of meaningnn and ascription of intentions, such objections lose much of their relevance.

2.3 How Objections to Grice’s Original Analysis Look if Seen from the Other Side It is possible to translate the objections to the sufficiency of the analysis into our reformulated framework by asking: may it be the case that A ascribes to U the three original Gricean intentions, but still there is no meaningnn? This question has different answers in different cases. I shall try to explore two of them. First, it should be noted that the fact that intentions identical to the three original Gricean intentions can happen to be ascribed by A to U for reasons independent of the assignment of meaningnn to U’s utterance does not constitute a real objection to an analysis of “U meansnn something by s” couched in terms of ascription of intentions. In the reformulated analysis, there is no claim that the only reason that A may have for ascribing those intentions has to be his or her being bound to do so because he or she takes U as meaningnn something. Of course, if A has reasons other than this one to ascribe the three original Gricean intentions to U, these will not be connected to any ascription of meaningnn.

    

77

So it is said that some people, in torturing a prisoner: (i) intend to achieve an effect E upon him (making him tell them the names of his companions) (ii) intend to get him to recognize that they intend to achieve effect E upon him (iii) intend effect E on the prisoner to be achieved on the basis of his recognition of their intention to achieve this effect upon him; but that nevertheless, in applying an instrument of torture to him, they do not meannn that he should tell them the names of his companions (Grice 1969, 93–4). This is correct and quite compatible with our reformulated analysis. In fact: (i) A, the prisoner, ascribes to U, the torturers, the intention to make him speak by torturing him (ii) A ascribes to U the intention to make him recognize they intend to make him speak by torturing him (iii) A ascribes to U the intention to make him speak on the basis of his recognition of U’s intention to make him speak by torturing him; but these ascriptions of intentions are independent of the torture itself, since they depend rather on verbal and non-verbal communication prior to or during the torture, so that no connection can be posited between the ascription of the three intentions and any ascription of meaningnn to the acts of which the torture consists. In our framework, this counterexample does not call for the addition of a new intention of the utterer’s, but for an explanation of the reasons, different from the attribution of meaningnn to the torture (which actually is not made), that entitle A to ascribe to U intentions which coincide with the three original Gricean ones. Second, it follows from our reformulated analysis that if A for any reasons is in a position in which he or she cannot ascribe the three original Gricean intentions to U relative to his or her utterance of s, A should not take U to meannn something by s. This is so because, if A took U to meannn something by s, A would be bound by an obligation which is impossible or unreasonable to fulfil. In the case of the counterexample proposed by Strawson (1964), U conceals one of the three original Gricean intentions from A, who is therefore not in a position to ascribe U one of the intentions. The fact that this case cannot be considered as a case of meaningnn, far from calling for the addition of new intentions to be ascribed to the utterer, already follows from the reformulated analysis—and thus confirms it, rather than undermining it. Let us briefly consider Strawson’s counterexample and contrast it with two clear cases, one of meaningn and the other of meaningnn.

78

    

(A) Meaningn. U, unknown to A, puts M’s handkerchief on the site of the crime in order to make A believe that M is the murderer. Since U does so unknown to A, nobody would assign his gesture meaningnn. In fact, A cannot ascribe to U even the intention (i) to make A believe that M is the murderer. (B) Meaningnn. Suppose U puts M’s handkerchief on the site of the crime in a public and emphatic way. In this case his gesture becomes, so to say, “symbolic” and it can be assigned meaningnn. In such a situation, however, it is quite possible for A to ascribe to U both the intention to make A believe that M is the murderer, the intention to make A recognize that U intends to make A believe that M is the murderer, and finally the intention that A should believe that M is the murderer on the basis of A’s recognition that U intends to make A believe that M is the murderer. (C) Strawson’s example. Strawson’s example is presented as a case in which the utterer really has the three original Gricean intentions, and which is nevertheless merely a case of meaningn. From the audience’s point of view, however, it appears as a case half-way between case (A) and case (B), in which some (but not all) of the original Gricean intentions can be ascribed. The story runs as follows. U puts M’s handkerchief on the site of the crime intending to make A believe that M is the murderer, and the situation is such that: U knows that A sees him, U does not know that A knows that U sees him, and U knows that A does not know that U knows that A sees him. In such a situation, A can ascribe to U the intention to make A believe that M is the murderer, but is not in a position to reasonably ascribe to U the intention to make A understand that U intends to make A believe that M is the murderer, and therefore cannot ascribe to U the more complex third intention either. Therefore, on the basis of our reformulated analysis, A should not ascribe meaningnn to U’s gesture. Our reformulated analysis predicts that a case such as Strawson’s example cannot be a case of meaningnn. However, the feature that makes this example not a case of meaningnn is not the absence of a further required intention on the utterer’s part, but the audience’s impossibility to ascribe to the utterer the intentions he or she ought to be assigned if he or she were taken as meaningnn something. This result is compatible with Grice’s proposal of including a negative condition in his analysis of meaningnn, that is, the absence of deceptive intentions (1969, 99). Cases of deception about the utterer’s intentions are, in fact, cases in which U puts A in a position in which A cannot ascribe a certain intention to U.

    

79

Another kind of objection—this time, an objection to the claim that the three original Gricean intentions are necessary conditions for an utterer to meannn something—can be exemplified by the story of Herod and Salome. Herod has his guards bring the head of St. John the Baptist to Salome on a salver. Is he meaningnn something? Or is it the head of St. John which meansn (as opposed to meansnn) that St. John is dead? This example is considered by Grice as a case of natural meaning, while others have maintained that it is a case of meaningnn in which, contrary to Grice’s analysis, one of the three original Gricean conditions is revealed to be superfluous (Neale 1992, 547–9; Schiffer 1972, 56–7; Sperber and Wilson 1995, 29). In our analysis, this is a case of meaningn and not of meaningnn, because Salome cannot ascribe to Herod all three of the original Gricean intentions, and so should not ascribe meaningnn to Herod’s behavior.² Though she can ascribe to Herod the intention to make her believe that St. John is dead and also the intention to make her recognize that he intends to make her believe that St. John is dead, she cannot ascribe to Herod the intention to make her believe that St. John is dead on the basis of her recognition of Herod’s intention to make her believe that St. John is dead. She just sees that St. John is dead, and knows that Herod knows she sees it. Therefore, she cannot think of Herod as intending her to believe that St. John is dead on the basis of any other reason, such as her recognition of his intention to make her believe that it is so. In our framework, it is possible to explain the slightly ambiguous character of this example, which has led it to be considered as controversial. If we proceed from the assignment of meaning to the ascription of intentions, we have first to assign a meaning to the utterer’s behavior. Thus, if the meaning attributed to the offer of St. John’s head to Salome on a salver is that St. John is dead, Herod is not ascribed either meaningnn or the corresponding set of original Gricean intentions. But suppose that Salome, from the macabre ritual of being offered St. John’s head on a salver, understands that Herod means that he has killed St. John just in order to keep the promise he had made to her. If she attributes such a meaningnn to Herod’s behavior, she will be bound to ascribe to Herod, not only the intention to make her believe that it is so and the intention to make her recognize that he intends her to believe that it is so but also the intention that she believes that it is so (that is, that he has killed St. John just in order to keep the promise he has made to her) on the basis of her recognition of his intention to make her believe that it is so. And these are ascriptions of

² Here I take for granted that whenever the addressee should not ascribe meaningnn to the utterer, nobody else should, and that therefore in any such case it is not correct to speak of meaningnn.

80

    

intention which she is in a position to make. It is therefore quite viable to regard Herod’s behavior as meaningnn that he has killed St. John just in order to keep the promise he has made to Salome.

3. The Cooperative Principle Let us now turn to a second main notion in Grice’s theory of language, the Cooperative Principle (henceforth CP): 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. (1975, 26)

In its received and most widespread interpretation, the CP is a principle which speakers should observe, and sometimes at least do observe. When they observe it, it may give rise to a particular kind of implicitly conveyed meaning, the conversational implicature, which consists in its own turn of an intention of the speaker. But the CP, together with the related notion of conversational implicature, can be also viewed from the point of view of the audience. From this perspective, the CP does not spell out what a speaker has to do in order to be cooperative, but what we are to think of a speaker whom we take as being cooperative. Besides, establishing whether an utterance has conversational implicatures does not involve checking whether the speaker really has the intention to convey them, but requires the adoption of an interpretive attitude which, as a consequence of the ascribed observance of the CP, ascribes that intention to the speaker. It should be noted that the CP, too, is related in some way to the historical context of ordinary language philosophy. It does not belong to that historical context: rather, it has been proposed by Grice in the framework of his polemics against some assumptions and practices proper to ordinary language philosophy, such as making appropriateness conditions part of the meaning of linguistic expressions. Much of what ordinary language philosophers claimed to be part of the meaning of a linguistic expression is, according to Grice, to be described in terms of inferences from the use of the expression in a given communicative situation in which the CP holds. But at the same time, it should be recognized that, while ordinary language philosophy has never been able to satisfactorily define what “ordinary language” means, Grice’s CP has implicitly provided such a definition. Since meanings characteristic of the ordinary use of language (as opposed to a formally constrained logico-semantic one) arise by inference when the CP holds, it is clear that the CP, however interpreted, sets out the conditions in which one can say that “ordinary language” is spoken. It is up to further

    

81

interpretation of the CP to decide whether ordinary language is just one language game among many, a game in which speakers observe the CP, or a pervasive dimension of human interaction.

3.1 Some Evidence for a Reading of Grice’s CP Which Focuses on the Audience The Cooperative Principle is presented as an expectation concerning the behavior of speakers: it is “a rough general principle which participants will be expected (ceteris paribus) to observe” (Grice 1975, 26). If we take the embedding of the CP in the context of such an expectation seriously, we may conclude that what sustains the participants’ reciprocal understanding in verbal interaction is not so much that each participant observes the principle, as that each is expected (ceteris paribus) to observe it. In the passage quoted above, it is not specified who is supposed to expect the participants to observe the CP. This indeterminacy becomes ambiguity in another passage, in which Grice is concerned with whether speakers really do observe the CP. This is how he formulates the question: . . . what the basis is for the assumption which we seem to make, and on which (I hope) it will appear that a great range of implicatures depend, that talkers will in general (ceteris paribus and in the absence of indications to the contrary) proceed in the manner that these principles prescribe. (1975, 28)

Now, does this question concern the “assumption which we seem to make,” that is, the reasons that we as participants have for assuming, rather than not assuming, that speakers observe the CP, or does it regard whether or not it is true that “talkers will in general . . . proceed in the manner that these principles prescribe,” and therefore whether we as theorists are justified in making such an assumption? In my opinion, this ambiguity can be resolved if we consider the argument which Grice gives for the rationality of the CP (1975, 30). In order to support the conclusion that observance of the CP is rational, Grice argues that participation in talk exchanges will be profitable “only on the assumption that they are conducted in general accordance” with the CP, and that “anyone who cares about the goals that are central to conversation/communication” must be expected to have an interest in participation in talk exchanges. Here again, the notions of assumption and expectation come in as a context in which the observance of the CP is embedded. What makes talk exchanges profitable is not (or not merely) their accordance with the CP, but the assumption that they are so conducted, and this assumption is here not a theorist’s assumption, but an assumption which participants themselves ascribe to each other. Moreover, this very assumption is based

82

    

on an expectation which each participant entertains insofar as he or she presumes that the other cares about the central goals of communication: the expectation that participants have an interest in communicating with each other. So, it is the participants in an interaction who assume of each other that the CP is observed, because they ascribe to each other an interest in communicating as well as some kind of rationality which drives them from such an interest to the assumption that the CP is observed.

3.2 Some Evidence against Conversational Implicatures Being Speaker’s Intentions Should conversational implicatures be conceived of as intentions which the speaker has or as intentions to be ascribed to the speaker by the audience? When Grice spells out what he calls the “general pattern for the working out of a conversational implicature” (1975, 31), he couches all its steps in terms of ascriptions of propositional attitudes to the speaker by the audience; the conclusive step ascribes to the speaker the intention that the audience thinks, or at least the willingness to allow the audience to think, something which corresponds to the content of the implicature. It would be superficial to conclude that an implicature is the intention or willingness of the speaker to let the audience think something. What the whole “general pattern for the working out of a conversational implicature” is up to is to show that a conversational implicature is something such that it is reasonable to ascribe to the speaker the intention to convey it. Another point in favor of the interpretation of conversational implicatures not as intentions that the speaker really has, but as intentions that the audience can reasonably ascribe to him or her, is Grice’s claim that a conversational implicature may consist in a disjunction of the various possible explanations which may be offered to preserve or repair the observance of the CP, the number of which may be open (1975, 40). It is obviously impossible for the speaker to actually possess an intention corresponding to such an implicature: how could one intend to convey an open, and therefore not wholly specified nor specifiable, list of disjuncts? But if we view CP and conversational implicatures from the other side, we find out that it is possible for the audience to ascribe to the speaker the intention to convey anyone of the disjuncts (each of which offers an explanation for the apparent violation of the CP or of a related maxim), or even any finite disjunction or conjunction of them. In such cases, further receivers (as happens, for example, in the interpretation of poems and other literary texts) may join a new member to the list of disjuncts which has already been worked out, by ascribing the intention of conveying it to the speaker, provided they discover it by following the same general argumentative pattern.

    

83

One objection to my reading of the notion of a conversational implicature may arise from the need to distinguish between the working out of a conversational implicature and the interpretation of symptoms. This distinction is drawn by Grice with reference to his example of the flouting of the second maxim of Quantity “Do not give more information than is required”: A wants to know whether p, and B volunteers not only the information that p, but information to the effect that it is certain that p, and that the evidence for its being the case that p is so-and-so and such-and-such. (1975, 33–4)

According to Grice, this example is not necessarily an example of the flouting of the second maxim of quantity, since B’s behavior may be undesigned. In this case, the audience may conclude that B feels uncertain about what he or she is saying, but this amounts merely to drawing a conclusion from a symptom—B’s giving too much information—, without any implication about what B intended to convey. It seems that, in order for the example to be a genuine case of conversational implicature, B’s behavior has to be designed, that is he or she must have the intention to convey the implicature (for example, to convey that p is not uncontroversial). But the role of the audience should not be overlooked. Who is to decide whether B’s behavior should be taken as a symptom or as endowed with meaningnn? Whether we face a case of conversational implicature or one of symptomatic interpretation depends largely on whether the audience takes the speaker’s behavior as designed or undesigned. This factor is mentioned by Grice himself, who uses expressions such as “B’s volubility may be undesigned, and if it is so regarded by A . . .” and “if it is thought of as designed, it would be . . . .” It is up to the audience to take the speaker’s behavior in either way. I do not want to claim, though, that the audience is completely free in its choice. For A’s interpretation of B’s utterance to be acceptable as the working out of a conversational implicature, it should be at least reasonably possible for A to ascribe B the intention to convey the implicature. I think that such a way of drawing a distinction between implicatures proper and inferences from symptoms is compatible both with Grice’s treatment of the example and with the foregrounding of the audience’s point of view I am proposing.

3.3 How the CP Works if Seen from the Other Side The received view of the CP as a principle observed or not observed by speakers has led to disputes about the limits of its application and this seems to undermine its use in discourse analysis. Is the CP at work in fiction or poetry? Is it at work when people quarrel? Does it still hold in cultures different from ours? Is it sometimes at least replaced by rules of politeness? In fiction we do not aim at saying something true; in poetry and in literature in general we do not observe

84

    

maxims of Manner such as those prescribing briefness or absence of ambiguity; when we quarrel or are in other ways competitive with each other, we do not cooperate. There seem to be cultures in which most people infringe the CP most of the time (Ochs 1976) and, even within Western culture, there are situations in which we do not follow the prescriptions spelled out by the Gricean maxims, but rules of politeness which recommend, for example, leaving our interlocutor open options or showing solidarity (Lakoff 1973; Leech 1983). In my opinion, the CP should be considered as a complex assumption that the audience makes about an utterer, in order to fully understand what the utterer meansnn. This view of the CP allows for ascriptions of its observance in any situation in which the audience wants to fully understand what the speaker meansnn. Of course, this does not mean that the audience is completely free to make such ascriptions or not. Like other ascriptions of intentions and other attitudes, ascriptions of observance of the CP are defeasible: they are made on the basis of some evidence which is taken as sufficient for practical purposes, but they can be cancelled if it turns out that there are particular reasons for doing so. So, to borrow an example from Grice (1975, 29), if you pass by when I am struggling with my stranded car and join me in tinkering under the hood, I’ll ascribe to you the intention to help me to get the car started again, but if from what happens subsequently it turns out that, say, you were just curious or even intended to damage my car further, perhaps in order to steal something from me, the previous ascription of intention is no longer valid. In a similar way, ascriptions of cooperativity to speakers who seem to address us are made and are operative unless particular reasons intervene, allowing for the working out of conversational implicatures which enlarge the audience’s understanding of what the speaker means. According to the view proposed here, the CP can be used as an interpretive device with respect to fiction and poetry (whose differences from everyday informative communication can be understood as conveying implicit meaning; see Pratt 1977) as well as in conflictual and competitive situations (in which participants need to understand each other, if not for any other reason than that this may help each of them in his or her strategic choices). As to the application of the CP to cultures different from ours, it should be noted that Ochs’s criticism based on Malagasy linguistic behavior, which is said to be regularly uncooperative, uses the CP as an ideal model to which Malagasy linguistic behavior is compared in order to be understood (Ochs 1976). It is in fact from the attempt to understand what is conveyed by the real or apparent infringement of the maxims, that Ochs can draw her conclusions about the Malagasy view of what is at stake in communication. Thus, far from demonstrating the inapplicability of the CP to Malagasy culture, Ochs’s analysis shows that reference to the CP may render linguistic data more significant for the anthropologist. Moreover, I think that the role of rules of politeness in communication is very different from that of the CP and its maxims. Grice maintained that sometimes we

    

85

work out non-conventional implicatures on the basis of principles other than the CP, but it is easy to see that most inferences we make on the basis of rules of politeness are in fact more like symptomatic interpretations than like conversational implicatures, in that it is not reasonable to ascribe to the speaker the intention to convey them. If we notice that somebody is behaving in ways not conforming to accepted rules of politeness for a certain kind of situation, instead of simply deeming him or her impolite, we may conclude that the situation is of a different kind from what we thought; or else, we may think that he or she is following the politeness rules of some other culture. In both cases, we make an inference, but do not ascribe to the agent the intention to communicate its content to us. Namely, we do not think that the agent has behaved in that way in order to let us know how the situation really is, or in order to make us learn the politeness rules of his or her culture. Only in conjunction with the assumption that the CP holds can politeness rules contribute to the working out of authentic nonconventional (and here, conversational) implicatures, which are (or at least can reasonably be) accompanied by ascriptions of intentions. For example, suppose I suspect that somebody is not speaking to me sincerely. If I assume that he or she is nevertheless cooperative, and if I am acquainted with some rule of politeness which prescribes that one avoids saying things that may hurt the addressee, I may infer that the speaker is behaving as he or she does because he or she is observing that rule of politeness, and that he or she is expressing concern for me and sets greater store on expressing concern than on telling the truth. I may conclude that he or she is conveying a message such as: “You may be hurt by some information I possess, but I care about you and do not want you to be hurt.” In fact, that is what one would understand in such a situation. Reference to the CP, and not only to rules of politeness, seems to be necessary to the justification of such an understanding. The heuristic role that the CP can play, especially when seen from the other side, cannot be supplanted by the proliferation of other situation-specific principles and rules directly addressed to speakers. In conclusion, then, the CP (seen from the other side) would seem to have a heuristic value in all kinds of interactional situations, and is therefore especially apt to be used as a tool in discourse analysis. The analyst is usually not the addressee of the text he or she analyzes, but is a receiver nevertheless. His or her understanding of the text is then optimized by a careful reference to the CP, to be adapted to the particular text and situation.

4. The Notion of a Person The interpretation of the CP proposed above can help Grice’s readers to bridge the gap between his philosophy of language and his theory of value (1991). In the last sections of this paper, I shall briefly illustrate Grice’s conception of value and some

86

    

of its connections with persons and rationality; I shall then attempt to show that the definition of a person which Grice proposes is structurally parallel to the notion of a cooperative participant in his philosophy of language, as this notion appears when the CP is viewed from the other side.

4.1 Value, Persons, and Rationality In his theory of value, Grice argues for a notion of absolute value, linked to a notion of autonomous finality, and involving the possibility of issuing value judgments which are not mere expressions of emotions or of interests but can be deemed correct or incorrect. It should be noted that philosophers belonging to the analytic tradition have often denied that possibility, with the exception of J. L. Austin, who classed value judgments as verdictive speech acts liable to be deemed correct or incorrect, together with factual judgments and various kinds of rankings and calculations (1975, 153). In order to argue for absolute value, Grice introduces the notion of a person, that is, of an essentially rational being. To introduce the notion of a person, Grice sketches a construction of it, which involves considering rationality, which human beings accidentally happen to possess, as an essential feature of a different kind of being, persons. Being essentially rational, the person has absolute value and can attach absolute value to what he or she evaluates. Since, in this way, absolute value is introduced into the world, values are no longer a subjective matter, but must turn out to be either correct or incorrect. The notion of rationality used by Grice here has little to do with the widespread notion of instrumental rationality, that is, rationality concerned with means–ends relationships. He is thinking of rationality as having strict and intricate connections with value and as distinctively marked by argumentative activity. This could be considered a recurrent theme in Grice’s thought. The connection with argumentation was already at work in the discussion of the rationality of the CP. Grice’s conception of rationality in his Essays of the “Logic and Conversation” series (1989, 3–143) remains closer to instrumental rationality, since it is related to a tendency to behave in ways suitable to gaining advantage from the conversation. But it appears clearly enough that such a tendency includes the readiness to work out conversational implicatures, a definitely argumentative activity. In his Lectures on the Conception of Value, Grice defines rationality as: a concern on the part of the creature which has it that its acceptances and perhaps (more generally) its attitudes . . . should be well-grounded, based on reasons, or . . . validated; a concern, that is, on the part of the reason-seeker that the attitudes, positions, and acceptances which he (voluntarily) takes up should have attached to them certificates of value of some appropriate kind. (1991, 82)

    

87

So, rationality is connected to demands for justification, and justification (to be attained by giving reasons, that is, in argumentation) is considered as an assignment of value or even, perhaps, as the very source of value, since what is justified is of value, and essentially rational beings seek value primarily by seeking justification. As to the notion of a person, it is introduced as a construction which involves ascription of rationality to a subject as an essential feature. Since it appears to be explicitly based on an ascription procedure, there is no need to turn it upside down to see it “from the other side.” However, two main queries arise. Firstly, are there conclusive grounds for deeming that a given subject is essentially rational and thus a person? Grice seems to admit that there is here an element of choice, of decision, even of “faith” (1991, 106), or—to use Judith Baker’s word (1991, 13–14)—of trust (which is connected to “faith” via the meaning of the Latin “fides”). This choice involves both an evaluation, and an assigning of value (in this case, the absolute value of being a person). Secondly, who is to carry out the ascription procedure? Grice seems to describe the transformation of human beings into persons as an operation carried out by an isolated creature on himself or herself. Therefore, it might seem that the transformation constitutive of persons cannot be described as an ascription procedure proper (which would involve at least two participants). Judith Baker, who understands Grice as speaking of a creature’s self-transformation, finds it necessary to proceed from the isolated transformation of one subject to the subject’s consideration of other subjects as persons (1991, 11–13). However, Grice speaks of this transformation—which he calls the Metaphysical Transubstantiation—as operated not by and on one and the same individual, but by and on the biological type Homo Sapiens: a species, and therefore a collective entity. He also uses the plural subject “they.” Thus, he can be regarded as referring to some collective, social transformation: one which is made possible only within the intersubjective relationship. If we want to be allowed to issue value judgments which are not merely subjective, we have to attain absolute value. If we want to attain absolute value (which, as we have seen, in Grice’s analysis coincides with being a person—an essentially rational being—and which persons can attach to what they value), we need to ascribe each other essential rationality.

4.2 A Retrospective Glance How are these views about the notion of a person related to the other major themes in Grice’s philosophy, such as meaningnn and the Cooperative Principle? I would like to explore this issue by way of a conclusion, and attempt to show that similarities and connections are easier to detect if one adopts the readings “from the other side” I have proposed.

88

    

First, I would like to recall that there are hints at a notion of value in Grice’s philosophy of language too. According to “Meaning Revisited” (1982), where Grice admits of the infinite regress of intentions, the intentions which characterize meaningnn can never be completely present in the utterer, since they are infinite in number. So his or her meaningnn something depends on an evaluation of the audience, who has to deem whether the utterer approximates enough to the ideal of meaningnn to be legitimately considered as meaningnn something. In the discussion of meaningnn above, I have not accepted that meaningnn involves infinite regress of intentions, because, in the perspective I have proposed, this does not seem to be required. Nevertheless, I would like to stress that Grice’s proposal to view ascription of meaningnn as an evaluative procedure is compatible with the “red thread” I have followed in reading him: what Grice admits is, finally, that it is up to the audience to accept an utterer as meaningnn something, and that the audience does so by issuing an evaluative judgment, which must have its own standards of correctness, but also involves a decision, the taking of an attitude toward the utterer. Reconsidering this idea of Grice’s in the light of his views on value, rationality and persons (1991), one might be tempted to develop it in the following way: taking someone as meaningnn something involves taking him or her as corresponding to an ideal; but taking someone as meaningnn something is at the same time taking him or her as endowed with subjectivity (a necessary component of being a rational creature); therefore, subjectivity itself might be considered as an ideal, and our access to it as something which is mediated by its intersubjective recognition, which is an evaluative procedure of ascription.³ Second, turning to the Cooperative Principle, what I have proposed about it in Section 3 amounts to considering it as a system of expectations we have with respect to those utterers we take to be rational subjects willing (or, at least, not unwilling) to communicate. There is no proof or conclusive evidence that someone is observing the CP, but it can still be a reasonable decision to consider him or her (ceteris paribus, i.e. unless there are specific reasons not to do so) as observing it. Thus, it is possible to claim that also in the case of ascription of observance of the CP (as in the case of meaningnn) an evaluative element comes in. Not only do we assess the speaker’s distance from the ideal model spelled out by the CP, but what we assign the speaker as a result of our assessment, the observance of the CP, is itself a value, since it is strictly connected with rationality. The parallel between such a reading of the CP and the construction of persons in Grice (1991) is therefore striking: we are in the same position with respect to considering someone as observing the CP, as we are with respect to considering him or her ³ I am sympathetic to such a view (which, of course, I am not ascribing to Grice, and for which I cannot further argue here), because I think that some of the ethical and political issues of our age cannot be solved by reference to a conception of the subject as coinciding with the individual (physical and psychological), but require more complex and value-laden intersubjective foundations to be granted to subjectivity.

    

89

as a person. We might even envisage Metaphysical Transubstantiation as an archetypal form of ascription of rationality on which all contingent ascriptions of cooperativity are modelled. Finally, I would like to comment on a remark of Grice’s, which I take to apply to all of the three areas of his philosophy explored in this paper. He writes: “Proof of the pudding comes from the need to eat it” (1991, 106). Such a cryptic remark can perhaps be best understood against the background of the claim, which I have put forward, that the internal coherence of Grice’s philosophy can be better appreciated in the framework of a consideration of it “from the other side,” foregrounding not intentions in the utterer’s mind, but ascriptive procedures on the part of the audience. In fact, in light of the reading of Grice’s philosophy proposed here, we may say that proof of subjectivity comes from the need to assign meaningnn, proof of cooperation comes from the need to fully understand one’s interlocutors, and finally, proof of human rationality comes from the need to give value judgments. It is these needs which lead to ascriptive procedures—ascriptions of meaningnn (with the related intentions), of cooperativity, of being-a-person (and therefore being of value)—and introduce each human being into a world in which there are meanings, people understanding each other, and last but not least, rational value judgments.⁴

⁴ This paper represents my views on Grice as I settled them soon after the San Marino conference “Paul Grice’s Heritage” (1998). I thank Patrick Coppock, Christopher Gauker, and Paolo Leonardi for their helpful comments and criticisms. Obviously, responsibility for the shortcomings of this paper is mine alone. Later on, I have discussed my views on Grice on several occasions, among which a meeting of the research group “Analytic Philosophy: History, Tendencies and Present Problems” (Bologna, December 1998; my contribution “What Is Grice’s Cooperative Principle?” was published in Knowledge and Meaning, ed. Diego Marconi, Vercelli: Mercurio, 2000) and two talks I gave in Spring 1999 in Baltimore and in Albany. I am very grateful to all those who have discussed my views on these occasions, especially Francois Cooren, Michele Di Francesco, Sandro Nannini, Bob Sanders, Paul Smolensky, and regret that the suggestions I have drawn from those discussions are not incorporated in this paper. I hope to take them into account in my further work on Gricean issues. This research has been made possible by a joint grant by the Ministero dell’Università e della Ricerca Scientifica e Tecnologica (MURST) and by the Università degli Studi di Trieste. The national project, on “Conoscenza e cognizione,” is coordinated by Paolo Parrini, Università degli Studi di Firenze; the local project, on “Linguaggio, comprensione e conoscenza,” is coordinated by myself.

5 Presupposition, Implicature, and Context in Text Understanding 1. Two Perspectives on Text Understanding It is by now widely recognized that discourse understanding involves more than the understanding of what is explicitly said. In order to understand discourse, or (as I shall say here) in order to understand a text, we must understand more than what is encoded in the text itself, and draw inferences. This broader comprehension, which is closely connected with contextual knowledge, is often described as the comprehension of what is presupposed and/or implicated by the text. There are two main ways in which we can conceive of the role of this broader comprehension (which I will call comprehension of the implicitly conveyed meaning) with respect to the overall understanding of the text. (i) Sometimes it may seem that the understanding of what a text presupposes or implicates, as well as the contextual knowledge involved in such understanding, are necessary conditions for a full comprehension of the text. If we do not know the circumstances in which a text has been written, or if an utterance is reported to us without any information about the circumstances in which and the goals for which it has been uttered, we may not be able to make sense of it. If we do not share the speaker’s pragmatic presuppositions (the assumptions he or she takes for granted in speaking) (Stalnaker 1973, 1974), we might misunderstand him or her. As to conversational implicatures, they depend on the assumption that the speaker is observing the Cooperative Principle (Grice 1989, 26) and therefore, in order to infer them, the hearer should already know whether, in the circumstances of the ongoing verbal exchange, the Cooperative Principle holds. According to Relevance Theory (Sperber and Wilson 1995), another pragmatic theory concerned with discourse understanding, in understanding a text we have to take into account the contextual premises which make the speaker’s contribution relevant. In all these ways, knowledge of, or at least beliefs about, what may in one word be called “the context” are Originally published in Modeling and Using Context. Second International and Interdisciplinary Conference, CONTEXT ’99, Trento, Italy, September 1999. Proceedings, ed. Paolo Bouquet, Luciano Serafini, Patrick Brézillon, Massimo Benerecetti, and Francesca Castellani, 324–38. Berlin: Springer, 1999. Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0006

, ,    

91

represented as necessary to the comprehension of the text. This might lead us to conclude that we can, and must, acquire knowledge of or beliefs about the context prior to, and independently of, our understanding of a text. (ii) Suppose, however, that we find ourselves in a situation in which we have little independent access to the context, as it happens in reading, in certain phone calls, or in those cases of face-to-face interaction in which we know little about our interlocutor and his or her possible aims. Should we despair of making sense of the text we are faced with? In such cases, it might be convenient to exploit all the details of the text in order to project as much of its context as we can. After all, many presuppositions have linguistic markers or triggers, and this enables receivers to detect them even in absence of text-independent information. As to implicatures, it could be claimed that the speaker’s observance of the Cooperative Principle must not be known in advance, but can be assumed in absence of evidence contrary to it, so as to allow for the working out of as many implicatures as possible. Finally, Relevance Theory admits of the possibility of inferring missing contextual premises, when the assumptions which are already available to the hearer do not make the speaker’s contribution relevant. In this perspective, context (or more precisely, the representation of context which is associated with the understanding of the text) is not something which has to be given independently of the text, but something constructed in the very process of text understanding. Although I do not want to deny that perspective (i) has its merits, here I am going to adopt perspective (ii), because I would like to outline a description of the text–context relationship which optimizes the chances of text understanding even in unfavorable conditions. In this framework, I would like to claim that presupposition and implicature play different roles with respect to text understanding and that therefore they should be considered as distinct phenomena. This runs contrary to most of the literature on the subject: presupposition and implicature belong, as it were, to two different conceptual frameworks and authors who use one of these notions do not use the other, so that only one of them does all of the work. Those authors who mention both notions have (since Karttunen and Peters 1979) identified presupposition with one kind of implicature, conventional implicature. I would like to claim that presupposition is different from both conversational and conventional implicature as to the role it plays with respect to what is asserted by a text and to its context.

2. On Text and Context I choose here to use the word “text,” following the semiotic (Hjelmslevian in particular) rather than the philosophical tradition, in a sense akin to the one

92

, ,    

recently specified by Stubbs with reference to the practice of discourse analysis: “By text, I mean an instance of language in use, either spoken or written: a piece of language behavior” (1996, 4). This definition of “text” leaves it open which size a text should have: a text (the relevant piece of language behavior) could well coincide with the utterance of one sentence, but might also consist of the utterance of more or, for that matter, less than one. The utterance of a syntactically complete and isolated sentence is therefore one case falling under the more general idea of the production of a text. As to the problem of text delimitation, it should be remarked that, whenever what is focused upon as a text is in turn a part of a larger episode of language behavior, we can (i) include relevant parts of this larger episode into the text focused upon, thus changing the delimitation of what is under consideration or (ii) consider the larger episode of language behavior in which our text is embedded as a part of the context. Choice (i) turns linguistic context into text, while choice (ii) considers linguistic context as a part of the context. As to context, I believe that insofar as we are concerned with its capacity of being that against which a text is evaluated (as to appropriateness and/or truth), it must be conceived of as “objective” or mind-transcendent (Gauker 1998). It is only with respect to something external to speakers and independent of what is focused upon as the presently considered text, that it makes sense to evaluate, or attempt to evaluate, that text as a piece of linguistic behavior. I will here conceive of objective contexts in an intuitive way, namely, as consisting of the set of facts which have to be taken into consideration by the participants if a given verbal exchange is to achieve its goals. One problem with this view it that evaluation may (or perhaps must) remain provisional, or defeasible. But this trouble is shared by all of our knowledge, which aims at objectivity, but is nevertheless persistently defeasible. Here, however, we will not be concerned with the function of context in text evaluation, but in text comprehension. In particular, in conformity to perspective (ii) outlined above, I want to specify the ways in which the information contained in a text can tell or show us something about its context. So we will be concerned with that representation of the context, relative to a given text, which can be worked out on the basis of that text in the process of understanding it. This representation of the context is, of course, directed at the objective context, but should not be confounded with it, since it has to be worked out by the participants, while the objective context transcends their cognitive processes.

3. The Dynamic Relation between Text and Context A text entertains a dynamic relation with its context. During text production, the addition of new speech acts to those already performed can be described as having context-changing effects (Gazdar 1981), so that the context at time 2 is different

, ,    

93

from the context at time 1 as regards the addition or elimination of some contents. On the notion of text I am using here, a text T1 is correspondingly changed into a text T2 as soon as new parts are added to it. The updating of the context is always relative to a new delimitation of the text, so that one-to-one correspondence between texts and contexts is preserved. However, the distinction (outlined above) between the objective context and the representation of it has to be taken into account. It might be thought that the objective context, being independent of the text, cannot be changed by it. In fact, nonverbal actions, bearing on the circumstances relevant to the goal of the exchange within which the text is produced, change the objective context, but they do not belong to the text either. However, it has to be conceded that the very occurrence of linguistic behavior changes the context, providing part of the context for subsequent text production. Moreover, it can be claimed that a text performs a context-changing action if the speech acts it contains have effects consisting of the bringing about of intersubjectively recognizable states of affairs (such as new obligations or rights or their cancellation). I have elsewhere called such effects “changes in the conventional context” (and following Gazdar 1981, I have defined illocutionary acts in their terms: 1987a, 1989a), where the “conventional context” may be construed as a specialized part of the objective context (insofar as we believe that human conventions too have their own kind of objectivity). Although the consideration of the ways in which the objective context is affected by changes can be an important issue, here we will be concerned with the changes which are produced in the representation of the context as a part of the process of text understanding. We will be concerned with (some) changes in the objective context only insofar as these play a role in inducing changes in the representation of the context. I shall try to describe three main ways in which the latter kind of change can be achieved: assertion, implicature, and presupposition. The proposed description will enable me to claim that presupposition cannot be identified with implicature.

4. Assertion I consider assertion as bringing about the addition of its content to the representation of the context. This view is partly inspired by the one proposed by Stalnaker (1978, 1998). But Stalnaker considers the context as a set of assumptions which the speaker takes as shared, while I am here drawing a distinction between the objective context and the representation of it. The representation of context associated with text understanding does not necessarily consist of assumptions actually made by the speaker, nor of assumptions which the speaker takes as shared by the participants. I view it, basically, as the information

94

, ,    

about the objective context which the text enables, and entitles, its receivers to work out. After a certain assertion has been made, both the speaker who has made it and is committed to its truth, and the hearer (unless he or she decides to challenge the speaker’s assertion) take the content of the assertion to be part of the representation of the context. It is the speaker who has the responsibility for this addition to the representation of the context and the sanction to which the speaker is liable if the resulting representation of the context later turns out to be inadequate is that he or she will be deemed to have said something false. However, it is in part also the hearer’s responsibility to accept the addition of the content of the assertion to the representation of the context. In fact, the hearer might well choose to challenge the speaker’s assertion. We could envisage the assertion as a proposal on the part of the speaker to add a certain content to the representation of the context, a proposal which the hearer might refuse, but which, if not refused, is effective by default (as has been claimed, albeit in a framework different from mine, by Perrault 1990). It should be noted that assertion, if it is considered as an illocutionary act (at least on my understanding of what an illocutionary act is), should involve a change in the objective context too, and more precisely, in what I have called above “the conventional context.” The speaker’s commitment resulting from assertion can be considered as the attribution of a new obligation to the speaker, since the speaker is then obliged not to contradict him or herself and to give evidence or reasons if his or her assertion is challenged. This obligation can be considered as a fact and, insofar as it is relevant to the goals of the conversation, it belongs to the objective context. Since no conventional change is unilateral (obligations assigned to one partner are usually countered by rights assigned the other, and vice versa)¹ it could be claimed that the hearer is modified in his or her turn by the acquisition of a right, which can be construed as the right to make the same assertion him or herself, or as the right to claim second-hand knowledge. These conventional changes justify the changes in the representation of the context associated with text understanding, but should not be identified with or reduced to them. The former changes can be described as changes in the set of modal predicates (“can,” “ought to”) to be attributed to the participants, while the latter consist of the addition of new content to the representation of the context. A trouble about assertion regards its relationship to “what is said” by a text. We are in need both of a notion of assertion and of a notion of “what is said”: we need the former in order to describe one kind of change in the representation of context, and the latter in order to contrast it with all the aspects of the overall meaning of a text which are understood by inference. Following Bach (1994),

¹ [See Essay 1, Section 2.]

, ,    

95

I shall assume that these are two separate notions, and that what is asserted (or, in general, what the content of a speech act amounts to) may draw on inferences from what is said.

5. Implicature The notion of implicature, proposed by Grice in 1967 (1989, 22–40), is well known. In order to characterize the way in which implicatures contribute to changes in the representation of the context, I will recall some of their salient features. Implicatures are invited inferences in which the inferred proposition bears no truth-functional relationship to any utterance contained in the text: when “p” implicates that q, the falsity of q has no consequence on the truth-value of p. So, for example: (1)

Mary is pretty, but intelligent

conveys by implicature that Mary, being pretty, is not likely to be intelligent, but is not false nor wholly unacceptable if this is false, since the truth-functional conjunction of “Mary is pretty” and “Mary is intelligent” can well be true. Likewise: (2)

Jane has two children

(as issued in the framework of a cooperative conversation about how many children certain people have) conveys by implicature that Jane has no more than two children; but is not false if this is false, since, if she has four children, it is still true that she has two. There are two main ways in which such inferences arise: (i) they can be invited by the fact that a certain word is used, which (because of linguistic conventions) has the function of inviting that inference (as is the case in (1) above, containing “but”: a “conventional” implicature); or (ii) they can be required in order to make viable an interpretation of the speaker’s linguistic behavior as conforming to the Cooperative Principle. So, in the case of example (2) (a “conversational” implicature), the implicature arises from the assumption that the speaker is conforming to the Cooperative Principle and, more specifically, is giving as much information as is required by the goals of the conversation. In case (i), Grice has suggested that the inferences which arise are connected with the performance of “non-central” speech acts, namely, further specifications of the central speech acts of asserting, asking, commanding (1989, 121–2, 362). In this vein, “but” may be taken as indicating an objection, or “therefore” an

96

, ,    

explanation. This view of conventional implicatures contrasts, however, with the idea also expressed by Grice that they might derive from the conventionalization of generalized conversational implicatures (1989, 39): reference to speech act notions does not seem to be necessary in this case. I believe that the connection of certain words conventionally suggesting implicatures with non-central speech acts is a puzzling fact, the role of which has still to be thoroughly explained, but I will not tackle this issue here. I shall distinguish conventional from conversational implicature only on the basis of the fact that the former is invited by the use of certain words. In case (ii), in which the Cooperative Principle is involved, the inferences can be drawn on the basis of what is said (the words uttered, considered as the starting point for all the inferences which can be drawn from the text) or on the basis of what is asserted (which may already involve inferences). In the former case, they contribute to the content of the assertion which is actually made; in the latter case, they associate with the assertion an additional content, which is conveyed together with the assertion but not as a part of its content. So, since the words used in: (3)

You are the cream in my coffee

(said to a human being) would give rise to a patently false assertion, if this utterance is to be interpreted as giving a cooperative contribution to the conversation, the asserted content must be somewhat different, although related (“You are my pride and joy”) (Grice 1989, 34). And in the following exchange: (4)

A: Where’s Bill? B: There’s a yellow VW outside Sue’s house

B’s contribution is a relevant answer to A’s question only insofar as it licenses the inference that if Bill has a yellow VW, he may be in Sue’s house (Levinson 1983, 102). I have couched my examples as inferences from the utterance of a sentence “p” to its implicature that q. This choice does not reflect one feature of implicature, that is, the fact that in the Gricean framework, what the hearer infers is primarily not the proposition that q, but the proposition that the speaker thinks that q. This is especially true for conversational implicatures, since the requirement for interpreting the speaker as conforming to the Cooperative Principle, on which they depend, is not that q holds, but that the speaker thinks that q. Here, however, we are concerned with text understanding, particularly with the ways in which a text enables its receivers to update their representation of the context. Implicature is relevant for us only insofar as, whether with or without the mediation of the proposition that the speaker thinks that q, it licenses inferences about facts in the

, ,    

97

objective context. Facts about what the speaker thinks are not necessarily relevant to the goals of an exchange and therefore are not always to be taken into consideration in the representation of the context. I take it that in examples like those I have quoted, the relevant inference which is licensed is about facts in the objective context and, therefore, is not generally about what the speaker thinks. Thus, implicatures are contributions or additions to the content of the speech act performed by the text unit. Their contribution to the update of the representation of the context is, so to say, on a par with that given by assertions (cf. Thomason 1990, 351–2). They convey information which either contributes to the information conveyed by assertions or supplements it. In the former case, the changes in the representation of the context associated with the text can be explained in the same way as for assertion. In the latter case, the speaker is not committed to the truth of the implicated content so strictly as to the truth of the content of an assertion (as is clear from the fact that, when what is implicated is false, the speaker is not responsible for saying something false). Correspondingly, the implicature counts merely as a suggestion, which makes a certain update of the representation of the context available to the participants.²

6. Presupposition Semantic presupposition was introduced as a relationship between an assertion and a proposition whose truth is a necessary condition for the assertion to have a truth-value (Frege 1960; Strawson 1950a). This account was meant to capture the ordinary intuition that when the presupposition of an assertion is false, the question whether the assertion is true or false doesn’t arise. But the account had also some flaws. It required abandoning standard two-valued logic in favor of a three-valued one. Besides, one of its claims is highly questionable: according to semantic presupposition theorists, if the utterance of an affirmative sentence has a presupposition, the same presupposition is shared by the corresponding negative sentence, so that, for example: (5a)

John has stopped smoking

(5b) John has not stopped smoking both presuppose: (6)

John used to smoke.

² [A similar idea was independently put forward by Saul (2002a).]

98

, ,    

To this, it can be objected that if John never used to smoke, an utterance of the negative sentence can well be considered as true: (7)

John has not stopped smoking (in fact he never used to smoke).

The relationship of the presupposition to the utterance of the positive sentence appears not to be identical to its relationship to the utterance of the negative sentence, since in the latter case the presupposition is cancellable. Since the 1970s, a pragmatic conception of presupposition has been preferred to the semantic one. According to this conception, presuppositions are the assumptions shared by speaker and hearer, which form the background of their ongoing discourse (Stalnaker 1973, 1974). Some if not all of these shared background assumptions have linguistic markers (or triggers). An utterance can be said to presuppose a proposition when it contains a linguistic element which functions as a presupposition trigger, and is therefore appropriate only if the associated presupposition is among the interlocutors’ shared assumptions. A problem for this approach is raised by the informative use of presuppositions. Often utterances which contain presupposition triggers are issued without assuming that the hearer already shares their presupposition or even knowing that he or she does not share it. In these cases, contrary to the theory’s predictions, no inappropriateness is felt, but the hearer “accommodates” the presupposition by adding it to his or her own background assumptions (Karttunen 1974; Lewis 1979). To describe this phenomenon in the framework of the pragmatic conception of presupposition is undoubtedly difficult and none of the answers which have been proposed (in terms of the speaker pretending to presuppose something: Stalnaker 1974, or in terms of the hearer being prepared to add the presupposition to the context without objection: Soames 1982) are fully satisfactory. In fact, a speaker would more successfully pretend to take something for granted by not mentioning it at all; and a hearer might well find an informative presupposition objectionable, without therefore considering the utterance which conveys it as inappropriate (Gauker 1998). The picture of presupposition I would like to outline aims at recapturing the original intuition that when the proposition presupposed by the utterance of a sentence does not hold, that utterance is in some way out of order. In the perspective adopted here (see Section 1), we are not interested in background assumptions which have no textual manifestation, but only in those presuppositions which are linguistically triggered. I propose to consider those presuppositions not as shared assumptions, but as assumptions which ought to be shared. If we admit, as we have done above, that conversations are governed by objective contexts (the content of which is selected by the goals of the conversation), and that only sentences whose presuppositions are satisfied by the objective context are appropriately assertible, it follows that a hearer, by deeming that a presupposition which is triggered by a text is

, ,    

99

not satisfied by the context, is considering the speaker not merely as being factually wrong, but as violating a normative requirement. Such a judgement on the speaker’s linguistic behavior would lead to a communicative breakdown. In fact, this is what happens when, faced with an assertion whose presupposition does not hold, we feel we do not know how to reply. Given the general tendency to avoid communicative breakdowns, in all the cases in which there is some possibility left that the objective context does satisfy the presupposition, the hearer’s default tendency will be to assume that it is so, namely, that the presupposition is in fact satisfied. What is the function of presuppositions, so conceived, with respect to the representation of context? Presupposed propositions have to be included into the representation of the context associated with a given text just because some utterance belonging to the text is in order only if the presupposition is satisfied by the objective context. Irrespective of whether the presupposed propositions are old or new information for the hearer, and more generally, irrespective of whether they initially belong to the speaker’s representation of the context, to the hearer’s, to both, or to neither, they must find a place in the representation of the context which is worked out in the process of text understanding. Thus the addition of the content of an assertion to the representation of the context is accompanied by the obligation to add to the representation of the context those presuppositions of it which happen not to be yet there. This obligation, which is not to be confounded with the speaker’s commitment relative to an overt assertion, concerns both speaker and hearer and protects the presupposed proposition from challenges, giving rise to the characteristic feeling that the presupposed proposition is, or is to be, “taken for granted.” The function of linguistic presupposition triggers is to indicate that speaker and hearer ought to take the presupposed content for granted. To say that a text has a certain presupposition means, therefore, that it conveys something as having to be taken for granted.

7. The Distinction between Presupposition and Implicature Pragmatic presupposition and implicature have often been considered as one and the same phenomenon. Within the framework outlined above, I will now put forward some remarks which give reasons for rejecting this claim.

7.1 Interactions with Asserted Contents We have described the function of implicatures with respect to the changes in the representation of the context as either that of contributing to the change brought about by an assertion or that of making a supplement to it available. The function

100

, ,    

of presuppositions seems to be quite different. They do not contribute to the content of assertions, as do implicatures such as that exemplified by (3), nor suggest supplementary information, as do implicatures such as those exemplified by (1), (2), and (4), but set requirements for the acceptability of utterances, the satisfaction of which depends on the objective context. This difference is confirmed by the following observations. We have said that the propositions presupposed by an utterance have to belong to the representation of the context and are therefore added to it when they are not yet there. Now, of course it is possible that they are already there, because they have just been asserted, and in this case the presupposition does not bring about any update. Nevertheless, the utterance which contains the presupposition trigger can still be said to carry the presupposition. So, in the following dialogue: (8)

A: Billie likes ice cream B: Susie likes ice cream too

B’s utterance can be said to presuppose that someone other than Susie likes ice cream, since its appropriateness requires this presupposition to be satisfied by the objective context. The presupposition of B’s utterance has no business updating the representation of the context, because it has already been said by A that Billie likes ice cream; but B’s utterance makes an appropriate contribution to the ongoing dialogue just because of this. If we wanted to interpret “too” as conventionally implicating, rather than presupposing, that someone other than Susie likes ice cream, we would have no explanation for its use by B, since it is clear that no suggestion is made to supplement the content of the assertion in any way (cf. Thomason 1990, 361). In fact, the utterance of a sentence like: (9)

Jane has not four children, nor three; she has two children

just lacks the implicature that Jane has no more than two children. There is no need to draw such an inference, and moreover, there would be no point in supplementing what is asserted by associating the implicature with it, since the same content has already been explicitly asserted. Likewise, the utterance of a sentence containing a word which invites a conventional implicature, such as: (1)

Mary is pretty, but intelligent

does not sound completely natural if a content corresponding to its conventional implicature has just been explicitly stated. Consider:

, ,     (10a)

101

Being pretty, Mary is not likely to be intelligent. She is pretty, but intelligent.

The “but” here seems at least misplaced. In fact, the following sequence of utterances makes by far more sense: (10b)

Being pretty, Mary is not likely to be intelligent. But she is both pretty and intelligent.

Here, however, the conventional implicature conveyed by “but” has changed, amounting now roughly to: (11)

The fact that Mary is both pretty and intelligent runs contrary to the assumption that, being pretty, she is not likely to be intelligent.

A related phenomenon which can be observed is the following. There are cases in which the content of a presupposition is linguistically formulated in the very utterance which triggers the presupposition (for example, in a subordinate clause). This happens regularly with factive verbs: (12)

John realized that he was in debt

contains, as a subordinate clause, the sentence “he was in debt,” corresponding to the presupposed proposition that John was in debt. Thus, the presupposed content is at hand and no inference is needed in order to grasp it. Also in the cases of itclefts and of definite descriptions, practically all of the linguistic material needed for making the presupposition explicit is at hand in the sentence which triggers the presupposition. But this does not make presupposition pointless: its point is not to invite additional inferences, but to convey that a certain content has to be taken for granted. In contrast, implicatures involve inference: conversational implicatures depend on a heuristic strategy relying on the Cooperative Principle, and conventional implicatures license inferences on the basis of the use of certain words, but in neither case is the content of the implicature linguistically encoded to such an extent as to make inference unnecessary. This would eliminate the implicature itself.

7.2 Implicatures and Background Assumptions Another consideration which might lead to identify implicature and presupposition is that sometimes at least, implicatures seem to coincide with background assumptions of the speaker. Thus in:

102

, ,    

(13)

A: Are you going to invite John to your party? B: No. I’m inviting nice people

it might be said that B’s comment is understood as relevant only on the background of B’s assumption that John is not nice. This assumption might therefore be considered as being “presupposed” by B’s reply to A. Now, it is indeed very likely that B is taking for granted that John is not nice. But does this contingent fact about B suffice to make his or her assumption a presupposition in the sense we have outlined above? Does the appropriateness of B’s utterance require the satisfaction of the presupposition that John is not nice by the objective context? Is “I’m inviting nice people” in any sense not appropriately assertible if this requirement is not met? It does not seem to be so. There is no normative requirement to be met in order to make the utterance appropriate, and therefore no normative requirement for the representation of the context to contain the proposition that John is not nice. In contrast, it is obviously necessary for a hearer to take B’s reply to be relevant, if he or she aims at a full understanding of its point, and this assumption of relevance is enough to suggest the conversational implicature that John is not nice as an integration of the explicitly conveyed information. The representation of the context will be updated by adding not only the information explicitly conveyed by B’s utterance but also the content of the conversational implicature. The content of an implicature, on our view, can happen to be “shared knowledge” because of previous experiences and interactions shared by the participants. This does not make the implicature useless, as would do the explicit assertion of its content within the same text. There is still some inferential work to be done, which will establish a relationship between the text and a certain content which happens to be a shared piece of knowledge, and remind the participants of it. But this does not suffice to turn the implicature into a presupposition.

7.3 Negative Sentences and Presupposition Cancellation The claim that presupposition really is implicature has sometimes been raised with particular respect to the presuppositions of negative sentences. It has been claimed that the relationship between the utterance of a negative sentence and its alleged presupposition is in fact an implicature, while the relationship between the utterance of a positive sentence and its presupposition can be reduced to entailment (Chierchia and McConnell-Ginet 1990). This claim is based on the fact, noted above, that: (6)

John used to smoke

, ,    

103

seems to be a cancellable presupposition of: (5b)

John has not stopped smoking

while in connection with: (5a)

John has stopped smoking

it does not seem to be deniable without contradiction. Moreover, from the negation of (6): (14)

John never used to smoke

(5b) seems to follow, just as it should be the case if the relationship between (5a) and (6) were one of entailment. Now, I do not think it is necessary to resort to implicature in order to explain the cancellability of (6) as a presupposition of (5b). Presupposition cancellation can be viewed as one aspect of the same process which, by default, leads to presupposition accommodation (Heim 1983, 401). In our terms: the representation of the context has to be internally consistent; two propositions contradicting each other should not be both added to it; if an utterance triggering a certain presupposition is introduced in a text at a point at which previous assertions have already added some content contradictory with that presupposition to the representation of the context, the obligation to include the presupposition in the representation of the context is suspended. So, when (5b) is introduced as a consequence of (14), it cannot carry the presupposition that John used to smoke and the appearance of entailment reversal is created. Such an explanation of presupposition cancellation is quite compatible with the fact that (5b), considered in isolation, has an intuitive relationship with (6), while nothing similar occurs in standard cases of entailment, and permits to describe this relationship as a presupposition triggered (in default conditions) by the verb “to stop,” that is, in the same terms in which the relationship between (5a) and (6) can also be described. It does not seem justified, therefore (at least, it is not economical), to consider the presuppositions of the utterance of a positive sentence and of its corresponding negative sentence as instances of two different relationships, one of which an implicature.

8. Concluding Remarks Concluding then, there are reasons for considering presupposition as a phenomenon distinct from implicature. Presuppositions play a specific role in the ways in

104

, ,    

which we project the representation of context from our acquaintance with a text. Presuppositions convey that a certain content has to belong to the representation of the context, irrespective of whether it does already belong to it or not and of whether inferences going beyond the rearrangement of linguistic material contained in the text are needed. The ways in which implicatures contribute to the representation of context are undoubtedly different: they aim at bringing about updates of the representation of the context, either by contributing to the content of assertions or by suggesting supplementary information, and typically require inferences going beyond the linguistic material contained in the text. The fact that in some cases what is inferred may already belong to the participants’ shared knowledge is not enough to make the implicature a presupposition. Moreover, resort to implicatures in order to explain the cancellable presuppositions carried by the utterance of a negative sentence, appears to be an unnecessary complication. Research on text understanding has the task to specify the strategies which enable the receiver to recover implicitly conveyed information and which may be used to justify such a recovery, distinguishing it from the working out of mere psychological associations not warranted by the text. These strategies are clearly connected with the ways in which implicit information is conveyed, namely, the ways in which it contributes to the representation of context. If, as I have tried to show, presupposition and implicature contribute to the update of the representation of context in different ways, for the aims of research on text understanding it will be useful to distinguish them from each other. Therefore, the development of a conceptual framework in which presupposition and implicature coexist and receive separate, complementary definitions turns out to be highly desirable.³

³ Thanks to Christopher Gauker and to an anonymous referee for comments on earlier drafts. [In order to be read at the conference “Context ’99,” this paper had to be put in final form before the conference itself. I now add my heart-felt thanks to the audience at the conference, especially François Recanati, for the discussion.]

6 Illocutionary Force and Degrees of Strength in Language Use 1. Introduction Research in pragmatics has shed light on a broad field of linguistic phenomena regarding “mitigation” or “attenuation” as well as “strengthening” or “reinforcement”¹ in language use (Bazzanella et al. 1991; Blum-Kulka et al. 1989; Caffi 1999; Dressler and Merlini Barbaresi 1994; Fraser 1980; Holmes 1984). This research field is of central importance for our conception of the use of language in communication: in tackling it, we are forced to abandon oversimplifications and idealizations, and to recognize the multi-dimensional nature of language use, ranging from cognitive to emotional facets, from actional to affective ones, from the social to the subjective. The aim is to account for language use in all its irreducible complexity. But there are risks in this enterprise. First, pragmatics might become like a Borgesian map: a description of language use so attentive to its multidimensionality and its innumerable, though still significant, details, that this description ends up by reduplicating that which it purports to analyze, thereby risking the loss of any informative and explanatory potential. Second, any attempt to analyze language use in all its stylistic complexity is liable to rely on ad hoc tools in an eclectic vein, and may therefore neglect basic theoretical issues. In this paper, I choose an intermediate route: I consider the relationship between the mitigation/reinforcement phenomena and speech act theory. Can mitigation and reinforcement be dealt with in a speech-act theoretical framework? I contend that they can and will substantiate this claim by analyzing examples of mitigation and reinforcement in terms of “degrees of strength” of the speech act involved (a notion first used by Holmes 1984). In doing so, however, I do not make use of the standard, mainstream version of speech act theory (Bach and Hamish 1979; Searle 1969, 1979; see also Tsohatzidis ed. 1994). I contend that some aspects of standard speech act theory are incompatible with the features of Originally published in Journal of Pragmatics 33 (12): 1791–1814. Amsterdam: Elsevier, 2001. ¹ The term “aggravation” has also been used (since Labov and Fanshel 1977). I prefer “reinforcement” because I do not believe that the upgrading of intensity is necessarily correlated with aggressivity or threat to face.

Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0007

106

       

linguistic action emerging from research on mitigation/reinforcement. Instead, I argue that a revised conception of the speech act (of Austinian inspiration) such as the one I have proposed in Sbisà (1989a) might be compatible with the results of the research on mitigation/reinforcement and even contribute to a better understanding of certain features underlying these phenomena. In this perspective, mitigation and reinforcement appear not as superficially adjoined to independently performed speech acts, but as closely connected to, or even identical with, the adjustment and tuning of the effects that the speech act brings about.

2. Why a Unified Account of Mitigation/Reinforcement and Illocutionary Force Is Desirable The term “illocutionary force,” a key term in speech act theory, is generally used to refer to the fact that in the uttering of a sentence, an illocutionary act of a certain kind is performed, that is, the utterance counts as a certain kind of move in verbal interaction: a command rather than a question, an assertion rather than an apology or a promise. I will now outline some of the reasons that make it desirable to give a unified account of such a phenomenon and of mitigation/reinforcement in language use. (i) On the one hand, force—a notion coming from Frege (1972, 1997), where it applies to assertions and yes–no questions—has been considered since Austin (1975) as one of the aspects of language use that make speaking the doing of something. It pertains to illocutionary acts (acts of doing something in saying something) just as meaning (sense and reference) pertains to locutionary acts (acts of saying something) (Austin 1975, 100). On the other hand, mitigation and reinforcement in language use are tied to matters such as the adjustment of the relationship between the interlocutors, the achievement of goals, the avoidance of undesirable consequences, and therefore to what we do in verbal exchanges. It would be reasonable to expect that there should be some relationship between illocutionary force and mitigation/reinforcement, namely, that these should correspond to modifications or adjustments of what is done in saying something. (ii) There are some features of the speech act that seem to be affected by variations in degree. First and most obviously, as was stated most clearly by Searle (1979, 4–5), each illocutionary force is connected with the expression of an inner state. Since inner states typically vary in intensity, in order to fully characterize a speech act that expresses some inner state of the speaker, it would be sensible to take into consideration not only the kind of state expressed but also its degree of intensity, and therefore mitigation/reinforcement phenomena. Second, there are speech acts the

       

107

performance of which is influenced, or influences, matters of power. Sometimes what entitles the speaker to perform a speech act is his or her status within the ongoing interaction or “local rank” (Diamond 1996), which can at least in part be negotiated (and thus upgraded or downgraded) during the interaction itself. Other speech acts can be performed only by speakers who are already endowed with the appropriate status, and in these cases the speaker can either appeal to such status openly or politely avoid exhibiting it. In both kinds of cases, illocutionary force and mitigating or reinforcing strategies work together in defining or redefining the kind and level of the speaker’s power as well as the extent to which it bears on the relationship between speaker and addressee. (iii) It is generally recognized that there are different ways of performing acts having the “same” illocutionary force: for example, there are different ways, with varying degrees of explicitness and directness, of requesting or apologizing or advising or undertaking commitments. It has also been recognized that these different strategies have different effects on the interpersonal relationship between the interlocutors: some may tum out to be disruptive, while others may mitigate the threat to “face,” avoid conflict, or be reassuring (Brown and Levinson 1987, 68–75; Geis 1995, 97, 181–3). Certainly, it is still possible to consider these different strategies as mere stylistic variations of an act whose essence remains untouched. But it cannot be denied that what the speaker does in speaking, changes (at least to some extent) depending on the different strategies employed. (iv) In research on the ways in which speech acts are performed in different languages and cultures (see in particular Blum Kulka et al. 1989), the linguistic indicators of mitigation and reinforcement (“upgraders” and “downgraders”) are considered in the context of the strategies for performing speech acts as in principle distinct from illocutionary force indicators. However, it has also been noticed (Blum Kulka 1985; Caffi 1999; Sbisà 1990) that sometimes at least, the relationship between mitigating or reinforcing devices and illocutionary force is closer. There are cases in which mitigation (e.g. the use of a particle like “please”) contributes to characterizing the speech act as a Directive (“Can you reach that book, please?” is more definitely a Directive than “Can you reach that book?”), while in other cases emphasis on the expressed inner state contributes to characterizing the speech act as an Expressive (“Oh, it is so pretty” is more clearly a compliment than “It is pretty”). A unified account of illocutionary force and of mitigation/reinforcement could explain these overlaps. (v) Finally, the connections established by a unified account encompassing both illocutionary force and mitigation/reinforcement would make the methodological framework of pragmatically oriented discourse analysis more economical and perhaps more illuminating.

108

       

I do not claim that any of the points (i)–(v) above gives conclusive evidence for the desirability of a unified account of illocutionary force and mitigation/reinforcement. But taken together, they strongly suggest that such an account is desirable and that its viability is worth exploring.²

3. Conditions for a Unified Account of Illocutionary Force and Mitigation/Reinforcement Phenomena In order for a theory of speech acts to provide a unified account of illocutionary force and mitigation/reinforcement, that theory must be able to satisfy at least the following conditions: (i) The conception of illocutionary force should be compatible with its being affected by variations in degree of strength. (ii) Descriptive tools for speech acts should be able to account for variations in degree of strength. (iii) It should be possible to describe mitigations and reinforcements by means of the same descriptive tools used to describe speech acts. I will now argue that the standard conception of illocutionary force does not meet condition (i). Since condition (i) is logically prior to conditions (ii) and (iii), this will be enough to show that the standard conception of illocutionary force cannot be included in any unified framework encompassing mitigation/reinforcement phenomena without modification. Subsequently, I will argue that a reformulated conception of illocutionary force could meet all three requirements and qualify as suitable for dealing with degrees of strength in language use.

3.1 The Standard Conception of Illocutionary Force Since Searle (1969, 46–9), the illocutionary act has generally been conceived as the act a speaker successfully performs when, uttering a sentence with a certain intention in certain circumstances, he or she gets the hearer to understand his or her intention. The speaker’s communicative intention determines what illocutionary act he or she should be taken to perform and therefore what illocutionary force his or her utterance may have. The effect characteristic of the illocutionary

² Such viability has been relied upon rather than discussed by Bazzanella et al. (1991) and Caffi (1999), who have distinguished between kinds of upgrading and downgrading devices on the basis of the relationship between the corresponding mitigation or reinforcement phenomena and different aspects of the speech act.

       

109

act (its “illocutionary effect”) has been identified with the recognition on the part of the hearer of the speaker’s communicative intention. The speaker’s communicative intention must be a definite one. A speech act must have (at least) one or other of the illocutionary points available (committing the speaker to the truth of the expressed proposition, trying to make the addressee do something, and so forth) (Searle 1979, 12–20, 29). It does not matter here whether recognition of the speaker’s communicative intention is achieved by recognizing the conventional meaning of the sentence uttered (as in Searle 1969, and in “direct” speech acts according to Searle 1979, 30–57) or by making inferences (as in Bach and Harnish 1979; Harnish 1994; Sperber and Wilson 1995; and in “indirect” speech acts in Searle 1979, 30–57): since in both cases, the illocutionary effect coincides with the recognition of the speaker’s communicative intention or illocutionary point, it is bound to be as discrete as they are.³ Aspects of the speech act which admit of degrees such as speaker’s status or speaker’s involvement are considered as marginal. According to Katriel and Dascal (1989), variations in degree of the speaker’s involvement do not modify his or her commitment to the act he or she performs. According to Searle and Vanderveken (1985), Directives display differences in degree as to the status that the speaker must have for performing them (from orders, which require the speaker’s higher status, to entreaties, which require the speaker’s lower status), but these differences affect the mode of achievement of the illocutionary point, not the illocutionary point itself. In no case is the “illocutionary effect” subject to variations in degrees of strength or describable in ways admitting of mitigation and reinforcement. Thus, the standard conception of illocutionary force is not compatible with its being affected by variations in degree of strength and cannot meet the abovementioned conditions for a unified account of illocutionary force and mitigation/ reinforcement phenomena.

3.2 A Reformulated Conception of Illocutionary Force Austin’s original conception of the illocutionary act displays at least some features which distinguish it from the standard view presented above. The one most relevant to us is that the illocutionary act is associated with three different kinds of effect (1975, 116–17):

³ One more reason for taking illocutionary force to be discrete is available to those linguists who consider it as tied to sentence mood (Sadock 1994; Sadock and Zwicky 1985): the sentence types of a language form a system and are mutually exclusive (Sadock and Zwicky 1985, 158–9).

110

       

(i) the securing of uptake; (ii) the production of a conventional effect; (iii) the inviting of a response or sequel. Effect (i) amounts to bringing about the understanding of the meaning and the force of the utterance and, unless it is achieved, the illocutionary act is not actually carried out. Effect (ii) amounts to the bringing about of a state of affairs in a way different from bringing about a change in the natural course of events: the act of naming a ship “Queen Elizabeth” makes it the case that this is the ship’s name, and that referring to it by any other name will be out of order, but these are not changes in the natural course of events. Effect (iii) amounts to inviting a certain kind of subsequent behavior; if the invitation is accepted, a certain further act by some of the participants will follow. In the standard, Searlean-Gricean sense the “illocutionary effect” is restricted to Austin’s effect (i), but Austin himself does not suggest that restriction. Rather, he seems to imply that effect (i) is a necessary step for the successful performance of the illocutionary act, which also comprises the achievement of effects (ii) and (iii).⁴ One might think that the last two effects should be dismissed as “perlocutionary,” but in Austin’s view they are clearly not such: perlocutionary acts are the bringing about of changes in the natural course of events (as opposed to effect (ii), which affects a conventional state of affairs) and consist of the actual eliciting of a response (as opposed to effect (iii), which is merely the inviting of a response). In my work on speech act theory (Sbisà 1984a [Essay 1], 1987a, 1989a) I have argued that Austin’s effect (ii), the production of changes not in the natural course of events but in conventional states of affairs, is an essential feature of the illocutionary act and can be detected not only in institutional acts such as the example provided by Austin (1975, 116) but also in illocutionary acts usually performed in verbal interaction (cf. von Savigny, 1988). The production of such a conventional effect is made possible by the hearer’s uptake of the act as bringing about that effect (as suggested by Austin), and conventionality appears therefore to be founded on intersubjective agreement.⁵ The conception of the illocutionary act according to which it has a conventional effect is thus committed to granting an active role to the hearer. Hence, to the standard conception of illocutionary force based on the speaker’s intention (which identifies the illocutionary effect with the hearer’s recognition of that intention), it

⁴ Strawson (1964) seems to have inferred from the fact that Austin (1975, 116) claims that unless a certain effect is achieved, the illocutionary act will not have been successfully performed, that for him the only effect essential to the illocutionary act is the achieving of uptake. On the basis of the whole context in which Austin’s claim occurs, I do not think this interpretation is fair. [For my criticism of Strawson, see Essay 11.] ⁵ I am not thinking of explicitly signaled agreement. If it should become manifest that default conditions do not obtain, implicit or even explicit negotiation will occur.

       

111

is possible to oppose a conception of illocutionary force according to which illocutionary acts have conventional effects, determined at least in part according to the hearer’s uptake. It might be useful to recall that the recognition of the hearer’s active role is independently supported by studies in verbal interaction which have emphasized the cooperation between the participants in determining the meaning of the interactional event and therefore the acts performed (Duranti 1988; Goodwin 1995; Streeck 1980). If these cooperative aspects of interaction are taken seriously, it is impossible to believe that the speaker determines what he or she does by having a certain intention. Indeed, a speaker may even turn out to have done (with his or her words) something he or she did not intend to do. It may be objected that in such cases, what happens is mere misunderstanding. But while there are cases in which the speaker engages him or herself in a repair procedure of the kind described by Schegloff (1992) (and even here, negotiation may come into it), there are also cases in which he or she just lets the hearer’s uptake pass.⁶ A conception of illocutionary force which admits of conventional effects in the sense outlined above allows for degrees of strength. What the speaker has done (the effect of the illocutionary act) is no longer bound to mirror a discrete intention of the speaker. Rather, since there are aspects of the interpersonal relationship that are settled on the basis of intersubjective agreement, the conventional effects of speech acts may be considered as affecting them, and as playing a role in their adjustment and fine tuning. Thus a conception of illocutionary force which does not identify it with the speaker’s communicative intention, but with the fact that utterances have conventional effects of certain kinds, meets condition (i) for a unified account of illocutionary force and mitigation/ reinforcement phenomena. As to condition (ii), regarding the tools to be used in the description of speech acts and their suitability for coping with matters of degree, I have proposed elsewhere (Sbisà 1984a [Essay 1], 1989a, 1992b) that conventional illocutionary effects may be described in terms of “deontic modality,” namely, as assignments or cancellations of modal predicates related to the necessity or possibility of actions with respect to norms to or from each one of the participants: for example, there are speech acts as an effect of which it can be said of the addressee that he or she “can” or “may” or “ought to” or “has to” do something.⁷ The lexicon of deontic ⁶ The objection rests on a view of communication which is only apparently common sense, the “expressivist” view (as Gauker 1994 calls it) going back at least to Locke or even to Augustine, according to which communication consists of the expression by the speaker of a preformed content and of the retrieval of the same content by the addressee. This view was attacked by Wittgenstein (1953) and ignored by Austin; research on interaction, highlighting the role of the hearer, opposes it at least potentially. ⁷ I am referring here to modality as the semantic domain relating to the expression of necessity and possibility. Deontic modality is concerned with necessity and possibility interpreted as the necessity or possibility of actions or modes of behavior with respect to some kind of social rule or norm (cf. Hugues

112

       

modality includes, besides modals and other verbs, a broad range of nouns (Lyons 1977, 825) such as “right,” “obligation,” “entitlement,” and “commitment.” The reason for choosing such a lexicon for the description of conventional illocutionary effects (and therefore of illocutionary acts and illocutionary forces) is that the assignment and removal of deontic modal values to and from agents is basically “conventional,” namely, dependent on social factors such as the relevant intersubjective agreement. Now, deontic modal notions are in many respects matters of degree: they display varying degrees of strength as well as qualitatively different shades of meaning. So, obligations can be legal or moral, and more or less binding; rights may be competencies, capacities, or authorities; the scope of their contents may vary too. Moreover, attributions of deontic modal values have intuitive connections with matters of status and power, which typically come in degrees: to be in a position in which one “can” do more things than someone else is typically at the same time to be in a “one-up” position, i.e., enjoy greater “power.”⁸ So in the analysis of verbal interaction, the lexicon of deontic modality offers considerable potential for capturing degrees and shades. The reformulated conception of illocutionary force can thus be held to satisfy condition (ii) for a unified account of illocutionary force and mitigation/reinforcement phenomena. As to condition (iii), that mitigation and reinforcement may be described by means of the same descriptive tools that are used to describe speech acts, I maintain that a great deal of the mitigating and reinforcing effects highlighted by research on interactional management, politeness, expression of emotions, and so forth, can be described completely or in part by means of the same deontic modal notions which I have proposed to use in the description of conventional illocutionary effects. The remaining part of this paper will be devoted to an attempt at showing that this is so, not by means of theoretical arguments, but by discussing examples. The discussion is not aimed at giving incontrovertible evidence in favor of the reformulated conception of illocutionary force I have sketchily presented. But I would like to show how this proposal can be put to work and what it is good for: in particular, how it can deal with mitigation and reinforcement phenomena.

and Cresswell 1996; Lyons 1977; von Wright 1972). In linguistics, deontic modality has been included in the wider notion of root modality (necessity and possibility depending on socio-physical forces or barriers) (Sweetser 1990). Narrative semiotics has used deontic modalities (“can” and “ought to” applied to doings) to describe the various kinds of “modal competence” that differentiate the respective positions or roles of agents and are presupposed and transformed by their performances (Greimas 1983; Greimas and Courtés 1979). ⁸ That this is so is even more obvious for the speakers of languages such as French or Italian, in which the meaning of the substantivized infinitive of the modal verb translating both “can” and “may” (“pouvoir,” “potere”) corresponds to one meaning of the noun “power.” Sociosemiotics (semiotics used to account for social facts: see Landowski 1989) has exploited this connection.

       

113

4. Cases of Mitigation and Reinforcement The examples I am going to discuss are drawn from a miscellaneous corpus of recorded conversations in Italian.⁹ They do not consist of isolated, clear-cut sentences (as is customary in philosophy), but of textual units, corresponding to conversational tums or functionally separable parts of these. Although cases of mitigation are more common and play a more important role in the polite management of interaction, the examples presented here will regard reinforcement too, because my aim is to test the descriptive potential of my approach in both directions. In discussing the examples, I focus on the conventional illocutionary effect, which I take to be the core component of the set of effects of the illocutionary act outlined by Austin (1975, 116–17) (I will call this the “core illocutionary effect”). In the description of the core illocutionary effect, I distinguish three main components: the deontic modal values which the speaker appeals to in order to perform the illocutionary act (in short, the speaker’s entitlement), the deontic modal values which the illocutionary act assigns to the addressee (e.g., rights, obligations, or cancellations of these), and the deontic modal values which are ⁹ The examples quoted are drawn from various contexts: TV programs, scientific work-groups, psychotherapeutic sessions, courtroom hearings, informal interactions. Some were recorded and transcribed by me, in part with the collaboration of my students, to whom I am very grateful. Others (examples 1, 2, 7, 11, 12, 13) were contributions provided by Carla Bazzanella and Claudia Caffi to the corpus we collected while preparing Bazzanella, Caffi. and Sbisà 1991 (see also Bazzanella 1994). In my research I have also examined transcriptions published by others (Gavioli and Mansfield 1990; Leonardi and Viaro 1990). Although the focus of my paper is not on the Italian language as such, I quote the examples in the original Italian because they are the source material which I have been working on, given my competence as a native speaker. The English translations are not pragmatic equivalents of the Italian examples, but aim at matching their syntactic and lexical-semantic characteristics, thus enabling the readers to identify the mitigating or reinforcing features in the original; punctuation is inserted only to make the text more readable. The Italian words to which the mitigating/ reinforcing effect can be traced back and their English counterparts are both underscored, but this does not imply that the English expression used to translate an Italian mitigation or reinforcement indicator plays an analogous role in English. I have used the following transcription conventions: Beginning of an overlap * or # (in both turns) End of an overlap + or/(in both turns) Latched utterances = Short pause (up to 2”) (–) Pause (between 2” and 5”) (—) Slightly rising or level intonation contour , Rising intonation contour ? Falling intonation contour . Lengthened syllables :, :: Incomplete words xLow volume (^ word) Emphasis bold Non-linguistic phenomena and comments ((xxxx)) Unintelligible words xxx Omissis (...)

114

       

correspondingly assigned to the speaker (the speaker’s commitment) (Sbisà 1984a [Essay 1]). Firstly, I will present examples of mitigation and reinforcement that affect the three components of the core illocutionary effect. Subsequently, I briefly consider cases of mitigation or reinforcement related to two dimensions of the illocutionary act which do not straightforwardly belong (in my description) to its core effect: the expressed inner state and the inviting of a response. My ordering of the examples, and the related distinctions among kinds of mitigation/reinforcement phenomena, differ from the set of distinctions proposed by Bazzanella et al. (1991) mainly because here I do not consider variations in indeterminacy/precision or diminution/augmentation of propositional content separately from variations in the core illocutionary effect. In fact, they contribute to these variations and are described by describing them. For the same reason, I shall not use the distinction between “bushes” (mitigation procedures which focus on the propositional content) and “hedges” (mitigation procedures which center on illocutionary force indicators) proposed by Caffi (1999). This distinction is undoubtedly useful when what is at issue are the linguistic means by which the mitigating effect is achieved, but the core illocutionary effect is affected by both kinds of procedure in closely intertwined ways. That is why “bushes” and “hedges” overlap, as Caffi herself points out.¹⁰ One limitation of my analyses is that the description of the effects of illocutionary acts is based on the analyst’s (i.e., my own) reception of the textual units considered. This might seem to neglect the hearer’s active role. Apart from the difficulty or even impossibility of extracting indicators of the addressee’s uptake from the turns immediately following the one under consideration (it often happens that the use of mitigation or reinforcement devices elicits some cumulative response later on, and not necessarily a verbal one), it should be noted that, although the analyst is never an addressee, he or she does act as a hearer (sometimes even a ratified one, as in the case of TV talk shows). For the aims of the present investigation, it is less relevant that we should know whether the illocutionary effects are actually ratified by the addressee or other participants, than that we should be able to motivate the proposed reading of the conversational turn. However, in many cases it was possible at least to make sure that the addressee’s or the audience’s reaction to the analyzed turn did not disconfirm the kind of reception elaborated on in the analysis of the turn itself. In other cases, notably examples (2), (10), (13), the effects I describe are merely purported, since the participant endowed with the highest conversational power in the context (the TV host, the therapist) skips or represses them. Exactly how the ¹⁰ As to the mitigation devices which Caffi calls “shields,” I must admit that the cases of mitigation in which they are involved largely escape my attempt to give a unified description of force and mitigation/ reinforcement. I believe that their mitigating effect cannot be captured by the description of variations internal to illocutionary effects because they involve shifts or displacements external to them, concerning the basic indexical features of the speech act. Maybe they are akin to “footing” (Goffman 1981).

       

115

principle of the hearer’s active role should be applied in discourse analysis is in need of further discussion.

4.1 Speaker’s Entitlement Unless there is evidence to the contrary, a speaker whose utterance is accepted as the performance of a certain illocutionary act is also taken to satisfy the felicity conditions for that act. Thus a speaker who gives an order appears as someone who has the authority to do so, a speaker who issues a judgment as someone competent to do so, a speaker who makes a promise as someone capable of doing so. Such retrospective definitions of the deontic modal attributes of the speaker may merely confirm them, but they can also refine or alter them (Sbisà 1984a [Essay 1], 1989a). Thus the “speaker’s entitlement” to perform the act can be considered as a component of its core illocutionary effect. In many cases, this entitlement involves some kind and degree of “power” (where “power” is connected with deontic possibility). There are different qualitative shades, from authority or mere influence, to legitimation, capacity, or competence (knowledge, the opportunity to acquire it, or mere linguistic and communicative competence). Each of these shades may be affected in tum by variations in degree: for example, authority may be absolute or subject to various kinds of restrictions; personal influence may be weaker or stronger and relative to a broader or narrower scope; competence may vary in range and reliability. A speaker may want to emphasize his or her entitlement to perform a certain speech act, for example by appealing to the sources of such entitlement: (1)

io non ho mai io sono una donna di teatro ex anche (–) ma di teatro (–) h ho notato in lui (–) ho seguito tutte le trasmissioni di Piero u- un fondo di umanità straordinaria. “I have never I am a woman of the theatre or rather ex, but of the theatre, I have noticed in him, I have seen all of Piero’s programs an extraordinary fund of humanity.”

In (1) a participant in a TV discussion reinforces the evaluation she is expressing (“ho notato in lui . . . un fondo di umanità straordinaria,” “I have noticed in him an extraordinary fund of humanity”) by listing her own credentials (her experience of the theatre, her acquaintance with the object to be evaluated). In this and similar cases, it would be an exaggeration to say that the speaker’s entitlement is strengthened, since the speaker possesses the status required for the performance of his or her speech act independently of the way in which the speech act itself is formulated. But the speaker’s entitlement is certainly emphasized by its

116

       

being openly appealed to. On the other hand, a substantial strengthening of the speaker’s entitlement may occur when his or her status or local rank is not precisely defined and he or she attempts to perform a speech act which requires a stronger kind or higher level of entitlement than previously granted. In the following example (from a TV discussion about the sale of weapons): (2)

ho visto anche violato il concetto di solidarietà internazionale, cioè veramente c’è una schizofrenia morale in uno stato che da un lato aiuta questi bambini aiuta le popolazioni del terzo mondo: e dall’altra parte le mutila per fare degli affari abbastanza schifosi. “I have also seen the idea of international solidarity violated that is to say there really is a kind of moral schizophrenia in a country which on the one hand helps these children, helps the peoples of the third world, and on the other mutilates them in order to do their rather disgusting business.”

the speaker, who has formerly proposed his contribution as a “testimony” (namely, as a commissive illocutionary act, to the performance of which he is entitled by his capacity to guarantee for the authenticity of the reported experience), attempts to issue judgments (verdictive illocutionary acts)¹¹ and therefore, to get his competence to issue judgments recognized. His attempt is underscored by the use of the adverb “veramente” (“really”), collocated exactly at the transition point between the testimony (“ho visto,” “I have seen”) and the attempted judgment (“c’è una schizofrenia morale,” “there is a kind of moral schizophrenia”).¹² On other occasions, speakers mitigate their entitlement by appealing to a lower status than they could claim in that situation. In the following example, the chairman of a work group attempts to direct the participants’ activities not by appealing to the authority he is endowed with, but by issuing informal suggestions and advice: (3)

Il mio suggerimento è che facciamo un brevissimo giro presentandoci ( . . . ) sarebbe opportuno che ci presentassimo per identificarci un po’. “My suggestion is that we very quickly go around introducing ourselves . . . it would be a good idea if we presented ourselves in order to identify ourselves a bit.”

¹¹ Commissives (Austin 1975, 157; Searle 1979, 14) commit the speaker to a certain course of action. Verdictives are defined by Austin (1975, 153) as consisting in the delivering of a finding, official or unofficial, upon evidence or reasons as to value or fact. ¹² The TV host does not ratify the attempted strengthening of the speaker’s entitlement, but interrupts the interviewee and redefines his contribution as mere “testimony” (“Grazie della sua testimonianza,” “Thank you for your testimony”). But other hearers (the TV audience) are in principle free to align themselves with the host’s uptake or deem it unfair.

       

117

The kind of entitlement required for suggestion or advice, that is, for the illocutionary acts which the speaker of (3) presents himself as performing, is weaker than he is in a position to appeal to. This does not hinder the speaker from achieving his perlocutionary goals: not only will his Directive be complied with, but he will also appear as a very kind chairman. The mitigation of the speaker’s entitlement may produce a shift from one kind of illocutionary act to another. For any kind of illocutionary act presupposing some “power” on the part of the speaker, there is a corresponding kind of illocutionary act, inviting similar responses, which requires only a lower level of power or even no power at all (apart from the possession of linguistic and communicative competence). For example, expressions of intentions can be described as promises which do not presuppose full capacity (and cannot therefore be equally binding), but still aim at reassuring the addressee, and protests can be described as similar to criticisms except for the specific competence which the latter require. Thus, as often happens, a woman interviewed in a TV talk show chooses to express an opinion (“Io credo che,” “I believe that”) instead of issuing a judgment: (4)

TV Host: Ma ce ne sono in natura? (–) Ecco scusi signora ma in questo tipo di: così: ce ne sono in natura? “But do they exist in nature? I mean, sorry madam but in this kind of thing, do they exist in nature?” Woman interviewed: Io credo che cominciano ad esserci nel senso che: no, nel senso che: gli uomini si stanno in qualche modo femminilizzando nel senso che eh stanno cominciando a conoscere: la loro parte sentimentale, che si sono per tanto tempo probabilmente negati “I believe they are starting to exist in the sense that, no, in the sense that, men are becoming in some ways more feminine in the sense that they are beginning to get to know their emotional side, which for so long they have probably suppressed.”

Finally, the entitlement to perform certain illocutionary acts may consist of the assignment of deontic modal values not related to possibility (and therefore to some kind of “power”) but to necessity (and particularly to obligation, as is characteristic of certain Expressives such as apologizing, thanking, and congratulating). Thus a participant in a TV discussion chooses to present his judgment on a very delicate issue not as the exercise of a competence, but as if his utterance were the fulfillment of an obligation (the duty to answer a question, or perhaps to react in a morally sound way to the event referred to):

118

       

(5)

TV host: Chi è *secondo lei?+ “Who according to you?” Participant: *Chi è, chi è+, e non lo so. Io devo dire, ( . . . ) non si tratta di omicidio di stato a mio avviso, ma certamente di un suicidio con grandi responsabilità istituzionali, credo di:: (–) credo di sì. “Who it is, who it is, oh I don’t know. I must say, . . . it’s not a state-organized murder, in my opinion, but certainly a suicide which the institutions are largely responsible for, I think, I think so.”

Even Directives may be formulated as presupposing an entitlement of the speaker related to necessity (typically, a condition of need). So in the following example, a psychotherapist bases his request for a detailed report (addressed to one of the family members in therapy) on his own condition of ignorance: (6)

Therapist: Ecco. Emilio. Lei deve mettersi un momentino nei miei panni. “Now. Emilio. You must put yourself in my place for a moment.” Family member: Sì. “Yes.” Therapist: Lei deve cercare di aiutarmi. Perché io non ci sono mai stata nella vostra famiglia e quindi quando “You must try to help me. Because I’ve never been in your family and so when” Family member: Lo so. Va bè. Io posso farle “I know. OK. I can make you . . .” (Leonardi and Viaro 1990, 112)

The obligation assigned to the addressee (“Lei deve,” “You must”), a feature which would make the speech act under consideration not merely a Directive, but specifically an Exercitive,¹³ is here presented as deriving not from the exercise of authority as in Exercitives, but from the therapist’s self-avowed state of need.

4.2 Modal Values Assigned to the Addressee The performance of an illocutionary act influences the deontic status of its addressee by assigning to him or her, or removing from him or her, some deontic

¹³ Exercitives are illocutionary acts characterized by the exercise of authority in favor or against some course of action (Austin 1975, 155) and typically assign obligations to their addressees (Sbisà 1984a [Essay 1]). Many Directives (attempts to get the hearer to do something: Searle 1979, 13–14) are Exercitives as well, but not all of them get the hearer to do something by imposing obligations on him or her.

       

119

modal value. For example, orders or advice assign obligations to their addressees; a promise assigns a right to its addressee; permission may be described as removing or preventing an obligation not to do something. Here too, variations in degree may occur. In principle, they should affect all kinds of assigned deontic modal values. Our examples here are limited to the mitigation or reinforcement of assigned obligations. The addition of the alternative “o no” (“or not”) to polar questions can be considered as a case of reinforcement of the obligation assigned to the addressee, since it emphasizes that he or she must choose between two and only two answers. Here are two examples (both from TV talk shows): (7)

Host: Ma tu pensi di avere la verità in tasca o no? “But do you think you have the truth in your pocket or not?”

(8)

Host: ecco secondo lei (–) Nicolas risponde (–) ai può dire quello che crede, ehm con la superlibertà ai requisiti, (–) secondo lei dell’uomo, sexy eh: o o no? “Yes in your opinion, does Nicolas correspond to, you can say what you think, with super-freedom, to the requisites, in your opinion of the sexy man eh or or not?”

The turn quoted in example (7) is openly aggressive (as is indicated also by the initial “ma,” “but”). In contrast, in example (8) the obligation assigned is qualified as an obligation to express personal commitment (“secondo lei,” “in your opinion”), and its reinforcement is counterbalanced by a concession of “superlibertà” (“super-freedom”), which, however, does not dispense the addressee from the obligation to answer the yes–no question in one of the two pertinent ways. Another case of reinforcement is found in example (6) above: the indicative mood in “Lei deve” (“You+formality must”), as opposed to the widespread use of the conditional mood for making requests, underscores the fact that the obligation assigned is a strict one. The mitigation of the assigned obligation by way of qualifying its content occurs (together with the mitigation of the speaker’s power) in example (3): (3)

Il mio suggerimento è che facciamo un brevissimo giro presentandoci ( . . . ) sarebbe opportuno che ci presentassimo per identificarci un po’. “My suggestion is that we very quickly go around introducing ourselves . . . it would be a good idea if we presented ourselves in order to identify ourselves a bit.”

Following the chairman’s directions will cost the participants very little effort indeed!

120

       

4.3 The Speaker’s Commitment Obligations concerning subsequent linguistic and non-linguistic behavior are assigned to speakers not only by Commissives but also by other illocutionary acts as a part of their core illocutionary effect. The assignment of deontic modal values is not a unilateral matter: a speaker who modifies the status of his or her addressee cannot remain unaffected by his or her own action. Thus, just as to a commitment of the speaker to do something, there corresponds the assignment to the addressee of a right to expect that behavior from him or her, to any assignment of obligations to the addressee, there corresponds a commitment on the part of the speaker not to interfere with the assigned task. Assertion, too, has been analyzed as involving a commitment on the part of the speaker (Brandom 1994; Searle 1979, 12–13). The speaker’s commitment is sometimes reinforced, especially as a reaction to the addressee’s lack of confidence in the speaker’s reliability. (9)

Judge: ma le ha dato una spinta a sta moglie o no. a sua moglie “But you gave this wife of yours a push or didn’t you, your wife” Defendant: no niente ne*ss-+ “no nothing no” Judge: *si+curo “sure” Defendant: sicurissimo “quite sure.”

(10)

per quel nostro intervento era stata pattuita una somma di quattrocentomila lire compenso che da lei, non abbiamo mai ricevuto “For that job we did a fee of four hundred thousand lire was agreed upon which we never received from her.”

In (9) (from a courtroom hearing) a defendant reinforces his commitment in answer to the judge’s expression of mistrust by using a superlative (“sicurissimo,” “quite sure,” lit. “very sure”). He is not merely expressing his confidence in the truth of the reply he has just given, but assuring the judge that his commitment to that reply is serious. Also in (10) (from a TV show), the addition of an informatively superfluous, but emphasized, “mai” (“never”) underscores the speaker’s commitment to what he asserts. The speaker’s commitment is often downgraded, so as to lower the cost of the speech act, thereby making it more easily defensible. There are two main strategies

       

121

for doing so. In the following example, from a psychotherapeutic session, the speaker uses both. (11)

Father: però logicamente siamo sempre lì nel merito (—) come giustamente io ho le mie idee e penso di essere dalla parte della ragione lei ha le sue idee è piuttosto dura “But obviously we are back to the same problem basically, like quite rightly I have my ideas and I think I’m in the right she has her ideas she’s rather hard” Little girl: ((cries out)) Father: ma molto dura almeno secondo il mio principio ( . . . ) “but very hard at least according to my way of seeing things . . .”

At first, the speaker doesn’t commit himself to the straightforward claim that his wife is hard, but to a weaker one (“piuttosto dura,” “rather hard”); then he switches to calling her “molto dura” (“very hard”), but relativizes his commitment to his own subjective perspective. In both cases, the criticism he levels at his wife is easier to defend than an unrestricted or absolute one. Sometimes, mitigated commitment parallels the mitigation of the speaker’s entitlement. Let us reconsider a turn from our example (4): (4a)

Io credo che cominciano ad esserci nel senso che: no, nel senso che: gli uomini si stanno in qualche modo femminilizzando nel senso che eh stanno cominciando a conoscere: la loro parte sentimentale, che si sono per tanto tempo probabilmente negati “I believe they are starting to exist in the sense that, no, in the sense that, men are becoming in some ways more feminine in the sense that they are beginning to get to know their emotive side, which for so long they have probably suppressed.”

Besides presenting her speech act as an expression of belief rather than an act of judgment, the speaker uses the modal adverb “probabilmente” (“probably”) to downgrade her commitment to the truth of the clause “che si sono per tanto tempo ( . . . ) negati” (“which for so long they have . . . suppressed”), and the adverbial expression “in qualche modo” (“in some ways”) to introduce vagueness into her expressed belief that men are becoming more feminine, thus making it easier to defend.¹⁴ ¹⁴ “Piuttosto dura” (“rather hard”) and “in qualche modo” (“in some ways”) might be classified as “bushes” (Caffi 1999), i.e. mitigation devices which focus on the propositional content. In my perspective, “bushes” offer one strategy for downgrading the obligations one might have to face as a consequence of one’s own speech act.

122

       

4.4 Degrees of Strength in Expressed Attitudes and Inner States Variations in degree affecting expressed attitudes and inner states may also be accounted for within the framework of the standard conception of the speech act. However, I consider them within the alternative framework I am proposing, in order to show their connections with the core illocutionary effect and its description in deontic modal terms. In my framework, the “expression” of attitudes and of inner states coincides with the speech act’s licensing the hearer to assign a certain inner state (with a certain degree of intensity) to the speaker. Degrees of intensity may come into the picture in two ways: they may affect either the inner state expressed or its expression, which may be foregrounded or backgrounded. Example (12) (from a psychotherapeutic session) is a case of the former kind, in which the intensity of an expressed inner state is reinforced: (12)

Therapist: ecco e che: (–) quando vi hanno detto di venire qui tutta la famiglia che cosa avete pensato “Well and what, when they told the whole family to come here what did you all think” Mother: io ho pensato the era la cosa giusta perché secondo me i bambini ne risentono tanto della situazione:: tesa in famiglia “I thought it was the right thing because in my opinion the children are so much affected by the tense situation in the family.”

By using “tanto” (“so much”) instead of “molto” (“much”), the speaker expresses, and invites hearers to ascribe to her, an intensified worry for her children. This effect is added to the core illocutionary effect of the performed speech act, but does not produce any substantial transformation of it: the speaker is still explaining her decision to initiate family psychotherapy. When reinforcement concerns the expression of an inner state as opposed to the inner state expressed, a closer connection with the core effect of the illocutionary act may result: the speech act as a whole may qualify as an Expressive. We have already met this phenomenon in example (4), in which the mention of the speaker’s inner state of belief (“io credo che,” “I believe that”) turns an act of judgment into the expression of an opinion. In example (5), in which the speaker’s entitlement is presented as a matter of obligation rather than competence or authority, a similar foregrounding of the expression of inner states can be noted: (5)

TV host: Chi è *secondo lei?+ “Who according to you?”

       

123

Participant: *Chi è, chi è+, e non lo so. Io devo dire, ( . . . ) non si tratta di omicidio di stato a mio avviso, ma certamente di un suicidio con grandi responsabilità istituzionali, credo di:: (–) credo di sì. “Who it is, who it is, oh I don’t know. I must say, . . . it’s not a state-organized murder, in my opinion, but certainly a suicide which the institutions are largely responsible for, I think, I think so.” The connection of the mitigation of the speaker’s entitlement with the foregrounding of the expression of inner states is not surprising, because both are features of expressive or behabitive illocutionary acts.¹⁵ The expression of not especially intense inner states is not perceived as a kind of mitigation, unless intense inner states are expected. As to the backgrounding of the expressive dimension of the speech act, it may coincide with an attempt to appear objective or formal, as in the following example (again from a psychotherapeutic session), in which the speaker, in complaining about his wife’s behavior, attenuates his involvement by using impersonal constructions: (13)

( . . . ) non riesce così a:: ad amalgamarsi e dire così::: questa cosa magari facciamola in questo modo e non in quest’altro: (–) cioè magari al momento ragiona pure e si dice sì facciamo questa cosa al posto di quest’altra::: poi ovviamente la si lascia a sè lei ritorna al punto di prima:: “she doesn’t manage you know to amalgamate herself and say you know this thing let’s do it like this and not like that, that is at that moment she may even be reasonable and one says yes let’s do this rather than something else, then obviously one leaves her to herself and she goes back to where she was before.”¹⁶

4.5 Perlocutionary Goals of the Speech Act The way in which a speech act is formulated may also manifest the degree of intensity of the speaker’s intention or desire to achieve a certain perlocutionary goal, and the attempt to achieve the perlocutionary goal may itself be made in

¹⁵ Expressives (Searle 1979, 15–16) have the illocutionary point of expressing some inner state of the speaker. Behabitives (Austin 1975, 160) consist in reactions to somebody’s behavior or to events affecting somebody. Although their definitions differ, in practice the two terms cover the same domain. I have stressed (Sbisà 1984a [Essay 1]) that these illocutionary acts do not presuppose any kind of “power” on the part of the speaker, but rather, a condition of obligation or debt. ¹⁶ The impersonal constructions used in this example might be described as “shields” (Caffi 1999). There may be a connection between the backgrounding of the expression of inner states and mitigation by means of “shields”: the only way to radically background the expression of the speaker’ s subjectivity is to conceal the enunciating subject from view.

124

       

weaker or stronger ways. One reinforcement phenomenon to be mentioned in this connection is insisting. Insisting is necessarily sequential and involves a series of attempts to achieve the same goal (basically, the elicitation of a certain response on the part of the addressee). The speech act is repeated and reformulated, and sometimes resort is had to explicit performative utterances. The increasing strength of the attempt to achieve the perlocutionary goal may be connected with reinforcements of the core illocutionary effect, since the reformulations may also strengthen the speaker’s entitlement or commitment or the modal values assigned to the addressee. In the following example (from a TV show acting out pretend courtroom hearings): (14)

Defendant: signor giudice io assolutamente non pago perché loro mi hanno offeso: la mia persona mi hanno: messo nel ridicolo con apprezzamenti diciamo: (-) volgari. (–) sicché (^non::)= “Your honor I am absolutely not going to pay because they have offended me, my person, they have made me look ridiculous with let’s say, vulgar comments, so I’m not” Plaintiff: =ma no= “but no” Defendant: anzi chiedo dei danni morali “in fact I demand moral damages.”

the initial refusal to pay is reformulated, by means of a performative utterance, as a demand to be refunded which presupposes a strengthened entitlement of the speaker and not only tries to deprive the plaintiff of his alleged rights but also assigns him an obligation. As to mitigation, it is not clear whether cases of mitigation of the desire or the attempt to achieve the illocutionary act’s perlocutionary object (the goal conventionally associated with the performed illocutionary act: Austin 1975, 118) can be detected at all. Sometimes, particularly in formal contexts, speech acts may be performed simply in order to fulfill one’s role, without any real interest in the achievement of their associated perlocutionary goals; but the absence of such an interest need not be linguistically marked and thus falls beyond the scope of our research. Moreover, if an illocutionary act is interactionally taken as successfully performed, it does invite the associated response, and whether the speaker actually cares to elicit it or not does not seem to influence any aspect of the illocutionary effect. Finally, speakers may have a wide range of additional perlocutionary goals beyond those conventionally associated with the illocutionary acts they perform, and speech acts may bring about an enormous range of perlocutionary sequels, whether intended or unintended. We may wonder whether this domain, too, is

       

125

affected by mitigation/reinforcement. In Bazzanella et al. (1991), the avoidance or intensification of conflictual perlocutionary sequels has been considered as belonging to the dimensions of the speech act that are liable to be mitigated or reinforced. In consideration of the role played by mitigation with respect to politeness, it might even be claimed that perlocutionary sequels are what mitigation is primarily about. I do not think that this is correct. Rather, a distinction should be drawn between the purposes that mitigation or reinforcement may serve and the mitigation/reinforcement phenomena themselves. Mitigation and reinforcement achieve the tuning of the interactional relationship by adjusting the core illocutionary effect or associated aspects of the illocutionary act. Mitigation strategies are often (or even regularly) aimed at downgrading conflictual perlocutionary sequels, but this outcome is mediated by the fine-tuning of the interactional relationship at the illocutionary level. Let us consider here a wider stretch of a conversational turn we have already quoted: (5a)

Chi è, chi è, e non lo so. Io devo dire, non sono:: d’accordissimo con quello che dice la figlia, la quale, la figlia di Lombardo, la quale rispetto naturalmente il il dolore, non si tratta di un xxx omicidio, non si tratta di omicidio di stato a mio avviso, ma certamente di un suicidio con grandi responsabilità istituzionali, credo di:: (–) credo di sì. “Who it is, who it is, oh I don’t know. I must say, that I’m not fully in agreement with what the daughter says, Lombardo’s daughter, whose naturally whose pain I respect, it’s not a xxx murder, it’s not a stateorganized murder, in my opinion, but certainly a suicide which the institutions are largely responsible for, I think, I think so.”

This turn contains a great deal of mitigating devices, some of which have been discussed above. The interactional problem the speaker is facing is how to give his interpretation of the event under consideration (the suicide of a Marshal of the Carabinieri in a mafia context), which comes close to an accusation against state institutions, while avoiding being scornful of other persons involved or of the institutions and exposing himself too much. The solution the speaker adopts involves: mitigation of the speaker’s entitlement (which is founded not on some kind of “power,” but on an obligation: “io devo dire,” “I must say”), foregrounding of expression of attitudes (“non sono d’accordissimo,” “I’m not fully in agreement”; “la quale rispetto naturalmente il dolore,” “naturally whose pain I respect”), restriction of the speaker’s commitment (by means of a litotes: lack of full agreement is mentioned rather than disagreement or even mere lack of agreement), relativization of the speaker’s commitment to truth to the speaker’s subjective perspective (“a mio avviso,” “in my opinion”), redefinition of the assertion made as an expression of belief (“credo,” “I think”). The overall mitigating effect on the degree

126

       

of conflict is brought about by means of the foregrounding of the expression of attitudes in a context of mitigated speaker’s entitlement and restricted or relativized speaker’s commitment. That is to say that this and similar cases of mitigation, though fundamentally aimed at the management of conflict (which is not itself a matter of illocutionary effect), can be described as based on variations in the degree of strength of the illocutionary acts involved, and therefore by reference to the deontic modal values mobilized by the core illocutionary effect.

5. Concluding Remarks I hope that the discussion of examples has shown that a wide range of mitigation/ reinforcement phenomena can be described in a revised speech-act theoretical framework envisaging illocutionary force as the language-triggered introduction of changes in the “conventional” level of the interpersonal relation. Such a conception of the speech act appears to possess a descriptive or even explanatory potential that the standard conception lacks. I would now like to append some brief remarks on the implications of the kind of analysis that I have been proposing for mitigation/reinforcement phenomena. (i) It has often been noted that in actual discourse, mitigation is overridingly more frequent than reinforcement, and perhaps more interesting for pragmatic theory. This might lead us to study mitigation as a phenomenon in itself. In my proposal, on the other hand, mitigation and reinforcement are in principle symmetrical: they are of the same nature, since both are to be explained as variations in the illocutionary effects of speech acts. This connection between mitigation and reinforcement is confirmed by the observation that certain reinforcements (for example, the foregrounding of the expression of attitudes) combine with mitigations (for example, the mitigation of the speaker’s entitlement) to yield a complex result. This can be so because both mitigation and reinforcement affect the illocutionary effect and can be described in deontic modal terms. (ii) Illocutionary acts need not exhibit a one-to-one correspondence with individual, syntactically complete sentences (as has also been noted by Geis 1995). Sometimes, one illocutionary effect is brought about by a sequence of connected sentences, while at other times, one sentence may give rise to a cluster of illocutionary effects. Against the standard conception of the illocutionary act as basically reflecting the communicative function of the sentence, it is consistent with the approach defended here to say that illocutionary effects are produced by textual units considered as structured wholes. This should not be taken to mean, however,

       

127

that each textual unit has to be internally consistent and homogeneous. It may well happen that the overall physiognomy of a textual unit emerges from a heterogeneous structure. In our example (5), discussed above, there is at least one feature which cannot be traced back to mitigation: “certamente” (“certainly”). Its collocation after the connective ma (“but”) and before the formulation of the speaker’s main claim (“un suicidio con grandi responsabilità istituzionali,” “a suicide which the institutions are largely responsible for”) hints at a reinforcement of the speaker’s commitment to that specific claim, which might seem inconsistent with the mitigating devices that precede or follow. Still, we feel that the speaker is not inconsistent. It is no coincidence that soon after the (rash) reinforcing device “certamente” (“certainly”) it becomes convenient, if not compulsory, for him to redefine his speech act as an expressive (“credo,” “I think”). The mitigating devices used counterbalance the reinforcement of the speaker’s main claim without cancelling it. (iii) As to the relationship between illocutionary force indicating devices and mitigation/reinforcement devices, the conception of mitigation/reinforcement proposed here supports the hypothesis of a fuzzy border between these two groups of linguistic expressions or strategies, with overlaps. However, it is not merely the case that some linguistic expressions are “multifunctional,” serving the distinct functions of mitigating/reinforcing speech acts and of facilitating the recognition of their illocutionary force (Blum Kulka 1985). At least in some of our examples, (2), (4), (5), (14), reinforcement or mitigation not only are associated with the use of certain linguistic expressions which also indicate illocutionary force but are part of the dynamics by which illocutionary force is determined. It is the overall physiognomy of a speech act that permits us to take it (and describe it) as having a certain kind of illocutionary force, and which at the same time sets the degree of strength or intensity of the various components of the illocutionary effect: you cannot have the one thing without the other. Each linguistic expression or textual strategy participating in the production of the illocutionary effect contributes in its own way to the overall physiognomy of the speech act. So if the task of discourse analysis is to identify the features of a text which motivate (or disconfirm) the reading that text is given, an analysis of illocutionary acts must appeal to all the features in the text which contribute to bringing about a particular illocutionary effect, whether they are standard illocutionary force indicators or indicators of a finer adjustment and tuning. (iv) Since mitigation and reinforcement have obvious connections with politeness, the view I have put forward suggests some refinements for politeness theory. One widely accepted idea about politeness is that it is strategically connected with matters of “face” (Brown and Levinson 1987; Goffman

128

        1967). But as used in pragmatic research, “face” is a highly ambiguous notion. Brown and Levinson (1987, 61–2) give a general definition of it in terms of wants that participants have, that is, in terms of psychological attitudes, while in their own (and others’) finer descriptions it turns out to be related to notions such as imposition, freedom from imposition, right, claim, debt, respect, competence, ratification or approbation of desires, legitimate expectation and violation of expectations, which are all either directly deontic or analyzable in deontic modal terms. If mitigation/ reinforcement phenomena are to be analyzed in deontic modal terms, as I have claimed, this ambiguity in the notion of “face” should be resolved in favor of the centrality of deontic modal notions as opposed to psychological attitudes. It would then be seen more clearly why and how variations in the degree of strength of speech acts play such a central role in politeness strategies. This might in turn induce us to recognize that the relationship between politeness strategies and illocution is closer than is usually admitted.¹⁷

¹⁷ A preliminary version of this paper was read at Boston University in spring 1999; I am grateful to Bruce Fraser and to the audience for comments and questions. Provisional results of my ongoing research on illocutionary force modification had formerly been presented at the Conference “L’efficacia del testo,” S. Marino, September 27–9, 1996; I thank the participants for the discussion. I also thank Claudia Caffi and Carla Bazzanella for all the discussions we have had on illocutionary force modification, and two anonymous referees for their stimulating suggestions and criticisms.

7 Speech Acts in Context 1. Context in the Received Speech Act Theory Speech act theory is one of the fields in the philosophy of language in which consideration of context was introduced earliest. Already for Austin, context was part of what philosophers of language have to elucidate, namely “the total speech act in the total speech situation” (Austin 1975, 148). In consideration of the close association between speech act and context, the way in which the context of a speech act is conceived contributes to what the speech act is supposed to be: for example, whether it is a genuine social action and in what sense. In this paper, I consider conceptions of context in the received speech act theory or authors influenced by it and discuss some main features that we may want a conception of context to possess. I also claim that the conception of context I favor enables us to describe speech acts as context-changing social actions.¹ In the initial chapters of How to Do Things with Words (Austin 1975), Austin shows that an utterance cannot be performative (that is, perform a social action having a conventional effect) unless it is issued in the appropriate circumstances. The identification of performative utterances with explicit ways of performing speech acts (Austin 1975, 69, 103) enables us to see that the contextual requirements for the felicity of performative utterances also hold for speech acts performed by using sentences that do not contain explicit performative formulas. The context of a speech act seems to be conceived by Austin as a cluster of actual states of affairs or events of various kinds, related to the issuing of an utterance and to its intended force. No sharp distinction is drawn between external states of affairs or events (present or past social behavior, facts about material objects) and psychological states or attitudes of the participants: at least in some cases (cf. Austin 1975, 29, 37–8), the participants’ attitudes and expectations are held to interfere with the successfulness of the speech act. The prevailing emphasis is, however, on whether Originally published in Language & Communication 22 (4), special issue on “Contexts of Social Action,” ed. Anita Fetzer and Varol Akman: 421–36. Amsterdam: Elsevier, 2002. ¹ By “social action” we can mean any action whose agent and patient are members of a society, but also, more intriguingly, any action whose performance needs a social environment. Hitting somebody may be a social action in the former sense, offending somebody is a social action also in the latter. I am concerned here with considering speech acts as social actions in the latter sense. It is not among the aims of this paper to discuss whether and how my claims about speech acts can be extended to nonverbal social actions.

Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0008

130

   

the contextual requirements for the felicity of speech acts are satisfied by the actual situation, as opposed to being merely believed to be satisfied. Searle (1969, 1979) agrees with Austin upon the fact that speech acts have felicity or successfulness conditions to be satisfied by the context. But the conception of context has changed. It is in Searle’s discussion of successfulness conditions for illocutionary acts (Searle 1969, 54–71) that context begins to be considered rather as a set of propositional attitudes of the participants than as a cluster of actual states of affairs: most of those conditions are formulated in terms of beliefs or intentions of the participants. So the way is opened to foregrounding a subjective or cognitive conception of context. A further step in this direction is taken by Searle (1979, 3–6). In his search for a precise definition of illocutionary classes, he focuses on three selected dimensions of the speech act (illocutionary point, direction of fit, expressed inner state) and correspondingly moves conditions on external social circumstances to a marginal position, as inessential to the illocutionary act. In the reformulation of speech act theory by Bach and Harnish (1979), deeply influenced by Grice’s intention-based and inferential view of communication, the success of the speech act (qua communicative illocutionary act) is defined in terms of the recognition of the speaker’s communicative intention by the hearer. Context is mentioned in the initial survey of the components of the total speech act (Bach and Harnish 1979, 3), but what is really invoked in descriptions of inferences by means of which the hearer recognizes the speaker’s communicative intention are “mutual contextual beliefs”: beliefs relevant to and activated by the context of utterance, or by the utterance itself, that are shared and believed to be shared by the participants (Bach and Harnish 1979, 5, 61). The sufficient conditions for a given utterance to constitute an illocutionary act of a certain kind are all couched in terms of expressed propositional attitudes and any reference to social situational factors has disappeared from the definitions of communicative illocutionary classes. Finally, in the currently most influential version of the inferential view of communication, i.e. Relevance Theory (Sperber and Wilson 1995), little room is left to illocutionary acts, since they seem to be reduced to hearer’s inferences to the way in which the speaker’s contribution is intended to be relevant. At any rate, the context upon which such inferences rely is of a wholly cognitive nature. In the history of speech act theory, context has also changed in function. Austin views felicity conditions as rules according to which the speech act is evaluated. The positive evaluation of a speech act’s felicity is the default option and involves the successful performance of the purported illocutionary act. Negative evaluations, or infelicities, have different consequences on the speech act depending on which rule is violated. So abuses (resulting from violations of the sincerity condition: for instance, when a speaker says “If I were you, I would accept that job” but does not believe that it is good for the addressee to accept the job) or

   

131

breaches (resulting from the violation of commitments to subsequent behavior: e.g., when a speaker says “Tomorrow I will buy you a present” but the following day buys nothing) do not impede the purported act from achieving its conventional effect: the piece of advice and the promise cited above are defective, but not thereby void. Other violations regard the appropriateness of the circumstances of utterance, e.g., the speaker’s authority in the case of orders or permissions, the previous behavior of the speaker in the case of an apology, the previous behavior of the speaker in the case of praise or thanksgiving, or, in promises, the suitability of the speaker for doing what he or she commits him or herself to do. Inappropriate circumstances may make the speech act misfire. If, when this happens, we still say that the speaker has done something, that is not the purported speech act: for instance, an order issued without authority may be a rude request, but not an order. Also utterances of declarative sentences may be infelicitous in this way (for instance, when the speaker is not in a position to make that assertion, or the object he or she intends to talk about does not exist) and so may not succeed in constituting assertions to be further evaluated in terms of truth or falsity. The evaluative role of context still seems to be taken for granted in Searle (1969), where felicity conditions provide a set of necessary and sufficient conditions for the appropriate performance of a speech act. But under Grice’s increasing influence it has appeared less and less plausible that any kind of speech act inappropriateness may prevent us from assessing the utterance of any declarative sentence as true or false. According to Grice (1989), truth-value should be independent of judgments of appropriateness or inappropriateness (and a fortiori of Austinian felicity) and inappropriate utterances are described as utterances whose conventional or conversational implicatures are false or misleading. In this framework context cannot retain the evaluative function that Austin intended to assign to it. In fact, the so-called “background” in Searle (1979) and mutual contextual beliefs in Bach and Harnish (1979) play a merely interpretive role. In Relevance Theory the most important dimension of contextual appropriateness, relevance, is so defined as to make it vacuously hold for any communicative act that does not prove altogether uninterpretable. Not by chance, the overt defense of a cognitive conception of the context of speech acts is accompanied, in Stalnaker (1999, 8), by the dismissive claim that inappropriateness is no theoretical category. The turn from context as actual states of affairs and events to context as consisting of propositional attitudes has influenced the very conception of a speech act. In Searle (1969), among felicity conditions there is also an “essential” condition, which requires of the speaker an appropriate intention such as, for promises, the intention to undertake an obligation. In Searle (1979), such intentions are considered as “illocutionary points” of speech acts, whose recognition coincides with the success of the act in question. So the role of one and the same intention shifts from satisfying a condition, albeit “essential,” to constituting what is communicated by the speech act. The internalization of felicity conditions has

132

   

undoubtedly favored such a shift, as well as the corresponding shift from conceiving speech acts as actions having conventional effects to conceiving them as utterances expressing communicative intentions (Sbisà 1995a; cf. also Rajagopalan 2000, 365). The adoption of a cognitive conception of context is not without connections with the gradual abandonment of Austin’s claim that language is action. But the success of the Searlean–Gricean view may be due to a greater extent to its higher degree of plausibility within the framework of received views of language and communication than to any evidence it may provide for the non-viability of the Austinian proposal. I consider it quite legitimate to take Austin’s ideas seriously and reconsider aspects of the conception of context that have not been duly appreciated or have even been dropped in the Searlean-Gricean development of speech act theory as well as by most authors influenced by it.

2. What Should the Context of a Speech Act Be Like? At least three questions about the context of a speech act are relevant to the aims of this paper: (i) Is the context of a speech act given (previously to the speech act or even the conversation it belongs to) or constructed (by the participants in the conversation, possibly with the contribution of the speech act itself)? (ii) Is the context of a speech act unlimited (without fixed borders, extendable in any direction) or limited (in a principled way)? (iii) Is the context of a speech act objective (consisting of actual states of affairs, actual events, etc.) or subjective and notably cognitive (consisting of intentional states either of the speaker or of both participants, notably beliefs)?

2.1 Given vs Constructed Context Most conceptions of context in the received speech act theory consider it as given: its content is independent of the speech act it is the context of and is established previously to the performance of that speech act. Austinian felicity conditions, ideally at least, have to be satisfied by speaker, addressee, situation of utterance, and so on before and independently of the performance of the speech act. Searle’s felicity conditions must hold before the speech act and independently of the words uttered to perform it, since they are to be necessary and sufficient conditions for those words to count as its successful and non-defective performance. Also in Stalnaker’s pragmatic view of assertion, the context set (the set of possible worlds

   

133

not excluded by the speaker’s presuppositions), albeit affected by previous discourse, is given with respect to the content of each newly issued assertion. Literature taking context into account in the framework of the sociolinguistic analysis of verbal interaction, but making only indirect reference or no reference at all to speech act theory, has argued strongly in favor of the constructed nature of context (for an overview, see Duranti and Goodwin 1992). Although I recognize that such literature has made fundamental contributions to shaping the notion of context, I will not rely upon it here. I would like to suggest that even in a philosophical, speech-act inspired perspective there is reason to doubt that the context of a speech act has to be already settled before that speech act is performed. We do not check out the felicity or appropriateness of a speech act before recognizing it as accomplished. Rather, we tend to take our interlocutors’ words as constituting felicitous or appropriate speech acts whenever no alarm signal (Goffman 1971) suggests to us they do not. Thus we tend to assume that felicity or appropriateness conditions are satisfied and to behave as if this were the case. Let us consider a piece of advice: “If you want to read an intriguing Italian novel, read La Coscienza di Zeno by Italo Svevo.” The receiver who takes this utterance as a piece of advice is led to believe that the speaker is well informed about Italian novels and in particular about the novel he or she advises the addressee to read. Or let us consider an order. If a man dressed as a policeman stops my car and says “Your driving license, please,” I take it as an order and take for granted that he is a policeman, thereby endowed with the authority to give such orders to drivers. I might not know, nor have the opportunity to check, whether the person who advises me is really competent and the speaker too might not have reflexively evaluated his or her own competence before speaking. I do not even wonder whether the man who stops my car is a real policeman (as opposed to a thief in disguise). The satisfaction of felicity conditions is assumed by default, that is, as a first option and without raising the issue explicitly. Like other default assumptions, this too is defeasible: it is suspended as soon as doubts arise and if it is discovered not to hold, the speech act is either evaluated as infelicitous or is re-described. Like felicity conditions, presuppositions too are taken to be satisfied by the context as a default option. This originates the phenomenon that Lewis (1979) calls “accommodation”: when the sentence uttered requires a certain presupposition, but the receiver does not yet share that presupposition, the receiver is led to include it among his or her background assumptions. Examples discussed in the literature, such as “We regret that children are not admitted to the commencement exercises” (Karttunen 1974), are no isolated cases. Talk in face-to-face conversation or in the media, as well as various genres of written communication, exploit presupposition accommodation to make background information accessible to receivers or even invite them to accept new information without further

134

   

discussion. If we overhear a conversation among acquaintances who say to each other “George has stopped smoking,” we assume that George (whether we are acquainted with him or not, and in case we are, whether we ever saw him smoking ourselves) used to smoke. Or suppose that we are not well acquainted with the geography of Asia: if we hear a TV speaker reporting that the king of Bhutan got married, we learn also the background information that Bhutan is a monarchy. But if the satisfaction of felicity conditions and presuppositions need not be taken into consideration by the participants in advance and instead, the receiver may infer what belongs to the context of a speech act from the speech act itself, it is misguided to consider the context of a speech act as something “given.” Rather, it should be conceded that the context of an interactional event is set up by its participants as the interaction proceeds.

2.2 Limited vs Unlimited Context It may seem otiose to wonder whether the context of a speech act should be conceived as limited or rather, as extendable in any direction and therefore potentially all-inclusive. It is often assumed that context should include “what is needed” for doing what we want to use it for (cf. Kaplan 1989, 591), be it interpretation or evaluation. In this perspective, it seems unnecessary to establish in principle whether the list of “what is needed” will stop at a certain point, that is, whether there are limits to what it may contain. But the role that the investigation of speech acts may assign to context changes to a remarkable extent, depending on whether context is conceived as limited or unlimited. An unlimited context may provide the means for interpreting a speech act (insofar as we are willing to accept that the resulting interpretation might turn out to be open-ended), but it is hard to see how it might yield an evaluation of a speech act. Either we give up assigning context any evaluative role, or context must be limited. Moreover, evaluating a speech act against its context differs significantly from evaluating it against a whole world only if not everything in the world belongs to the context. In Austin (1975), the context pertinent to a speech act is limited, because felicity rules single out those aspects of the situation against which the felicity of the speech act is to be evaluated. A further kind of delimitation comes into play when, in discussing truth, he claims that in order to evaluate an assertion as true or false (but also as exaggerated, rough, etc.) we have to consider the situation in which it is made, including the participants’ goals: the assertion “France is hexagonal” may count as true if made in view of a certain goal (considering from how many sides an army could invade France) and as false if made in view of another (describing the shape of the borders of France in

   

135

reasonable detail; cf. Austin 1975, 143–5).² Thus, the goals of the conversation determine against which (and how fine-grained) aspects of the same object or event the truth/falsity of the speech act is to be evaluated. The very fact that those aspects may be different on different occasions shows that the situatedness of the speech act goes hand in hand with the delimitation of its context. Searle (1969, 57) includes “normal” communication conditions among the requirements for the successful performance of a speech act, on a par with those features of the situation or attitudes of the participants that are made pertinent by specific felicity rules. Since, as Searle himself admits, there is no exhaustive characterization of what is normal, there seems to be no end to the details one might want or need to add to the context of a speech act. In fact, in Searle (1979, 130–1), context is explicitly characterized as indefinitely extendable and appears potentially all-inclusive: every speech act has meaning only against a set of background assumptions, which are indefinite in number and whose explicitation (tending to bring with it other assumptions) is endless. Other pragmatic views of linguistic communication consider context as always extendable, albeit in fact empirically delimited. In Stalnaker’s pragmatics (Stalnaker 1999), context as background information (or what the speaker believes is, or intends to treat as, background information) is delimited on every occasion by the presuppositions a speaker happens to make. In Relevance Theory (Sperber and Wilson, 1995) what counts as context is delimited by the empirical fact that certain assumptions are present or activated in the mind of a speaker or receiver and (when these do not suffice to make an utterance relevant enough) is enlarged by activating further contextual assumptions. Also conceptions of context used in sociolinguistic research tend to be open-ended, as if it were the case that the more is brought into the picture, the better a speech act can be understood. But there are reasons for seeking a more principled way of delimiting the context of a speech act than is provided by any of these views. First of all, the context-dependence of meaning not only is constantly experienced by sociolinguists in the field but also has been noticed and discussed by philosophers of language as regards indexicals, demonstratives, descriptions, domains of quantification, and even whole propositions (for an overview, cf. Bianchi 1999). This context-dependence must have some source, which may be traced back to the fact that language use is always situated. In fact, language is regularly used in such a way that there is an agent who counts as being responsible for its use within the framework of some activity, and such situatedness of speech acts requires their context to be limited. Secondly, if a speech act is produced and understood in a context and is therefore a situated event, it seems reasonable to think that it should be evaluated with respect to that context. Again, in order to yield a

² [See also Sbisà in preparation.]

136

   

definite evaluation of a speech act (in terms of felicity/infelicity, appropriateness/ inappropriateness, truth/falsity), context must itself be delimited, that is, it should be clear or ascertainable what belongs to context and what does not. An all-inclusive context would make situatedness disappear and reduce the speech act to Austin’s caricature of the constative utterance as “the ideal of what would be right to say in all circumstances, for any purpose, to any audience, etc.” (Austin 1975, 146). In the light of these considerations, all conceptions of context that do not include delimitation criteria or in which context is delimited merely by empirical facts about the assumptions of the participants may be deemed unsatisfactory. Greater effort needs to be put into providing an account of the delimitation of context, if we want to make a serious attempt at putting situatedness into our picture of the speech act.

2.3 Objective vs Cognitive Context I turn now to our third question, concerning the objective or cognitive nature of context. A context has an objective nature in the sense relevant here if it is conceived as determined, not by the content of the participants’ intentional states, but by relevant states of affairs occurring in the world, of which participants might not even be aware. Objective context in this sense has also been described as “mind-transcendent” (Gauker 1998). Cognitive context consists of intentional states of one or more participants and qualifies as “cognitive” since these states involve representations. As has been seen above, reference to context in speech act theory has shifted from context as actual states of affairs (therefore, objective) to context as a set of propositional attitudes (therefore, cognitive). Cognitive conceptions of context are now widespread, and some arguments in support of them appear quite forceful (cf. Penco 1999), but I would like to challenge the idea that they are suitable for accounting for speech acts as social actions. This is most clear if we consider the role of context in the evaluation of speech acts. If context consisted of intentional states, the felicity of a speech act would demand too little of the agent: it would demand of him or her that he or she should have certain intentional states, but not that the representations involved should yield a correct grasp of the situation, and not even that he or she should have made sure this is so. As a necessary condition for successfully commanding me to shut the door, it would suffice that you believe that the door is open; in order for me to be in a position to apologize successfully for ruining your book, it would suffice that I believe that I have inadvertently ruined your book. But speaker’s beliefs hardly suffice to get the interlocutor to take a speech act as successfully performed. If I know that the door has already been closed, I also know that the imperative sentence you utter cannot bind me to close it. If you know that your

   

137

own cat has ruined your book, you cannot accept my utterance as a felicitous apology. Not even shared beliefs are enough. The command “Shut the door,” although felicitous under any other point of view, fails to be binding when the door is already closed, whatever the participants’ assumptions about the door’s state might have been when it has been issued. If we both are convinced that I have ruined your book, even though it was the cat’s fault, my apology appears to us to be appropriate, but the conventional effect characteristic of apologizing, namely repairing a fault of the speaker, cannot be achieved when the fault does not in fact exist. As soon as it is discovered that there was nothing to apologize for, the whole situation has to be retrospectively redefined. If we are not interested in the mere appearance of appropriateness, but in actual appropriateness and therefore in the successful bringing about of conventional effects, we have to take context into account not as a set of cognitive states and intentional attitudes, but as a bunch of external circumstances, either material or social. This suggests that the study of speech acts as social actions needs to rely on an objective conception of context. The case for objective context can be made stronger if we consider the requirements that are posed to the context of a speech act by linguistically indicated presuppositions. For a speech act to be appropriate, is it enough that the presuppositions triggered by the sentence used are assumed to be true by the speaker? Or to be both true and shared by the interlocutor? Should they actually be shared by the participants? Should they actually be true? It is apparent that my belief that France has a king cannot make my utterance of a declarative sentence about the present king of France a felicitous assertion. If other participants in the conversation share my belief and do not mind speaking of the king of France, this may create an appearance of appropriateness, but does not keep our conversation from being deeply out of order. If we consider such phenomena as flaws of the act of making an assertion, the background against which we evaluate assertions must be objective rather than cognitive. The use of presuppositions to convey information that has already been mentioned above provides us with a further argument in support of objective conceptions of context. Among the strategies that have been proposed for explaining informative presupposition in the framework of a cognitive conception of context, there is the claim that, in informative uses of presupposition, the speaker pretends that a certain assumption is already shared by the addressee (Stalnaker 1974). But this account is unconvincing. As observed by Gauker (1998), the most natural way to pretend that an assumption is already shared would be not to convey it at all. Moreover, it is a mystery how we shift from pretended to actual appropriateness: why should pretense invite the interlocutor to actually share the pretended presupposition? This transition continues to remain a mystery even in the more articulated account offered by Stalnaker (1998). Another strategy proposes a broadening of the definition of presupposition that lets presupposition include, beyond shared assumptions, also assumptions that the receiver is merely disposed

138

   

to accept (Soames 1982). But then what about those informative presuppositions about which the receiver has no previous disposition at all, or even those that the receiver is not disposed to accept (for example because he or she believes they are false)? Suppose I firmly believe that Bhutan is a Republic: either I reject your talk about the king of Bhutan as inappropriate (your utterance about the king of Bhutan, in spite of my opposite belief, does presuppose that Bhutan has a king) or I might be led to doubt my memory and thus align myself with you in assuming that Bhutan is a monarchy. Informative presupposition presents itself to the receiver as an assumption he or she should accept (Sbisà 1999a [Essay 3]), irrespectively of his or her previous disposition towards its acceptance. Adopting an objective conception of context enables us to account for informative presuppositions in the following way. When S utters a declarative sentence p and p (because of some presupposition indicator it contains) presupposes q, if q is not satisfied by the objective context, S’s utterance cannot be a felicitous assertion. So if we take it that S’s utterance is a felicitous assertion, then we also have to assume that q is not merely believed by S (or believed to be shared, etc.), but is satisfied by the objective context. But we do assume by default that S’s utterance of a declarative sentence constitutes a felicitous assertion. Therefore we are entitled, and indeed even committed, to assume that q. Thus informative presuppositions are retrieved by default from the linguistic indicators included in an utterance precisely because they are conditions to be satisfied by the objective context. The way in which they are retrieved explains why the receiver stores them as presuppositions, that is, as information taken for granted, even if they provide new information. It also explains why they provide information about the world. In the framework of a cognitive conception of context, the receiver’s assumption that the speaker’s declarative utterance is a felicitous assertion could only entitle him or her to conclude that the speaker thinks that q (and thinks the receiver thinks that q).

3. Some Alleged Incompatibilities I have proposed so far that the context of a speech act should be conceived as constructed, limited, and objective. But are these features compatible with one another? Evidence in support of the view that context is constructed applies primarily to cognitive context. Participants in an interactional event may tacitly agree or come to an agreement about the definition of the situation they are in, as well as upon the definition of the situation they are talking about: it is a matter of jointly selecting one among several available frames, or of using and understanding contextualization cues. But it may be held that all this amounts to forming and sharing attitudes, particularly beliefs. How would participants be able to construct

   

139

new objectively existing states of affairs just by forming or sharing the belief they are the case? In order to be constructed, then, context must be cognitive as opposed to objective. One way to cope with this alleged incompatibility would be to admit that both kinds of context play some role with respect to speech acts, but that cognitive context may be constructed, while objective context cannot. This solution requires separate roles to be assigned to the two kinds of context: the interpretation of speech acts should be taken to depend on cognitive context and their evaluation on objective context. But this separation is not clear-cut: some aspects of utterance interpretation rely on objective facts such as who is speaking or what is the time, and furthermore, objective context is involved in the understanding of informative presuppositions, since it is only as requirements to be satisfied by the objective context that presuppositions can convey new information. Another way to dispel incompatibility is to claim that not only cognitive context but also certain features of objective context may be constructed by the participants. There are different kinds of states of affairs in the world, and some of them are conventional or institutional (cf. van Gelder 1998).³ The occurrence of a conventional or institutional state of affairs may be analyzed as depending on some kind of practical (rather than cognitive) agreement among the relevant social participants (Sbisà 2001a [Essay 6]; 2002b [Essay 8]⁴). So participants’ agreement may be held to play a role with respect to what is to count as being (objectively) the case when matters of right, entitlement, obligation, authority, etc. are at issue. Of such states of affairs it would not be inconsistent to claim both that they can be constructed by the participants in an interactional event and that they may belong to the objective context of that same interaction. An example of such a state is somebody’s being endowed with some kind of authority (which, for instance, entitles him or her to give certain kinds of orders to other people). Authority does not exist apart from its recognition by relevant social participants. In formal situations and in connection with institutional roles (such as being a policeman) the relevant social participants and the procedure for attributing authority are well defined and separate from everyday informal interaction (and so, contrary to what has been suggested in Section 2.1 above, the authority that enables a policeman to issue felicitous orders is to be counted as “given”). But in informal situations, it might be up to the participants to decide whether or not to recognize the utterance of a certain imperative sentence as a felicitous order and thus grant its utterer the corresponding authority. People may thereby succeed in redefining their role simply by issuing a speech act that presupposes that role: the

³ It has been admitted that there are institutional facts insofar as constitutive rules bring them into existence (Searle 1969). But while a conventional element is certainly present in everything institutional, not all conventional matters can be traced back to institutions. ⁴ [Sbisà 2002b was originally cited as “in press.”]

140

   

default assumption of the satisfaction of felicity conditions, if not challenged, leads directly to the relevant agreement and thus to fixing a conventional framework, which counts as an element of the objective context of that very speech act. A further problem is whether context can be both limited and objective. We have seen above that there are no clear criteria for the delimitation of context, especially when context is conceived as objective. One of the most natural candidates to the role of mind-transcendent element enabling us to account for the delimitation of context is the fact that not all actual states of affairs in the world are relevant to what the interlocutors are up to, that is, to the goals (broadly intended) of the interactional event. It has been proposed to conceive of context as constrained by all and only those aspects of the world that it would be helpful for the participants to consider in order to achieve their goals, and thus as specifiable by a set of sentences that is delimited in a principled way (Gauker 1994, 225–47; 2003⁵).⁶ Such a proposal may both account for the situatedness of speech acts and provide the mind-transcendent background against which they should be evaluated (as felicitous or appropriate, but also as true or false). It might be objected that if goals contribute to the delimitation of context, objective context is in fact made to rely on a subjective element and, inasmuch as goals involve representations, on a cognitive one. The problem might not surface when groundlevel practical goals are focused upon which do not in themselves need to be linguistically represented. But it is doubtful whether there is always such a nonrepresentational element underlying our ordinary, often negotiated, conversational goals. Be that as it may, it still holds that once goals are fixed, they can select the features of the world that are objectively relevant to their achievement and about which the same participants who agree about the goals may be wrong in many ways. Finally, constructed context is often characterized as extendable in any direction, so that a suspicion of incompatibility may also arise between conceiving context as constructed and as limited. But insofar as additions or refinements are still obliged to conform to delimitation criteria, the constructed nature of some parts of the objective context does not give rise to potential all-inclusiveness. The dependence of objective context on goals may help explain how, even when constructed, it avoids being unlimited. Enlargements of the context that correspond to a change in the goals of the interactional event simply yield another context. Consider, for instance, topic shift. A topic-shifting utterance contributes to the construction of

⁵ [In the original paper, published in 2002, Gauker 2003 was cited as “in press.”] ⁶ It should be noted that Gauker’s conception of context is not aimed at solving problems in speech act theory. It provides the framework for a logic of assertibility in context, in which the validity of inferences is dealt with not in terms of truth-value, but of assertibility in a context (non-epistemically defined), and thus it becomes the main instrument of his criticism of received approaches to semantics. Here, we are not directly concerned with such foundational issues. But it might be instructive to recall that a polemic attitude towards core semantic notions was far from foreign to Austin.

   

141

context, since it makes matters that were not relevant become such. But what happens is not an enlargement of the current context: it corresponds to a shift in the orientation of the participants (what they are focusing on, what they are up to) and therefore in their goals (broadly intended). So we get a limited context again, albeit a different one.

4. Speech Acts and Context Change Context change has sometimes been used to define speech acts and in particular to substantiate the intuitive (and Austinian) idea that they are actions. Actions typically involve results, such as the bringing about of a new state of affairs, the conservation of an old one, or the substitution of a new state of affairs for an old one. Some authors have claimed that speech acts can be described as bringing about such changes in the context (Ballmer 1978, 1981; Stalnaker 1978; Gazdar 1981; see also Sbisà 1987a, 1989a, ch. 2). Gazdar (1981, 68–9) has defined speech acts as functions from contexts into contexts: they apply to a context and transform it in regular ways depending on their illocutionary force. Levinson (1983), in reviewing such a proposal, has deemed it interesting, but has raised doubts about its possibility to develop into a full-fledged theory. Nobody has taken his challenge up, probably because (as I have noted in Sbisà 1995a) mainstream Searlean–Gricean speech act theory tends to identify acts with gestures rather than with the bringing about of effects under the responsibility of an agent, and therefore most speech act scholars are not interested in the effects of speech acts (in fact, they tend to be happy with the reading of Austin (1975) given by Strawson (1964) and Searle (1969, 47), which considers reception as the only effect essential to illocutionary acts and disregards the fact that Austin quite clearly thought of conventional effects, produced thanks to reception but going beyond it, as equally essential). Nevertheless, or perhaps for this very reason, I think the contextchanging function of speech acts is worth reconsidering. A good starting point to realize how the objective or cognitive nature of context affects the description of speech acts as context-changing actions is Stalnaker’s analysis of the speech act of assertion (1978). After an assertion is made, its content is added to the participants’ presuppositions, and therefore to the cognitive context. The context has changed; the set of possible worlds among which participants may want to distinguish from now on has been reduced. At first sight, Stalnaker’s idea is simple and convincing, but some doubts arise nevertheless. Firstly, it is difficult to transpose context change so conceived to non-assertive speech acts. If I say “Please close the door,” what information is added to the context? Is it a piece of information about what I want you to do? But this would reduce the command to a speech act that gives information about the speaker’s

142

   

desires. Secondly, a perplexity may arise as regards whether Stalnaker’s context change belongs to the illocutionary or perlocutionary dimension of the speech act. Presuppositions are cognitive states of the participants, basically beliefs (albeit sophisticated ones: beliefs that the interlocutor shares a certain belief of the speaker). But producing belief is a typical perlocutionary act. So Stalnaker’s context change does not account for speech acts qua illocutionary acts. Furthermore, how do assertions achieve Stalnaker’s kind of context change? It is up to the receiver to accept an assertion as true or reject it. Gricean cooperative maxims are of no help since they say that the speaker ought to try to say something true (by avoiding saying what he or she believes to be false or what he or she has no good evidence for), but they do not say that the speaker ought to be right. So, from the assumption that the speaker observes the cooperative maxims, we may derive only the further assumption that he or she has attempted to say something true, not that his or her assertion actually is true. Therefore, the receiver may assume that the speaker is cooperative but disbelieve his or her assertion. So, the speaker is not entitled to deal with the content of that assertion as henceforth presupposed. Stalnaker also recognizes another kind of context change, that he deems to be “commonplace” (Stalnaker 1978, 88). An assertion may be said to change the context just because it is made, that is, according to Stalnaker, by the very fact that certain words are uttered. Something new has happened in the world and the participants in the conversation are both aware it has. This too makes cognitive context change: participants now share the assumption that something has been said. Stalnaker deems this kind of context change to be commonplace because, he says, it is as if a goat walked into the room. I think that Stalnaker’s commonplace kind of context change is more interesting than he believes. Consider nonassertive speech acts. It definitely makes sense to say that “Please close the door” changes the context by adding to it “Mary has requested John to close the door.” But somebody’s having made a request is no such trivial a state of affairs as the presence of a goat. It consists of a change in the entitlements and obligations of the participants with respect to each other. This change is brought about by convention, that is on the basis of an agreement among the relevant social participants: it is the conventional effect of an illocutionary act as opposed to mere reception or a perlocutionary consequence. And it is a change in the objective context, since entitlements, obligations, and the like are (once brought into existence by human agreement) objective features of interactional situations, not merely cognitive states of the participants. So if we admit that the very utterance of a sentence affects the context, we may see that speech acts are not merely expressions of communicative intentions to be recognized, but full-fledged actions yielding results. This perspective can be extended to assertion too, since an assertion entitles its addressee to assert the same to further addressees and commits the speaker not to contradict him or herself and perhaps to give, on request, evidence

   

143

or reasons. Thus, the objective conception of context turns out to be compatible with a context change view of speech acts.

5. Concluding Remarks Context is a matter of what the interlocutors are doing or attempting to do. There is undoubtedly an underlying subjective element, which was perhaps already implicit in Austin’s way of delimiting the context pertinent to a speech act and is explicitly present in Gauker’s definition of the objective context. Gauker’s contribution is helpful in that it points out that the delimitation of context not only is a matter of relatively sophisticated, conventional felicity conditions but may depend in a very straightforward way on non-verbal action and the extralinguistic conditions of its success. The presence of such a subjective element in the delimitation of context does not keep context itself from being objective. An objective, limited context is what is needed for situating speech acts and evaluating them as situated. This conception of context should not push us to disregard the results about construction of context, definition of situations, frames, contextualization cues, and so on achieved by the sociolinguistic analysis of verbal interaction. These results may be held to concern primarily cognitive context. But even objective context can be negotiated, constructed, and changed, insofar as goals may be negotiated or shifted (even non-verbally) and conventional or institutional states of affairs such as attributions of rights, obligations, entitlements, commitments depend on the agreement of the relevant social participants. In the view I am proposing, contexts are continuously shifting, but at each moment of an interaction it is possible to evaluate the performed speech acts against the context set by the goals of the interaction considered so far. Contexts are also continuously changing, not only because non-verbal actions or events make actual circumstances change but also because speech acts themselves bring about changes in the conventional features of the context, notably those regarding rights, obligations, entitlements, and commitments of the participants. So, a conception of context as at least in part constructed, limited, and objective enables us to describe speech acts as context-changing social actions.⁷

⁷ I am grateful to all those with whom I have discussed the notion of context in recent years, among which are Claudia Bianchi, Roberto Festa, Christopher Gauker, and Carlo Penco. I would like to thank also two anonymous referees for their critical comments and suggestions and Anita Fetzer and Varol Akman for encouraging me. This research was made possible by a joint grant by the Ministero dell’Università e della Ricerca Scientifica e Tecnologica (MURST) and by the Università degli Studi di Trieste. The national project, on “Mente e linguaggio,” is coordinated by Paolo Leonardi, Università di Bologna; the local project, on “Credenza, conoscenza e contesto,” is coordinated by myself.

8 Cognition and Narrativity in Speech Act Sequences 1. Introduction One of the recognized faults of received speech act theory (Searle 1969, 1979) is its inability to account for sequential phenomena in language use (noted for example by Levinson 1983 and Schegloff 1988, and even recognized by Searle 1992). There is no speech act analysis of conversation as such: speech act assignments seem to be tied to a sentence-level analysis. But when faced with such shortcomings of speech act theory, should we definitely put aside the idea that in using language we perform actions? This was Austin’s original insight—although its subsequent developments have stressed the speaker’s intentions rather than the production of effects which should intuitively characterize action. Whoever still finds the idea attractive that in using language we perform actions should reconsider the cognitive and interactional dynamics by which speech act sequences are produced and understood. This paper aims at offering a contribution to such a reconsideration. By the way, I am not only convinced that the connection between language and action is sound but also that it has not been sufficiently exploited by standard speech act theory. I will try to foreground some features of speech acts as actions which are often neglected and supplement speech act theory with a conception, inspired by narrative semiotics, of an action as a move in a sequence. In my reconsideration of the cognitive and interactional dynamics of speech act sequences, I will reassess the role of action (as opposed to that of cognitive representation), but also concede a role to a factor that may be called cognitive, which contributes both to our understanding of actions and to the successful performance of those linguistic actions that have been called illocutionary acts. I will argue that: (i) the cognitive component in speech act sequencing is secondary to and dependent on action; (ii) one cognitive factor relevant to the understanding of speech act sequences is the so-called “narrative schema,” proposed by A. J. Greimas (1983) in the framework of narrative semiotics, that analyzes sequences of events in three main steps: Manipulation, Action, and Sanction. Originally published in Rethinking Sequentiality (Pragmatics & Beyond New Series 33), ed. Anita Fetzer and Christiane Meierkord, 71–98. Amsterdam: John Benjamins, 2002. Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0009

      

145

2. Speech Act Sequencing and the Production of Effects In order to make the analysis of speech act sequences viable, we should consider what kind of a bond can make a sequence out of a few speech acts uttered in a row. Much depends on what we understand as the output, or final state, of a move in a speech act sequence, namely, how we define the effects of speech acts. In dealing with this problem, I will introduce the Austinian idea of a “conventional” effect characteristic of illocutionary acts. Then I will consider the recognition on the part of the analyst of the effects brought about by illocutionary acts, and the role that speech act sequencing is to play in such recognition.

2.1 What Effects Do Illocutionary Acts Bring About? When considering a sequence of moves, it is reasonable to view the output of one move as coinciding with the input for the next. For example, once the move of one chess player modifies the setting of the chess board from an initial state At₀ to a resulting state At₁, state At₁ becomes in turn the initial state to be modified by the following move of the other chess player so as to yield a further resulting state At₂ (see Figure 1). If we were to understand the dynamics of speech act sequences this way, we should first try to achieve a clear understanding of what kind of a state may constitute the output of each move. This involves taking into consideration the effects of speech acts and asking ourselves what kind of states they consist of. Speech acts are generally thought to have cognitive effects, since they bring about the hearer’s recognition of the meaning and force of the speech act. This is usually held to be the only effect essential to the illocutionary act. But an effect

A t0

M1

A t1

M2

A t2

Figure 1

146

      

which is merely the cognitive reflex of the speaker’s intention does not seem to be a suitable input for the next move in the sequence, at least not in the sense exemplified above. If the recognition of the meaning and force of the speech act were the only output of the illocutionary act, it should coincide with the initial state to be affected by the subsequent move, but this is clearly not the case. A communicative intention has been manifested and recognized, period: if anything else happens, it would seem to fall under the category of perlocution and therefore be suitable for being tackled by psychological or sociological tools rather than by speech act theory. The cognitive state produced in the hearer by the speech act contributes to the speech act sequence only indirectly, by creating the environment in which a new communicative intention and therefore a new speech act will arise, addressed by the hearer to the former speaker (see Figure 2). The ties between each move and the following one are internal to each participant and therefore the moves are not connected with one another in a series. I hold a different view of the effects of illocutionary acts which, to some extent, goes back to Austin (1975, 116–17). According to this view, inasmuch as they are illocutionary acts, it is essential to speech acts to have conventional effects (such as assignments of obligations or entitlements) (see also Brandom 1994; Gazdar 1981; Sbisà 1987a, 1989a). According to Austin, the illocutionary act is associated with three different kinds of effects (1975, 116–17): (i) the securing of uptake; (ii) the production of a conventional effect; (iii) the inviting of a response or sequel. Effect (i) amounts to bringing about the understanding of the meaning and the force of the utterance and, unless it is achieved, the illocutionary act is not actually carried out. I would like to emphasize that, pace Strawson (1964), Austin does not imply that this is the only kind of effect that any illocutionary act must achieve; instead, there is an implication that this is a necessary step for the successful performance of the illocutionary act, which also comprises effects (ii) and (iii). Effect (ii) consists of bringing about a state of affairs in a way different from bringing about a change in the natural course of events: the act of naming a ship “Queen Elizabeth” makes it the case that “Queen Elizabeth” is the ship’s name, and that referring to it by any other name is henceforth out of order, but these are not changes S

H

communicative intention

recognition

recognition2

communicative intention2

communicative intention3

recognition3

Figure 2

      

147

in the natural course of events, since they involve conventional elements (notably norms, and therefore obligations). Austin says too little about the effects of illocutionary acts that can be so characterized. However, it is intuitive that matters such as the assignment of an obligation are not changes in the natural course of events: what should be done does not depend in any simple way on what is the case. Moreover, obligation (as well as commitment, right, entitlement, and the like) is a matter that can be established or modified by convention (law, social consensus, or intersubjective agreement) and is therefore very plausibly connected to the “conventionality” of the illocutionary act. This sense of “conventionality” should not be identified with the linguistic conventionality of the performative formula as implied by Strawson (1964). Effect (iii) amounts to inviting a certain kind of subsequent behavior; if the invitation is accepted, a certain further act by some participant will follow. In my work on speech act theory (Sbisà 1984a [Essay 1], 1987a, 1989a, 2001a [Essay 6]), I have argued that Austin’s effect (ii), the production of changes not in the natural course of events but in conventional states of affairs, is an essential feature of the illocutionary act and can be detected not only in institutional acts such as the example provided by Austin (the naming of a ship) (1975, 116) but also in illocutionary acts usually performed in verbal interaction (see also von Savigny 1988). The relevant conventional states of affairs basically consist of the possession or lack of obligations or rights on the part of some agent. In my reading of Austin, the conventionality of illocutionary acts is grounded in the conventionality of effects of kind (ii), and the conventionality of these effects is grounded in their dependency on some kind of agreement between the interlocutors.¹ I am not thinking of explicitly signaled agreement (which may nevertheless sometimes occur): the default situation itself (the absence of any manifestation of disagreement) is constitutive of the interlocutors’ agreement. Whenever the hearer does not challenge the felicity of the speaker’s illocutionary act, he or she may be counted as accepting it as felicitous under the characterization that the speaker has provided. Whenever the speaker does not reject the hearer’s uptake, he or she agrees that he or she has done what he or she has been taken to do. In case the hearer happens to have reasons for challenging the felicity of the speech act, he or she may still accept the speaker’s characterization of its force but deem it infelicitous, or take it as having a different force from that suggested by the speaker’s characterization, so as to make it felicitous (in this case it will be up to the speaker to accept or reject the hearer’s uptake). When no agreement between hearer and speaker is achieved, the illocutionary act does not count as having been successfully performed and fails to produce its conventional effect. Imagine an order issued without authority: the hearer accepts the speaker’s characterization of the speech act as an order (after all, the speaker has used the imperative mood and an

¹ [See especially Essay 10 and Essay 11.]

148

      

authoritative tone), but once he or she discovers that the speaker does not possess the relevant authority, he or she no longer accepts the order as successfully performed and the purported order turns out not to be binding for him or her. The fact that the bringing about of the conventional effect needs an agreed-upon uptake explains why the Austinian conventionality of illocutionary acts goes hand in hand with their being liable to infelicities and even voidability or annulment. If the effects of illocutionary acts are so conceived, especially if illocutionary acts have conventional effects in the sense explained above, there is indeed some intersubjective reality that can work as both output and input of the moves of a speech act sequence. Any speech act in a speech act sequence may be thought of as bringing about an effect on the conventional setting of the interpersonal relationship between speaker and addressee as it ensues from the preceding speech act (if there has already been one). The modified conventional setting provides the input to be acted upon by the following speech act. Thus the production of conventional effects is just what makes speech act sequencing possible.

2.2 The Role of Sequencing in the Production of Effects I turn now to one of the consequences of the account of illocutionary acts I am proposing as regards the applicability of speech-act theoretical notions to the analysis of discourse. If, while analyzing a certain conversation, we want to tell what conventional effect has been brought about by a certain utterance, and if that conventional effect depends on an agreed-upon uptake, that uptake must be taken into consideration. This creates a closer link between each element in a speech act sequence and its preceding or subsequent one than has ever been suggested by standard speech act theory. Firstly, we must consider the response of the hearer and the kind of uptake it manifests, and secondly, we must also consider (at least) the speaker’s response to the hearer’s response, in order to see whether there is agreement between speaker and hearer about the illocutionary act and its felicity and therefore what conventional effect, if any, has been achieved. Consider the utterance of an imperative sentence. If it succeeds in assigning the addressee an obligation (as it should, provided it is taken as a felicitous order), the next move in the sequence will act upon this obligation, for example by either accepting or refusing to comply with it. But how can the analyst tell whether the utterance has succeeded in assigning that obligation to the addressee? If the addressee says “OK, I’ll do it as soon as possible,” or “Sorry, I cannot do it,” we may say that yes, he or she behaves as someone who agrees that he or she has been assigned an obligation. But what if the addressee replies: “Don’t bother me,” or “I’ll do it if I want to”? He or she has replied to the speaker’s utterance, so that a second step in the speech act sequence has been taken. But no authority to assign obligations to the addressee seems to be accredited to the speaker by such replies,

      

149

even implicitly, so that it is apparent that the hearer does not agree that he or she has been assigned an obligation, and therefore the speech act cannot be said to have succeeded in assigning an obligation to him or her. We could say that the speech act has been taken as a rude request, or as an illegitimate claim to authority, or as a mere expression of will or desire. If the speaker does not initiate a repair procedure, he or she will be counted as accepting the hearer’s refusal of his or her authority and the failure of his or her own purported order. For the sake of simplicity, I am not considering here other participants whose responses may be relevant too, for example ratified hearers who are not themselves addressees; indeed, their alignment with the addressee rather than with the speaker or vice versa may in many cases provide a decisive element of intersubjective agreement. In these cases, one of the interlocutors may be said to be wrong about the force of the speech act. Simplifying a bit, the construction of a speech act sequence determines at each of its steps (assuming the next step does not withdraw agreement), the effect brought about by the previous one. Thus, we have a two-fold relationship between the conception of the illocutionary act as having conventional effects and the availability of a speech-act theoretical analysis of speech act sequences: the fact that there are conventional effects (as has been argued above) makes sequencing possible, but sequencing also makes it possible for an illocutionary act to achieve its conventional effect and even enables the analyst to tell what illocutionary acts have been performed by the participants.

3. The Dynamics of Speech Act Sequences I would now like to exemplify the account of speech act sequences outlined above by discussing a sequence from a recorded conversation.² The participants, A and

² I have used the following transcription conventions: Beginning of an overlap * or # (in both turns) End of an overlap + or/(in both turns) Latched utterances = Short pause (up to 2”) (–) Pause (between 2” and 5”) (—) Slightly rising or level intonation contour , Rising intonation contour ? Falling intonation contour . Lengthened syllables :, :: Incomplete words xLow volume (ˆ word) Emphasis bold Non-linguistic phenomena and comments ((xxxx)) Unintellegible words xxx Omissis ( . . . ) Punctuation marks in the English translation are only meant to make it readable.

150

      

B, are both women; A is the hostess and B the guest. B is going to record an interview of A; the recording has already begun, but not the interview itself. A and B are in A’s kitchen but A, as it turns out, wants to have B sit more comfortably in the living room. (1)

1 A: scu:sami B, “sorry B” 2 B: niente “it doesn’t matter” 3 A: speta che ((noise of chairs being moved)) (—) a a ti verrà uno scric *(ˆ uno) eheheh+ “wait, I’ll just . . . you’ll get a stiff neck (a) eh” 4 B: *perché eheheh+ “why eh” 5 A: (xxx) ma (–) no: ((quickly)) se volevi andavamo in salotto che stavamo tanto #più comode/ “but, no, if you’d wanted we could have gone into the living room we’d have been much more comfortable there” 6 B: #indiffe/rente (–) portiamo le robe. (–) “it’s all the same, let’s bring the stuff through . . .” 7 A: sì dai (–) ti porto io (ˆ questo) “yes, do let’s. I’ll carry this for you” 8 B: prendi la maniglia “take hold of the handle.”

In this example we find both responses that manifest by-default agreement and responses that display a refusal to take the speaker’s act at its face value, sometimes introducing an uptake that diverges from the speaker’s own expectations. Among the replies that manifest by-default agreement by taking the preceding speech act at its face value, there are: 2 as a standard reply to 1 as an apology, 7 as a reply to the second part of 6 “portiamo le robe” (“let’s bring the stuff through”) taken as a proposal, and 8 as a reply to 7 taken as an offer of help. In 8, B implicitly accepts A’s utterance “ti porto io (ˆ questo)” (“I’ll carry this for you”) as an offer to carry her recorder, by telling A how she is to carry it. In none of these cases is the hearer’s uptake further challenged by the speaker. As to responses (and responses to responses) that do not manifest agreement, in the central part of the sequence (from turn 3 to turn 7) at least two interesting cases can be found. B, in turn 4, interrupts A with a question that manifests her taking the second part of A’s turn 3, henceforth 3b “aa ti verrà uno scric (ˆ uno) eheheh” (“you’ll get a stiff neck a eh”) as an assertion or prediction (the first,

      

151

truncated part of 3 does not seem to elicit any response). Apparently, 3b bears the mark of prediction (future tense) and a prediction is an act of judgment (a Verdictive in Austin’s terminology),³ a kind of speech act in response to which it is natural to ask for evidence or reasons. B’s uptake of 3b as a prediction hinders 3b from counting as an explanation of why B should be expected to desire to move elsewhere (giving the grounds for a possible offer to move), or as a warning to B about the consequences of her sitting uncomfortably. But B’s turn 4 (“perché eheheh,” “why eh”) is in turn unsuccessful as a question: A doesn’t answer it, rather, she seems to reject its legitimacy or appropriateness by saying, after some words too unclear to be transcribed, “ma (–) no” (“but, no”) and going ahead to issue an utterance unrelated to 4. If 4 does not count as a successful question, it is also hindered from making assignments of force to preceding turns on the basis of its being a question. So, no force of assertion or prediction is assigned to 3b. A slightly different case is the relationship between the second part of turn 5, henceforth 5b “se volevi andavamo in salotto che stavamo tanto più comode” (“if you’d wanted we could have gone into the living room we’d have been much more comfortable there”) and B’s reply to it in 6. Here it is possible to observe how sequencing selects one reading for an illocutionarily ambiguous conversational move. 5b might be an offer: the undertaking of a commitment to do something that is desirable for the addressee; or merely a proposal: an act granting the addressee the right to do something with which the speaker agrees to cooperate. But B, in 6, denies having any desire or interest in moving by saying “indifferente” (“it’s all the same”), so that the reading of 5b as an offer does not become effective. The way in which B behaves here is consistent with her ruling out, by issuing turn 4 (discussed above), the reading of 3b as giving grounds for a possible offer to move. Utterance 3b does not give grounds for making an offer, since no offer can be made. Utterance 5b, however, manages to become effective as a proposal. The second part of B’s turn 6 (portiamo le robe. “let’s bring the stuff through”) is quite appropriate as a reply to a proposal to move. So, turn 6 can be viewed as operating a selection among the forces that 5b might have, that of an offer (which is rejected) and that of a proposal (which is accepted, by accepting the proposal itself). At the beginning of turn 7 (“sì dai,” “yes, do let’s”), A ratifies what has been done up to that point. A further problem is whether A has managed to do anything with her utterance 3b “a a ti verrà uno scric (ˆ uno) eheheh” (“you’ll get a stiff neck a eh”). Does a process of positively selecting an illocutionary force take place here too? To see this, we have to consider not only turn 4 and the first part of 5 (henceforth 5a) that

³ Verdictives are, in Austin’s definition (1975, 151, 153) “the delivering of a finding, official or unofficial, upon evidence or reasons” as to “something—fact, or value—which is for different reasons hard to be certain about.” It is because of such a connection with evidence or reasons that the speaker issuing a verdictive is exposed to being asked why he or she has done so.

152

      

rejects 4 as a reply to 3b, but also its second part 5b, and realize that 5b works at least in part as a third position repair,⁴ clarifying A’s intention to make B feel comfortable, so that 3b, as a result, counts as an expression of concern for B’s comfort, indirectly motivating the proposal to move. In all these ways, the conversational sequence contributes to determining what is brought about by the speech acts that compose it. It is once we view the conventional effects of each move as settled, that the conversational sequence can be seen as proceeding by means of moves each of which operates on the output of the preceding one. It could be objected that this analysis is incomplete: there are features in example (1) that cannot be traced back to illocutionary force, and even less to conventional illocutionary effects defined as matters of rights and obligations. For example, it would be quite pertinent to describe A’s behavior as too polite and B’s behavior as impolite (because it expresses uneasiness for A’s attempt at being exaggeratedly polite). But at least some of the facts that would justify this description are to be looked for in the chain of retrospectively ratified conventional effects I have tried to illustrate, as well as in the negotiation processes that shape them by blocking some assignments of illocutionary force and letting others through. I have said that utterances actually produce conventional effects only if there is some kind of agreement between speaker and hearer about the conventional effect to be produced (according to my reading, this was the rationale of Austin’s often misinterpreted claim that the securing of uptake is necessary to the successful performance of an illocutionary act). We have seen how this conception can be put to work in the analysis of speech act sequences. But the nature of the agreement we have been dealing with is still unclear. Is this agreement a cognitive matter, namely, the sharing of a mental representation (to be intended, most plausibly, as the instantiation of identical or very similar representations in the minds of the participants)? Or does it depend primarily on action and even consist of action coordination? We might be tempted to say that illocutionary acts succeed in producing their conventional effects when the hearer’s representation of these effects matches the speaker’s. In this conception, the relevant agreement is a cognitive fact. The speaker must have in mind some representation of the intended effect; the hearer understands the speaker correctly only if he or she constructs a mental representation of the effect intended by the speaker which matches the speaker’s own representation. If the two representations do not match, and the participants realize that this is so, the issue can be solved by negotiation. Negotiation is

⁴ A third position repair is a conversational turn that reformulates a preceding turn by the same speaker (considered as occupying the first position) in response to a turn by the interlocutor (occupying the second position) that has given the turn in the first position a reading the speaker is not willing to accept. The three positions are not necessarily occupied by adjacent turns (Schegloff 1992).

      

153

successful if it leads the hearer to form the correct representation of the effect intended by the speaker, or in any case, if it gets speaker and hearer to work out matching representations. But if the sharing of a representation of the conventional effect were the necessary condition for its production, we could never legitimately consider any such effect as achieved, and no analysis of speech act sequences as chains of conventional effects would be viable. Indeed, what evidence may we ever have for a representation being shared by two participants, or more precisely, for the matching of the representation entertained by one participant with the one entertained by the other? Mental representations (whatever their role in, say, psycholinguistic hypotheses) are not things that can be examined and compared. We might try to infer what mental representations are entertained by the participants from what they say and do, but even such evidence is never final. On the one hand, equivalent representations might for idiosyncratic reasons give rise to different utterances or stretches of behavior; on the other hand, the smooth connection between utterances by different interlocutors and even the use of the same linguistic expressions might hide differences in the corresponding mental representations. There is no way to be sure, and therefore, since the production of conventional effects depends on agreement, there should be no way to tell whether any conventional effect has been produced. But in our analysis of speech act sequences, we can tell that a certain conventional effect is actually achieved whenever subsequent steps in the speech act sequence do act on it. As the analysis provided in the preceding paragraph has shown, it is the observation of verbal behavior (utterances, silences . . . ) that entitles the observer to say that there is agreement between the participants about the effect of a certain speech act. Since behavioral evidence is the only kind of evidence available,⁵ what’s exactly the point of supplementing the remark to the effect that the participants converge as to their way of dealing with a speech act with the further claim that they share a mental representation of its effect? Such a claim does not seem to play any substantial role with respect to the assignment of force. Thus, insofar as assignment of force is concerned, resort to mental representations is at best otiose. Let us consider the issue from the point of view of the participants themselves. There might well be the same mental representation of what has been done both in the speaker who utters the imperative sentence and in the addressee who replies “Sorry, I cannot do it.” But this is not what enables the speech act actually to assign an obligation to the addressee. What is relevant to this effect is that the addressee, by apologizing for not being able to comply, implicitly agrees that he or she has been assigned an obligation. Otherwise, he or she would not offer an apology in ⁵ In discussing examples from conversations in this paper, I focus on verbal behavior, but of course, the observation of verbal behavior may be usefully completed by the observation of its non-verbal accompaniments, such as gestures, non-verbal actions, and omissions.

154

      

order to cancel the debt he or she owes the speaker for not complying. He or she is in a position to apologize for not being able to comply exactly because (or insofar as) he or she owes something to the speaker for not complying with the obligation that the speaker has just assigned to him or her. By apologizing for not being able to comply, the addressee has dealt with the speaker’s utterance as assigning him or her an obligation. We may further imagine that the speaker does not say or do anything that implies a rejection of the hearer’s way of dealing with his or her speech act. This again does not coincide with his or her having the same mental representation as the hearer. But it is when the behavior of the speaker converges with (or fails to diverge from) that of the hearer in this way that the conventional effect of the illocutionary act is achieved. And therefore it is the observation of such behavior that enables the analyst to tell whether the participants agree about the obligation-assigning conventional effect of the speech act being examined. So, I conclude that the agreement which determines the actual effect of the speech act is not primarily a cognitive matter: it does not consist of mental representations or of other psychological states, but of interactional processes. The way the speech act is dealt with by the hearer (provided the speaker does not reject it) grants that a certain effect has been brought about and becomes the source of the cognitive appreciation of it as well as of all the possible associated metapragmatic judgments (whether those of a participant who later on reports on what has happened or those of an analyst). Of course, this does not mean that decisions such as taking the speaker’s utterance in a certain way aren’t made possible in their own turn by underlying psychological processes. But first, psychological facts as such do not seem to play any part in the ratification of the conventional effects of speech acts; therefore, they need not be included in the analysis of the speech act sequence as a chain of effects. Second, the underlying psychological processes that enable speaker and hearer to produce language and to coordinate their action by means of it, might in their own turn be something different from the forming and computing of language-like mental representations, that is, states of the mind/brain representing states of affairs in much the same way in which the sentences of a language may be said to do so.

4. The Role of Narrativity in Speech Act Sequencing The claim put forward above that the speech act sequence as a chain of conventional effects comes into being independently of its cognitive appreciation does not rule out every cognitive factor from speech act sequencing and its analysis. So far, I have considered speech act sequences as if they were unstructured. In order for a series of speech acts uttered in a row to be considered as making up a sequence, I have posed the sole requirement that the effect of each speech act on the conventional aspects of the relationship between the participants serves as the

      

155

initial state to be acted upon by the following speech act. But this says nothing about where a sequence begins or ends; it does not explain why certain co-occurring non-verbal actions count as part of the chain while others do not; and it is not always enough to tell what other speech act a certain speech act is responding to. The analysis of speech act sequences is in need of some kind of criterion for discerning a structure in apparently unstructured and undelimited series of moves. Moreover, participants themselves might have to possess and apply such a criterion, insofar as responding to a certain move involves collocating it in a sequence and identifying a sequence is no trivial matter. At this point, I propose taking into account a factor that might be considered as cognitive, which contributes to our understanding of action in the framework of structured sequences of events. Narrative semiotics (more specifically, the semiotic methodology for analyzing the narrative dimension of texts, elaborated by A. J. Greimas and his school: Greimas 1983; Greimas and Courtés 1979) has provided an analysis of action (considered as an agent’s bringing about of a state of affairs) which contextualizes any event that is to count as somebody’s action between an initiative move (Manipulation) and a reactive move (Sanction). Such a three-fold structure, called the “narrative schema,” can be viewed as a part of the competence by which we understand sequences of events in general and should therefore apply to speech act sequences too. I have already claimed elsewhere (Sbisà 1989a, ch. 7; 1998) that it can be applied quite successfully to the analysis of speech act sequences.⁶ According to the narrative schema hypothesis, the first move in a sequence creates the conditions out of which the agent’s action will arise. For example, an agent loses something or is robbed of something, and therefore becomes desirous to find it again; or else, somebody assigns the agent a task, and he or she thereby acquires the obligation to carry it out. The subsequent move in the narrative schema is the performance of the action, namely, the carrying out of the envisaged “narrative program” (finding the lost or stolen object; performing the assigned task). Such a performance consists of the bringing about of a state of affairs and presupposes that, previously to the performance itself, the agent is assigned the necessary “competence.” In fact, the agent will find the lost object only if he or she remembers where it has been lost, or will manage to fulfill the task assigned to him or her only if he or she first appropriates, say, a magic tool. The third move in the narrative schema is a reactive or sanctioning move. After the agent has carried out his or her performance, whether successfully or not, his or her behavior will receive an implicit or explicit evaluation, in consideration of the extent to which ⁶ It should be noted that my use of tools drawn from narrative semiotics, such as the narrative schema, is basically heuristic. I do not think that narratives “really are” or “should be” structured (whatever this might mean) exactly like narrative semiotics makes them. But narrative semiotics proves able to yield insightful regimented paraphrases of narrative texts (as well as of narrative aspects in texts of any kind).

156

      

it meets the requirements set by the manipulating move. Sanction may take a variety of forms: it may consist of an explicit assessment and reward (or punishment), but also of any reaction which implicitly manifests an assessment of what the agent has done or failed to do. As is known by all those who have used narrative semiotics in the analysis of narratives, the sanctioning authority is formally the same “actant” as the manipulating authority (the “Destinator”), but can be manifested in discourse by an actor (i.e., character) other than the one manifesting the manipulating authority.⁷ Sanction may be positive or negative; when it is negative, it may be so to various degrees. It might consist merely of some criticism of some aspect of the agent’s performance, of a radical rejection of the purported action as not responding to the initial manipulation, or of a refusal to recognize the agent supposedly responsible for the performance as really responsible for it. Such an analysis of (narrated) action can be used to make the following general point, which may be taken to hold also as regards the analysis of social interaction: whenever we want to consider a piece of behavior as an “action,” we should expect to find the manipulating move that has stimulated the agent into behaving that way, as well as the sanctioning move which assesses the performance that has been carried out. It should be noted that in the case of speech act sequences, if the description I have given of the way speech acts produce their effects is sound, the sanctioning move is not a mere comment on (or reward of ) something which has already been done independently of the sanctioning move itself, but contributes to defining and making effective the very speech act it sanctions. Should we expect the sequential positions of Manipulation, Action, and Sanction to be regularly occupied by different kinds of speech acts? A regular association has sometimes been claimed to hold between kinds of speech acts and sequential positions. This is the case, for example, with “initiative” and “reactive” speech acts as discussed by Weigand (1989, 1994) in the context of a dialogically oriented reformulation of speech act theory. I share some of Weigand’s intuitions, for example the idea that illocutionary acts should be defined and identified not in isolation, but with reference to a dialogic sequence. For Weigand, though, there exist inherently initiative and inherently reactive speech act types: whether an illocutionary act is initiative or reactive depends on its linguistic form. While I admit that some linguistic forms (such as the interrogative form or the imperative mood) are more suitable for initiating a sequence and others (such as indicative sentences) for sanctioning a previously performed action, in my view the narrative schema can in principle be applied at any point of any sequence for heuristic aims, that is, for the sake of clarifying the connections between the speech act we want to consider as Action and the preceding and subsequent ⁷ Narrative semiotics distinguishes actants qua syntactic roles in the narrative from actors qua semantically loaded agents. Actants and actors need not stand in a one-to-one correspondence to each other.

      

157

moves in the sequence. This does not mean that there are not preferred points for such an application: the narrative schema is most useful when such preferred points can be detected. But the fact that it can in principle be applied at any point accounts for the availability of multiple readings of what appears to be the same sequence or even the same speech act. We can better make sense of some of the intriguing complexities of example (1) if we consider the various readings at issue as related to the construction of different Manipulation–Action–Sanction (henceforth: M–A–S) sequences. Let us consider the relationship between A’s utterance 3b “a a ti verrà uno scric (ˆ uno) eheheh” (“you’ll get a stiff neck a eh”) and B’s reply 4 “perché eheheh” (“why eh”). Let us focus on either of these moves as filling in the Action slot in a M–A–S sequence (see Figure 3). (i) If we focus on 3b as filling in the Action slot, what performs the function of a manipulating move is the non-verbal situation, that is, the fact that B is sitting in A’s kitchen and A as a hostess owes B as much comfort as possible. So 3b appears as an attempt of A to fulfill her obligations as a hostess (by showing concern for B, or perhaps, by warning B about the negative consequences of her sitting uncomfortably). In many ways, it is an inappropriate attempt: basically, it is not enough to repair A’s debt. It is no surprise that B’s utterance 4 sanctions it negatively, not even recognizing it as such an attempt: in fact, in 4 B asks A to provide reasons for her remark 3b, and asking for reasons is clearly not the kind of response that an attempt to fulfill one’s role as a hostess may elicit. (ii) If we focus on 4 as filling in the Action slot, we have to take 3b as the manipulating move. So, 4 appears as the action of asking for evidence or reasons, prompted by 3b taken as a verdictive speech act of prediction. But the next move, A’s utterance 5a “ma (–) no:” (“but, no”) sanctions 4 negatively by failing to accept it as the performance of an action appropriately responding to the manipulating move in 3b. The negative sanction consists of A’s implicit refusal to consider 4 as a felicitous question: B does not answer it or in any way recognize her obligation to do so; A gives up defending the right to receiving an answer, which she would have, were her question felicitous. This indirectly undermines the assignment of the force of a prediction to 3b. m

a

s(–)

NV

3b

4

5a

m

a

s(–)

Figure 3

158

      

The resort to the narrative schema, with respect to the sequence described in (ii), suggests that 3b as a manipulating move is not exhausted by B’s uptake and, moreover, with respect to the sequence described in (i), adds to the previously outlined analysis a reason for connecting the utterance of 3b with A’s desire to act appropriately as a host, which constrains its possible readings. It should be noted that 4 remains fairly effective as a negative sanction of 3b even though, as an Action in the sequence described in (ii), it receives a negative sanction in its own turn and is therefore unable to redefine the force of 3b in a positive way. Another pair of moves that is worth discussing are 5b “se volevi andavamo in salotto che stavamo tanto #più comode/” (“if you’d wanted we could have gone into the living room we’d have been much more comfortable there”) and 6 “#indiffe/rente (–) portiamo le robe.” (“It’s all the same, let’s bring the stuff through”). We have already seen that 6 selects for 5b the force of a proposal rather than that an offer. Here too, however, we can focus either on 5b or on 6 as filling in the Action slot (see Figure 4). (i) If we focus on 5b as filling in the Action slot, we must once again consider the Manipulation slot as filled in by the non-verbal situation in which B is sitting in A’s kitchen and A as a hostess owes B as much comfort as possible. The action might be either offering or proposing to B to move to the living room. Turn 6 gives a selective but essentially positive sanction to 5b as a proposal. (ii) If we focus on 6 as filling in the Action slot, 5b plays the role of the manipulating move. B is invited to do something, and has to accept or reject the proposal. In fact, she accepts by making a further, related proposal. The way in which 6 accepts the proposal made in 5b is not fully explicit, but since 6 is sanctioned positively by 7, it can be counted as a successful proposal acceptance. That is, in 7, A not only accepts the proposal made by B in 6 “portiamo le robe.” (“let’s bring the stuff through”) but also accepts that proposal as a satisfactory response to the manipulating move 5b. Here, the interesting contribution provided by the narrative schema is that, since 5b cannot be construed as manipulated by 4, the narrative schema suggests m

a

s(+)

NV

3b+5b

6

7

m

a

s(+)

Figure 4

      

159

that it should be connected to the same non-verbal situation to which also 3b had to be connected if focused upon as filling in the Action slot, as in its reading (i) illustrated above. So 5b appears as an attempt on the part of A (stronger than 3b) at fulfilling the role of hostess. This might lead us to a closer consideration of the ties between 3b and 5b. If B had not interrupted A’s turn 3 by uttering 4, would A have gone ahead immediately, offering to B to move to the living room? It seems very likely. Moreover, the first part of turn 6 “indifferente” (“it’s all the same”) responds to 3b as much as to 5b, insofar as in both these utterances B’s interests or desires are at issue. This might suggest considering 3b plus 5b as one complex move aimed at fulfilling the role of the hostess, by which A does not succeed in making an offer to satisfy one of her guest’s desires, but succeeds at least in making a proposal motivated by her concern for her guest’s comfort. It might be objected that the heuristic use of the narrative schema offers the analyst too many possible alternatives: if any three speech acts in a row can be understood as Manipulation–Action–Sanction, we are still considering speech act sequences as unstructured. To this objection, I reply that while the multiple applicability of the narrative schema to speech act sequences shows that there is no real structure of speech act sequences separate from the interactional play of mutual responses amounting to agreement, this interactional play must itself rely on some abstract form according to which sequences are constructed. It might be the case that the response given to a certain speech act depends at least in part on the sequential slot in which the receiver collocates the speech act he or she responds to. The narrative schema as a form of the human imaginaire should perhaps be considered as prior to that kind of action coordination which enables human beings to achieve conventional effects. The perspective proposed here stresses that conversational turns do not constitute sequences simply by being produced one after another in the same physical circumstances. A sequence may consist of steps distant in time, so that the position next to a certain speech act need not coincide with the conversational turn immediately following it (cf. Schegloff 1992). This is one more reason why the analysis of speech act sequences needs to rely on a structured model. Given a certain move and its identification as filling in one of the slots of a M–A–S sequence, not everything is suitable for filling in the neighboring slots. If the conversational turns immediately preceding or immediately following are unsuitable, the analyst has to consider more distant conversational turns or non-verbal acts, or even events or situations, as candidates for the Manipulation or the Sanction position. In order to orient themselves in the interaction, participants too must rely on some appropriate competence, and it would not be surprising if this were the same competence that the analyst can successfully appeal to in an explicit way.

160

      

5. The Narrative Schema and the Understanding of Insisting The multiple applicability of the narrative schema enables the analyst to explore the relationships into which each move in a sequence enters when considered as filling in the Action slot and the ways in which these relationships affect its conventional effects. Thus, the narrative schema helps us to figure out what room a conversation grants to the strategic choices of the participants. But when the analysis aims at spelling out what chain of effects has actually been realized in a given conversational episode, multiple applicability has to be supplemented with criteria for choosing one preferred application. It is in conjunction with these criteria that the narrative schema enables receivers (whether participants or analysts) to identify those M–A–S sequences that are most relevant to the interactional event. One criterion for choosing where to apply the narrative schema focuses on the Action slot directly: sometimes, the move that is to fill in the Action slot is manifestly salient and the M–A–S schema merely suggests to look for a preceding manipulating move as well as a subsequent sanctioning one. But in other cases, and perhaps more often, contextually preferred applications of the narrative schema are triggered by the sanctioning move, thanks to the recognizability of its closing function. In such cases, the action is indirectly identified as the move being sanctioned. The manipulating move is then identifiable not only because it precedes the move in the Action slot but also thanks to the correlation between Sanction and Manipulation (an action can be sanctioned just insofar as it is expected to respond to whatever fills in the Manipulation slot). Thus, sanctioning moves work as a kind of narrative punctuation, facilitating both the delimitation of M–A–S sequences and the identification of the action. In all cases in which participants, or analysts, need to recognize when, in a complex conversational episode, the main action is actually performed, it might be a promising strategy to look for a move suitable for being understood as Sanction and apply the narrative schema from that point backwards. This strategy proves especially helpful when it is not at first sight clear what is being done in a sequence or whether any performance of an action is achieved at all, as is often the case in sequences where insisting occurs (Sbisà 1998). I will now consider one such sequence: our example (2) will consist of two related stretches (2a) and (2b) from one and the same conversational episode, a family therapy session,⁸ in which at least one of the participants insists on the pursuit of his conversational goal. My aim in discussing this example is not to ⁸ Example (2) is drawn from the corpus that was collected by Carla Bazzanella, Claudia Caffi, and myself while writing Bazzanella, Caffi, and Sbisà 1991. Part of the conversational stretch I comment upon here is quoted also by Caffi (1992, 292–3), who discusses it with particular respect to the involvement displayed by speaker C.

      

161

contrast a semi-formal conversation with the totally informal one in example (1), or to contrast a multi-party conversation with the two-party one in example (1) (although, as is obvious, these features influence the way in which these sequences are organized). I would like to show how insisting can be understood and described with reference to the narrative schema. The participants in the conversation from which example (2) is drawn are: the therapist (A), the wife (B), the husband (C), and the daughter (D). A has to collect information about the willingness of the family members to undertake the therapy and addresses a question (turn 1 from stretch 2a) at all the members of the family (particularly the husband, C, and the wife, B). The ensuing multi-party interaction comprises at least two two-party interactions: that between A and B (mostly smooth) and that between A and C, in which insisting occurs. B and C make various attempts to relate directly to each other, but the mediating role institutionally belonging to the therapist prevents any full-fledged two-party interaction between them from arising. We will be concerned with interaction between A and C in particular. I begin by presenting the first stretch: (2a)

1 A: ecco e che: (–) quando vi hanno detto di venire qui tutta la famiglia che cosa avete pensato “well and what, when they told the whole family to come here what did you all think” 2 B: io ho pensato che era la cosa giusta perché secondo me i bambini ne risentono tanto della situazione:: tesa in famiglia ((laughter)) “I thought it was the right thing because in my opinion the children are very much affected by the tense situation in the family” 3 (—) 4 A: ((laughter)) e lei che cosa ha pensato (–) cioè del fatto che sua moglie::: “and what did you think, I mean of the fact that your wife . . .” 5 (—) 6 C: ((rapidly and in a loud voice)) si: diciamo che avendo:: è logico che ne risentano anche loro della situazione che diciamo così si è venuta a creare: situazione che secondo me così era dovuta alla accentuazione da parte della madre così dentro hh diciamo a questa questione mentre secondo me i bambini dovrebbero essere lasciati fuori: “yes, let’s say that having it’s obvious that they too suffer a lot from the situation that we might say has arisen: a situation that in my opinion was due to the mother’s accentuating let’s say within this question whereas in my opinion the children should be left out of it” 7 A: hmm:: e quindi lei che cosa pensa del fatto di venire qui tutti insieme? “hmm and so what do you think about the fact of all coming here together?”

162

      

At the beginning of this conversational stretch the imposition of a task takes place. A’s question in 1 is aimed at eliciting information from the family members as to their willingness to undertake the therapy and as a result of A’s asking that question, whose legitimacy is not overtly challenged by anybody, the family members are assigned an obligation to state or otherwise reveal their attitude about the therapy. While the response provided by B in turn 2 does reveal B’s attitude and is immediately accepted by A as an adequate reply to his question (as is shown by A’s laughter at the beginning of turn 4 as well as by the fact that he turns to another participant, C, who has not replied yet), the lack of compliance on the part of C (apparent in the silence in 3) triggers a series of insisting moves on the part of A. So while the series of moves 1, 2, and 4 constitutes a simple M–A–S sequence, the interaction between A and C does not display an overt M–A–S articulation. Already A’s turn 4 is a reformulation of turn 1, addressed at C alone, and may therefore be considered as a first instance of insisting. However, the real problem between A and C arises only later on, after C’s turn 6. That turn is a dispreferred response to 4, as is already made apparent by the pause in 5. It is a dispreferred response in an especially radical sense. It is not an expectation-frustrating reply, nor a refusal to reply; rather, it fails to take 4 at its face value, that is, as an information-seeking question (itself a reformulation of 1). C’s turn 6 displays also features that would permit some receiver to take it as a criticism or even a reproach leveled by C at his wife B: “situazione che secondo me così era dovuta alla accentuazione da parte della madre” (“a situation that in my opinion was due to the mother’s accentuating”) and “secondo me i bambini dovrebbero essere lasciati fuori:” (“in my opinion the children should be left out of it”). How is C’s turn 6 received by A? In turn 7, after a hesitation, A reformulates his initial question again. This may count as an implicitly negative sanction of 6, showing that A doesn’t accept it as a reply to his questions in 4 and in 1. The negative assessment implicit in A’s turn 7 is only mitigated by the suggestion, due to the adverb “quindi” (“so”), which evokes the drawing of a conclusion, that 6 might turn out to be, although not itself a reply to 4 and 1, at least a premise to the real reply. It should be noted that since 7 sanctions 6 only qua expected to reply to A’s question, the features of 6 by which it attempts to affect the relationship between C and B are dropped and remain ineffective. Maybe C was trying to criticize or reproach B, but he doesn’t succeed in redefining the interactional situation as one in which he possesses authority over B or competence to judge her behavior. By reformulating his question (and therefore insisting), A shows that C has not answered it yet, and that whatever else he was trying to do was not relevant to the current conversation. The second conversational stretch (2b) begins several turns later, but still at a point of the conversational episode at which C has not provided an answer to A’s initial question. I have selected this stretch as a complement to stretch (2a) because there the insisting starting in stretch (2a) comes to an end.

       (2b)

163

1 A: e quindi lei (–) appunto (—) che cosa ha pensato quando sua moglie le ha detto che dovevate venire qui? “and so you, exactly . . . what did you think when your wife told you that you all had to come here?” 2 C: bè ovviamente visto che io (–) così (–) sinceramente (–) diciamo (–) vorrei salvare questa cosa anche se mia moglie pensa che non la pensi come lei ma io:: tengo molto ai miei figli cioè non:: (–) non non vorrei che loro diciamo così dovessero soffrire. “well obviously seeing that I, well, to be honest, let’s say I’d like to save this thing even though my wife thinks that I don’t think the same way about it as she does but I am very attached to my children, in other words I wouldn’t like them let’s say to have to suffer” 3 B: ma così stanno già soffrendo “but like this they’re already suffering” 4 A: e:: un attimo e quindi lei dice per il bene dei bambini:: “and just a moment and so you say that for the children’s sake” 5 C: diciamo così che se ci fosse un qualcosa che riuscirebbe a farci capire dove stiamo sbagliando:: ((quickly)) bè in fondo in fondo secondo me io sono sempre dell’idea e poi l’ho sempre detto a lei che forse nei nostri colloqui tra me e lei ci vorrebbe un terzo interlocutore che ovviamente ci farebbe capire:: dove stiamo sbagliando dove::: non ci rendiamo conto che ovviamente le cose non possono andare in un modo potrebbero andare in un altro::. “let’s say that if there were something that would enable us to understand where we’re going wrong well basically I think I’ve always been of the opinion and I’ve always said to her that perhaps in our conversations between me and her there should be a third person who could obviously enable us to understand where we’re going wrong where we don’t realize that obviously things can’t go a certain way they could go in another” 6 B: bè ma scusa e? “Well, excuse me!” 7 A: aspetti che sentiamo D ((turning to the couple’s daughter)) “wait a moment let’s hear what D has to say.”

In conversational stretch (2b), A insists on asking C the very same question that he has formulated at the beginning of (2a): 1 is a reformulation of that question and also 4 may be so taken. A also keeps taking C’s responses as premises to the real answer: in both 1 and 4 he uses the connective “quindi” (“so”) as if C’s turns were premises to a conclusion, implicitly urging him to draw it. The two turns issued by C, 2 and 5, are both ambiguous and it is no surprise that each receives two different responses, by A and B, respectively.

164

      

Turn 2 is taken by A as an argumentative premise to a forthcoming answer to A’s initial question and by B as a (somewhat infelicitous) expression of intention with respect to the couple’s children. Turn 5 is taken by B as reporting (falsely) C’s usual attitude towards the matters at issue and receives no direct reply from A. In both cases deviant side sequences may arise, in which C’s contributions would count as manipulating moves addressed at B and might enable C to play a further sanctioning role with respect to B’s reactions. A prevents these side sequences from arising: notice “e:: un attimo” (“and just a moment”) in 4 and “aspetti” (“wait a moment”) in 7. Thanks to the authoritative position in which he is collocated by his institutional role, his uptake of C’s turns prevails over B’s. So, 2 is actually dealt with as an argumentative premise to a forthcoming answer. As to 5, I propose to take the fact that A in turn 7 passes over to the next participant as showing he accepts 5 as that response on the part of C he had been trying to elicit all along. Let us consider C’s turn 5 and A’s turn 7 in greater detail. What makes 5 more suitable for counting as a reply to A’s initial question than any turn previously uttered by C? In turn 5, C grants (by presupposition) that he too might be wrong and declares a long-standing agreement with the idea that external help might be useful: in so doing he might be taken either as accepting to undertake the therapy or as merely reporting a usual attitude of his. Under the former interpretation, by uttering 5 he does perform the task assigned to him by turn 1 in (2a). Under the latter interpretation, he might still be said to perform that task insofar as his reported attitude entails an attitude towards therapeutic aid (and therefore family therapy). However, whether 5 actually counts as C’s reply to A’s initial question substantially depends on the reception it achieves. In fact, A’s turn 7 seems to ignore 5, but at a more attentive consideration, we may notice it displays two features closely related to its reception. First, A, by uttering the imperative “aspetti” (“wait”), stops B, who in 6 has just started protesting against what she takes to be a false report by C of his own attitudes, therefore undermining the effectiveness of her reception of 5. This favors the other possible reading according to which C, in uttering it, accepts to undertake the therapy. Second, 7 is de facto a final move and its very position enables it to play the role of a sanctioning move. If there is no longer a need for A to insist, it means that something in C’s behavior can now be counted as constituting the performance, whether correct or incorrect, of the task assigned to C by A’s initial (and reiterated) manipulating move. From the discussion of example (2), some morals about insisting as a sequential phenomenon can be drawn. Insisting appears to involve: (i) the attempt to get the manipulating function of a certain move to be recognized by the addressee, by reiterating and reformulating it; (ii) the implicit negative sanction of those responses that do not constitute yet the performance of the action the manipulating move aims at eliciting:

      

165

whenever the manipulating move is reiterated, this suggests that the assigned task either has not been recognized as assigned or, anyway, has not been fulfilled, or not fulfilled completely; (iii) a final sanctioning move that implicitly identifies the move it sanctions as the performance, whether correct or incorrect, of the elicited action. So in insisting, even if the sequence in which it occurs does not consist of three moves, the M–A–S structure is at work and helps us realize what is going on. It is the narrative schema which enables us to recognize the borders and the steps of the insisting sequence, however many side sequences or attempts at them are embedded between the initial manipulating move and the move that, thanks to the sanction it receives, eventually counts as filling in the Action slot. It may also be noted that by insisting, a speaker attempts at giving the speech act sequence a tendentially rigid M–A–S structure, one that does not allow for multiple applications of the narrative schema. This follows from the fact that insisting just consists of trying to get the interlocutor to accept a certain move as a manipulating move and therefore perform the task it imposes. Receptions of speech acts belonging to the sequence, that would exploit or enhance applications of the narrative schema other than the preferred one, have to be neutralized by the speaker who wants to achieve this goal. Since it is natural to expect that as soon as the manipulating moves manage to elicit the action, a sanction will follow, in insisting the mere cessation of the production of reiterated manipulating moves may count as a final sanction and help identify the move that counts as filling in the Action slot.

6. Some Conclusions In this paper I have contended that sequentiality is a fundamental dimension of speech acts. First of all, no illocutionary act can produce its conventional effect apart from its reception, which is manifested and even constituted by the response it is given. Moreover, there seems to exist some general form of our understanding of sequences of events which contributes to the structuring of speech act sequences and therefore, indirectly, to the production of conventional effects by a given individual speech act. My view presupposes an interactional conception of communication according to which communication is made possible by the participants’ affecting one another by the conventional effects of their speech acts. I have attempted to show that the implicit or by-default agreement about what each participant has done, required for validating conventional effects, does not involve any petitio principii sending us back to cognitive representations of actions, because the participants’ way of reacting to each other’s moves is enough for constituting or

166

      

failing to constitute the required agreement. I have thus claimed that in speech act sequencing, action itself plays a more basic role than cognition. However, this does not rule out other kinds of cognitive contributions to the structuring and the effectiveness of verbal interaction (or of social interaction in general). I have suggested that whenever a move in a speech act sequence is focused upon as constituting an action, it is also framed between a preceding, eliciting, or manipulating move and a subsequent sanctioning one. I have attempted to show that the articulation of sequences of events into the three steps of Manipulation, Action, and Sanction, proposed by narrative semiotics, can be applied to speech act sequences quite successfully and contribute to descriptions of both individual speech acts and sequential structure. I have outlined two main kinds of application of the narrative schema. On the one hand, M–A–S sequences can be applied wherever we want to explore the sequential relationships in which a speech act enters or may be considered as entering, in which case multiple application is possible and even desirable in order to shed light on interpretive potentialities and how participants exploit them or negotiate about them. On the other hand, it is sometimes relevant to detect one preferred application of the narrative schema and univocally identify the moves playing the role of Manipulation, Action, and Sanction, respectively. Apart from those cases in which the salience of one move makes it a suitable candidate for filling in the Action slot, it is Sanction which is most easily recognizable, because of its closing function. Even when sequences do not display any overt M–A–S articulation, as in the case of insisting, the closing function of one move may support the identification of the move it responds to as the Action, that is, the performance of the task assigned to the relevant agent by the initial manipulating move. Fruitful application of an analytic device is no proof that it actually belongs to the participants’ competence, rather than merely to the analyst’s metalinguistic equipment. Since, however, in the kind of speech act analysis outlined above the analyst’s way of dealing with speech acts is just an extension or shifting of that of the participants (the analyst, moreover, must rely on the way participants deal with each other’s speech acts in order to tell what counts as having been done by whom), it appears reasonable to suppose that if a given tool proves fruitful, it is because the participants use it themselves, although most often tacitly. Thus my analysis of speech act sequences supports also the suggestion, inspired by narrative semiotics, that we make sense of what happens or is done around us, and of speech act sequences in particular, thanks to the general form provided by the narrative schema.

9 Two Conceptions of Rationality in Grice’s Theory of Implicature 1. Implicature Paul Grice launched the word “implicature” for use within his theory of speaker meaning, in order to account for aspects of speaker meaning not contributing to the truth-conditions of the sentence uttered by the speaker. Surprisingly enough, he never explicitly defines “implicature”; he provides examples, but gives a definition only to one kind of implicature, the one in which he is most interested: conversational implicature. This is his definition: A man who, by saying (or making as if to say) that p has implicated that q, may be said to have conversationally implicated that q, provided that (1) he is to be presumed to be observing the conversational maxims, or at least the Cooperative Principle; (2) the supposition that he is aware that, or thinks that, q is required in order to make his saying or making as if to say p consistent with this presumption; and (3) the speaker thinks (and would expect the hearer to think that the speaker thinks) that it is within the competence of the hearer to work out, or grasp intuitively, that the supposition mentioned in (2) is required. (Grice 1989, 30)

The definition presupposes the following formulation of 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 1989, 26)

While such a principle has the form of a precept or a rule (imperative mood, second person), its use in the theory of conversational implicature is never directly

Originally published in Rationality of Belief and Action. Proceedings of the International Philosophical Conference Held in Rijeka, May 27–8, 2004, ed. Elvio Baccarini and Snježana Prijić-Samaržija, 233–47. Rijeka: University of Rjieka, Faculty of Arts and Sciences and Croatian Society for Analytic Philosophy, 2006.

Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0010

168

      ’  

normative. It is the assumption, or presumption, that the speaker is observing the Cooperative Principle and not the real observance of it, nor the (possible) obligation, or commitment, of the speaker to observe it, that comes into play in the definition of conversational implicature quoted above. I have argued elsewhere that the Cooperative Principle is no actual rule but a presumption that receivers make (and that it is rational for them to make) for the sake of giving as full an interpretation as possible of what they are told (Sbisà 2001b [Essay 4]), quite apart or even beyond any independent evidence regarding the conformity of the communicative behavior of the speaker to the principle. Also the conversational maxims that articulate the Cooperative Principle (as applying to assertions or other information-oriented speech acts) with respect to the four classic categories of Quantity, Quality, Relation, and Manner are cast in imperative form, but their use in utterance interpretation goes mostly through the assumptions of the receiver as regards their observance or open violation. In Grice’s definition of conversational implicature, clause (1) apparently requires to be satisfied by the speaker, but at a second glance we realize that if the speaker is merely “to be presumed” to observe the conversational maxims, the requirement is actually posed on the receiver, who should have good reasons for presuming of the speaker that he or she is observing the conversational maxims. Since, on a default view of what is “normal” in conversational interaction, a good reason for presuming that the speaker is conforming to the conversational maxims is already provided by the absence of perceivable alarm signals, situations in which the receiver does not notice any indication that “normality” is suspended are in general situations in which the speaker is to be presumed to be observing the conversational maxims. Clause (2) spells out a condition that the speaker should satisfy (his thinking that q), if his utterance (his saying or making as if to say p) is to be consistent with his observance of the conversational maxims. Clause (3) is again to be satisfied by the speaker and is concerned with his beliefs or more generally attitudes towards the hearer. Saul (2002a) has pointed out that the three clauses make the applicability of the definition depend on both hearer and speaker, so that unmeant implicatures and cases in which the speaker attempts to conversationally implicate something but fails to do so are no conversational implicatures. The distinction between saying that p and making as if to say that p, which features in the definition, is meant to make the definition applicable to both the main brands of Gricean conversational implicature: those implicatures that supplement the content of the utterance in order to make the utterance count as a cooperative move in the current conversation, and those that are triggered by the “flouting” (or explicit violation) of a conversational maxim and consist of a complete re-reading of the utterance (again, as a cooperative move in the current conversation). This distinction is a weak point in Grice’s theory of conversational implicature and criticism of it has opened the way to alternative, competitive readings of the alleged flouting of the conversational maxims and therefore new pragmatic analyses

      ’  

169

of rhetorical figures (Carston 2002; Sperber and Wilson 1995). I will not be concerned with this distinction here; not much of what I try to say hinges on it. The definition of conversational implicature sets conditions for an implicature’s being conversational, provided that the speaker is implicating something. From it, we indirectly gather that an implicature consists for Grice in an assumption q, of which a speaker is (at least) aware and that the speaker communicates (or makes available: see Saul 2002a, 245) to the hearer not by saying that q, but by saying or making as if to say that p. We may then say (with a bit of imprecision, but it has become common usage in pragmatics) that q is an implicature of p.¹ In consideration of the function that implicatures (both “conventional” and “conversational”) are designed to play in the Gricean analysis of language and communication (Grice 1989, 3–21), we may say that implicatures are for Grice non-truth-functional inferences: sentences that may be derived from the fact that a certain sentence has been uttered, without their truth-value affecting the truthvalue of that sentence or depending on it. Clearly, if the inferential relation is not guaranteed by truth-value preservation as in deductive reasoning, it must rely on other grounds. For Gricean conventional implicature, such grounds are given by convention (specifically, conventions regarding the performance of “non-central” speech acts such as objecting, concluding, and the like), while for conversational implicature, they are given through reference to the Cooperative Principle and its maxims, as specified in clause (2) in the definition cited above. The most frequent criticism of Grice’s conception of conversational implicature is that the Cooperative Principle does not always hold. It does not hold in all cultures (for an early criticism along these lines, see Ochs 1976), it does not hold in all contexts (for an early criticism along these lines, see Lakoff 1973) or for all aims (how far can cooperation go in the context of a conflict?). I believe that, if understood as a presumption that receivers make (and that it is rational for them to make) for the sake of giving as full an interpretation as possible of what they are told, the Cooperative Principle may enjoy universal applicability. But I will not be concerned here with the issue of the universality or cultural relativity of the Cooperative Principle. I will be concerned with a distinct, albeit related, issue: the alleged rationality of implicature and particularly of Gricean conversational implicature.

2. Is Implicature Rational? Both post-Gricean and neo-Gricean approaches to linguistic communication, while accepting some version of the Cooperative Principle, have supplemented, ¹ In Grice’s usage, “implicature” is the fact that the speaker implicates something, while what is implicated is called “implicatum.” Usage has shifted and “implicature” is more and more often used to indicate also or only what is implicated. The latter is the standard usage in Relevance Theory, where both implicatures and explicatures are contents as opposed to acts.

170

      ’  

criticized, and modified the Gricean conception of implicature in various ways. Relevance Theory, the post-Gricean approach par excellence, has claimed that our understanding of linguistic utterances is governed by the Relevance Principle, originally inspired by Grice’s maxim of Relation “Be relevant,” which has then reached the following two-fold formulation: (i) Human cognition tends to be geared to the maximization of relevance. (ii) Every act of ostensive communication communicates a presumption of its own optimal relevance. (Sperber and Wilson 1995, 260) According to Relevance Theory, utterance processing always comprises an inferential component, governed by the cognitively built-in presumption of optimal relevance. Neo-Griceans (see in particular Levinson 2000) have claimed that the Relevance Principle does not suffice to explain all kinds of implicit communication and has to be replaced by two or more principles inspired by other conversational maxims of Grice’s, each with its distinct function and sphere of influence. In Relevance Theory utterance processing always involves contextual assumptions (shared or mutually accessible knowledge), while neo-Griceans emphasize Grice’s conception of Generalized Conversational Implicature, according to which some implicatures at least are communicated thanks to reference to the Cooperative Principle and its maxims, but without direct involvement of further cognitive material from the current context. Whatever the differences, on all versions implicature is supposed to be a “rational” matter. For Grice, it is conversational implicature in particular which is rational. According to him, it is rational to follow the Cooperative Principle and assume it is followed by others, therefore communicating and understanding conversational implicatures. Moreover, conversational implicature is “calculable,” and calculations are quite obviously a rational matter. It is harder to tell whether or to what extent Grice would also count conventional implicature as something rational. Grice’s “rational” approach to conversational implicature has influenced his followers and critics. So, it is rational speakers and hearers (or speakers and hearers qua rational) that, according to Relevance Theory, conform to the Relevance Principle to yield inferential interpretations of explicit and implicit utterance meaning (including implicatures). Also in Levinson’s analysis of Generalized Conversational Implicatures, these depend on the application of rational principles and rules to utterance processing. I will take it for granted that Gricean conversational implicature is to be considered a rational matter and explore why. The exploration will show that two distinct conceptions of rationality are at play in Grice’s thought and that only one of these has been picked up by post- and neo-Gricean approaches.

      ’  

171

3. Grice on the Rationality of Conversational Implicature Two distinct hypotheses about why conversational implicature is rational can be extracted from Grice’s work. (i) Conversational implicature is rational insofar as it is part of rational communicative behavior to maximize understanding (gather as much information as possible, etc.). (ii) Conversational implicature is rational because it is calculable. Hypothesis (i) traces the rationality of conversational implicature back to the rationality of the Cooperative Principle. Grice (1989, 29–30) says that he would like to think of the Cooperative Principle not merely as something that we do in fact follow but as something that it is reasonable for us to follow (hence, we may say, something “rational”) and outlines two possible arguments in support of this idea. According to one line of reasoning, conversational cooperation can be described as quasi-contractual: each interlocutor’s response to a move of the other interlocutor enhances mutual expectations of cooperativity, makes them more specific, and gives to each interlocutor reasons to believe that his or her interlocutor is cooperative. It is hard to see, though, why this should prove the observance of the Cooperative Principle to be rational. Moreover, this description of conversational cooperation seems to fit only some types of exchanges. The second line of reasoning seems to work better: according to it, the rationality of conversational implicature depends on its capacity of maximizing understanding (and we may say, it is in fact rational to maximize understanding or to squeeze as much information as possible from one’s interlocutor’s words). Surprisingly enough, Grice drops this line of reasoning too, declaring himself uncertain about its soundness as well as about some of the concepts involved in it. Hypothesis (ii) traces the rationality of conversational implicature back to its calculability. After all, calculation is a rational activity. Calculability involves, according to Grice, the availability of an inferential path—a series of inferential steps—leading from the utterance of a sentence to its implicature. This inferential path is outlined by Grice as follows: He has said that p; there is no reason to suppose that he is not observing the maxims, or at least the Cooperative Principle; he could not be doing this unless he thought that q; he knows (and knows that I know that he knows) that I can see that the supposition that he thinks that q is required; he has done nothing to stop me thinking that q; he intends me to think, or is at least willing to allow me to think, that q; and so he has implicated that q. (Grice 1989, 31)

172

      ’  

Hypothesis (ii) is, in my opinion, quite plausible both as a claim about the rationality of conversational implicature and as an interpretive claim about Grice’s theory of it. But before accepting it and drawing further consequences from it, we should stop to clarify the nature of the calculability requirement, which we have so far left largely indeterminate.

4. The Nature of the Calculability Requirement While the form of the inferential path proposed by Grice for conversational implicature is clear enough for our current aims, its nature is problematic and this might interfere with the project of interpreting the rationality of conversational implicature as based on its calculability. We may want to know whether the Gricean inferential path for the calculation of a conversational implicature is meant to have the nature of an actual psychological process (a psychological necessary condition for an utterance to carry a conversational implicature) or that of some kind of abstract, logical model giving the grounds for the recognition of the implicature as conversational. Are conversational implicatures characterized by the fact that the hearer, in order to understand the implicature, actually goes through a certain (Gricean or otherwise defined) inferential path? Must the speaker actually go through the same inferential path while planning the implicature-carrying utterance? Are only those implicatures “rational” that are actually calculated by the hearer and planned by the speaker through a certain kind of inferential path? We may consider these questions as applying to actual or ideal hearers. In the case of actual hearers, it is clear enough that a hearer might fail to calculate the implicature and, conversely, not everything a hearer infers from an utterance is an implicature. Actual hearers make every kind of mistakes.² But of course, it must be possible to determine what implicature a certain utterance (in a certain context) should be recognized to carry, independently of whether actual hearers manage to derive the implicature. Were it not so, it would be impossible to distinguish correct from incorrect implicature comprehension. If, instead, we take into consideration only those hearers who succeed in grasping the implicature, we implicitly introduce a normative criterion. Hearers who understand well approximate ideal, “rational” hearers. So claims about how conversational implicature is derived are in fact claims about how hearers approximating ideal rational hearers derive it, and therefore, are deprived of any empirical character. They are claims about how the implicature should be derived.

² A research project on textbook comprehension at which I am still working has yielded rich data as regards comprehension failures involving presuppositions and implicatures [for the results, see Sbisà 2007b; 2007c, 160–90].

      ’  

173

But after all, as Saul (2002a) has correctly noticed, the actual working out of the implicature by the hearer is not included in the conditions for conversational implicature laid down by Grice (1989, 30–1). A hearer might grasp the implicature without going through exactly that inferential path or even without any inferential path at all. According to Grice himself (1989, 31), conversational implicature may well be grasped intuitively: calculability is not actual calculation. For an implicature to be calculable and therefore conversational, Grice only requires that the intuition be “replaceable by an argument.” As to the speaker, observation of everyday life tells us that actual speakers do not in general plan their implicature-carrying utterances by consciously anticipating the inferential path of the hearer. It is often quite legitimate to attribute to a speaker the intention to implicate something whose relation to the utterance is characteristic of conversational implicature, even in absence of evidence to the effect that the speaker has gone through the corresponding inferential path. Actual calculations by the speaker cannot therefore be a realistic requirement for an implicature to be conversational (and calculable) and if we make the rationality of implicature depend on its calculability, they cannot be a requirement for its rationality either. It could be objected that Grice, while clearly not concerned with the psychological processes on behalf of the hearer, requires of the speaker—in clause (3) of the definition of conversational implicature quoted in Section 1 above—that he or she thinks that the hearer is able to work out the implicature. It seems therefore that the speaker must at least be aware of the inferential path leading from the implicature-carrying utterance to what it implicates. If in order for an implicature to be conversational, the speaker must think that the hearer is able to understand that the supposition that the speaker thinks that q is required, shouldn’t the speaker be also aware that the supposition that he or she thinks that q is required? If an individual, in order to evaluate another individual’s ability to grasp an assumption, must represent this assumption to him or herself, the answer must be positive. But it is not so clear that this must be so. It might be enough that the speaker be willing to grant to the hearer the ability to grasp whatever assumption is needed for a thorough understanding of the utterance. Moreover, we might take it that the requirement posed by Grice onto the speaker need not be satisfied by his or her having an occurrent thought about the hearer’s abilities, but by a general attitude towards him or her. The speaker might, so to say, deserve to be attributed the relevant thought because his or her general attitude towards the hearer is of the kind one has towards a fellow thinking being, typically able to grasp assumptions of the kind of the one which is required on that specific occasion. So, we may escape commitment to counterintuitive claims about the psychological implementation of calculations in the speaker’s mind. Of course, it may also be said (as is done e.g. by Relevance Theory) that utterance processing, both in production and in reception, consists of sub-personal

174

      ’  

calculations, of which speaker and hearer need not be aware and that they do not control at all. But sub-personal calculations aren’t such clear a hallmark of rationality as personal-level, conscious ones. So, if in order to save the psychological implementation of implicature calculability, we make it sub-personal, the support to the rationality of conversational implicature that may come from reference to calculability is weakened considerably. Rather, I believe that in conversational implicature, the unavoidability of going through reference to the speaker’s subjectivity (expressed by Grice in clause (3) of his definition) depends on something subtler than an inclination to apply a psychological reading of calculability to the speaker. Conversationally implicating something, as well as understanding a conversational implicature, is a behavior that requires consideration of one’s interlocutor as a thinking being capable of and keen to cooperation, who in turn considers his or her interlocutor as a thinking being capable of and keen to cooperation. This reading of Grice’s condition (3) for conversational implicature is fully compatible with supporting the rationality of conversational implicature by reference to calculability. In fact, what emerges from the above exploration of the nature of the calculability requirement is that the calculability of an implicature is important not because it defines the way in which the implicature is actually planned and understood, but because it guarantees for the rationality of the implicature, testifying that the attribution of that implicature to the speaker who has issued a certain utterance can be supported by argument. We can thus take as confirmed that conversational implicature is rational because it is calculable.³

5. Two Ideas of Rationality The two hypotheses about why conversational implicature can be deemed to be rational, which we have extracted from Grice’s work in Section 3 above, make implicit reference to two distinct conceptions of rationality and so do also the two conceptions of calculability (as requiring actual psychological implementation or as mere replaceability of the grasping of a conversational implicature by an argument) that we have discussed in Section 4. I call these two conceptions of rationality “instrumental rationality” and “argumentative rationality.” I will try to show that our second hypothesis about the rationality of conversational ³ Wayne Davis (1998) argues that conversational implicatures are not in fact calculable. But, firstly, here I am claiming that they are rational (according to Grice) because they are calculable (according to Grice). Secondly, the interpretation of calculability I give might escape Davis’s argument anyway, since in my view, the step in the inferential path he calls the “determinacy requirement” (his main reason for rejecting the calculability of implicature) is no longer essential to the role the inferential path has to play.

      ’  

175

implicature (that conversational implicature—defined along the Gricean lines—is to be considered rational because it is calculable), and particularly the reading of it that envisages calculability as replaceability by an argument, rely upon the argumentative conception of rationality. Whoever accepts hypothesis (ii) as true should prefer the argumentative conception of rationality over the instrumental one. Let us illustrate the contrast between the two conceptions and their association with our two hypotheses. For instrumental rationality, I adopt the following definition: (IR) A course of behavior is rational if it is characterized by the agent’s nonaccidental use of effective means, or of means he or she believes to be effective, for achieving his or her goals.

This conception of rationality, widespread and pervasive (but perhaps it is a whole family of conceptions), is typically concerned with means–end relations. It is instrumental rationality that Grice relies upon in his attempt to qualify implicature as rational by establishing the rationality of the Cooperative Principle, as in our hypothesis (i) about why conversational implicature is rational. Grice attempted to show that the Cooperative Principle is something that it is rational for speakers to follow by considering their interest in participating in talk exchanges that will be profitable only on the assumption that they are conducted in general accordance with the Cooperative Principle (1989, 30). In such a perspective, the Cooperative Principle is assumed to hold, and conversational implicatures are retrieved, because it is profitable to do so: because, for example, this enables the receiver to squeeze more information from the speaker’s utterance, and when that does not yield more information about the world, it yields at least more information about the intentions, beliefs, and other attitudes of the speaker him or herself. Since the availability of the conversational implicature mode of communication depends on the talk exchange being conducted in general accordance with the Cooperative Principle, the speaker’s choice to communicate through conversational implicature may also be taken to inherit the same kind of rationality as the adoption of the Cooperative Principle. As already said above (Section 3), Grice abandoned the project to support the rationality of conversational implicature on the basis of means–end relations. But post-Gricean and neo-Gricean approaches have read the rationality of conversational implicature as instrumental and have retained this feature in their reformulations of the Gricean framework. For Relevance Theory, human cognition tends not merely to sufficient results, but to optimal ones (Sperber and Wilson 1995, 266) and these have to be achieved by optimizing relevance, which is itself a balance of costs and benefits and therefore a matter of instrumental rationality. For Generalized Conversational Implicature Theory, language use suffers from a “bottleneck” problem: we most often need to convey more information

176

      ’  

in a given stretch of time than the linear production/reception of speech allows for (Levinson 2000, 6); and the heuristics that generate generalized conversational implicatures provide us with the means for doing so. When implicature is viewed as instrumentally rational, it is natural enough to consider it as a means for the optimization of communication (in various possible senses) and to this aim it is quite reasonable, or perhaps even mandatory, to assume that it has to do with how communication actually functions, that is, with the processes actually occurring in the minds of speaker and hearer. So instrumental rationality is fully compatible with, or even requires, a view of the calculability of conversational implicatures as psychologically implemented in actual calculations. It is not a problem for instrumental rationality that there is often no introspective evidence of the relevant calculations, because these may well be subconscious. Under the instrumental conception, sub-conscious mental processes can nevertheless count as rational, if they succeed in providing effective and economic means to the given ends. Argumentative rationality does not focus on means–end relations, but stresses the connection of rationality with argumentative reasoning, that is, reasoning aimed to provide support to the acceptance of a belief or the making of a decision. Such an idea of rationality plays a remarkable role in Grice’s posthumous writings, such as Lectures on Value (Grice 1991) and Aspects of Reason (Grice 2001). The following definition of it can be extracted from Lectures on Value: (AR) Rationality is a concern that one’s moves are justified and a capacity (to some degree) to give effect to that concern. (Grice 1991, 82–3)

In this definition, an important role is played by reference to justification. The argumentative character of rationality is one and the same thing with its function as a source of justification. The relation between argumentation and justification, as I view it, is double: on the one hand, justification is a practice of citing one or more assertions in support a certain conclusion or decision, which is the typical or even the basic move in arguments; on the other hand, an argument is good insofar as its steps and their connections justify the conclusion. The connection between argumentation and justification explains, in turn, why argumentative rationality is connected to value: justification is a matter of value and being justified is one and the same thing as being of value. It is interesting at this point to recall that argumentative rationality plays an important role in Grice’s defense of absolute value (1991, 80–91; 114–20; cf. Sbisà 2001b [Essay 4]). There is absolute value, according to Grice, insofar as there are essentially rational beings, whom he calls “persons,” who have absolute value and can attach absolute value to what they evaluate. Humans turn themselves into such essentially rational beings when they consider rationality, which they possess contingently and accidentally, as their essential property. What I call argumentative rationality comes into the picture because a rational being is for Grice a being

      ’  

177

who seeks justifications for his or her choices. A person, or essentially rational being, is a kind of being who does not behave like this only accidentally or on occasion, but in whom seeking for justifications is essential: such a being, therefore, essentially possesses argumentative rationality. When considering argumentative rationality as a contingent property of humans, we must say that somebody who refuses to provide a justification for a move of his or her, or fails to provide appropriate reasons or evidence in support of one of his or her claims, is not, on that occasion, rational. Argumentative rationality, by the way, does admit of borderline cases, in which for example there is the concern for giving justifications but the justification provided is under some respect flawed (Grice 2001, 6–14). But we may also, with Grice (1991), consider humans as essentially rational, even in those circumstances or stages of development at which argumentative rationality is not contingently present. We thus recognize that the justification of one’s moves remains a task for every human being who wants to take him or herself seriously as a person. The connection between argumentative rationality and the calculability of conversational implicature is never directly focused upon by Grice. But it should be recalled that he shows a preference for explaining implicatures, whenever possible, as conversational as opposed to conventional. He even adds that the choice of dealing with an implicature as conventional⁴ is in need of “special justification” (1989, 39). Indeed, only conversational implicature is calculable and therefore inherently rational in the argumentative sense. Explaining a certain part of the meaning conveyed by an utterance as a conversational implicature, rather than as part of what is said or conventionally implicated, requires that there be an inferential path leading to that implicature and thus an argument in support of it. If argumentative rationality is a value (as we have seen, Grice collocates it at the very root of values), calculability makes conversational implicature preferable as an analyst’s explanatory strategy for the implicit communication of a certain thought. This explanatory strategy is also the most charitable one with respect to the rationality of speakers and hearers, since it assumes that the fact that they communicate and understand a certain thought has an (argumentative) justification. The Modified Occam’s Razor, the well-known and debated principle according to which senses should not be multiplied beyond necessity (Grice 1989, 47), turns therefore out to be not merely a matter of achieving theoretical economy, but of preferring, whenever possible, argumentatively justifiable meaning assignments. The request to give “special justification” to analyses of implicit meaning considering a certain implicature as conventional suggests that all interpretive, analytic moves are according to Grice in need of some justification. The advantage of conversational implicature is that its attribution is supported by an argumentative path, while conventional implicature does not by itself provide a justification of the inferential connection it consists of. ⁴ [Because conventionalized.]

178

      ’  

The reading of calculability as replaceability by an argument is well fit in the framework of argumentative rationality. As we have already remarked, it is quite counterintuitive to maintain that conversational implicatures, in order to be such, must be consciously planned and understood by speaker and hearer through the inferential path outlined by Grice (or a similar one). In fact, approaches maintaining that conversational implicature must be actually worked out through the implementation of such an inferential path admit that this happens at the subpersonal level of unconscious cognitive workings. But argumentative rationality, being connected to justification, requires personal-level involvement. Thus a conception of calculability as necessarily implemented in the minds of the interlocutors is not suitable for the framework of argumentative rationality, in which alone it makes sense to claim that conversational implicature is rational because it is calculable. Under the argumentative conception of rationality, instead, calculability can be conceived as availability of an argument in support of the assignment of the implicature to the speaker’s utterance. What the rationality of conversational implicature requires of speaker and hearer is only that they should be willing to justify their understanding of the implicature and capable (to a certain degree) to provide such justifications by replacing their intuitive grasping by some more or less complete version of the relevant inferential path.

6. Concluding Remarks Our exploration of the rationality of Gricean conversational implicature leaves several questions open. First of all, is the contrast between instrumental and argumentative rationality genuine? Are these two conceptions of rationality incompatible or may they be reconciled? Both conceptions of rationality are represented in Grice’s work. We have already cited his argumentation-based definition of rationality and shown that his theory of conversational implicature presupposes a commitment to argumentative rationality. But in his philosophical psychology (1989, 283–303), as well as in his analysis of practical reasoning (2001), Grice makes use of the received folkpsychological connections among belief, desire, and behavior and seems therefore to consider the rationality of behavior as instrumental: an agent behaves rationally if he or she does what he or she believes to help the achievement of what he or she desires. Also in his theory of value (1991), when he introduces the demand for absolute value (to which his argument for absolute value attempts to respond), Grice seems to consider it as a “rational” demand just because answering it would be instrumental to the satisfaction of the human aim of validating acceptances and attitudes objectively. Sure, the concern for such objective validation is in turn integral to argumentative rationality, but this only shows that in Grice’s thought, argumentative and instrumental rationality can but coexist.

      ’  

179

At least insofar as the explanation of behavior and the analysis of practical reasoning are concerned, Grice seems to suggest that the instrumental facet of rationality should be dealt with as a sub-case of the argumentative. In fact, when an agent justifies his or her behavior as a means for achieving a certain goal, the instrumental rationality of the behavior coincides with the argumentative rationality of the subject. But not all possible justifications or validations of acceptances and attitudes are of this form. So, even if instrumental rationality were a sub-case of argumentative rationality, the distinction survives. That such a distinction should be drawn, however, is strongly suggested by the comparative consideration of the two conceptions. Instrumental rationality, as already noted above, does not presuppose self-awareness, while concern for justification does. Instrumental rationality bears no relationship to values, or rather, has effectiveness as its only value, so that it is of value only derivatively, while argumentative rationality may be claimed to be itself a value. Finally, instrumental rationality cannot raise questions about goals, except insofar as they are the means for other goals. That is a main source of dissatisfaction with instrumental rationality, which argumentative rationality may prove able to cope with. So we can safely enough admit that, within Grice’s thought but also beyond it, a distinction between argumentative and instrumental rationality is not otiose. I turn therefore to further and more central queries. Is the outlined distinction actually relevant to the theory of linguistic communication and to what extent? What are its consequences? Instrumental rationality inspires the view of the calculability of conversational implicatures as involving actual psychological implementation. In this perspective, conversational implicature is seen as the effect of actual efficient processing. Inferential paths are taken to spell out the form of the causal chains through which speaker and hearer come to utterance production and understanding. Argumentative rationality inspires the view of conversational implicature as rational just because calculable. According to this view, conversational implicature is meaning made available by the speaker to the (argumentatively rational) hearer, that the hearer should understand. The point of spelling out inferential paths is just offering argumentative support to the attribution of implicatures, but these need not be actually calculated through them. We face then, not so much a theoretical dispute or competition between Grice’s philosophical project and the post- or neo-Gricean cognitive projects, but two kinds of project with different aims and means (as Saul has remarked in her 2002b, 371). Post- or neo-Gricean cognitive projects cannot replace the kind of philosophical consideration of linguistic communication offered by Grice. Rather, the two kinds of project are best seen as complementary. The distinction between instrumental and argumentative rationality we have been discussing yields some suggestions for both. On the one side, the normative features of the perspective inspired by argumentative rationality provide a powerful means to the analysis of discourse. In

180

      ’  

conversational practice as well as the practice of reading, it is often difficult to distinguish implicit meaning from mere psychological associations or fanciful assumptions of the hearer that the hearer projects onto the speaker’s utterances. When non-psychologically understood, the Gricean calculability requirement offers us a criterion at least insofar as conversational implicature is concerned, and in general, a model for the monitoring of discourse comprehension. Conversational implicature attribution must be supportable by argument, and by an argument following an inferential path of a certain kind. Likewise, also other kinds of meaning that are made available but not explicitly formulated should be attributed to an utterance or text only with some justification, possibly spelled out in the form of a motivated inferential connection. On the other side, in cognitively oriented approaches to linguistic communication, it should no longer be taken for granted that the processes actually leading to comprehension mirror the rational justifications we may give for our understanding of an utterance or text. Research on utterance processing might need to free itself from the model originally offered by the reflection on the inferential paths that justify conversational implicature attribution. Efficient processing may not be isomorphic with what is and remains, in the perspective of the reasoning subject, good argument.⁵

⁵ This research has been made possible by a joint grant by the Ministero dell’Istruzione, dell’Università e della Ricerca and by the University of Trieste (national project: “Representing and Reasoning,” coordinated by Carlo Penco, University of Genova; local project: “Assertion, Inference, Justification,” coordinated by myself). I thank the audience at the International Conference “Rationality in Belief and Action” in Rijeka, May 27–8, 2004 for comments and questions. I had formerly presented papers on related topics on other occasions, among which the 10th Meeting of the Società di Filosofia del Linguaggio (Rimini, September 19–21, 2003) and “Mind and Language—Seven Symposia” (Bologna, October 15–18, 2003); I thank those audiences too.

10 How to Read Austin 1. Four Assumptions to Be Disputed In this paper I consider what contributions are offered to pragmatics (in a broad sense, ranging from philosophy to interdisciplinary research into language use) by John L. Austin’s famous booklet How to Do Things with Words (1975) (henceforth in this paper: HTW). It might seem the question is otiose: everybody who has ever been concerned with pragmatics knows that HTW proposes first the performative/constative distinction, then the distinction between meaning and force and the distinction between illocutionary and perlocutionary acts. But if we reflect on those distinctions, it may appear that it is not so clear what their point is. A lot of literature has been produced for decades on force vs meaning or on perlocution vs illocution, but has not shed final light on the job these distinctions were originally designed for. It is also uncertain what relationship there is in HTW between those distinctions and the performative/constative distinction, from which they appear to stem. And if we even superficially consider Austin’s claims, terminology and definitions, as distinct from subsequent developments of speech act theory, we easily realize that they have hardly ever been put to use either in theoretical debates or in analyses of discourse and conversation. Why is this so? First of all, HTW is a misleading book, the structure of which is easily misunderstood. There have been early misreadings of it, which have been extremely influential. Moreover, Austin’s original proposal has often been conflated with subsequent versions of speech act theory, that of John Searle in particular. I believe therefore that there is still room for a reassessment of the contributions provided by HTW to pragmatics; obviously, this reassessment requires a more careful reading of the book. As a contribution to such a re-reading, I will discuss four assumptions belonging to the “received reading” of HTW: (i) that HTW was written in pursuit of performatives; (ii) that for Austin, performing an illocutionary act amounts to producing the uptake of the meaning and force of the utterance;

Originally published in Pragmatics 17 (3): 461–73. Amsterdam: John Benjamins, 2007.

Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0011

182

   

(iii) that for Austin, performing a perlocutionary act amounts to using language to produce consequences; (iv) and finally, that Austin’s theory unduly excludes “non-serious” speech acts. These assumptions are incorrect and prevent us from understanding the actual contributions provided by the volume.

2. In Pursuit of Performatives? Let us start with the first assumption: that HTW was written “in pursuit of ” performatives. People have often thought that in HTW, Austin proposes the performative/ constative distinction, which he subsequently abandons in favor of the less clearcut locutionary–illocutionary–perlocutionary distinction. For example, Max Black writes in his review of the first edition (1962) of HTW: “The late John Austin’s William James Lectures might well have borne the subtitle ‘In Pursuit of a Vanishing Distinction’ ” (1963, 217). Even a supporter of Austin’s conclusion that the distinction does not hold because assertions are performatives too (hence, illocutionary acts), like L. W. Forguson, seems to assume that Austin conducted his inquiry in pursuit of the performative/constative distinction and then changed his mind (Forguson 1966). But there is no trace of a change of mind in the various sets of notes forming the manuscript.¹ The earliest set of notes (1951) already contains the locutionary–illocutionary–perlocutionary distinction and considers assertions as illocutionary acts. Moreover, some early preparatory notes hint at “happy go-lucky” uses of performative utterances, which Austin disapproves of and which are characterized by an exclusive opposition between having meaning and performing an action. We may say that in HTW, from the very start the performative/constative distinction plays an instrumental role. The analysis of performative utterances provides a first approach to actions performed in speech and therefore to the kinds of things that are done by illocutionary acts. Constatives are just a straw man, to be replaced with an analysis of assertion as an illocutionary act (such as the one outlined further on in the volume, in Lecture 11). Confirmation of this reading is scattered throughout the initial lectures, in hints that make sense only in the light of the main thesis which will be revealed later in the book, i.e., that all speech should be considered as action. ¹ The manuscript from which the printed text of Austin’s William James Lectures was drawn is conserved in Oxford at the Bodleian Library. It comprises various sets of notes written from 1950 to 1955, including those used by Austin for lecturing at Harvard in 1955. I consulted it several times (e.g., while collaborating with James O. Urmson on the second revised edition of the volume) and most recently in 2005.

   

183

So, HTW appears as a complex argument in support of the claim that all speech should be considered as action and, more specifically, that speech can be described as the performing of actions of the same kind as those performed by means of performative utterances. This complex argument has the form of a proof by contradiction: the thesis proposed at the beginning is the opposite of the intended one and its refutation serves as a proof of the intended thesis. To be more schematic: in HTW we have a Hypothesis (P) “Some utterances are performative,” a Thesis (A) “All of our utterances perform actions of the same kind as those performed by performative utterances” and a proof by contradiction, which goes approximately like this: Suppose that A does not hold (not A): Since some utterances are performative (P), it should be possible to distinguish them sharply from constative utterances (if P and not A, then P/C); But the performative/constative distinction is flawed, and therefore not A does not hold; So, the supposition that A does not hold is false, and therefore, A holds. Of course, the correctness of this argument depends on the correctness of the sub-arguments against the performative/constative distinction. I recall that there are two, the former consisting of an attempt to show that assertions too have felicity conditions (HTW, 45–52, 137–9), the latter claiming that any utterance can be reformulated as an explicit performative (see, e.g., HTW, 67–8, 85, 134–6). They are far from unproblematic, but it is not my task here to evaluate them; I only want to point out their function within Austin’s project. The main point of HTW, as it emerges from the reading I am proposing, is the discovery of illocution as a dimension of language use, which is fully general (therefore involving criticism of the ideal of assertion as purely constative), and intermediate between saying and producing extralinguistic consequences.² We are going to have a closer look at what exactly was so “discovered” in the discussion of the other assumptions belonging to the received reading of HTW, the second one in particular, namely, that performing an illocutionary act amounts to producing the hearer’s uptake of the meaning and the force of the utterance.

² Although Austin’s discovery of illocution has been acknowledged in the subsequent development of speech act theory, the initial bias towards reading HTW as proposing and self-defeatingly discussing the performative/constative distinction is still in the background and has influenced the reception of what Austin says about illocution, often making his criticism of assertion go unnoticed. That it was among Austin’s goals to criticize the ideal of assertion as purely constative has been recently stressed by Crary (2002).

184

   

3. Illocutionary Acts and Their Effects Illocutionary acts are characterized by most authors in speech act theory as acts, whose effect consists in the production of the hearer’s uptake. This idea is to be found in a paper by Peter Strawson, “Intention and convention in speech acts” (1964), where uptake is considered as the only effect that every felicitous illocutionary act must have. An analogous idea can be found in early work by John Searle, who defines the “illocutionary effect” as “the hearer understanding the utterance of the speaker” (1969, 47). But that is not an idea of Austin’s. Of the uptake effect, Austin writes: Unless a certain effect is achieved, the illocutionary act will not have been happily, successfully performed. . . . I cannot be said to have warned an audience unless it hears what I say and takes what I say in a certain sense. . . . So the performance of an illocutionary act involves the securing of uptake. (HTW, 116–17)

Now, according to Strawson, Austin’s words “Unless a certain effect is achieved” imply that no other effect than uptake is necessarily connected to the performance of an illocutionary act, i.e., a felicitous illocutionary act may have uptake as its only effect. This reading of Austin has allowed for the reinterpretation of illocutionary force as some kind of Gricean speaker meaning. As we know, this reinterpretation has been developed in mainstream versions of speech act theory such as those of Searle (1969, 1979) and Bach and Harnish (1979). But Austin merely meant uptake to be a necessary condition for the successful achievement of the core effect of the illocutionary act, which according to him, goes beyond mere understanding. He describes such a core effect thus: The illocutionary act “takes effect” in certain ways, as distinguished from producing consequences in the sense of bringing about states of affairs in the “normal” way, i.e. changes in the natural course of events. . . . (HTW, 117)

This passage of Austin’s has often been disregarded by speech act theorists and is in fact rather mysterious. What sort of an effect is this, which does not consist of bringing about states of affairs in the “normal” way? Note that the “normal” way is by natural causation: that is, through the production of changes in the natural course of events. So what is the “non-normal” way? Austin gives only one example, and not a very helpful one: the illocutionary act of naming a ship. The effect of such an illocutionary act is, first, that the ship is given a name and second, that certain subsequent acts (such as referring to the ship by a different name) will be out of order.

   

185

The example may seem not to be representative: in fact naming a ship is not a common, everyday illocutionary act such as promising, commanding, warning, or apologizing. Nevertheless, we can use the example to characterize the kind of effect that, according to Austin, illocutionary acts “take.” The effect of the naming of a ship consists of a change not in the natural course of events but in norms, that is, in something belonging to the realm of social conventions; and it is an effect that comes into being thanks to the hearer’s uptake, that is, insofar as it is clear to the relevant audience that the speaker has (felicitously) named the ship. It is not rash to identify such an effect with the “conventional effect” which the illocutionary act, as an accepted conventional procedure, is designed to produce. In fact Austin’s felicity condition A1, the first in his list of felicity conditions for performative utterances, reads: There must exist an accepted conventional procedure having a certain conventional effect, that procedure to include the uttering of certain words by certain persons in certain circumstances. . . . (HTW, 14; my emphasis)

It could be objected that this condition applies to performative utterances and not to illocutionary acts. But in commenting on illocutionary acts in the second part of HTW, Austin clearly shows that the doctrine of Infelicities expounded in the first lectures should apply to them (see e.g. HTW, 106). This is further evidence that Austin maintained that illocutionary acts have conventional effects. We may say that Austin was wrong in generalizing from the naming of a ship example. But we have to admit that his claim about the conventional character of the core effect of illocutionary acts is put forward as a general claim. This suggests that appropriate support for Austin’s claim that illocutionary acts are conventional should come from the conventionality of their effects, rather than from the conventionality of the means by which they are performed. In the debate over the conventionality of illocutionary acts, attention has been paid only to the means for performing the illocutionary act and not to the nature of its effect. Since linguistic means are only trivially conventional, there seemed to be no reason for calling “conventional” those illocutionary acts which do not involve extralinguistic rituals or social institutions. But we can make better sense of Austin’s claim that all illocutionary acts are conventional if we focus attention on the conventionality of effects. In the reading of HTW I am proposing, all illocutionary acts are conventional primarily because they all have conventional effects. This in turn raises various problems. First, we should be able to describe the conventional effects of all types of illocutionary acts, but for many of them such a description is difficult and Austin has not dealt with this descriptive problem exhaustively, although several suggestions can be gathered from his work. Second, we should also reflect on what makes an effect “conventional.” I will not discuss

186

   

the first problem here, although I have ideas about how to handle it.³ As to the second problem, I would like to propose two conditions that an effect should satisfy in order to be conventional. First, it should be “defeasible” (that is, liable to turning out null and void under certain conditions). Non-conventional, material or “natural” effects either are produced or not: The states of affairs they consist of can be further modified, but not just annulled. Second, the effect should be achieved in some way other than by producing a change in the natural course of events. The most obvious (if not the only) candidate is the production of an effect thanks to the agreement on the part of the relevant participants on which act it is that has been performed. By “agreement” I do not mean explicit, metacommunicative agreement; in most cases implicit or even tacit agreement is enough. These two conditions can be traced back to Austin: the former to his insistence, in his preparatory notes for HTW, on the idea of “making undone”; the latter to the very fact that he deems uptake to be a necessary condition for successful performance. In his preparatory notes, Austin refers more than once to Aristotle’s remark that since what is past is not capable of not having taken place, nobody can make undone what has been done: on this view, the effects of our actions are irreversible.⁴ But, Austin notices, the acts performed by means of performative utterances (therefore, illocutionary acts) appear to be an exception, since they may turn out to be null and void if certain conditions are found not to be satisfied. This does not mean that what was actually done did not really take place, or that nothing at all was done by the performer of the infelicitous act: in fact, we have ways of re-describing null acts in terms of the performance of different acts and we may say that it is those acts which really took place. But the discovery of infelicity may make the illocutionary act undone insofar as the bringing about of its conventional effect is concerned. The words were uttered, the conventional effect was supposed to be there, it was even acted upon for a while; but once we discover the infelicity, we realize that the conventional effect never really came into being. Subsequent agreement about the nullity of the act overcomes the former, provisional agreement about its performance. It should be noted that illocutionary effects are also cancellable, that is, they can, once achieved, be abrogated or superseded by acts bringing about opposite conventional effects (for example, an assertion can be withdrawn, or what was prohibited may be permitted). But what Austin’s remarks on “making undone” attribute to illocutionary acts, namely

³ The search for ways of describing the conventional effects of illocutionary acts has inspired much of my work in speech act theory. See Sbisà 1984a [Essay 1], 2001a [Essay 6], 2006c. ⁴ “It is to be noted that nothing that is past is an object of choice, e.g. no one chooses to have sacked Troy; for no one deliberates about the past, but about what is future and capable of being otherwise, while what is past is not capable of not having taken place; hence Agathon is right in saying: ‘For this alone is lacking even to God, To make undone things that have once been done’ ” (Nichomachean Ethics, 1139b 5–11, trans. in Aristotle 2009).

   

187

that which I have called “defeasibility,” goes beyond the cancellability of achieved effects, since it affects the conventional effect in its making, and therefore the act of bringing it about. As to uptake, since it involves the agreement of the participants (often implicit or even tacit) upon what has been done, its role with respect to the conventionality of illocutionary effects is to distinguish them as non-normal, conventional effects from normal, causal ones. Dependence on uptake is characteristic of the former. If illocutionary acts are successfully performed only when they produce the hearer’s uptake, this is because (i) illocutionary acts have conventional effects and (ii) these come into being thanks to the participants’ agreement. So, in Austin, the uptake requirement is not in competition with the alleged conventional nature of illocutionary acts, as in the received reading of HTW: rather, the opposite holds.

4. What Is a Perlocutionary Act? The next assumption I want to discuss is that performing a perlocutionary act amounts to using language to produce consequences. This assumption does not misrepresent Austin’s notion of perlocution, but is misleading nevertheless, in two main ways. It may be taken to imply that all production of consequences beyond the production of the hearer’s uptake is perlocutionary. Or it may be taken to imply that, in order to perform a perlocutionary act, a speaker must intend to produce certain consequences: in other words, that his or her speech act must have a perlocutionary goal. Both these assumptions yield incorrect interpretations of Austin’s theoretical proposal. Let us examine the former issue more closely. If the assumption that performing a perlocutionary act amounts to using language to produce consequences is meant to imply that all production of consequences beyond the production of the hearer’s uptake is perlocutionary, this is again a misunderstanding of Austin, complementary to the misinterpretation of the illocutionary act I have just discussed. In fact, if (with Austin) it is admitted that illocutionary acts have conventional effects, these may have further consequences at the conventional level, and therefore, not all consequences of a speech act are perlocutionary. Moreover, the conventional effect of the illocutionary act, although not precisely a “consequence” (but an outcome or result), itself goes beyond the production of the hearer’s uptake. The border between illocution and perlocution runs between conventional effects and changes in the natural course of events: The former are cancellable and defeasible, the latter are not, and if they are brought about by means of a speech act, their production amounts to a perlocutionary act. In fact, the distinction between illocution and perlocution is at all possible only if a difference between natural and conventional effects is accepted. The assumption that performing a perlocutionary act amounts to using

188

   

language to produce consequences may also be taken to imply that, in order to perform a perlocutionary act, a speaker must intend to produce the relevant consequences. Is then the perlocutionary act to be identified with a speech act having a perlocutionary goal? Of course not, for two reasons. Also unintended consequences are considered by Austin as constitutive of perlocutionary acts (HTW, 106, 118). And conversely, having a perlocutionary goal is not sufficient for a speech act to be successful as the perlocutionary act achieving that goal. If a speech act has a perlocutionary goal but fails to achieve it, there is no corresponding perlocutionary act. At this point, an objection can be raised to the internal consistency of the notion of a perlocutionary act. How can a speaker ever perform a perlocutionary act, if what is needed for a perlocutionary act to be performed is not an act of the speaker, but one of the listener? This objection (due to Gu 1993) is misguided but interesting, since it draws attention to what we may call the “invisibility” of the perlocutionary act, that is, the fact that it does not consist of any physical or psychological gesture of the speaker (such as a bodily movement or the formation of a mental state). Insofar as the speaker’s behavior is concerned, there is nothing more in the perlocutionary act than there was in the illocutionary or for that matter, in the locutionary. Still, Austin insisted in presenting it as an act of the speaker. What does the perlocutionary act consist of, according to his view? The source of most misunderstandings about Austin’s conception of the perlocutionary act lies in his conception of action. Austin has been read in the framework of the mainstream view of action, according to which actions basically are, or can be identified with, gestures of an individual, intentionally performed. I surmise that Austin was presupposing a different framework. His view of action, as it can be reconstructed from HTW as well as from his papers in the theory of action (Austin 1956, 1957, 1966), was non-reductionist and description-related. By this I mean that according to him, there is no reduction of the different effects of one and the same gesture to one and the same basic action, rather, descriptions of these different effects pick out different actions. We perform more than one action with one and the same gesture, insofar as that produces more than one effect.⁵ So, Austin seems to have considered agency as justly ascribed responsibility: there is agency whenever it is fair to ascribe to an agent responsibility for a certain outcome.⁶

⁵ The only attempt to develop a view of action of a non-reductionist, description-related kind I am aware of is Goldman 1970. ⁶ The notion of responsibility I am using here to characterize Austin’s view of action does not by itself amount to moral responsibility: it is meant to apply to actions independently of their being blameworthy or blameless. A similar use of “responsibility” has been made, also in connection with Austin’s philosophy of action, by Yeager (2006, 83ff.).

   

189

It is interesting to notice that while Austin’s speech act theory did not evolve significantly after 1955, between 1955 and 1959 he worked more on the theory of action. While it has been often thought that his speech act theory is no more than an incomplete outline because he died before really developing it, I would say that his truly incomplete work is that regarding the theory of action. This incompleteness affects his speech act theory indirectly, because a speech act theory must presuppose some analysis of action. We may say that in the sense of action which I have attributed to Austin, a perlocutionary act can certainly be an action of the speaker, even if no gesture of the speaker specifically corresponds to it. If agents “do” whatever they may be justly ascribed (at least partial) responsibility for, a perlocutionary act is performed whenever the speaker is (at least partially) responsible for some act or state of the listener. And if this holds for the perlocutionary act, the whole locutionary–illocutionary–perlocutionary distinction should not be read as a distinction among gestures (whether physical or psychological or both), but rather, as one among the kinds of effects for which speakers may be ascribed responsibility.

5. Did Austin Exclude “Non-Serious” Cases? I come now to the fourth assumption I have to comment upon, namely, that Austin’s speech act theory unduly excludes “non-serious” cases (HTW, 21–2, 104). Jacques Derrida has claimed that the possibility of non-seriousness in language use is ubiquitous (in fact, we never know whether a speech act is actually seriously performed) and that this undermines the whole speech-act theoretical enterprise. According to Derrida (1972), Austin overlooked this possibility or, more precisely, tried to sweep it under the carpet by excluding non-seriousness from consideration. So his speech act theory relies on flawed foundations. When Derrida leveled this criticism at speech act theory, his most natural respondent could only be John Searle, who replied that “non-serious” cases can indeed be studied by speech act theory or by an appropriate extension of it, and that Austin merely accidentally failed to do so (Searle 1977). It is interesting to notice that Searle shares at least one of Derrida’s presuppositions, namely, that Austin should have dealt with “nonserious” cases.⁷ But is it true that Austin unduly excluded “non-seriousness” (whether by mistake or by accident)? He excludes it just insofar as the serious/non-serious distinction is not a distinction between types of illocutionary acts, nor between

⁷ [Searle also assumes, like Derrida, that Austin was keen to deal with them in terms of speaker intentions vs lack of speaker intentions.]

190

   

felicitous vs infelicitous illocutionary acts. Since his aim is to focus on the illocutionary level of the speech act, inasmuch as he excludes “non-seriousness” from consideration he is justified in doing so. At the same time, precisely because the serious/non-serious distinction is not a distinction between types of illocutionary acts, Austin does not need to exclude “non-seriousness” from the concrete possibilities which are open to language users: speech acts belonging to any illocutionary type may be performed either seriously or in any one of several “nonserious” ways. So, he recognizes the ubiquitous possibility of non-seriousness, or as he prefers to call it, “aetiolation,” but does not consider it as a threat to his project. It is just another dimension along which the uses of language may be studied. Derrida’s attack aims at a critique of the speaking subject. He wants to convey that it is not legitimate to assume that there is a well-identifiable speaking subject, having beliefs, desires, intentions, and thus actually acting. Without such a subject no talk of speech acts would be legitimate either. But his attack does not really affect Austin’s speech act theory, because speaker intention is not central to it. If actions are not gestures (including the formation of an intention as a psychological component), then undecidability or even lack of speaker intention does not prevent us from speaking of actions. It is to be noted that John Searle is closer to Derrida than to Austin in believing that speech act theory needs a robust conception of the speaking subject. His debate with Derrida is about whether such a robust conception is possible. So he defends a speech act theory based on speaker intentions, that is, his own. The contrast between Derrida and Austin, on the other hand, is about whether a robust conception of the speaking subject is needed at all. Derrida believes that since no robust conception of the speaking subject is viable, speech act theory is undermined. But Austin’s outline of a speech act theory does not rely on any such conception. In concluding this section, I would like to mention that if Austin had developed a theory of aetiolation himself, he would most likely not have employed an umbrella-notion of “pretending” or “doing as if ” (which is Searle’s proposed solution). From his preparatory notes, it appears that he conceived of aetiolation as affecting features of the speech act such as its source—for which he uses the more specific word “authority,” possibly in the sense in which later on, Erving Goffman will use “author” (e.g., in his 1981)—, addressee, sense, and reference, by concealing, suspending, shifting, or weakening them. In a brief note, presumably written in 1951, Austin sketchily suggests to view quotation as aetiolation of authority, soliloquy as aetiolation of address, acting as aetiolation of both, joking as aetiolation of sense, and fiction as aetiolation of reference.⁸ Nowhere does he

⁸ [See on this Sbisà in preparation.]

   

191

suggest that an aetiolated speech act is therefore a non-act or a merely pretended act. Rather, the resulting image of aetiolation may remind us of some of the processes which, in interdisciplinary research into language use, have recently been studied under the label of “mitigation.”⁹

6. Some Implications of Austin’s Claims After our explorations in HTW, what can we say about the contributions it offers to pragmatics? We have seen that the performative/constative distinction is exploited in the proof by contradiction of the claim that all speech is action and that the analysis of performatives serves to introduce the conventional actions that are performed in speaking, that is, illocutionary acts. We have not tackled the meaning/force distinction directly (this would be a complicated matter, because in HTW nothing clear is said about meaning—which is perhaps the main defect of the book), but we have characterized the illocutionary dimension of language use as conventional in an unusual sense, that is, because of the conventionality of the effects it produces. What has been deemed a hasty mistake, namely, the conventionality of all illocutionary acts, appears now as a thoughtful philosophical claim. We have pointed out that the perlocutionary act presupposes action as responsibility rather than intentionally performed gesture, and have attempted to show that, accordingly, Austin’s speech act theory does not presuppose a robust conception of the speaking subject. Of course, we should also recognize that the claims I have attributed to Austin may have various shortcomings. Is Austin’s proof by contradiction that all speech is action sound? Is his view of action correct? Above all: how can his idea that all illocutionary acts have conventional effects apply to assertion? I cannot tackle such issues here, because my aim has not been to discuss whether Austin was right, but merely, to clarify his claims. Instead, I would like to try to say what are the implications of the reading of HTW I have proposed: why we should be concerned with the claims Austin makes, why we should consider them seriously and discuss them or try to test them by putting them to use. Austin’s claims have implications both for research into language use and for philosophy. Their main implications for research into language use are:

⁹ Mitigation attenuates the impact of the speech act on the receiver as well as its costs for the speaker. It has been studied by several linguists and sociolinguists since 1980, including Fraser (1980) and Caffi (1999). It has been said to suspend or weaken various aspects of the speech act, among which (according to Caffi) the relationship of the speech act to its deictic origin.

192

   

(i) that the role played in speech act theory and speech-act inspired discourse analysis by explicit performative utterances should be reassessed; (ii) that such a reassessment makes it viable to give non-trivial analyses of any text or discourse in terms of illocution. It has sometimes been thought that Austin’s theory is applicable only when explicit performative utterances can be used, or that utterances have illocutionary forces only insofar as these forces can be made explicit by performative verbs. Since explicit performative utterances are rarely used and the force of an utterance rarely corresponds to one and only one specific performative verb, it has often been deemed that the applicability of Austinian illocutionary acts is limited. But if, in considering illocution, we focus on its effects rather than on the means for achieving them, we see we can recognize the performance of an illocutionary act not only (classically) from the fact that it can be made explicit by a performative verb but also from the kind of effect it brings about. So, we might want to say that illocutionary acts are conventional because their effects are based on intersubjective agreement and not because they can be performed by means of explicit performative utterances. Rather, they can be performed by explicit performative utterances just because they are conventional in our sense. Only for acts having conventional effects, can saying make it so. Conventional effects, based on intersubjective agreement, concern matters liable to defeasibility and cancellation such as rights, obligations, licenses, commitments. So we can find ways of describing illocution other than the lexicon of performative verbs, that is, directly describe the effect (often vague or hybrid) that the utterance “takes” on the local statuses of the interactants in an interactional sequence. This makes it viable to analyze the illocutionary dimension in any text or discourse, including informal conversation, written texts, fictional discourse, and discourse in the context of different cultures. While performatives are rarely used in informal conversation, implicit negotiation about what each participant can do and should do or should expect from other participants is constantly present and draws on the dimension of illocution as I have characterized it. In writing, illocutionary effects are made available to any reader who takes the role of the addressee. In fictional discourse, most illocutionary effects affect fictional entities. Cultural differences may involve differences in the gamut of admissible illocutionary effects as well as in the procedures for bringing them about. Illocutionary effects may be associated with utterances smaller or larger than sentence size, they may be indicated with varying degrees of precision by linguistic illocutionary indicators and selected and made operative by the hearer’s uptake (as manifested in verbal replies or non-verbal responsive behavior), and they may also display varying degrees of intensity or strength, often in connection with politeness strategies. Illocution also provides a mediation between speech and many of its psychological and behavioral effects (which are

   

193

perlocutionary). The distinction between illocution and perlocution enables us to keep both the conventional and the psychological or behavioral score of an interaction and monitor their reciprocal influence. Certainly all these matters are constantly studied in many fields of pragmatic research, but their connection with the illocutionary dimension of speech often goes unnoticed, while a greater awareness of it might, sometimes at least, be illuminating. Turning to philosophy, I would like to outline three philosophical implications of Austin’s HTW, which are highlighted by the reading I have proposed. They concern: (i) action and agency; (ii) the social contract; (iii) ontology. The conceptions of action and agency underlying Austin’s speech act theory grant an unusually wide role to the ascription of responsibility. If responsibility is central to agency, identifying an action requires first identifying a state of affairs (its effect) and then correlating it with an agent (who is made responsible for it by his or her behavior and circumstances). This broadens the range of things we “do,” which cannot always be reduced to bodily movements made with an intention in mind. Illocution sheds light on the nature of the social contract. Indeed, social order and social structure avail themselves of illocutionary acts or, in other words, of the delicate mechanisms producing local tacit agreement, which enable human creatures to create “conventional” states of affairs. The conception of illocution I have presented implies that local tacit agreement on what conventional states of affairs are in force is basic and does not require further foundations. As soon as it is seen that it is one of the main business of language to let human beings coordinate their behavior and converge on assigning commitments or obligations and legitimate expectations to each other, thus creating conventional states of affairs, the social contract is no longer a mystery. Finally, although Austin does not put forward metaphysical claims about “what there is,” it is clear that if we admit of the conventional effects of illocutionary acts, we are committed to some kind of ontological pluralism. Conventional effects of illocutionary acts are states of affairs which may be said to exist, but by definition are not reducible to material, natural states of affairs. Moreover, an analysis of aetiolation as shifting or weakening features of the whole speech act might suggest to admit of fictional or virtual entities, which exist but, so to say, in weakened ways. All this, I grant, is highly controversial, both by itself and in connection to the cognitive turn that for independent reasons has gained so much ground in linguistic pragmatics as well as in the philosophy of language (if not in philosophy tout court). But I hope it is also stimulating enough to be worth further consideration and discussion, and induce a reassessment of Austin’s contributions.

194

   

A final comment. In the various ways I’ve been trying to illustrate, Austin gives us insights into the capacity of mankind for creating shared environments through language, not as a matter of transmitting anything from one head to the other or of causally influencing each other’s mental states, but as a matter of establishing situations and roles and attributing local statuses to participants. Herein lies the power of human civilization as opposed to “state of nature”: the power which alone makes it possible, on occasion, for someone weak and without weapons to be listened to and even obeyed, the power which makes it possible to conceive and pursue things such as social equality or solidarity and equal opportunities for genders, all of which would not be conceivable in a “state of nature” ethology. To acknowledge in theory and investigate such power is at the same time to foster it and defend it against the risk of regression into forms of social life based on brute force and coercion.¹⁰

¹⁰ This paper was read as a plenary lecture at the 9th International Pragmatics Conference, Riva del Garda, July 10–15, 2005. I thank the audience and particularly Mitch Green, Rob Stainton, and Rukhmini Bhaya Nair for questions and comments. I have discussed my views on Austin’s philosophy of language on various occasions with several colleagues, among whom I would like to thank in particular Anita Avramides, Christopher Gauker, Andreas Kemmerling, and Paolo Leonardi. I thank Anita Avramides, Robyn Carston, and Paolo Leonardi for commenting on a draft of this paper. I am very grateful to Jean Austin who kindly agreed I could cite from notes belonging to the manuscript of HTW not included in the published edition. This research was made possible by a joint grant of the Italian Ministry for Education, University and Scientific Research and the University of Trieste (PRIN 2003). The national project, on “Representation and Reasoning,” was coordinated by Carlo Penco, University of Genova; the local project, on “Assertion, Justification and Inference,” was coordinated by me.

11 Uptake and Conventionality in Illocution 1. Introduction One of the classical problems in speech act theory is whether, and if so, how illocutionary acts are conventional. This paper addresses this problem and claims that the conventionality of illocutionary acts resides, first of all, in the conventional status of the effects they bring about. This amounts to a change in perspective, since the conventionality of illocutionary acts has traditionally been traced back to the conventionality of the means by which they are performed. There is one aspect of the received view that should be retained: that is, the indispensability (already stated by Austin) of uptake as a condition for the successful performance of illocutionary acts. But, while the role of uptake in illocution has usually been taken as evidence of the intention-based nature of illocutionary acts as opposed to their alleged conventionality, I reinterpret it as stemming from the conventional status of the effects of illocutionary acts: it is in order for an utterance to produce conventional effects that it needs to be understood as bringing about those effects. Before coming to my own proposal, however, I will outline and discuss the views of the role of uptake in illocution put forward by such influential authors in speech act theory as Strawson and Searle, in order to show how the currently mainstream perspective has originated and the extent to which it has departed from Austin’s initial project. Since Strawson’s view of the role of uptake goes hand in hand with the introduction of some kind of a divide between two main sorts of illocutionary acts, one of which is conventional in the full sense of the word, the other intention-based, I will also devote a brief discussion to the various ways in which such a divide has been formulated and assess its overall significance.

2. The Role of Uptake in Illocution As is known, according to Austin:

Originally published in Lodz Papers in Pragmatics 5 (1), Special Issue on Speech Actions, ed. Iwona Witczak-Plisiecka and Maciej Witek, 33–52. Warsaw: Versita Open/Berlin: de Gruyter, 2009.

Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0012

196

    

Unless a certain effect is achieved, the illocutionary act will not have been happily, successfully performed. . . . I cannot be said to have warned an audience unless it hears what I say and takes what I say in a certain sense . . . the performance of an illocutionary act involves the securing of uptake. (Austin 1975, 116–17)

This passage clearly takes the (successful) securing of uptake (and, with it, the actual occurrence of uptake) to be a necessary condition for the performance of an illocutionary act. Peter F. Strawson (in his 1964) correctly pointed out that the fact that illocutionary acts have the audience’s uptake as a necessary condition is revealing of their peculiar nature. However, he used reference to this passage of Austin’s to support his reinterpretation of Austin’s illocutionary acts in terms compatible with Grice’s (1957) theory of non-natural meaning, that is, as being characterized (at least in the most ordinary cases) by the presence of a complex intention of the kind described by Grice. So, his emphasis on the indispensability of uptake for illocutionary success assumes that the role of uptake in illocution is to be understood in connection with the intention-based nature of illocutionary acts. Besides, Strawson seems to make the further assumption that Austin, in writing “Unless a certain effect is achieved, the illocutionary act will not have been happily, successfully performed” (1975, 116) meant that the effect he is focusing upon, namely, uptake, is the only effect essentially connected with the performance of an illocutionary act.¹ In Strawson’s reinterpretation, an illocutionary act may achieve its audience’s uptake and (therefore) succeed in being an illocutionary act of a certain type, for example a warning, without thereby achieving any other effect.² He appears to find it completely irrelevant to the issue that Austin also admitted of at least one other kind of effect as “connected” with or “brought in” by the illocutionary act (cf. Austin 1975, 115–16): The illocutionary act “takes effect” in certain ways, as distinguished from producing consequences in the sense of bringing about states of affairs in the ‘normal’ way, i.e. changes in the natural course of events. (Austin 1975, 117)

¹ Strawson admits of cases in which even this may be doubted, but concludes that “the analysis of the aim of securing uptake remains an essential element in the analysis of the notion of the illocutionary act” (1964, 158). ² [A possible argument in support of Strawson’s reading might be offered by the former of the two “omissis” from my quotation of Austin (1975, 116–17): “This is not to say that the illocutionary act is the achieving of a certain effect” (1975, 116). Might that mean that it is wrong to conceive of the illocutionary act as the achieving of any effect apart from uptake (including conventional ones)? If so, Austin would be contradicting his next point (1975, 117). But I think that sentence is used as a disclaimer to block pragmatist-like readings of illocution in general, and particularly (since Austin seems to be thinking of an effect of a specific kind: a “certain” effect), to reject the hypothesis that the illocutionary act might be the achieving of uptake, which is the effect to which the passage is devoted.]

    

197

This further assumption made by Strawson is also to be placed in the framework of his project of reinterpreting illocutionary force, insofar as possible, in terms of Gricean speaker meaning. Given the well-known divergences between Austin and Grice with respect to several philosophical problems (such as, to limit ourselves to issues in the philosophy of language, the relationship between felicity and truth),³ it should be clear from the very beginning that any project aiming to make sense of Austin in a Gricean framework, however valuable from a theoretical point of view, could not lead to a reliable clarification of Austin’s own views and might therefore miss some essential feature of the notions he wanted to introduce. This is (very sketchily) how Strawson argues in support of the reinterpretation of illocution he aims at: (i) the uptake to be secured involves the understanding of the force of the utterance and therefore the recognition on the part of the audience of what the speaker intended to do in issuing it; (ii) if this recognition is necessary for the speaker’s performance of the illocutionary act to be successful, that is, for the speaker to actually perform that act, it is because her utterance expresses a complex intention of the type analyzed by Grice as speaker meaning, which has to be fully transparent (that is, the speaker has to intend it to be recognized, and to be fulfilled thanks to its recognition); (iii) the illocutionary act can therefore be traced back to (one kind of) speaker meaning. Intuitively, this line of reasoning appears to be more cogent if the production of uptake is taken to exhaust the effects that the illocutionary act brings about. Only with this additional assumption can the illocutionary act be completely reduced to one kind of speaker meaning. If the assumption that uptake is the only effect essentially connected with the performance of an illocutionary act is not made, the second step of Strawson’s reasoning may be challenged, that is, the indispensability of uptake might be traced back to circumstances other than the transparent character of Gricean meaning intentions. So, insofar as Strawson’s line of reasoning is taken to be convincing, the idea that uptake is the only effect essentially connected with the performance of an illocutionary act is itself implicitly reinforced. Strawson’s reading of the role of uptake in illocution was deeply influential on the subsequent developments of speech act theory. First of all, as we have seen, it was integral to his reinterpretation of illocutionary acts in terms of speaker meaning, which was then discussed and accepted with amendments by Searle

³ There is clearly an attack on Austin as regards this theme in Grice (1989, 3–21).

198

    

(1969, 42–50) and renewed, in the framework of an inferential theory of speech acts, by Bach and Harnish (1979) (to mention two classics). Secondly, it allowed for a reformulation of the Austinian claim that illocutionary acts are conventional that opened the way to a divide between convention-based and intention-based illocutionary acts (which in turn implies that not all illocutionary acts are, strictly speaking, conventional). We will discuss this kind of development below. Thirdly, Strawson’s assumption that uptake is the only effect essentially connected with the illocutionary act contributed to making the very idea of an effect or result brought about by the illocutionary act, other than its very understanding, fade away. Strawson himself admitted of a core intention to produce an effect on the audience, other than uptake, as a part of the Gricean complex and fully transparent intention. But the role a Gricean core intention has is, arguably, that of a perlocutionary goal and is therefore not specific to illocution. John Searle, after discussing and criticizing Grice’s conception of speaker meaning and its applicability to illocution, further stressed the role of uptake, that is, the hearer’s understanding the utterance of the speaker, by making it not only a means for fulfilling the core intention within a complex Gricean intention but also the content of the core intention itself: In the case of illocutionary acts we succeed in doing what we are trying to do by getting our audience to recognize what we are trying to do. But the “effect” on the hearer is not a belief or response, it consists simply in the hearer understanding the utterance of the speaker. (Searle 1969, 47)

This formulation certainly avoids the risk, inherent in any reference to Grice’s speaker meaning, to import perlocutionary goals (such as getting the hearer to believe something) into the illocutionary intention. But there is a price to be paid. The limitation of the effect of the illocutionary act to uptake is further emphasized: for Searle (1969), uptake is the only effect to which an illocutionary act must aim in order to be such, while any other effect is deferred to perlocution. Bach and Harnish (1979) take this same route, referring by “illocutionary effect” to the fulfillment of the illocutionary intention, that is, the achievement of uptake. For them too, whenever an intention the speaker has in issuing her utterance requires more than uptake in order to be fulfilled, it is to be considered as pertaining to perlocution. Thus, such effects as sustaining a social or conversational situation and keeping it within mutually expected bounds, or as changing the terms or the course of the conversational exchange (Bach and Harnish 1979, 103), are characterized by them as perlocutionary. The result of these developments in speech act theory was an impoverishment of the heuristic potential of the notion of illocutionary act in the analysis of verbal interaction. No doubt also because of this, after the initial enthusiasm of the 1970s, so many scholars doing research on verbal interaction turned to other approaches.

    

199

3. The Divide between Conventional and Intention-Based Illocutionary Acts As is known, Austin maintained that illocutionary acts are conventional acts (Austin 1975, 103, 107, 119). Partly on the basis of his reading of the role of uptake in illocution, Strawson (1964) argued against Austin that they are not necessarily such and that convention does not play any essential role in illocution. His line of reasoning about this problem may, with some simplification, be synthesized as follows: (i) Austin has claimed that the illocutionary act is conventional “in the sense that at least it could be made explicit by the performative formula” (Austin 1975, 103); (ii) if Austin were thinking of the conventionality of language itself (that is, of the fact that the words of which the performative formula consists have conventional meaning), his remark would be trivial; (iii) however, the performative formula, thanks to its explicitness, secures the recognition on the part of the audience of the speaker’s intention to perform a certain illocutionary act; (iv) the conventionality of the illocutionary act, insofar as it is an essential feature of it, can thus be traced back to the transparency of Gricean intentions; (v) in all cases in which illocutionary acts appear to be “conventional” in the minimal sense indicated by Austin, they are indeed intention-based; (vi) if some illocutionary acts require extra-linguistic conventions, they may be taken to be conventional, but they form a particular class and their conventionality cannot be considered as a general feature of all illocution. Strawson’s conclusion that only some illocutionary acts are conventional, while others are (merely) intention-based, is made possible by two interpretive moves, one of which draws on the reinterpretation of uptake we have commented upon in Section 2, while the other amounts to assuming that Austin, in speaking of the conventionality of illocutionary acts, was concerned with the conventionality of the means by which such acts are performed. This assumption is natural enough, since “conventional act” is easily understood to mean “act performed conforming to a convention,” that is, performed by means of some conventionally established behavior, and some passages of Austin’s give support to this reading (even beyond the one cited by Strawson: cf., e.g., Austin 1975, 105). However, such an exclusive focus on the conventionality of the means by which the act is performed is never declared by Austin explicitly, so that it may well be the case that Strawson’s reading is missing something that for Austin was pertinent. We will come back to this below. For the moment, I would only like to observe that, as in the case of

200

    

the tacit assumption that uptake is the only effect characteristic of the illocutionary act, here too whoever takes Strawson’s line of reasoning as acceptable is committed to the assumption that what is at issue when the conventionality of illocutionary acts is discussed, is the conventionality of the means by which such acts are performed. Strawson’s proposal to distinguish between convention-based and intentionbased illocutionary acts inaugurated a conception of illocution in which illocutionary acts are no longer held to be essentially conventional and conventionality can no longer be invoked as their distinguishing feature. This encouraged intention-based views of illocution, by which I mean (very broadly) views of illocution in which at least one of the necessary conditions for performing an illocutionary act is the speaker’s having an intention of a certain kind. More specifically, by attributing to the illocutionary intention the transparent character of Gricean speaker meaning, Strawson (as is correctly recognized by Harnish 2009) anticipated further developments of speech act theory, such as, in particular, the inferential theory of illocutionary acts of Bach and Harnish (1979), which also affirms a divide between intention-based or “communicative” illocutionary acts and conventional ones. John Searle appears to have resisted at least some of the suggestions put forward by Strawson. His early speech act theory (1969), largely drawing on a view of language as a rule-governed activity, is in many aspects correctly described as fostering a convention-based view of illocution (cf. Harnish 2009, 12) and does not endorse the divide between conventional and intention-based illocutionary acts. However, I contend that Searle was not foreign to the turn of speech act theory towards a (broadly) intention-based conception of illocution. The notion of intention plays a central role in the analysis of the structure of illocutionary acts in Searle (1969); in particular, most of the rules for the performance of the speech act of promising (1969, 57–62) are formulated in terms of speaker intentions or other attitudes. Only his most general conditions (1) and (9), requiring “normal input and output conditions” and the availability of semantic conventions for illocutionary force indicators, are stated in terms external to the speaker; condition (4) is stated in terms partly external to the speaker, but only because it is concerned also with an attitude of the addressee. Conditions (2) and (3) are concerned with what the speaker “expresses” and “predicates” (which, in this framework, can be taken to be matters of attitude). Conditions (5), (6), (7), and (8) all require intentional states of the speaker. Referring to speaker intentions and other attitudes as that which is to satisfy the conditions for the successful performance of an illocutionary act may be considered (as I have done in Sbisà 2002a [Essay 7]) as marking a shift from an objective conception of the context to a cognitive one (or from externalism to internalism, as Harnish 2009 puts it). What characterizes an objective conception of context is that illocutionary successfulness (by which I mean the fact that the speaker

    

201

actually accomplishes a certain illocutionary act: cf. the use proposed for the expression “success conditions” by Bach and Harnish 1979, 55) depends on how the world actually is as opposed to what attitudes are present in the speaker’s mind and possibly in the minds of the audience. This was, with good approximation, Austin’s position, at least as regards those rules the violation of which engenders misfires or other flaws threatening illocutionary success. Austin also envisaged a sincerity rule, prescribing that the speaker should have psychological states and attitudes appropriate to the procedure she is invoking, but violations of that rule are for him mere “abuses,” which presuppose the act as performed nevertheless (Bach and Harnish 1979, 55 would reserve the label “felicity conditions” for conditions of this kind, since they are not required for success in performance, but for the non-defectiveness of the performed act). Searle agrees with Austin that insincere illocutionary acts, e.g., insincere promises, are performed nevertheless (1969, 62), but would not allow for performances devoid of certain other intentions; to this aim, he reformulates the sincerity condition for promising as follows: S intends that the utterance of T will make him responsible for intending to do A. (Searle 1969, 62)

So, although he retains traces of Austin’s preference for an objective context, Searle is on his way towards foregrounding a context consisting of intentional states of the speaker as that against which the illocutionary successfulness of the speaker’s utterance is to be evaluated. His occasional oscillations in the direction of the world external to the speaker do not express the core tendency of his thought, which is revealed, rather, in his main addition to the Austinian rules: the “essential condition,” that is, for promising: S intends that the utterance of T will place him under an obligation to do A. (Searle 1969, 60)

What is essential for an illocutionary act of a certain type to be performed is whether the speaker has the right type of intention. The intentionalist or internalist turn in Searle’s work is confirmed, as far as I can see, in his further work. Harnish (2009, 18) finds Searle (1979) even more externalist than Searle (1969), observing that he lists a large number of dimensions according to which illocutionary acts can be described and classified and that among these, several are social in character and therefore external to the speaker’s mind (Searle 1979, 10–16). But I believe it is a more significant fact, to the aims of an assessment of the general direction of Searle’s work, that all of the three features of the illocutionary act which he chooses as the grounds for his classification, namely: the point or purpose of the type of act; the direction of fit between words and world; and the expressed psychological states (1979, 5), are matters of

202

    

intentional states of the speaker and are therefore internalist in character. The first and main feature, the illocutionary point, is declared by Searle himself to reformulate the “essential condition” of Searle (1969), which as I have already said, is satisfied by an intention of the speaker.⁴ So, it could be claimed that, after all, Searle converges with Strawson on the fact that intentions play the central role in illocution. But as regards the divide⁵ between conventional and intention-based illocutionary acts, Searle could not accept it in the forms proposed by Strawson (1964) or by Bach and Harnish (1979), which are incompatible with his intent to accommodate conventional aspects, insofar as possible, within the basically intention-based framework. However, in this connection too (whether because of direct influence or by following an independent but parallel path) he is not insensitive to the issues originally raised by Strawson. In fact, he too ends up isolating institutional speech acts from other illocutionary acts, by assigning them all to the newly introduced class of Declarations instead of scattering them throughout illocutionary classes as Austin had done (Searle 1979, 16–20). Strawson’s conventional illocutionary acts are utterances which form part of a wholly convention-governed procedure and include such examples as: marrying, redoubling, giving out, pronouncing sentence, bringing in a verdict. Conventional illocutionary acts according to Bach and Harnish (1979, 108 ff.), besides being made possible by conventions, are all in various ways connected with institutions and include: voting, resigning, acquitting, marrying, christening, bequeathing, calling a runner out, finding a defendant guilty. Approximately the same acts fall into Searle’s class of Declarations, so that it can be reasonably held that the three groups, albeit characterized in different ways, are coextensive. Thus, it seems that the tendency to adopt, in some form or other, one and the same divide has spread around in the field of speech act studies beyond differences in theoretical framework and in the characterizations given to members of the two groups. This tendency is further confirmed by the presence of analogous divides, even without reference to Strawson’s proposal or to Searle’s classification, in the work of other philosophers who have been concerned with illocutionary acts. I would like to refer in particular to James O. Urmson who, in his (1977), expresses his negative assessment of what he takes to be Austin’s change of mind from the theory of the ⁴ This point should be discussed further. It could be objected to my reading of Searle that in his (1969, 66–7), the essential conditions for some other types of illocutionary acts are stated without reference to the speaker’s intention. But those pages of Searle’s display conditions for the successful performance of illocutionary acts in an abbreviated form. And there is, in my opinion, ample evidence scattered throughout Searle’s works that the performance of an illocutionary act involves, for him, the speaker’s actually having an intention corresponding to the illocutionary point of his utterance (cf., e.g., Searle 1989, 548). ⁵ It might be objected that Strawson admits of a spectrum from a maximum of conventionality to an absence of conventions (Harnish 2009, 16). But, albeit admitting that Gricean intentions and conventional requirements may combine with one another in certain cases and to different degrees, he theorizes these two factors as distinct and even opposed in nature.

    

203

performative utterance to speech act theory and claims that performative utterances are a matter of conventional action, not a linguistic phenomenon, while speech acts are a linguistic phenomenon, which falls short of being action in the way in which performatives are such. So separated from “real” performatives, allegedly central cases of illocutionary acts such as asserting, telling, and asking do not appear very different from what Hornsby, in her discussion of the locutionary–illocutionary distinction (1988, 41–4), calls indicatively saying, imperatively saying, and interrogatively saying, that is, acts which consist (roughly speaking) of the expression of truth-conditions with a mood.⁶ If we reflect on all these divides and wonder what it is that they have in common, we notice that they place on the one side performances (speech acts performed according to conventions; speech acts that, by convention, bring about new states of affairs; performative utterances regulated by non-linguistic conventions); whilst, on the other side, there are uses of language that merely “express” or “communicate” (intention-based speech acts; speech acts expressing various attitudes of the speaker; communicative speech acts). Theories admitting of such divides deal with those utterances that undeniably perform actions as with a “special” case (performatives, Declarations, conventional illocutionary acts), different from the regular cases of linguistic communication. Strawson makes this point with some diplomacy, as follows: Acts belonging to convention-constituted procedures . . . form an important part of human communication. But they do not form the whole nor, we may think, the most fundamental part. It would be a mistake to take them as the model for understanding the notion of illocutionary force in general. (Strawson 1964, 167–8)

Urmson (1977) is more polemical, but the substance is the same. In these perspectives, when speech happens to be really action, it is not so because of its nature but in virtue of particular properties that, depending on the specific theory, belong either to the speaker (Searle 1989, 554), to a certain utterance type (which satisfies a convention: Bach and Harnish 1979, 109–10; Strawson 1964; Urmson 1977), or to a certain situation (whenever the changes brought about by the illocutionary act are credited to perlocution). The acceptance or accommodation of what is basically the same divide in different theoretical frameworks suggests that there is an underlying philosophical problem or perplexity: something that Austin wanted to question and attempted to disregard, but which appeared (and still appears) to most philosophers as something impossible to give up. I surmise that behind the divide between conventional and intention-based speech acts ⁶ [Hornsby, however, keeps acts of indicatively, imperatively, and interrogatively saying distinct from full-fledged illocutionary acts.]

204

    

there lies the contrast between saying and doing, reborn within the very context of speech act theory. Of course, in a speech-act theoretical framework “saying” is never “pure” saying (something to be taken into consideration quite apart from circumstances, purposes, audience etc., as Austin says in characterizing the “ideal” of the constative utterance: 1975, 146): it is always at least expressing attitudes, or making and inviting context-sensitive inferences. But albeit expressing attitudes or making and inviting inferences are activities, they are cognitive, psychological activities, not actions bringing about effects on states of the world. We might think that their contrast with full-fledged actions is grounded in the reality of the facts investigated by speech act theorists. But there is no final reason to exclude the opposite, namely, that the contrast might be ultimately in the philosopher’s own eye. Thus we may want to distinguish, with some oversimplification and disregarding for the occasion more finely-cut distinctions, between theories of language as a tool for communication and theories of language as a vehicle of action. In theories of language as a tool for communication, communication is most often represented as a kind of mind reading, speech acts as the expression of communicative intentions, illocutionary force as mood. Power relationships and matters of social organization, both at the macro- and at the micro-social level, are left aside and deferred to other disciplines. Austin appears to have sought a theory of language as a vehicle of action. He seems to view the widespread tendency to rely on such notions as expressing and communicating as a hurdle for his project, and indeed avoided using them in his descriptions of how speech works. This point could be otherwise made by claiming, as Petrus (2005) has done, that Austin’s illocutionary acts are not a matter of “communication” at all. Of course, that which is ordinarily called “communication” has to be dealt with within such a perspective too, but its focus is on action and, in the case of illocutionary acts, on conventional action. In this perspective, there is no reason to introduce the divide between conventional and intention-based illocutionary acts: rather, it would be crucial to give an account of illocution as essentially conventional.

4. The Conventionality of Illocutionary Effects I would like now to go back to Austin and consider a neglected aspect of his view of the conventionality of the illocutionary act. I take it that while the conventionality of the means, on which Strawson focuses in establishing the divide between intention-based and conventional illocutionary acts, is more apparent and may be symptomatic of the conventionality of an act, an act should be said to be conventional not so much on the basis of the means by which it is performed, but on the basis of what it does, namely, the kind of effect it is its job to bring about. It is arguable that for Austin, the illocutionary act was to be characterized as

    

205

having a conventional effect (Sbisà 1989a, 2007a [Essay 10]; cf. also Doerge 2006). On this basis, we can make sense of Austin’s claim that illocutionary acts are essentially conventional in a way other than the one suggested by Strawson and outline a new account of their conventionality. Since it is not uncontroversial that Austin held illocutionary acts to have conventional effects (as we have seen in Section 2, starting from Strawson 1964 and Searle 1969 it has become customary to believe that he too took the securing of uptake to be the only effect essentially connected with the illocutionary act), I would like to cite two sources in support of my reading, namely, Austin’s description of the illocutionary act’s “taking effect,” and his formulation of rule A1 for performative utterances. After considering each of these sources, I will put forward a sketchy account of what it is for an effect, and therefore for the act bringing it about, to be conventional.

4.1 The “Taking Effect” of the Illocutionary Act Austin (1975, 116–18) distinguishes three effects that, he says, are connected with the illocutionary act, the first of which is the securing of uptake. The third, inviting a response, is not relevant for the aim of an essential characterization of illocution, because Austin himself says that it belongs only to certain kinds of illocutionary acts. The second effect is introduced as follows: The illocutionary act “takes effect” in certain ways, as distinguished from producing consequences in the sense of bringing about states of affairs in the “normal” way, i.e. changes in the natural course of events . . . (Austin 1975, 117)

I admit that this passage is rather mysterious. The characterization of the effect at issue is mainly negative and, therefore, says little about it. The “normal” way of achieving an effect is through the production of changes in the natural course of events, that is, by natural causation. But it is not said what the “non-normal” way of achieving an effect consists of or what kinds of effects, presumably consisting of the bringing about of states of affairs, can be achieved that way. Austin gives only one example of an illocutionary act’s “taking effect,” and not a very helpful one: the naming of a ship. The effect of such an illocutionary act is, intuitively, that the ship receives a name, but Austin illustrates its effectiveness by remarking that certain subsequent acts such as referring to the ship by another name “will be out of order” (1975, 117). The example is clearly one of a conventional act, and not a common, everyday illocutionary act such as commanding, warning, or apologizing. One conclusion that can be drawn from this is that the “taking effect” Austin is illustrating only characterizes conventional acts and (since not all illocutionary acts are conventional) is not a general feature of illocution. This might explain

206

    

why Strawson (1964) did not pay attention to this passage. However, Austin does not introduce “taking effect” as the peculiarity of certain illocutionary acts only (as he does in the case of the third kind of effect he specifies, the inviting of a response), but as if it were a general feature. Moreover, it would not be fair to interpret him, or anyone else, by using criticism of their views to support the interpretation. After all, Austin might have chosen this example in his lecture precisely because it is a clear, uncontroversial case of the kind of effect he invites his audience to focus upon (this is not the only case in which he proceeds this way: performative utterances too are introduced by him as an uncontroversial case of speech that is action, while the goal is clearly to extend this idea to far less uncontroversial cases). So, we can use the example to characterize the kind of effect that, according to Austin, illocutionary acts “take.” The effect of the naming of a ship consists of a change not in the natural course of events but in norms, that is, in something belonging to the realm of social conventions: a new norm is enacted, as it can be seen from the assessments of people’s relevant behavior that may stem from the norm. The moral to be drawn is simple: to extend this idea to other illocutionary acts, we should look at changes in states of affairs belonging to the same level of reality as norms. Such are, for example, assignments or withdrawals of rights and licenses, assignments of obligations or waivers, undertakings of commitments. Indeed, a large number of illocutionary act types (or varieties) have effects that can be described in such terms (cf. Sbisà 1984a [Essay 1]; Doerge 2006) and it is moreover possible to describe this way the effects of illocutionary act tokens (cf. Sbisà 2001a [Essay 6]). Even for a variety of illocutionary act that is explicitly invoked by Strawson to show that not all illocutionary acts are conventional, that is, warning (1964, 153), there is, quite intuitively, a state of affairs which is brought about and can be described in terms similar to those introduced above: it is a state in which the addressee, or anybody else in the community, is no longer allowed to take the speaker as responsible for some mishap or trouble, related to the content of the warning, in which the addressee might incur. To clarify Austin’s passage on “taking effect” more thoroughly, however, it should also be explained how it is possible to bring any effect whatsoever about in a way other than natural causation. The most plausible explanation for this aspect of Austin’s claim is that he had in mind the contrast between a natural chain of causes and a human intervention which may introduce a break in it. When an effect is conventional, the state of affairs it brings about cannot exist without some kind of human intervention or decision. Somebody has to take a certain attitude towards the utterance, take it a certain way, accept that it has brought about a certain state of affairs or agree upon what has been brought about. So, the naming of a ship does not cause the ship to have a name by initiating a simple chain of natural causes, because in order to achieve an effect of the kind at issue—the creation of a norm—it is at least necessary that what the speaker does and says be socially accepted as having that effect.

    

207

4.2 From Conventional Procedures to Conventional Effects In explaining how performative utterances work, Austin spells out the rules they must follow in order to be free from those defects he calls “infelicities.” This is the first rule he specifies, rule A1 (the violation of which would make the alleged performative utterance null and void): There must exist an accepted conventional procedure having a certain conventional effect, that procedure to include the uttering of certain words by certain persons in certain circumstances. . . . (Austin 1975, 14, see also 26–8)

Here Austin, in requiring that there should exist a conventional procedure in order for the performative utterance to be at all such, explicitly specifies that the relevant kind of procedure is to include a “conventional effect.” Thus, he suggests that it is typical of the procedures of which performative utterances are part, that they are not mere rule-governed sequences of words or gestures, but give rise to effects of a conventional kind. It might be objected that rule A1 belongs with performative utterances as opposed to illocutionary acts, and cannot therefore be cited to support the claim that Austin took it that illocutionary acts have conventional effects. But Austin intended to transfer his analysis of the rules for the successfulness of performative utterances to illocutionary acts in general, as he says or implies on several occasions. If there is any need to provide evidence, here is some: (i) Defects of illocutionary acts are called by Austin infelicities (or “unhappinesses”: cf., e.g., 1975, 106). Since infelicity has formerly been introduced as the typical illness of performative utterances, this (at least) suggests that illocutionary acts and acts performed by issuing performative utterances belong to a same kind; (ii) That a certain act can be performed by using the verb that names it in an explicit performative formula is used by Austin as a criterion for distinguishing types of illocutionary acts (1975, 150). (iii) The types of acts that are cited as examples of acts performed by means of performative utterances appear also in the final list of illocutionary acts (Austin 1975, 151–62). (iv) Last but not least, the theme of the indispensability of uptake is already present in the first chapters of Austin (1975), when it is said that performative utterances can achieve the conventional effect envisaged by the procedure to which they belong only if they are so received by the relevant audience (1975, 22–3). If rule A1 is meant by Austin to apply to illocutionary acts in general, then he is also committed to the claim that every illocutionary act involves a conventional

208

    

effect. It is quite natural to identify the effect “taken” by the illocutionary act in the “non-normal” way we have discussed above with the conventional effect of the procedure in which the illocutionary act consists.

4.3 From Conventionality to the Securing of Uptake So far, I have argued that, according to Austin, illocutionary acts have conventional effects. Now I have still to discuss in greater detail in what sense these effects are “conventional,” and how it is that they are brought about. Here Austin has little to say, apart from an interesting set of handwritten notes⁷ which insist on the peculiarity of actions performed by means of performative utterances to be liable to annulment in ways barred to standard physical actions. I have commented on these notes in Sbisà (2007a) [Essay 10], reaching the provisional conclusion that there is a particular kind of defeasibility (due to the discovery of infelicities of the “misfire” kind) which may be considered a hallmark of conventional effects. Here, I will shift from the interpretation of Austin’s writings to a simple theoretical proposal which I take to be respectful of the conception of the illocutionary act he has outlined, but, of course, goes beyond it. We have seen in Section 3 that, in commenting on a passage of Austin’s in which Austin (apparently focusing, in this case, on the means by which illocutionary acts are performed) comes close to defining what it is for an act to be conventional (“conventional, in the sense at least it could be made explicit by the performative formula,” Austin 1975, 103), Strawson argued that a mere appeal to linguistic conventions would be irrelevant and that the use of performative formulas has, rather, to be connected with the indispensability of uptake and with the Gricean transparency of illocutionary intentions. Partially accepting Strawson’s remark, but keeping closer to Austin’s perspective, I would say that if we admit of the illocutionary act’s conventional effect, there is an easy connection we can make between conventionality and availability of an explicit performative formula: explicitness helps produce that social agreement without which the act’s conventional effect cannot be brought about. The explicit performative formula forces such an agreement by drawing on the use of a lexical element which, since it belongs to the language of the social group within which the act is designed to be performed (and the utterance designed to perform it is issued), cannot avoid being understood by any competent member at hearing distance. From this, of course, it does not follow that the illocutionary act is not (essentially) conventional. Quite to the contrary, explicitness helps just because there is a conventional effect that is to be brought about. I suggest that the feature of the conventionality of illocution that

⁷ Conserved in Oxford together with the manuscript of How to Do Things with Words.

    

209

Austin’s remark about the availability of performative formulas should be taken to highlight is that the bringing about of conventional effects depends on agreement about their coming into being among members of the relevant social group. This is why the securing of uptake is required. This is also why Austin, in the few examples he makes, shows an inclination towards considering actual uptake (as opposed to the mere intention to secure one) as the standard requirement (1975, 2–23, 116). This way of understanding the conventionality of illocutionary acts does not assign a primacy to explicit performative utterances and, moreover, does not require that the means by which the illocutionary act is performed be conventional in a strong, extra-linguistic sense. It does not assign a primacy to explicit performative utterances, because what is indispensable for the achievement of the conventional effect of the illocutionary act is that the effect which the utterance is designed to produce be indicated clearly enough to be identifiable and possibly agreed upon. It does not matter how it is indicated: this job may be done by performative formulas in the canonical form (first-person present indicative active of a performative verb), by ritual performative formulas (established by extralinguistic convention, and displaying various ad hoc linguistic forms), by illocutionary force-indicating devices of various kinds as well as by content-based and context-based “indirect” strategies. Whether we need inference or mere pattern recognition in order to understand these indications is an important issue which I cannot discuss here. I am particularly interested in the hypothesis that the understanding of illocutionary force may be based on pattern recognition, for two reasons. The existence of shared patterns of conventional action (which are cultural facts and can be expected to be linguistically encoded at least up to a certain point) could account for the “accepted conventional procedure” of Austin’s rule A1 without binding us to an obsessively rule-governed view of how illocutionary acts are performed.⁸ Moreover, patterns may be cognitively processed in different ways, for example by means of Gestalt-like mechanisms, but also, if needed (as in the case of unfamiliar patterns, gravely incomplete display, and other complications), inferentially, which would assign a legitimate role to inferential theories of illocutionary force understanding.

5. How to Achieve a (Conventional) Illocutionary Effect So, I propose to read Austin’s remark on conventionality as availability of a performative formula, cited in Section 4.3, as suggesting that what is revealing of the conventionality of the illocutionary act (understood as the conventionality ⁸ It seems to me that the understanding of the conventionality of illocutionary acts proposed by Millikan (1998a) in her own naturalized framework goes in this direction.

210

    

of its effect) is the need to secure uptake. A conventional effect is such insofar as it comes into being by being agreed upon by the relevant members of a social group. This agreement is made possible by the securing of uptake. That is, it is because we understand that the speaker’s utterance has the force of a promise that we take it that ceteris paribus (namely, unless infelicities leading to annulment are discovered) she is committed to a certain further course of action. It is because we understand that the speaker’s utterance belongs to the procedure of naming a ship that we take it that ceteris paribus the ship has been given that name. It is because we understand that the speaker’s utterance has the force of a warning that we take it that ceteris paribus she should not be attributed responsibility for whatever may happen to us in connection with the warning’s content. Under this respect, there is no substantive difference between so-called “conventional” institutional or ritual cases and informal, conversational ones. There are differences in the degree to which certain more specific requirements about the form of the uptake are relevant: for example, in institutional and ritual cases it is more important that members reach awareness of how the act performed should be named (was that marriage? did the president resign? was that the closure of the session? was that a formal promise or only an emphatic expression of intention?), while in informal, conversational cases the problem is not so much to understand how the act should be named, but to have a sense of the setting of rights, obligations, and the like that it is designed to bring about, and to be ready to act upon it. Of course, there are many other open questions about uptake and its impact on the conventional effects of the illocutionary act. What exactly should the “uptake” of a certain illocutionary act (token) consist in? Whose uptake should be secured? What happens to the illocutionary act, and to its conventional effect, if uptake is not achieved, or if actual uptake does not correspond to the speaker’s expectations? What about misunderstandings or instrumental misinterpretations? Can uptake be multiple (different participants may take an utterance in different ways) and what happens to be “done” in such cases? I maintain that uptake need not be present in an explicit linguistic form, or as a full-fledged thought in the mind of the addressee. Of course it should always be possible to make explicit, by means of further linguistic utterances, how a certain utterance has been taken: we often do so, for example, in reporting conversations. But in many cases uptake consists in a tacit agreement, that is, either is made manifest in the audience’s response (insofar as people act and speak upon what they take has been done up to that point) or holds by default (if there is no reason to doubt that the relevant participants have heard and no second-turn or thirdturn repairs have been initiated). As to the relevant participants, this would be a field for applied speech act theory. It depends on the kinds of illocutionary acts and on the structure of the social context whether the speaker should care to secure the addressee’s uptake (as in the case of warnings), or the uptake both of the addressee and of certain ratified

    

211

bystanders (as in the case of many ceremonial performatives), and when the uptake of ratified bystanders may supplant the failure to achieve the addressee’s own uptake. In various kinds of semi-formal social interaction, moreover, there is a recognized “director” (the host in TV talk shows, the psychotherapist in group therapy sessions) who is granted more conversational rights than other participants, not only in allocating conversational turns but also, as the analysis of stretches of such conversations easily shows, in ratifying or refusing to ratify other participant’s illocutionary acts. A major theoretical problem is what happens if uptake is not achieved, or in case of multiple and contrasting receptions. For theories taking illocutionary acts to be intention-based, the reply is easy: what is relevant to the performance of the illocutionary act is that the speaker intends to achieve the hearer’s uptake. Also Petrus (2005), while defending the conventionality of illocutionary acts and proposing to connect it with uptake, argues that it is mistaken to assign the primacy to actual uptake, which might be arbitrary or just wrong, and fosters a normative view of uptake as the way in which, in consideration of the utterance and its context, the utterance should be taken. In the perspective proposed here,⁹ since uptake (whether actual or reasonably “secured”) decides, in the immediate context at least, what has been done by an utterance (what illocutionary effect it brings about, what illocutionary act has been performed in issuing it), it would be crucial to elaborate principled ways to establish what is to count in actual cases as the required interpersonal or social agreement. I would like to stress, in conclusion, that these open problems and complications about how to achieve a (conventional) illocutionary effect, on which I cannot dwell longer here, witness the ample applicability, to formal situations as well as to informal ones, that a perspective on illocution grounding conventionality in uptake may have.¹⁰

⁹ Backed by my own research on samples of natural conversation, see, e.g., Sbisà 1992b, 2002b [Essay 8]. ¹⁰ Ancestors of this paper were read at the University of Cincinnati in 2002, at the University of Sheffield in 2005, and at the International Conference “Tra pragmatica e linguistica testuale. Ricordando Maria-Elisabeth Conte,” Bergamo, September 10–11, 2008. I am grateful to the audiences for the discussion. I would like to thank Mike Harnish for allowing me to read the draft of his paper “Internalism and Externalism in Speech Act Theory” [Harnish 2009], to many points of which I still owe replies, and Iwona Witczak-Plisiecka for her editorial work. This research has been made possible by a joint grant by the Ministero dell’Istruzione, dell’Università e della Ricerca and the University of Bologna. The national project, on “Aspects of Truth,” is coordinated by Diego Marconi, University of Torino; the local project, on “Correspondence and the Semantic Conception of Truth” is coordinated by Paolo Leonardi.

12 Illocution and Silencing 1. Speech Acts and Power Relationships Hardly ever has speech act theory been perceived and used as an instrument for the critical analysis of power relationships. When I was young, I was confident speech act theory had such a potential for criticism, which I hoped would become manifest with time. But, I have to admit, this did not happen. People now discussing power relationships in speech and through speech do not seem to realize that illocution and perlocution would be useful concepts. Here, I do not want to inquire why it is so nor plead for a change of direction: I have done this elsewhere (Sbisà 1984a [Essay 1], 1989a, 2001a [Essay 6], 2002a [Essay 7], 2006c). Rather, I will discuss an isolated, tricky case in which the notion of illocution has been used in connection with matters of power: the debate on free speech and pornography.

2. Pornography and Speech Acts: A Debate Catharine MacKinnon was the first to claim that pornography should not be protected as “free speech”, claiming that it is not a matter of speech (of expression of opinions), but of action and particularly of two interconnected kinds of action that should be banned, that is, subordinating women and silencing them (MacKinnon 1987, 1993). Langton (1993, 1998) has assimilated subordinating and silencing to illocutionary acts, observing that illocutionary acts affect people’s rights. She argues that subordinating someone affects that person’s rights and that silencing can be described as depriving people of the right to perform speech acts. McGowan (2003) has proposed a new sub-class of illocutionary acts, acts that change the bounds of conversational permissibility or “Conversational Exercitives,” to account for the illocutionary force of pornographic discourse. If illocutionary acts are actions, not mere speech, pornography cannot be protected as mere “free speech.” Several objections have been leveled at the proposal to strengthen the case against pornography by depicting pornographic discourse as performing Originally published in Language in Life, and a Life in Language: Jacob Mey—A Festschrift (Studies in Pragmatics 6), ed. Bruce Fraser and Ken Turner, 351–7. Bingley: Emerald/Leiden: Brill, 2009.

Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0013

  

213

illocutionary acts. It has been claimed that pornography is just speech after all (Jacobson 1995), or that it may perform different acts according to context (Saul 2006). It has been claimed, by way of a compromise, that illocution is communication, but has political repercussions anyway (Hornsby 1994). I believe Langton (1993) is right in taking both pornography and illocutionary acts to affect people as to their rights (as well as obligations, legitimate expectations, or commitments). Nevertheless, it is unclear whether all changes in such matters should correspond to the core effect of some illocutionary act. Even if both pornography and illocutionary acts have such effects, it does not follow that pornography brings those effects about qua illocution. In fact, quite apart from the legal and political problem of whether prohibiting or limiting pornography amount to limiting “free speech,” it is an open question what explanation should be given to its alleged “subordinating” and “silencing” functions. I will revert to this question in my conclusions; in the meantime, I prefer to address a more limited one, namely, how to account for a person’s being “silenced.” Here, what is at issue is not whether it is pornography or illocution or both that make certain speech acts “unspeakable” for a woman (to use Langton’s word). What is at issue is, rather, how to describe the dynamics producing such a condition. I think that this question has some priority with respect to the broader question of whether and how pornography subordinates and silences: the reply to the former may shed light on the latter.

3. The Case of the Ineffective Refusal Let us consider a case in which a woman is “silenced”: the case of the ineffective refusal (Langton 1993, 320–1). A man sets out to have sex with a woman. She says (shouts, or whispers): “No!” The man does not mind and goes ahead. The encounter results in an intercourse that, according to the woman, amounts to rape. The status of the woman’s “No!” and the reasons of its ineffectiveness may be traced back to the “silencing” effect of pornography, since prejudice fostered by pornography contributes to hindering a woman’s refusal of sex from being recognized and complied with. But whether or not it is the fault of pornography, what exactly has happened and wherein lies the wrong? According to Langton, the refusal’s ineffectiveness may involve two different kinds of failure. If the man understands that the woman is refusing to have sex with him, but goes ahead nevertheless, perhaps because, due to exposure to pornographic discourse, he has acquired the belief that possessing a woman against her will is most exciting, the failure is perlocutionary: the perlocutionary object of the refusal is not achieved. But the refusal itself is illocutionarily successful: it is felicitous and recognized as such. When the man does not even pause to consider that the woman has refused to have sex with him, the failure is illocutionary: it affects the very act of refusal.

214

  

This is, according to Langton, the crucial case of a woman’s being silenced. But how exactly should this illocutionary failure be described? According to Langton (1993, 1998), this case of illocutionary disablement may be assimilated to Davidson’s example in which an actor on the stage, who has already shouted for fire as a part of the play, attempts to actually shout for fire because the stage is going on fire and is not taken as issuing a serious warning (Davidson 1984, 269–70). She also suggests that the man takes the woman’s refusal as insincere (since, he believes, women enjoy being raped) (Langton 1998, 275). Let us discuss these accounts separately (they are not equivalent). In the insincerity account, the man does take the woman’s “No” as a refusal, albeit a flawed one. The kind of infelicity to which insincerity is to be traced back is the abuse of a procedure (Austin 1975, 14–18) and does not keep the act from being actually performed, albeit making it one that it was not fair to perform. By accounting for the man’s behavior as due to the belief that the woman’s refusal is insincere, the refusal’s ineffectiveness is assimilated to a perlocutionary failure, prompted by the perceived abuse. So perhaps this description is not wholly appropriate to the crucial case of “silencing.”¹ In the play-acting account, the woman is represented by the man as performing the illocutionary act of refusal non-seriously. Austin used the term “aetiolation” for this kind of situation, which he assimilated to infelicities but did not class with them (1975, 22). As is known, Derrida (1972) criticized Austin for this (missing, perhaps, Austin’s rationale for keeping infelicities and aetiolations separate), and Searle elaborated on the problem proposing a pretense-like view of aetiolation with respect to the case of fiction (Searle 1979, 64–72). Both Searle and Davidson tend to conflate non-serious performance with insincerity (if play-acting is “pretense,” it has an element in common with insincerity: the speaker is not in the appropriate psychological state). But Austin wanted to save the distinction. In aetiolated speech, some component of the speech situation is overtly modified, displaced, or suspended: for example, the speaker may not coincide with the subject who is responsible for the speech act, as in quotation and in play-acting, or no attempt may be made to achieve the apparent perlocutionary object of the speech act, as when a command occurs in a poem, or when the speaker is joking (Austin 1975, 104; see Sbisà 2007a, 469 [Essay 10, Section 5]). A “non-serious” refusal is a refusal performed in a weakened way, for example with a suspension or displacement of what would be its regular perlocutionary goal. Again, from an Austinian point of view the failure of the woman’s refusal is perlocutionary, and we are missing an illocutionary account of the inefficient refusal case.

¹ [Here I avoid considering the trickier case in which the man believes that women are always insincere when refusing to have sex. This case goes some way toward denying women in general the capability of being sincere in refusing to have sex, which would amount to epistemic injustice. See below, Section 4.]

  

215

A third explanation is available, which also is explored by Langton (1993, 1998), although not clearly distinguished from those we have already discussed. According to this explanation, the woman’s refusal misfires (Austin 1975, 18, 34–5). It is meant as a refusal, it is serious and not abusive, but the woman is not recognized as in a position to refuse: the circumstances are therefore not appropriate and the refusal is void. It does not succeed in creating an obligation for the recipient. This is genuine illocutionary failure and, therefore, the crucial case of a woman’s being silenced. But this third explanation has a paradoxical consequence: we may conclude from it that the man is right in acting as he does (Jacobson 1995). If the refusal misfires, if it does not succeed in binding the recipient, if therefore the recipient is not bound, why should he not go ahead? There has been no refusal. So what is wrong?

4. What Is Wrong in Silencing? I would like to consider here the replies to this puzzle offered by Jennifer Hornsby and Miranda Fricker. Hornsby (1994) (see also Hornsby and Langton 1998) analyzes the dynamics of illocution as intentional speech, rather than as conventional action, and highlights the role of uptake in communication (Austin 1975, 116–17; Searle 1969, 47), introducing the notion of “reciprocity.” According to her, language relies on a mutual capacity for uptake, which involves a minimal receptiveness on the part of language users in the role of hearers. Thanks to this reciprocity, the speaker is taken to be performing the very illocutionary act that, in being so taken, she is performing. When reciprocity obtains between people, they are such as to recognize one another’s speech as it is meant to be taken. The wrong of pornography is that it undermines reciprocity, so that hearers are no longer capable to grasp the illocutions that women are trying to perform. That’s why the woman’s refusal of sex is ineffective: lack of reciprocity prevents the hearer from taking her “No” as a real refusal. Since uptake is necessary to the completion of an illocutionary act, the refusal itself is flawed. Miranda Fricker examines the case of the ineffective refusal in the framework of her discussion of “epistemic injustice,” taking it to be an example of “testimonial injustice,” that is, of credibility deficit affecting a speaker due to identity prejudice (here, gender prejudice) in the hearer (Fricker 2007, 9–29, 140–2). According to Fricker, the silenced woman sustains an extreme form of testimonial injustice: since women drastically lack credibility for certain subject matters in the social environment, her “No” fails to register as testimony, that is, is not recognized as a source of knowledge. The wrong consists in the prejudice-based undermining of the woman’s capacity as a knower, and therefore of her human status. Fricker (2007, 141–2) claims that this account of the ineffective refusal case is more plausible than that of Hornsby, because it requires less erosion of the woman’s

216

  

human status before the episode occurs. In fact, on Hornsby’s account, the question of the woman’s credibility does not even arise, since she is for first not attributed the communicative intentions her words are designed to express. In my opinion, Fricker is right in insisting that there is injustice in failing to acknowledge the woman’s refusal, and that this kind of injustice may occur in the framework of the recognition of communicative competence. But the injustice is not exactly “epistemic.” Fricker seems to take for granted the insincerity account of the refusal’s ineffectiveness, according to which the man may be depicted as claiming to know what the speaker desires better than the speaker herself. The woman says “No,” perhaps she does so because she had a repressive education and is ashamed to confess her desire, perhaps she even believes herself to be sincere, but indeed is not: unknown to herself, her desire is to have sex. Or, even, to be raped. As Langton notes (1993, 324–5), this is one of the ways in which rapists may construe the situation in order to justify themselves. There is an element of epistemic injustice in this construal, since the man does not recognize that the woman knows what she wants or does not want to do. Driven by prejudice, he fails to recognize the claim to knowledge which is implicit in the refusal and thus undermines the woman’s capacity as a knower. But this epistemically oriented narrative does not tell the whole story about the refusal’s illocutionary failure. As an illocutionary act, refusal is exercitive, not verdictive: it is an act of making a decision, not of issuing a judgement (Austin 1975, 151, 155–6; Sbisà 1984a [Essay 1]; Sbisà 2006c, 164–5; see Langton 1993, 325). In our case, it is not a matter of producing knowledge about one’s own sexual desire, but a decision not to make oneself available for sex. (Consider that a woman might refuse to have sex not because she does not desire the man who is wooing her, but because she has no anti-conceptional protection available: she refuses against her desire, but her act is not thereby undermined; insofar as it is an Exercitive, it is not even an abuse of the procedure). Exercitives presuppose that the speaker has authority over the field at issue: in the case of a refusal to have sex, the speaker—the woman—must be in a position to decide about her own availability for sex, if she is to refuse felicitously. When the refusal is not taken as valid, what is undermined is not so much the woman’s capacity as a knower, as her authority over her own body. What prejudice contributes to this is the idea that the woman’s body is not and need not be under her personal, autonomous control. The idea that the woman’s body is there for men to use it. When this prejudice about women is accepted, only another man having authority over the woman (her father, brother, or husband) can refuse the use of a woman’s body to a man. For both Hornsby and Fricker, the wrong done to the woman lies not so much in taking her refusal as a misfire (or as non-serious or abusive), but in undermining her subjectivity and competence as a member of the linguistic community or as a knower. My proposal to broaden Fricker’s account to cover the exercitive

  

217

nature of the act of refusal may allow us to bridge the gap between accounts focused on the undermining of subjectivity and competence and the core problem of the alleged illocutionary failure of the refusal. It is because prejudice denies a woman jurisdiction over her own body that her refusal is seen by the man as a misfire and therefore, very likely, as some other kind of action such as hardly credible expression of feelings or polite play-acting or even sexual game: cooperative assumptions of the kind described in Grice (1975) regularly prompt us to attribute a different illocutionary act to the speaker whenever we believe that the illocutionary act indicated by the speaker’s words misfires. But if denying a woman jurisdiction over her own body is recognized to be wrong by other social agents (the woman herself; an anti-rape mutual help association; a law or human rights declaration, etc.), then prejudice cannot say the last word about the misfiring of the woman’s refusal. There is at least conflict between two different assessments of the woman’s authority over herself, which is something very different from actual non-satisfaction of felicity conditions leading to annulment. We may say that the refusal is felicitous after all and that its felicity is unjustly disregarded. So, the rapist makes the woman a threefold wrong: he denies she has jurisdiction over her own body, he fails to see he is bound by her refusal, and, last but not least, fails to comply with it.

5. Conclusions What is then the role of pornography in making a woman’s refusal of sex ineffective? I believe it consists of shaping and transmitting gender prejudice. It is the tacit acceptance of certain sorts of prejudice about women, what they are, what function they should serve, what they want and are likely to enjoy, and so on that sets the stage for episodes sharing the structure of the example we have discussed. This is what has been impressively, but misleadingly represented as the core effect of alleged illocutionary acts of “subordinating” and “silencing.” That which in pornography produces subordination and silence is not a specific type of force (as it would be if pornographic discourse contained commands or prohibitions addressed at women, or such acts as depriving someone of a right or role or ranking someone low), but implicit meaning (presupposition and implicature) and its persuasive effects (as is recognized in Langton and West 1999). If, because of the way some pornographic story goes (and it does not matter if it is fictional: fiction may well communicate something implicitly about the real world), it comes to be taken for granted that women are not in a position to decide about their own bodies, women are at risk of seeing their decisions about their bodies unjustly dismissed as misfires. It should be added at this point, echoing Fricker’s claim that testimonial injustice causes “deep and wide harm to a person’s psychology and practical life” (2007, 145), that also being systematically denied

218

  

acknowledgment as to one’s entitlement to perform a certain illocutionary act does damage a person’s capacity to reclaim that entitlement. Thus, after all, pornography does “subordinate” and “silence” women, and does this by exploiting the illocutionary level of speech, that is, presupposing for the felicity of its illocutionary acts (everyday ones such as reporting, describing, commenting, praising . . . ) something that once presupposed, hinders women from being accepted as satisfying the felicity conditions of certain speech acts they may want to perform.

13 The Austinian Conception of Illocution and Its Implications for Value Judgments and Social Ontology 1. Introduction In this paper I present the Austinian notion of illocutionary effect and discuss some of its philosophical implications as to value judgments and social ontology. Illocutionary acts are characterized by most authors as acts, whose effect is the production of the hearer’s uptake. They aim to be understood or, more specifically, to make the hearer understand the speaker’s communicative intention. This idea is to be found in early work on speech act theory by Strawson (1964) and Searle (1965, 1969). It seems both historically fair and theoretically relevant to mention that it was not Austin’s idea. Indeed, reconstructing Austin’s perspective on illocution can help us see that speech act theory was originally designed to have philosophical implications that have failed to be made explicit ever since. In Section 2, I explain how the notion of illocutionary act as the expression of a communicative intention aiming at its own recognition became part of the mainstream version of speech act theory and what I believe Austin meant to say about illocution with his famous remark about the role of the hearer’s uptake.¹ In Sections 3, 4, and 5 I focus on what the resulting notion of a conventional illocutionary effect contributes to some issues in discourse analysis and in philosophy, particularly concerning value judgments and social ontology.

2. Austinian Illocutionary Effects In introducing three ways in which illocutionary acts are, according to him, connected to effects, Austin writes: Originally published in Ethics & Politics 16 (2): 619–31. Trieste: EUT, 2014. Reprinted in Moral Realism and Political Decisions. Practical Rationality in Contemporary Public Contexts, ed. Gabriele De Anna and Riccardo Martinelli, 135–50. Bamberg: University of Bamberg Press, 2015. ¹ [Section 2 of this Essay summarizes the Austinian view of illocutionary acts and effects which I expounded and argued for in previous writings, including Essays 1, 7, 10, and 11.]

Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0014

220

    

Unless a certain effect is achieved, the illocutionary act will not have been happily, successfully performed. . . . I cannot be said to have warned an audience unless it hears what I say and takes what I say in a certain sense . . . the performance of an illocutionary act involves the securing of uptake.” (1975, 116–17)

Strawson, in his paper on “Intention and Convention in Speech Acts” (1964), reads Austin as implying that no other effect than uptake is necessarily connected to the performance of an illocutionary act. This reading enables him to reinterpret Austin’s illocutionary force as a kind of speaker meaning (in the sense given to this expression by Grice 1957), that is, the intention of the speaker to achieve a response thanks to the recognition by the audience of her intention to achieve it. Searle shares Strawson’s view, but adds to it that the response aimed at should not be thought of as the actual formation of a new belief or other attitude in the audience, which would be a perlocutionary effect as opposed to an illocutionary one, and that the conventionality of language should be recognized as contributing to the understanding of the speaker’s intention on the part of the audience. He then defines the “illocutionary effect” as the understanding of the meaning and force of an utterance (1969, 47). Strawson’s reading of Austin’s notion of uptake later influenced the inferential approach to speech act theory of Kent Bach and R. M. Harnish (Bach and Harnish 1979), and continues to be influential. However, as I have already argued elsewhere (Sbisà 2009a [Essay 11]), Austin merely meant uptake to be a necessary condition for the successful achievement of the core effect of the illocutionary act, and just after the passage we have cited, he describes thus the second effect to which he holds the illocutionary act to be connected: The illocutionary act “takes effect” in certain ways, as distinguished from producing consequences in the sense of bringing about states of affairs in the “normal” way, i.e. changes in the natural course of events. (Austin 1975, 117)

It is clear enough that he means this effect to be produced by all kinds of illocutionary acts (while the third effect he mentions, the inviting of a response, is explicitly attributed to some kinds only). The problem with the illocutionary act’s “taking effect” is that Austin’s characterization of it is vague and more negative than positive (saying what it is not rather than what it is), and exemplified by one case only, the formal and ceremonial act of naming a ship, from which it appears difficult to generalize. I think that the key to understanding what this mysterious effect consists of, lies in considering what way of bringing about a state of affairs is to be contrasted with the “normal” way, that is, with the introduction of a change in the natural course of events. When we do something that interacts with a natural chain of causes and effects, our contribution and its effects come to belong to that chain, on a par with

    

221

its other members: the efficacy of what we do is by natural causation. But in the case of the naming of a ship, the illocutionary effect is that the ship is given a name, so that certain subsequent acts (such as referring to the ship by another name) are out of order. Such an effect is not the output of a natural causal chain, nor of a change we introduce into such a chain. It occurs in a social frame and needs the audience to recognize that a naming procedure was successfully performed. Austin may have been wrong to rely on this example alone, as if it were intuitive that we can generalize from it. But that Austin maintained that illocutionary acts have conventional effects, is already shown by his Rule A1 for the felicity of performative utterances (I recall that illocutionary acts are the kind of actions that performative utterances perform): There must exist an accepted conventional procedure having a certain conventional effect, that procedure to include the uttering of certain words by certain persons in certain circumstances. . . . (Austin 1975, 14, my italics)

Thus, according to Austin, the core effect of an illocutionary act is “conventional”; that is, it is made possible by the social frame and brought about thanks to that kind of agreement between speaker and audience about what is being done, which we may call uptake. The conventionality that according to him is common to all illocutionary acts appears, then, to pertain primarily to effects. It should be noted that in the debate over the conventionality of illocutionary acts, attention has been paid mainly to the means by which the illocutionary act is performed, while the nature of its effects has been neglected. This has led to recognizing as conventional acts only those illocutionary acts whose performance is explicitly and rigidly regulated by extralinguistic conventions, while other illocutionary acts have been analyzed, following Strawson’s suggestion, as intention-based. But if all illocutionary acts are conventional for Austin, it must be so because they all have conventional effects. It is this sense of “being a conventional act” that has to be taken into account in order for us to grasp what Austin wanted to contend, or if we think there is something to be learnt from his conception of illocution.² Of course, what makes an effect conventional is still an open question, which I will have something to say about in Section 2.1. I will then move on (in Section 2.2) to another matter not treated in sufficient detail by Austin: the problem of how to describe the conventional effects of illocutionary acts. ² Ruth Millikan, who defends the conventionality of illocutionary acts, does so on the basis of her conception of conventionality as the repetition of patterns (Millikan 1998b, 2005; see also Kissine 2013, 179–81). Although she focuses, as usual, upon the means by which the act is performed, the resulting view of illocution may be compatible with the Austinian view I favor (for a view based on Millikan’s, which seems to accept some of my points about the nature and dynamics of illocution, see Witek 2015b [cited as “forthcoming” in the original paper]).

222

    

2.1 When Is an Effect “Conventional”? In his preparatory notes for How to Do Things with Words, Austin seems to be worrying about a possible contradiction between his doctrine of infelicities (the flaws in performative utterances that can make them “null and void”) and the received principle that nothing that was done can be made undone (the source of which is in Aristotle’s Nicomachean Ethics, 1139b5–11 (cf. Sbisà 2007a, 465fn [Essay 10, Section 3]). He realizes that, when an utterance designed to perform a certain illocutionary act turns out to be “null and void,” the act that it purported to perform does not hold and its alleged effects vanish. In a way, it seems that something which was done is indeed rendered undone. In fact, by-default agreement about some act and its effects gives provisional reality and efficacy to states of affairs (e.g., a couple’s being man and wife) that later on may be discovered not to hold (e.g., if it turns out that one of the spouses was already married). This provisional, by-default reality is puzzling, but seems to be typical of “conventional” effects. It corresponds approximately to the property of “defeasibility” pertaining, according to Hart (1949), to the ascriptions of rights and of responsibility: the liability to being annulled in particular circumstances. When an illocutionary act is “null and void,” not everything in the speaker’s performance is rendered undone: something was done in any case, and namely, the (flawed) execution of a procedure, and there are effects, though these might be different from those which the execution of that procedure is designed to bring about (e.g., legal responsibility for bigamy). The discovery of infelicity makes the illocutionary act undone insofar as the bringing about of its conventional effect is concerned. In this sense, defeasibility can be seen as the hallmark of conventionality for actions and their effects. Obviously, if the effect brought about by a certain action is to be defeasible, it must be produced in a way which admits of annulment. This way cannot therefore consist in the causal modification of a material object. Austin’s claim that illocutionary acts “take effect” in a way other than by interacting with a natural chain of causes and effects, together with his claim that they cannot be successfully performed (and therefore “take effect”) unless the audience’s uptake is secured (see Section 2 above), suggest that the way in which conventional effects can be brought about is precisely through by-default agreement among the relevant participants as to their being brought about. The “uptake” to be secured by illocutionary acts is, therefore, an agreement of the participants (often implicit or even tacit) upon what has been done. When this agreement occurs (or can be presumed to occur by default, in the absence of any sign of disagreement), the illocutionary act has been successfully performed and its conventional effect has been achieved (in the defeasible mode explained above). There are a whole series of further problems as regards the securing of uptake: notably, whether the actual bringing about of uptake is required or the mere effort of doing all that would

    

223

reasonably lead to actual uptake, and whether it is the actual uptake or intended uptake that determines the illocutionary act performed. I cannot address these problems here, but I would suggest that there are many shades to them. For example, a patently wrong uptake will usually not be intersubjectively shared, and will therefore remain ineffective, while a plausible, not contested actual uptake can reasonably be recognized as selecting the illocutionary act that is actually performed with an utterance displaying vague or ambiguous illocutionary force indicators.

2.2 How to Describe Illocutionary Effects The claim that all illocutionary acts have conventional effects brings to the fore the problem of how to describe illocutionary effects. Quite obviously, and apart from any theoretical motivation and argument, this claim will be plausible only if, for any illocutionary act, one can describe an effect belonging to the conventional kind. In my discussions of this topic, I have proposed that illocutionary effects be described in terms of the distribution of deontic roles among the relevant participants in the situation (1984a [Essay 1], 2006c, 164–7). Interpersonal relations can be described in psychological terms, both cognitive and emotional, but they also have a deontic dimension, relative to what members are allowed or authorized or obliged or committed to do with respect to each other or possibly to third parties too. My proposal is that the illocutionary effect is a change in these aspects of the interpersonal relation, which I call “deontic” inasmuch as they are connected with what one can or cannot or should or should not do (one might also call them “normative,” see Witek 2015b³).⁴ The illocutionary effect is bilateral, since a change in the deontic role of one participant requires a corresponding change in the deontic role of some other one. The variety of illocutionary effects, so intended, can be described on an empirical and linguistic-phenomenological basis, without any pre-conceived symmetry or constraint. The resulting typology is certainly weaker than a theoretical, principled one, but is perfectly suited to the aims of exemplifying conventional effects and of providing discourse analysis with empirically grounded heuristic tools. In such descriptions, we can use the lexicon of modal verbs (“can,” “should,” “ought to”), other deontic verbs (e.g., “oblige,” “commit,” “entitle”) and nouns (e.g., “obligation,” “commitment,” “entitlement,” “duty,” “debt,” “right,” “license”). Not all states that can be described in such terms are the direct ³ [In the original paper, published in 2014, Witek 2015b was cited as “forthcoming.”] ⁴ John Searle too recognized and discussed the deontic dimension of social facts in his The Construction of Social Reality, from a perspective based on his class of Declarations and on “count as” rules (1995, 100–10). He concluded, however, that reference to deontic aspects was not particularly useful for his project.

224

    

effect of an illocutionary act; some may be the effect of non-verbal procedures approximately equivalent to an illocutionary act of a certain kind, others may be deontic-level consequences stemming from illocutionary effects. Here are examples of conventional effects of illocutionary acts, described in terms of the deontic states produced: Marrying: bilateral rights and obligations Naming: semantic rule and social norm (holding for all participants) Promising: commitment (for speaker) vs right (for addressee) Apologizing: exemption (for speaker) vs empowerment (for addressee) Advising: bilateral non-strict obligations Warning: obligation (for addressee) vs exemption (for speaker)⁵ The typical effects of Austin’s illocutionary classes can also be described in these terms, approximately as follows: (i) Verdictives: license to act upon the judgment (for addressee) vs commitment to truth or correctness (for speaker); (ii) Commissives: right to expect a certain kind of behavior from the speaker (for addressee) vs commitment to perform (for speaker); (iii) Exercitives: obligation and (possibly) rights and powers (for addressee) vs commitment to support the addressee’s deontic state (for speaker); (iv) Behabitives: license to act upon the speaker’s expressed state (for the addressee) vs satisfaction of a task or debt (for the speaker); (v) Expositives: rights, obligations, etc. as above, affecting relations internal or relevant to discourse or conversation. Note that by focusing on conventional effects, we find ways of describing illocution other than the lexicon of performative verbs and thus provide the analyst with subtler tools for all those illocutionary acts, the force of which is the result of the combination of a number of indicators and is therefore complex or hybrid. It goes without saying that this way to describe illocution is not bound to assume one-to-one correspondence between sentence type and (intended) illocutionary act. Though not relevant to the topic of this paper, it should be stressed that the way of describing illocutionary effects presented here also preserves the illocution– perlocution distinction, which is sometimes under threat from those approaches that treat all effects of speech acts (other than uptake) as perlocutionary ones ⁵ [In this quick exemplification of illocutionary types, I did not introduce the distinction between participants as actors and as “actants” (see Essay 1). Still, I was and am convinced that a full analysis of illocutionary effects should make that distinction and perhaps elaborate further on it.]

    

225

and therefore external to the illocutionary act. Distinguishing illocution from perlocution is not a matter of contrasting an act with its effects or consequences, but regards instead the kind of effects (material or psychological on the one hand, conventional on the other) that are taken into consideration. Once this is clarified, using the illocution–perlocution distinction in the analysis of discourse and conversation may become easier and more fruitful. Indeed, it should be possible to recognize the mediating role of illocutionary effects between speech and its psychological and behavioral effects.

3. Philosophical Implications for the Value–Fact Distinction If, as I have tried to explain, the production of conventional effects is ultimately grounded in local tacit agreement, various consequences follow as to the foundations of the human social world. Illocution appears as the locus of the “social contract.” The very agreement that something has been done by words—that, after certain words are uttered, the relation between the interlocutors changes in some way—is an instance of “social contract” which might (in principle) be conceived of as radically bottom-up (while of course it is most often the renewal of a previously established social frame). Thus illocution (with language, which makes it possible) enables us to regulate our living together, building up shared environments and (hopefully) reducing the need to resort to brute force and coercion in order to coordinate with one another. The conception of illocution as presented here has also implications for the philosophy of action, since it comprises the non-trivial assumption that performing an action is equivalent to making oneself responsible for its effect (a certain state of affairs in the world); for the conception of value and the value–fact distinction, which is in question whenever the assertive or descriptive use of language is distinguished from or compared to the use of language in assessments, value judgments, and other value-laden speech acts; and for social ontology, since it appears to introduce new kinds of entities into our social world. In this and the remaining sections of this paper, I will touch briefly on some of the implications concerning the value–fact distinction and social ontology. How is moral judgment to be described in this Austinian framework? In his conclusion to his (1975), Austin suggests that “good” is to be considered as at least in part an illocutionary force indicator (1975, 163–4). This reminds us of the theories of ethical language put forward by C. L. Stevenson (1944) or R. M. Hare (1952), in which “good” is said to express approval or prescribe conduct. But is illocutionary force the same as Hare’s “neustic,” or is Austin’s view of moral language a sophisticated variety of emotivism? There are hints at different ideas in Austin’s conclusive chapter. He says that his classification of illocutionary acts, albeit tentative, is enough to “play old Harry” with the “fetish” of the fact/value

226

    

dichotomy (1975, 151). But how could the mere shift of the problem of the analysis of “good” from locutionary meaning to illocutionary force have this effect? If facts are still represented in the “locutionary content,” while values are the effect of choices and preferences and therefore belong with illocutionary force, is the fact/value dichotomy questioned at all? In principle, the fact/value dichotomy can be questioned in two ways: by assimilating values to facts or facts to values. One of these ways is naïve realism about values. Those who embrace it are admirable, because they display unshakable faith in the reality of the Good. But their position is liable to be borrowed and exploited by fundamentalism, since, with naïve realism about values, there comes the assumption that one can know them with at least the same degree of certainty with which we know facts. The other way stresses the human and actional component in factual judgment. We are keen to recognize that value judgments are actions: they involve choice. They are not so arbitrary or subjective as emotional reactions, since they involve the application of criteria, but still there is choice in them, at two levels at least: as to the criteria to be adopted, and as to the appreciation of the relevant details of the situation which is subject to evaluation. But are factual judgments so different? Don’t we use criteria in them too (for example, when the judgment results from an inference)? Moreover, factual assertions, too, depend as to their content on our appreciation of the relevant details of the event or situation which is reported or described. Austin’s way of questioning the fact/value dichotomy is the latter. It is not so much expressed by his attempt to present matters of value as matters of force (as opposed to meaning), as by the way in which he speaks of value in his classification of illocutionary acts. There, he groups all judgments together in the class of Verdictives, which are “the delivering of a finding, official or unofficial, upon evidence or reasons as to value or fact” (153), need not be final (152), and are concerned with something “which is for different reasons hard to be certain about” (152). The last feature is perhaps the most interesting one. Why should acts of judgment address only matters which are for some reason hard to be certain about? Insofar as our knowledge of events and situations in the world is based on our judgments (that is, on verdictive illocutionary acts), Austin depicts it as something not effortless, but involving actual search for evidence, adoption of criteria, or reasoning, which are kinds of active behavior on the part of the speaker, with some degree of choice as to what is to count as evidence, which criterion is correct, or what reasons are good ones. Different agents might reach different findings: we might then compare those findings and prefer the soundest and best grounded one, or, at least, the one which appears as such to us. But the same picture, with slight recontextualization and small adjustments, may be taken to hold for value judgments. It is not clear how Austin intended to deal with those statements of fact that are not about something difficult, unclear and the like. What he clearly wants to stress, though, is that whenever we issue a judgment we

    

227

are engaging in a complex activity. Non-verbal perception may be “direct” and be both phenomenologically understood and verbally reported as the perception of the real object.⁶ But judgment, albeit a source of knowledge, is not a passive reflection of reality. For Austin, our knowledge is no mirror of anything: unlike mirrors, it involves an active stance. It is brought about by action, precisely by verdictive illocutionary acts. If Verdictives, so characterized, may also be concerned with values, should we conclude that value judgments too produce knowledge? Can we say there is knowledge of values, not in the trivial sense in which one can describe some people’s axiology, but intending to admit value judgments as production of knowledge? Austin does not tackle this issue. But in the Austinian perspective we are elaborating, I think we can go so far as to say that there should be room for knowledge of values, insofar as not only Verdictives about facts but also Verdictives about value can be correct or incorrect. What it is for a value judgment to be correct, however, need not be defined in the same way as what is for a statement of fact to be true.

4. Philosophical Implications for Social Ontology Be it as it may with values, Austin’s view of illocution appears to support realism as to deontic states, since those deontic states that illocutionary acts are designed to bring about must be real ones if the act is to be an action at all. Indeed, without an effect, there would be no action. This view, once again, highlights action, and namely, the active production of deontic states by social agents. As we have seen in Section 2, the key to such production is intersubjective agreement. While a stone may be there even if nobody realizes it is there, a state of right or obligation cannot exist unless there is some kind of agreement about its being the case. On certain occasions, this agreement may be cognitive and conscious, but it need not always be so. It is enough for it to be implicit, for example, in the coordination of the lines of conduct of the relevant agents. If a command is complied with without protest or hesitation, it is safe to assume that the state of obligation stemming from it has been agreed upon (indeed, it has been acted upon). If, in complying with a command, an agent protests against being obliged to perform that action, this too reveals the basic acceptance of the speaker’s authority and the binding force of her illocutionary act. Lack of worry about one’s non-compliance may indicate a failure to take the received command seriously or perhaps the refusal to take it as a command at all. Non-verbal, by-default agreement suffices for the deontic state to be brought about and become a component of the situational context of the

⁶ [See Austin 1962 and Sbisà in preparation.]

228

    

current interaction, from which it might even be inherited by other contexts, thus becoming a largely trans-contextual feature of the interpersonal relation among the agents involved. It should be noted that the production of deontic states follows both a forwardlooking and a backward-looking direction. Looking forward, the output of an illocutionary act is a deontic state, which comes into existence thanks to the illocutionary act (as part of its performance). Looking backward, the accomplished illocutionary act and its outcome presuppose the satisfaction of certain conditions about the agent’s status before the performance. If the agent did not have the presupposed status, but the illocutionary act is accepted as such, her status is somehow reassessed and redefined. This is one of the phenomena that Lewis (1979) dubbed “accommodation” and has perhaps not been studied in sufficient depth with respect to illocutionary acts, where it displays paradoxical features. The rise of a leadership (which did not exist at all before the first command of the new leader was recognized as such and obeyed) may be an example of “accommodation”: it is certainly an example of the influence of presuppositions on social relations. The feeling is that it is from the recognition of that command on, and because of the presuppositions of commanding, that the agent starts enjoying authority over his or her addressee. The status of the agent is changed from that moment on, thanks to the presuppositions or felicity conditions of his/her act, which should have been true before that act. Many social realities can be described as sets of deontic states of the agents involved: property, marriage, contracts of employment can all be almost completely specified by listing the rights, obligations, or commitments, and the like that are assigned to their participants. Even complex institutions can be described in terms of what it is that those who participate in them can do or have to do, or can expect others to do, or must not do, and the like. Roles (in a family, in a peer group, or professional ones) often involve a weaker deontic state, that is, the kind of duty that corresponds to other agents’ legitimate expectations. Deontic states such as having a right or an obligation are represented in language as the possession on the part of an agent of “the right to . . .” or “the obligation to . . .”: we speak of rights, obligations, duties, licenses, etc. as of objects of a particular kind, which we may call “deontic objects.” Here, I make no attempt to discuss whether this way of speaking is literal enough to amount to a shared assumption that rights, obligations, duties, licenses, and the like exist as a peculiar category of objects. Laws, for example, are most certainly thought of as something that really exists: perhaps, as existing normative, and therefore deontic, objects. So, one may want to consider deontic states as consisting of the attribution of a deontic object to an agent. It is to be noted that the ability to bring about deontic states (therefore creating deontic objects, if you like) is the basis of our capacity for creating shared environments through language and is therefore central to culture and civilization.

    

229

5. A Puzzle and Its Proposed Solution I conclude by indicating a limitation from which this fascinating perspective suffers. Are deontic states (and objects) steady and permanent enough, as one would expect of the building blocks of our social and cultural reality, or are they constantly liable to cancellation because there might be infelicities in the illocutionary procedures producing them or the speaker-agent carrying out the relevant illocutionary procedure might fail to secure uptake? Certainly, these states and objects appear in the perspective illustrated as being dependent upon human interaction (as well as framing it). It would seem that they cannot come into being, or survive, without the support of intersubjective agreement. Thus, Austin’s notion of illocution might be deemed inadequate for providing social ontology with secure foundations, or at the very least, it is not enough if what we are searching for are agent-independent objects. It is to be noted that deontic objects may even be observer-independent, but are not, and cannot be, agentindependent. Let us recall, though, what we have said above about the Austinian perspective on value judgments. Verdictives produce legitimate claims to knowledge (which, by default, can be taken as expressing knowledge) if they are grounded in the speaker-agent’s recognized competence, but a perfectly felicitous Verdictive may still be wrong. If it is a judgment about facts, it will be either true or false. And as to value judgments, Austin seems to admit of some objective correctness/incorrectness for them too. Now consider an issue of human rights. Imagine a social group in which a child is believed and dealt with as not endowed with any right. The way she is spoken to and about never comprises any right-granting illocution: she is never addressed with permissions, concessions, promises, apologies, or wishes; moreover, no matter what she says (imagine she speaks, nevertheless), she is never taken as performing Verdictives, Exercitives, or Commissives. Has she, then, no right? If by-default intersubjective agreement around her is that she is to be dealt with that way, can this be accepted as a reason to say that so-called human rights do not hold in her case? Of course not. Indeed, the notion of human rights is designed to apply precisely to such cases and to help protect people in such conditions. It can apply, I surmise, because judgments about justice are Verdictives and therefore liable to be correct or incorrect. A judgment to the effect that the way those people deal with that child conforms to justice would be clearly incorrect (whatever they may believe about their own behavior and their reasons). We may conclude that, in fact, the child does have human rights, or perhaps, more precisely, that human rights should be attributed to her. The former way of putting it assumes that her rights are already there, albeit unrecognized. The latter way amounts to claiming that she has a right to human rights. I would prefer the latter way of putting it, as it matches better not only our intuitions about justice but also the perspective on

230

    

rights as deontic objects developed above. It is we, in fact, who both recognize the child’s right to human rights and issue a Verdictive about her deontic state that is both felicitous and (hopefully) correct. The moral is that illocutionary uptake can be accepted and theorized as the basic source of deontic states and objects, without falling into a counterintuitive (if not dangerous) sort of relativism, providing that the correctness/incorrectness of Verdictives is not reduced to a mere matter of intersubjective agreement. This is one reason for having not just one level of assessment of speech acts, but two: in Austin’s terms, this means a felicity/infelicity assessment and an assessment aimed at detecting (objective) correctness/incorrectness. Defeasibility is limited to cases of infelicity, while error and injustice are the targets of our continuous efforts to redress and improve our relations with the world we live in as well as with each other.⁷

⁷ [This paper was read in Bamberg in 2013 at a joint seminar of the Universities of Trieste (Italy) and Bamberg (Germany). I am grateful to my colleagues Riccardo Martinelli and Gabriele De Anna who organized the event, and to the audience for the discussion.]

14 Varieties of Speech Act Norms 1. Introduction The aim of this paper is to explore the field of speech act norms, shedding some light upon their variety, in particular as regards the different roles they play in the dynamics of illocution, that is, in the interactional mechanisms that make it possible for the utterance of one interlocutor to bring about an illocutionary effect, recognized by the other interlocutor. I propose a three-fold distinction between constitutive rules, maxims, and objective requirements. In my perspective, constitutive rules are those speech act norms which, when complied with, enable us to perform the acts they define; they organize procedures or routines that are repeatable and recognizable from one occasion to another and whose function (the production of illocutionary effects) is only exerted against a background of intersubjective agreement. Maxims encode regulative advice for optimal speech act performance in the perspective of the participants in the current verbal interaction. Objective requirements set standards of assessment (specific to each illocutionary act type) that the speech act performed should meet, irrespective of the perspective of the participants, in order to qualify as correct (or proper, or good, or if suitable, true). Since Wittgenstein (1953) and his language games, many have considered speech as a rule-governed activity. Even philosophers and linguists with no sympathy for Wittgenstein’s perspective admit that speech acts follow rules or obey to norms. While at first sight one might think that the rules or norms applying to speech all belong to one and the same kind, this paper argues that it is not so. Linguistic rules such as rules of syntax or lexical semantics and rules of polite verbal behavior do not function in the same way as the rules or norms concerning the performance of illocutionary acts. Moreover, even the latter (upon which this paper will focus attention) do not all work in the same way. When paying attention to differences, one might come to doubt whether all the rules that play a role in accounts of how speech acts work are indeed norms and, therefore, whether there is indeed a single, all-encompassing object of inquiry such as the field of speech act norms. In this paper, I will not discuss in a general Originally published in Normativity and Variety of Speech Actions (Poznań Studies in the Philosophy of the Sciences and the Humanities 112, Special Issue, ed. Maciej Witek and Iwona Witczak-Plisiecka), 23–50. Leiden: Brill, 2019.

Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0015

232

    

way whether all rules are also norms, but I presuppose that “speech act norms” are a field worth investigating and within it, examine those speech act rules that play some kind of normative role with respect to one or other of the aspects or phases of speech act production and understanding (among which I take as salient the performance and recognition of the illocutionary act). I do not assume, however, that all norms have to be formulated as rules and that is why I prefer the label “speech act norms” for my current object of inquiry. Distinctions between the functions that speech act norms may play can be grounded in remarks concerning illocutionary act dynamics and, most importantly, in the description of what happens when the norm is violated. Indeed, when a norm is violated some penalty follows, and the kind of penalty is generally correlated with the function served by the norm. I intend to keep my discussion on norms and their functions separate from the issue of the origin and nature of norms (both in general and with respect to speech acts).¹ I think that the analysis of the norms applying to speech acts as they appear to our consciousness and practical experience is a significant task in itself, regardless of other, more directly explanatory attempts, which in any case presuppose the explananda and their properties.

2. Constitutive Rules Taking inspiration (and borrowing the name) from the speech-act theoretical tradition, I call a first kind of speech act norms “constitutive rules.” Constitutive rules are widely recognized as rules without which a certain act type would not exist and performances of acts of that type could not occur. Acts that are so constituted need not be speech acts, but all speech acts, qua illocutionary acts, must have constitutive rules. The most classic reference for constitutive rules in speech act theory is John Searle’s chapter on constitutive and regulative rules (1969, 33–42). Searle claims that constitutive rules “do not merely regulate, they create or define new forms of behavior,” creating as it were their “very possibility” (Searle 1969, 33). Thus the activities to which these rules apply are logically dependent upon them. While regulative rules establish how something should be done, constitutive rules establish how one can do something and thus, in the case of speech acts, when a certain utterance amounts to the performance of a certain illocutionary act. Searle takes constitutive rules to comprise all the conditions, individually necessary and jointly sufficient, for the felicitous and non-defective performances

¹ A recent discussion of this issue, not unrelated to the speech-act theoretical perspective developed here, is due to Witek 2015b. In the same vein, Witek 2019 argues for a naturalistic account of some normative aspects of illocutionary acts.

    

233

of an illocutionary act and, after specifying the conditions for promising, specifies also those of various other illocutionary act types (Searle 1969, 54–71). It is instructive to compare what he says about constitutive rules with Austin’s discussion of how saying can “make it so” (Austin 1975, 7, 13–38), which also involves the specification of a set of rules, primarily meant to account for performativity in explicitly performative utterances, but actually extendable to the performance of illocutionary acts and therefore functionally similar to Searle’s constitutive rules.² Austin’s rules, however, are cast in a generic form, with variables to be filled in to yield the rules for specific illocutionary acts, for example: There must exist an accepted conventional procedure having a certain conventional effect, that procedure to include the uttering of certain words by certain persons in certain circumstances . . . (Austin 1975, 14)

Obviously, the actual constitutive rules of an illocutionary act should specify the procedure that is required to exist and its effect. Thus, Austin’s rules are not themselves constitutive rules of any illocutionary act, but are templates for sets of such rules. Austin distinguishes three main kinds of rules (or rule templates), which he calls A, B, and Γ: A rules are concerned with the existence of a procedure and the appropriateness of participants and circumstances (they require the existence of a procedure and the appropriateness of persons and circumstances to that procedure), B rules with the execution of the procedure (they require its correctness and completeness), and Γ rules with the sincerity and consistency of the participants. However, Austin does not present these rules as (templates for) jointly sufficient conditions, but leaves the performance of illocutionary act tokens open to unforeseen forms of defeasibility. Thus, one of the main differences between his speech act theory and Searle’s resides in the alleged completeness of the Searlean set of rules as opposed to the advertised incompleteness of that of Austin. It is indeed in the effort to be exhaustive that Searle posits rules requiring, for example, normal conditions of communication to obtain, introduces the “essential” condition, and a further condition we may call “meaning-intention” condition requiring the speaker to entertain a Gricean meaning intention in her utterance of the linguistic means designed for the performance of the act she intends to perform (Searle 1969, 60–1; cf. Grice 1989, 219). The rule requiring normal communication conditions has no counterpart in the Austinian set, because of Austin’s general attitude as regards what is “normal,” which according to him should not be

² That Austin intended those rules to apply to illocutionary acts is clear from the fact that he takes the defects and failures in performing illocutionary acts to be infelicities (that is, defects due to the violation of one or other of the rules in question) (1975, 105–6 and 148) and from his appeal to the same rules while discussing assertion as an illocutionary act (136–8).

234

    

defined positively (as fulfilling a finite set of conditions, which would never be exhaustive), but presumed by default (unless special conditions obtain). Moreover, while in Austin rules of the A and B forms apply to a speaker’s invocation of a procedure, in Searle the invocation of the procedure is itself turned into a condition of successfulness of the procedure by requiring (in the essential condition) the speaker to intend to achieve the procedure’s effect and (in the meaning-intention condition) to utter the words that are part of the procedure.³ Beyond the differences between the two philosophers, their accounts converge on granting the power to make the act null and void (and therefore suspending or “making undone” its effect) (Sbisà 2007a, 465–6 [Essay 10, Section 3]) to a subset of rules comprising, for Austin, rules of the forms A and B, and for Searle, all his rules but the sincerity rule. In their respective speech-act theoretical contexts, these rules are indeed such that success in the act’s performance depends upon compliance with them. They are therefore meant as “constitutive” in the clearest and strongest sense of the word. Other rules that have been viewed as constitutive may not have the abovedescribed property. I am referring in particular to Searle’s sincerity condition (1969, 60), to Austin’s rules (or rule templates) Γ1 and Γ2 (1975, 15, 39–41), and to the Knowledge Rule put forward by Williamson as the norm of assertion (2000, 243). According to Austin, the procedures for performing illocutionary acts specify the psychological state that the speaker invoking them should be in, but noncompliance with this kind of requirements is an abuse of the procedure which cannot make the act null and void (Austin 1975, 16). Likewise, Searle admits that an insincere promise may be a successful promise nevertheless: it does commit the speaker, even if she lacks the intention to do what she promises (Searle 1969, 62).⁴ He agrees that an insincere promise is somewhat defective, but not in such a way as to hinder the promise from doing its job, namely, committing the speaker. Also rules of Austin’s Γ2 kind, requiring appropriate subsequent conduct, do not make the act null and void in case of violation: it would be absurd to absolve breakers of promises and of other speech-act related commitments because of their very non-compliance. According to Williamson’s norm of assertion, the speaker must have knowledge of what she asserts

³ It is controversial whether Austin and Searle are externalist or internalist about the conditions for illocutionary act performance and which approach is preferable (for different readings and opposite evaluations, see Harnish 2009; Sbisà 2002a [Essay 7]). I would argue, however, that the essential condition and the meaning-intention condition at least introduce internalist requirements into Searle’s set of constitutive rules, while Austin’s rules of the A and B kinds only pose externalist requirements. I also find the externalist approach more suitable to dealing with constitutive rules, since in principle, their violation should be publicly detectable. ⁴ Searle does not thereby become an externalist. Even if the intention to do what one promises is not indispensable to making a successful promise, the intention to commit oneself to doing something remains such, since it is required by the essential condition.

    

235

(Williamson 2000, 243). This means that whoever asserts something without knowing that things are so is thereby committing a violation and is subject to criticism. But Williamson grants the utterance to be an assertion nevertheless, and he is certainly right: also in those conditions, the speaker will be committed to the truth of the content of her utterance. According to Williamson’s understanding of constitutive rules, the decisive trait for a rule to be constitutive is whether it contributes to the definition of the act it applies to. Now, sincerity rules (Searle’s sincerity condition, Austin’s Γ1 rule) and Austin’s consistency rule Γ2 are certainly part of the definitions of illocutionary act types. All these rules can be called “constitutive,” if by this we mean that they contribute to defining illocutionary act types. But their violation is never a fatal flaw in the procedure of performing an illocutionary act token: it does not lead to suspending or annulling the illocutionary effect. So, an agent can well perform her illocutionary act successfully (insofar as its effect is concerned) without abiding by them. I will say that rules like these are constitutive in the weak sense, while only a subset of them, those fixing the requirements which must be complied with to ensure the success of the illocutionary act, are constitutive in the strong sense I prefer to use here. Constitutive rules may interact or overlap with both linguistic rules and norms of politeness. They overlap with linguistic rules when they fix the utterance of certain words or of words in a certain syntactic construction as part of the procedure for performing illocutionary acts of a certain type. From the point of view of the rules of language, this amounts to assigning to a certain form of utterance, or to a certain word or expression, the role of an illocutionary indicator. As to politeness norms, they affect the performance of speech acts, just as they affect all interactional behavior. For example, they may impose or recommend the use of indirect forms for the performance of illocutionary acts that might turn out to be “face-threatening,” such as (most famously) requests (Brown and Levinson 1987; for an overview of research on requests, see Walker 2013; for the role of face-work in social interaction, see Goffman 1967, 1971). When the indirect forms that are motivated by politeness become standard ways to perform a given type of illocutionary act in a given language and culture, it would be reasonable to say that they are now part of its “accepted conventional procedure” (Austin 1975, 14): consider, for example, idiomatic formulas for requesting such as “Would you mind . . . ?” or for apologizing such as “Sorry.” Moreover, there are types of illocutionary acts whose function in social interaction is to cope with politeness requirements, repairing or preventing threat or damage to face, or fostering the participants’ positive face. While on the one hand it is a norm of politeness (for example) that participants in a speech event of a certain kind should foster one another’s positive face, on the other, in order for kinds of illocutionary acts which help to achieve this aim, such as wishes or congratulations, to be

236

    

performed (or even to be merely available for performance), there must be constitutive rules for them.⁵

3. Maxims The second kind of speech act norms I would like to distinguish are maxims. They encode advice for optimal communicative behavior from the point of view of the subjects involved. Since, in introducing conversational maxims, Grice refers to Kant (albeit on a somewhat different topic; see Grice 1989, 26), it might be interesting to recall the use Kant made of maxims in his “Appendix to the Transcendental Dialectic” (1998, 603). There, Kant introduces maxims of Reason, sharply distinguished from the categories which are constitutive of judgment and therefore of knowledge, to express the guidelines that Reason follows in its attempts to optimize knowledge. In the sense outlined above, Grice’s conversational maxims stemming from the Cooperative Principle, and the Cooperative Principle itself (Grice 1989, 26–7) are maxims. They serve the optimization of communicative behavior in two main ways: when they directly inspire the speaker’s utterances (Grice 1989, 26) and when speaker and hearer rely on the assumption that they hold in projecting or deriving conversational implicatures (Grice 1989, 32–3). Among the rules or conditions specified by Searle and Austin, Searle’s sincerity condition and Austin’s rules Γ1 and Γ2, which (as I have already argued in Section 2) do not function as constitutive rules, can be taken to be maxims. Maxims are regulative, not constitutive.⁶ Contrary to a recent interpretation, which expresses a recurrent temptation (see Bermejo-Luque 2011), I do not take the Cooperative Principle to be “constitutive of meaning.” In Grice’s perspective one can well say something without abiding by the Cooperative Principle: this happens, for example, when a philosopher discussing perceptual knowledge says “This post-box looks red to me” in circumstances in which she can see very clearly that it is red (Grice 1989, 6, 224–47). Here there is truth-conditional meaning, but no cooperativity, at least in the standard sense in which ordinary conversation is cooperative. But I do not take the Cooperative Principle to be constitutive of conversation either.

⁵ [I have developed these considerations on illocution and politeness in Sbisà 2021b, 164–6.] ⁶ Regulative rules are prescriptive. An anonymous referee has objected to my treatment of Grice’s maxims as “maxims” (in my sense), arguing that Grice’s maxims are descriptive. But Grice conceives of the CP and the maxims specifying it as formulating how it is rational to behave in verbal interaction, and therefore how participants in a conversation should behave (1989, p. 29). By the way, if conversational maxims were merely descriptive, then they would depict how speakers actually behave; but it is not the case that we always follow the maxims.

    

237

Conversation is not a rule-governed activity in the sense in which an activity such as a card game is rule-governed and even less in the sense in which acts such as commands, warnings, promises, apologies are governed by constitutive rules. It is a matter of fact that there are conversational exchanges: that people talk to each other is just a fact the occurrence of which does not depend on definitions and labels. Moreover, when a covert and non-repairable violation of a conversational maxim is discovered, the penalty for the speaker consists of criticism, damage to reputation, or loss of reliability, but does not include making the conversational episode “null and void.” This does not make sense with conversations, since there is no single conventional effect to be associated with conversation as such (while there are conventional effects brought about by single conversational moves qua illocutionary acts). People actually and thoroughly not cooperating with one another cannot be said to have a conversation: they just talk past each other, which, again, is a matter of fact and not the effect of (or penalty for) noncompliance with rules. The sincerity rule associated with illocutionary acts by Searle (1969; 1979) and the illocutionary type-specific versions of Austin’s rule Γ1 are analogous in content (at least when applied to speech acts of the assertive family) to Grice’s first maxim of Quality “Do not say what you believe to be false” (1989, 27). That the sincerity rule cannot be constitutive in our strong sense, as are other speech act norms apparently belonging to the same set, is recognized, as reported above (Section 2), both by Austin and Searle. Austin is explicit in distinguishing his A and B rules from his Γ rules, which read as follows (and again, are ruletemplates rather than rules for specific illocutionary acts): Γ1: where, as often, the procedure is designed for use by persons having certain thoughts, feelings, or intentions, or for the inauguration of certain consequential conduct on the part of any participant, then a person participating in and so invoking the procedure must in fact have those thoughts or feelings, and the participants must intend so to conduct themselves, and further Γ2: must actually so conduct themselves subsequently. (Austin 1975, 15)

The main difference between the A and B rules and Γ rules is that failure to comply with the former may lead to failure to perform the act, while failure to comply with the latter generates defects that make the act liable to certain kinds of criticism (as an abuse of the procedure or as a performance followed by inconsistent behavior). By calling the latter kind of rules “maxims,” I mean to imply that the procedure for achieving a certain conventional effect is optimally performed only if it complies with them too. In Searle (1969, 60) the sincerity condition is included among necessary and sufficient conditions for the successful and non-defective performance of illocutionary acts. Apparently, according to Searle, this condition functions as a

238

    

constitutive rule. So, on the one hand, the mental state that should accompany a certain illocutionary act is part of the very definition of that illocutionary act, but on the other, violating the sincerity rule does not undermine the performance of the illocutionary act: insincere assertion, deceptive promise, etc. are assertion, promise, etc. nevertheless. But the sincerity norm cannot be both a constitutive rule and a maxim. Searle’s solution of the puzzle is to propose a reformulation of the sincerity condition that can work as a constitutive rule while allowing insincere promises to be successfully performed promises: for promises, he replaces the condition “S intends to do A” with the condition “S intends that the utterance of T will make him responsible for intending to do A” (1969, 62). An insincere speaker may comply with the reformulated condition, while she obviously does not comply with the original one. It should be noted that Grice’s approach to sincerity norms is more radical, since in his theory all connections between sincerity and constitutive rules are severed. His super-maxim of Quality requires that the speaker should attempt to provide the addressee with good quality information, namely attempt to say something true (Grice 1989, 27). The more specific maxim recommending sincerity is derived from this general one: indeed, if you say something you do not believe to be true, you are certainly not attempting to say something true. Being sincere, then, amounts to avoiding one kind of hurdle on one’s way towards saying what is true. Austin’s Γ2 rule has no direct counterpart among Grice’s maxims and one may wonder whether we need it as a maxim, since subsequent behavior is in general mandated, directly or indirectly, by the illocutionary effect of the performed speech act. I think it is reasonable to keep it among speech act maxims, however, as consistency with the deontic states assigned by the illocutionary act to the speaker is not automatic, but contributes to optimizing the performance of the speech act from the point of view of the participants. In some respects, maxims of politeness such as “Don’t impose” or “Show solidarity” (see Lakoff 1973), insofar as they apply to speech acts, may also be included in this group of speech act norms: they advise us to select, from among the alternative forms for performing a certain illocutionary act, the form which is most suitable to the face-wants in the situation.

4. Objective Requirements A third kind of speech act norms is what I would like to call “objective requirements.” These are normative standards for “accomplished utterances” (Austin 1975, 140–1), or complete speech act tokens, which take into consideration both their force and their meaning. In proposing a characterization of this kind of norms, I will start from a reassessment of what Austin says about “correspondence with facts.”

    

239

Austin considers the idiom “correspondence with facts” as significant but not to be taken too literally (Austin 1979, 117–33). He does not assume one-to-one correspondence between true propositions and facts, nor does he posit a truthmaker for every sentence or proposition (123). His theory of truth links statements (or assertions) to “historical situations” in the world by means of “demonstrative conventions,” which he conceives of as distinct and even opposed to the “descriptive conventions” in virtue of which sentences represent types of situation (121–2). Thus the “historical situation” with regard to which the assertion is to be assessed is not identified by means of its descriptive characterization by the sentence used in making the assertion. “Correspondence with facts,” therefore, means—in the case of assertions—that the pertinent situation in the world, demonstratively identified, is as the assertion says it is: that it is correct to speak of it as the speaker has done, in the light of the facts, but also of certain elements of the context, among which are the speaker’s goals in making the assertion (Austin 1975, 145). Austin considers “It is true that p” and other formulations of the assessment in the dimension of truth and falsity as giving an assessment of the overall correctness of a statement, or other assertive speech act. I speak of “overall” correctness, because the assessment targets both the locutionary and the illocutionary dimension of the assertion: obviously, it is concerned with its meaning, but it also presupposes the felicitous performance of the act of asserting (in our terms, the satisfaction of its constitutive rules; see Austin 1975, 138, 145). It is (so to say) the “total speech act” that has to be the “right thing” to say and therein do within the “total speech situation” (cf. Austin 1975, 52, 148), with respect to the “historical situation” referred to. And what makes it the “right thing,” is not merely the participants’ belief (or even justified belief ), but also, and primarily, how the world actually is. Austin’s theory of truth has been criticized by Strawson (1950b) and others (see, e.g., Pitcher ed. 1964), mainly for not managing to avoid all the mistakes that correspondence theories are liable to make. I shall not discuss this issue here, nor shall I tackle the issue of the nature and functioning of what Austin calls “demonstrative conventions,” an aspect of his theory that he failed to fully develop, despite its being most central, and one which has often been misunderstood (see for example Johnson 1992; Vision 2004). What I am interested in here is the role Austin assigned to the assessment of speech acts as regards their “correspondence with facts.” Indeed, according to Austin, speech acts other than assertion or statement are also liable to be assessed in the dimension of “correspondence with facts,” that is, to use the terms adopted in this paper, have to comply with their objective requirements. He introduces the topic thus: Can we be sure that stating truly is a different class of assessment from arguing soundly, advising well, judging fairly, and blaming justifiably? Do these not have something to do in complicated ways with facts? (Austin 1975, 142)

240

    

Austin discusses briefly how this kind of assessment applies to acts of judgment (which he calls “Verdictives”) such as estimating or pronouncing, acts of exposition (which he calls “Expositives”) such as arguing and concluding, acts concerned with social behavior (or “Behabitives”) such as praise, blame, and congratulations, and acts of exercising authority (or “Exercitives”) such as warning and advising or even naming and appointing. In my opinion, developing his suggestions in a fullblown speech-act theoretical perspective would involve more effort in the direction of specifying how an assessment considering “correspondence with facts” applies to speech acts of any kind of illocutionary force. For example, when we leave the field of cognitively oriented speech acts to which assertion or statement belong, evaluative terms other than “true” and “false” come in, and the objective requirements they embody differ from truth in ways that need to be further explored. Estimates and assessments (albeit belonging to Verdictives, cf. Austin 1975, 153) are not so much “true” as “right” or “fair”: being right, or fair, is their way of being correct. In order to deem them right or wrong, fair or unfair, we have at least to consider: the speech situation, the situation in the world to which the speech act refers, the pertinent criteria of judgment and their application, possible application precedents, and, finally, the aims for which the estimate or assessment was contributed. But we also have “good” and “bad” advice, “merited” and “unmerited” blame and (I would like to add) “just” or “unjust” orders. In each case what is at issue is not whether the speech act was felicitously performed (a matter of constitutive rules, in the sense of our Section 2), nor whether it was performed optimally in the subjective perspective of the participants (a matter of maxims, in the sense of our Section 3), but whether the speaker was right in performing that speech act for those aims in that context, given how things are in the world. A complete survey of forms of assessment of speech acts of all kinds of illocutionary forces might tell us whether and how speech acts of the various illocutionary classes have to “correspond with facts” in order to comply with their standards of correctness. The requirements or standards of overall correctness for a speech act can be called objective, since they are to be complied with objectively. In order for an assertion to be true, it does not matter what the speaker or the receiver believe, it does not matter even what its evaluator believes: it matters how the assertion, as made in a certain speech situation with a certain descriptive content, actually relates to the historical situation to which it refers. Truth and falsity (as well as other values concerning satisfaction of the pertinent standard or failure to satisfy it) are therefore “mind-transcendent”:⁷ any of the participants in the speech situation might be wrong about whether a certain assertion is true or false, or a certain piece of advice good or bad, and so on. For Austin, certainly, a speech act

⁷ I am borrowing this word from Gauker (1998).

    

241

always occurs in a context, and so does the assessment of a speech act according to its pertinent objective requirements (cf. Austin 1979, 127, 129). In taking the speech situation into account, such an assessment has to consider facts such as the ongoing activity, the participants’ state of knowledge, and the aims of the speaker in issuing the speech act. In this process, the contextual character of the application of objective requirements does not undermine either their objectivity or the objective nature of the assessment: when evaluating correctness (in the case of assertions, truth), if the aims for which the speech act is performed have to be taken into consideration, this is precisely because it is in its speech situation that the speech act must meet, or fail to meet, its pertinent standard.⁸ Disagreements among participants about how to “objectively” assess the same speech act may also be taken to undermine the objective character of the requirements. But there is indeed no clash, because meeting the requirement at issue “objectively” means merely that only one of the disagreeing participants (or at most one) can be right in her application of the relevant assessment criteria.

5. Three Examples Let us now exemplify the proposed three-fold distinction by applying it to a small sample of illocutionary act types: promise, advice, and congratulations.⁹ (i) Promise (a) Constitutive rules: the promised feat must be a future feat of the speaker or to be performed under the speaker’s responsibility; its performance must be in the powers of the speaker; it must also be in the interest of the addressee, or at least be believed by speaker and addressee to be such; some linguistic form must be used such as to make the act of promising recognizable: the explicit performative “I promise that . . . ,” a performative gloss (“That was a promise”), or words clarifying that the speaker is not simply expressing an intention but actually committing herself. ⁸ While Austin’s discussion of the assessment of assertions such as “France is hexagonal,” “Lord Raglan won the battle of Alma,” or “All swans are white” (1975, 142–5) has been a source of inspiration for contextualists (e.g., Carston 2002; Recanati 2004; Travis 2000, 2008), it should be kept in mind that his original considerations focus on the context-dependency of truth/falsity assessments (or in Austin’s terms, of assessments in the dimension of correspondence with facts), not on the context-dependency of expressed propositions [see Sbisà in preparation]. ⁹ [In these examples, I focus on the various kinds of norms and take it for granted that the illocutionary act consists in bringing about a conventional effect on the relationship between the relevant participants. The conventional effects of promising, advising, and congratulating are not dealt with in detail here.]

242

     (b) Maxims: the speaker must have the intention to perform the promised feat; the speaker must behave consistently with the obligation she has undertaken (unless this becomes impossible for reasons beyond her control).

(c) Objective requirements: the promise is objectively correct when the speaker was right in promising that addressee, in that situation, for those aims and with those expected consequences, to perform that feat, and the promise was, therefore, a righteous action. Comments: If it turns out that the feat mentioned in the utterance is not in the speaker’s powers or is not to be performed under the speaker’s responsibility, or that it cannot be performed in the future at all, the speech act cannot be a promise, even if it presents itself as a realization of the procedure of promising. But due to the general tendency to accept speech acts at their face value in default conditions, if it is merely uncertain whether it is or is not in the speaker’s powers to perform the promised feat, the belief that it is so is tacitly accepted or “accommodated” by the audience, thus letting the illocutionary act take effect.¹⁰ Also the belief that the promised feat is in the interest of the addressee may be accommodated by members of the audience other than the addressee, provided the latter does not belie it. Compliance with the maxims is assumed by default if there is no reason to doubt the speaker’s good faith. Of course, time will show whether the maxim concerning consistency is actually complied with. As to objective requirements, they apply to the accomplished promise as its standard of correctness, provided that the action of promising was actually performed. They are concerned both with its locutionary and illocutionary aspects, possibly also covering those perlocutionary effects that are reasonably expected. It should be noted that insincere promises would fail both compliance with the maxims and satisfaction of objective requirements (to the extent to which they are in fact misleading), while sincere promises, albeit complying with the maxims, may still fail to meet the objective requirements. (ii) Advice (a) Constitutive rules: the speaker must have authority over the addressee with respect to the field of activities with which the piece of advice is concerned; the speaker must have competence about the relevant features of the situation, especially those potentially relevant to the achievement of the addressee’s goals; ¹⁰ Constitutive rules, successful performance, and the “accommodation” phenomenon (Lewis 1979; Stalnaker 1999) are closely linked [see also Essay 3]. Accommodation is discussed in Section 7.

    

243

it must be open to the addressee to choose one among various lines of conduct; the speaker’s words must be a realization of the procedure of giving advice, indicating a line of conduct or criteria for choosing it and clarifying that it is best suited to the addressee’s goals, given the constraints and requirements of the situation; when not clear enough from the situation (as when the speaker is officially in charge of giving advice to the addressee) or from the content of the speaker’s utterance, some additional linguistic form must be used to make the act of advising recognizable, such as the explicit performative “I advise . . .” or a performative gloss (“This is my advice”). (b) Maxims: the speaker must believe that it will be good for the addressee, in view of his goals, to behave as advised; the speaker must behave consistently with the piece of advice provided, for example neither hindering nor pre-empting the addressee’s compliance with it. (c) Objective requirements: the accomplished piece of advice is “good” advice if it is apt to help the addressee to achieve or approximate his goals in a manner conforming to the other possible constraints and requirements of the situation. Comments: If it turns out that the speaker has no relevant authority or no relevant competence or that there is no point in advising since the addressee has no choice, the speech act, even if it presents itself as a realization of the procedure of giving advice, cannot be the giving of a piece of advice. But in default conditions the speech act may be taken as successfully performed advice, accommodating the relevant beliefs about the speaker’s competence or the availability to the addressee of more than one course of action. The speaker’s authority may also be accommodated, in informal settings at least. Compliance with the first maxim is assumed by default and compliance with the second is standardly expected. While we may assess a piece of advice as “good” or “bad” already on the basis of our knowledge of the addressee’s situation and our consequent expectations, whether or not it was “good” advice will often become clear over time. (iii) Congratulations (a) Constitutive rules: the addressee must be the author of a certain achievement; the speaker must be in such a relationship to the addressee, that she has an obligation to recognize that achievement;

244

     the words to be uttered must be suitable to fulfilling this obligation: for example “Congratulations on . . .” or some phrase acknowledging the addressee’s achievement. (b) Maxims: the speaker must have a positive attitude towards the addressee’s achievement; the speaker must subsequently behave in ways consistent with the appreciation demonstrated. (c) Objective requirements: congratulations are objectively correct if they are not merely socially expected but actually merited.

Comments: If it turns out that the alleged achievement of the addressee did not occur or is something which the speaker should not be congratulating him on, the speech act does not bring about the effect of an act of congratulating. But in default conditions, any audience of an act of congratulating not previously informed about the achievement of the addressee will accommodate the belief that one such achievement has occurred. The speaker’s obligation to recognize the addressee’s achievement also depends on the speaker putting herself in the position of being so obliged (that is, her obligation toward the addressee can be accommodated). Compliance with the sincerity maxim is assumed (in default conditions), so that the act of congratulating may even be said to “express” appreciation. But usually nobody cares about what the speaker really thinks of the addressee’s achievement, because the point of congratulating lies more in social recognition than in actual appreciation. The consistency maxim licenses certain expectations of the audience, which may be confirmed or disconfirmed with time. As to objective requirements, there is a subtle difference between an act of congratulating being due and therefore appropriate, that is, conforming to its own constitutive rules, and the congratulations being actually deserved. Social conventions about what people should be congratulated on and the kind of relationship holding between speaker and addressee play a far more important role in the former assessment than in the latter.

6. What about Assertion? Since assertions too are speech acts and have illocutionary force, one should expect that the distinctions put forward in this paper apply to them too. We have no room here to tackle the issue of norms of assertion in depth, so I shall give just a few very informal hints of how things appear from the point of view of our distinctions.

    

245

A plausible constitutive rule (in the sense adopted here) would be that the speaker must be in a position to make the assertion. I would consider as a separate constitutive rule the condition that the situation referred to must be one about which an assertion (with that purported content) can be made. While the latter rule appears to me unproblematic, the former gives rise to some problems. As to the “position” required of the speaker, I am inclined to agree with Williamson (2000) that it should involve knowledge. Indeed, the best way to be in a position to make an assertion is to know how things are. In our framework, though, this has the undesired consequence that any time a speaker makes a sincere assertion which then turns out to be false, because she did not actually know how things were (but only thought she knew), she violates the constitutive rules of assertion and therefore has failed to make a real assertion, according to our definition of constitutive rules (her assertion is only apparently such, but fails to produce the expected binding and licensing effects). False assertions, however, are assertions nevertheless, in that they, at least, are binding for the speaker, and addressees acting upon them do so legitimately. In order to get out of the impasse, one can either follow Williamson and weaken one’s conception of constitutive rule, or keep the stronger conception of constitutive rule and weaken the constitutive rules for assertion (cf. Maitra 2011, 280–4). However, weakening knowledge into belief will not do. Belief in the truth of what is asserted is already the object of the maxim of sincerity for assertion, and we have already seen that sincerity rules cannot be constitutive in the strong sense. The only way to weaken Williamson’s Knowledge Rule (see above, Section 2) which is compatible with our framework is to take the speaker to be in a position to assert something when she (because of circumstances and personal competence) has publicly recognizable good chances to produce an objectively correct assertion, namely, an assertion that is true. One might even conceive of adopting an ad hoc technical sense of “knowledge,” to be used to refer to socially or intersubjectively ratified entitlement to make assertions (based for example on acquaintance with data, reasoning ability, recognized methodological competence in the specific field) thereby formulating knowledge for the use of the addressees (for a notion of knowledge so oriented, see Kusch 2002). Maxims for optimal assertions are less problematic. They should require (at least) belief that what is asserted is true and subsequent behavior (both verbal and non-verbal) consistent with the assertion made. Objective requirements, too, come as no surprise: assertions are evaluated in the dimension of truth and falsity. I would like to point out, however, that is it one thing for the uttered sentence to satisfy abstract truth-conditions, but quite another for the assertion made in certain circumstances to be the right speech act to be performed on that occasion. If an assertion is to be true as an assertion (and not as the expression of an opinion or guess), it must be true not by chance, but because it is made by a competent and therefore knowledgeable speaker upon evidence or reasons (one might say with

246

    

Ball 2019 that in asserting we incur an obligation to know the truth of the proposition). This takes us back once again to the Knowledge Rule. It might be said that the Knowledge Rule sets an objective requirement for assertions, alongside with the requirement of truth. One could also add that the Knowledge Rule may influence the maxims for optimal assertion, suggesting that these should include the recommendation that, in order to be fully sincere in asserting, the speaker should be willing to do everything reasonably within her power to make a true assertion. But all these are mere speculations in need of further discussion, and I shall therefore leave aside this topic now and return to themes concerning speech act norms in general.

7. Are There “Rules of Accommodation”? In discussing the examples in Section 5, we saw that the dependency of the performance of the illocutionary act (and of the achievement of its illocutionary effect) upon the fulfillment of the constitutive rules for the relevant illocutionary act type has also a reverse application in actual situations, since in default conditions, there is a general tendency to accept speech acts at their face value: this can be done only if the fulfillment of constitutive rules is tacitly accepted or “accommodated” by the audience. In his well-known paper “Scorekeeping in a language game,” David Lewis (1979) describes this phenomenon and introduces a new kind of rules, which he calls “rules of accommodation.” These rules have been introduced into speech-act theoretical investigations as norms that are applied by speakers and audience in producing and understanding speech acts (cf. for example McGowan 2004). Lewis represents conversation as a game, the moves of which affect “scores” marked on a “scoreboard.” The “kinematics” of the scores, their moving up and down, or entering and exiting from the scoreboard, may be seen as analogous to what other authors writing around the same time began to describe as “context change” (Ballmer 1978, 1981; Gazdar 1981; Isard 1975), as Lewis too mentions in a footnote. Indeed, Lewis includes among the abstract objects that constitute the scores presupposed propositions and deontic roles of the participants (the latter under the name of “permissibility facts”), both components of the context of a speech act. What counts as correct play at each time depends on the scores (the context) at that time, and correct play makes the scores move (the context change). In addition to this, Lewis posits a tendency of the conversational score to evolve in ways that make whatever is done count as correct play. That is accommodation: a matter of context change again, but adapting the scores (the context) according to which a move is to be evaluated to that conversational move, rather than the other way around. As any move in a game and any change in the scores are described by Lewis as connected to

    

247

rules, accommodation too is governed, according to him, by rules of accommodation, of the following form: If at time t something is said that requires component sn of conversational score to have a value in the range r if what is said is to be true, or otherwise acceptable; and if sn does not have a value in the range r just before t; if such-and-such further conditions hold; then at t the score-component sn takes some value in the range r. (Lewis 1979, 347)

Lewis uses accommodation to describe both cases of context change such as those brought about by explicit performatives or formal permissions and prohibitions (which a speech-act theoretical consideration of language use would rather explain in terms of the conventional effects of illocutionary acts, as in Sbisà 1984a [Essay 1], 2002a [Essay 7]), and cases in which, as Witek points out (2015a, 17), the context-changing process adjusts the source score of an act so as to make the act a “binding and felicitous” illocution of a certain type. My issue here is not with the former cases (since I am assuming a speech-act theoretical, non-reductionist framework: see Witek’s considerations in his 2015a, 20), but with the latter. And my concern is not whether to accept that accommodation phenomena exist: of course they do.¹¹ My concern is, instead, whether we need specific rules governing the accommodating context changes, and whether these should be considered as speech act norms mandating those context changes. The framework within which Lewis theorizes about accommodation is influenced by the notion of language game and by the comparison between speech activities and games. It focuses on activities, as opposed to acts which are defined at least in part by their effects, and is therefore distinct from a speech-act theoretical framework which considers the speech act as part of a procedure designed to produce a conventional effect. Witek (2015a) has attempted to bring together the speech-act theoretical framework (and, with it, the context-change perspective) and Lewis’s “scorekeeping” analysis of conversation: he uses the scorekeeping metaphor to deal with the felicity conditions of illocutionary acts as well as with their illocutionary effects. Following Lewis, he distinguishes appropriateness rules (corresponding in part to constitutive rules as described in this paper, Section 2, and in part to what I have called maxims, Section 3), which govern the relationship between scores and moves, from kinematics rules which govern the relationship between moves and scores (and which in our terms are constitutive rules, cf. Section 2, since they establish what conventional effect is achieved by what procedure). He also admits of accommodation rules, which

¹¹ I have myself made lengthy explorations of both informative or persuasive presuppositions (since Sbisà 1979) and the “backward effects” of illocutionary acts on the deontic roles of their speakers (since Sbisà and Fabbri 1980). See Sbisà 1984a [Essay 1], 1999a [Essay 3], 2006c.

248

    

govern the context-changing process by adjusting the source score of an act in order to make the act a successful and appropriate illocution of a certain type (Witek 2015a). He focuses on the accommodation of those components of the score that correspond to the initial position of the speaker (her deontic role: competence, authority, state of debt, and the like), making them be such as required by the type of the illocutionary act that the speaker’s utterance is designed to perform. Here is the form of accommodation rules as revised by Witek: If at time t speaker S makes a binding illocution I, and if the felicity of I requires Austinian personal presupposition p [that is, a certain deontic role of the speaker] to be part of the illocutionary score relative to which I is evaluated, and if p is not part of the score just before time t at which I is made, then—ceteris paribus and within certain limits—p becomes part of the score at t. (2015a, 18)

While maintaining that the phenomenon at issue is quite real, that is, that the initial deontic role of the speaker is in fact often retroactively adjusted in correspondence to the recognition of her illocutionary act as successfully performed, I do not find Witek’s reformulation wholly satisfactory. In adapting Lewis’s formulation, originally concerned with “saying something” and its being “true, or otherwise acceptable,” to illocutionary acts and their conforming to constitutive rules, it squeezes the dynamics of illocution in a way that renders invisible the distinction between the issuing of an utterance and the making of the illocutionary act (which amounts to the achievement of its conventional or “binding” effect) as well as the role of the securing of uptake in the achievement of the illocutionary effect. But leaving aside the details in the formulation of the rule, why should there be such a rule at all? Witek appears to assume, with Lewis, that since there is a change in accommodation too, there must be specific rules governing that change. But while any change may be represented as a function (here, a function from nonaccommodated scores to accommodated scores), it is not the case that every function should be expressed as a rule, or that anything expressed in the form of a rule should act as a norm. In fact, I would argue that there is no need for rules of accommodation, since the dynamics of illocution is enough to explain why the initial context is retroactively adjusted in certain cases. Once the speaker secures uptake—either by uttering words unmistakably recognizable as (part of) the performance of a certain illocutionary procedure, or by otherwise getting the audience to recognize it as such a performance—and unless there is reason to question the satisfaction of one or more of the act’s constitutive rules, the illocutionary effect is achieved and the deontic role required as the initial position of the speaker is attributed to her as the grounds for her act. No rule is needed to achieve the latter effect apart from the constitutive rules governing the illocutionary act type in question.

    

249

Accommodation, then, is a peculiar way in which constitutive rules can be made to function and not a phenomenon governed by specific rules of its own. Such a peculiar way of functioning is governed by general principles, one of which concerns pattern recognition (a pattern can well be recognized from the presentation of some of its parts) and the other the by-default recognition of other minds or subjects. These principles are not norms, but general characteristics of human cognition. It is indeed quite obvious that a pattern that is partially presented may be completed by the observer if the part presented suffices to make it emerge. If the pattern is articulated in temporal terms, the presentation of its central and final parts will be completed by the initial one. As to the recognition of other minds or subjects and the attribution to them of mental states, meaning intentions, and actions, all these processes typically function on a default basis, ceteris paribus or in absence of defeaters. Finally, even if one chose to keep track of the phenomena described above by formulating accommodation rules, these would not act as a new kind of speech act norms, since they would not posit conditions to be satisfied (beyond those already established by constitutive rules) nor advise or mandate appropriate behavior (not already advised or mandated by the dynamics of illocution). Indeed, the motivation underlying accommodation phenomena lies in the fact that the rejection of the speaker’s attempted illocutionary act as a misfire (or of her utterance as otherwise void, uncooperative, or inconsistent) would jeopardize the interactional relationship between the interlocutors, an undesirable outcome in many respects, both emotional and rational. The dynamics of illocution is also enough to explain why there are cases in which accommodation affects the cognitive states of the interlocutors (presupposition accommodation changes the beliefs of the participants, not the facts) and cases in which it affects features of the actual world (albeit at the deontic or institutional level). The difference between these two kinds of accommodation—which Witek (2015a, 20–1) refers to as “subjective” and “objective” accommodation—is more a question of the kinds of contextual elements to be accommodated than of the nature of the accommodation procedure. If the initial state to be accommodated is of a kind that can be introduced into the real world by (undefeated) intersubjective agreement, as are the deontic roles of agents, accommodation is—in absence of defeaters—“objective.” If a defeating situation is the case, the speaker has no access to the accommodated deontic role, even against the beliefs of the participants (which are false). While informal leadership in a peer group can develop by accommodation (once members start accepting a given member’s imperatives as orders: there are no defeaters in this situation), no accommodation can make a thief dressed like a policeman become a real policeman, even if some may be temporarily misled into believing that person to have actual authority to give certain kinds of orders. If the initial state to be accommodated is of a kind that cannot be introduced into the real world by (undefeated)

250

    

intersubjective agreement, accommodation is “subjective” (although one must admit that it remains objectively motivated, since what is needed to make the utterance felicitous is that the accommodated content be actually true). Think for example of the accommodation of the belief that I have a sister, as presupposed by the utterance of “I cannot come to the meeting. I have to meet my sister at the airport”: obviously, the accommodation does not make any non-existent sister of the speaker exist, but merely makes “The speaker has got a sister” part of the common ground for that conversation. However, the utterance of “I have to meet my sister at the airport” is really appropriate and makes a felicitous assertion (and, moreover, gives a felicitous account of the speaker’s absence from the meeting) only if the speaker really has got a sister, while if she has no sister, the accommodated belief notwithstanding, the utterance is liable to be rejected as infelicitous.

8. Are Speech Act Norms Conventional? The conventionality of speech acts is one of the oldest and the most debated issues in speech act theory (since at least Strawson 1964). I would also like to emphasize that it is made particularly complex by the difficulty to identify which notion of conventionality is really pertinent to speech acts, as well as by a basic uncertainty about what it is that is supposed to be either “conventional” or “nonconventional” in speech acts (the procedure for performing it? the illocutionary effect? cf. Sbisà 2007a [Essay 10], 2009a [Essay 11]). I cannot discuss this issue thoroughly here: I would merely like to examine how it interacts with the threefold distinction regarding speech act norms that has been argued for in this paper. First of all, what does it mean for a norm to be “conventional”? Some authors have made sharp distinctions between norms and conventions as if these were two mutually exclusive notions (cf. Pagin 2015, who does not take constitutive rules to be norms). I have already explained, though, why I prefer to use “norms” as a general term for our field of inquiry, thereby encompassing certain rules that many have taken to be “conventional.” So, let us assume that the question is legitimate. As to its possible answers, we may say that a norm is conventional when it is not inescapably imposed on us by some kind of natural or rational necessity, but when its normative powers as well as the choice of its content depend on social agreement (which may vary across times and cultures). This fits with the idea that the procedure by which an illocutionary act is performed may be “conventional” (which would mean, then, that it is fixed by norms depending on social agreement). It may also fit with Millikan’s notion of convention as the repetition of procedures motivated by reference to precedents, already applied to illocutionary acts (Millikan 1998b, 2005; Witek 2015b, 2019), which suggests describing the relevant social agreement as agreement about the precedents and

    

251

the repetition relation. It is also worth pointing out that there are norms, the observance or violation of which is related to achieving or not achieving the conventional effects of a speech act. We shall bear these aspects in mind in the rest of this section. In consideration of our three-fold distinction between constitutive rules, maxims, and objective standards, I would suggest that constitutive rules can be taken to be conventional, while maxims and objective standards (even if they apply to conventional acts) cannot. Constitutive rules (in the sense illustrated here in Section 2) fix the requirements which must be complied with to ensure the success of the illocutionary act. Since success in performing the act depends upon compliance with constitutive rules, if something “goes wrong” with them the performance may turn out to be ineffective and its alleged effect may be null and void. Now, the very liability to annulment manifested by the effects of illocutionary acts confirms their conventional nature, since only conventional (agreement-depending) states of affairs can be annulled or “made undone.” But if we assume that only conventional means can bring about conventional effects (as Austin seems to think: cf. 1975, 119, where he admits of non-conventional means for perlocutionary effects only), then we see that the procedures for performing illocutionary acts must be conventional themselves. This means that the content of our constitutive rules will consist of procedures that are not determined by natural inclinations or rational motivations but are established by social agreement as repetitions of precedents. In this sense, constitutive rules can be taken to be conventional. Some maintain that constitutive rules (also called “constitutive norms”) are not conventional: according to Williamson (2000, 239), they possess a necessary character which appears to exclude any potential choice (taken to be essential to conventionality). There is some truth in this remark. Once a set of constitutive rules for a certain act is fixed, nothing can be an act of that kind if it is not subject to those rules (and, in our framework, if it does not observe them). Thus the constitutive rules for a certain act actually impose upon all those who intend to perform acts of that kind. But one might not want to play that game. Or the social agreement upon that set of rules and its effect may change. Or it may not hold for a different social group. Not all conceivable sets of constitutive rules are compatible with all cultural frameworks. Let us imagine a culture that does not believe people to be able to control their future behavior and therefore take responsibility for future action: it would lack any procedure corresponding to our promise (while it might have weaker forms of commitment, say, expressing intention). Or let us imagine a strictly egalitarian society in which people in need of help would perform acts of request, but orders could not exist. Incidentally, anthropologists have sometimes noticed procedures in distant cultures that we lack (cf. for example Duranti 1992 on the Maaloo exchange in the Samoa Islands). Williamson himself seems to find the conventional view of constitutive rules compatible with his theory insofar as they

252

    

are necessary only for the game or language they apply to: if conventions change, then that game or language turns into another (Williamson 2000, 266). What I would suggest is that the constitutive function, far from being incompatible with conventionality, presupposes it. As to speech act norms other than constitutive rules, I agree with Pagin (2015) that they are not conventional. Certainly, the speech act norms which we have called “maxims” (Section 3) are not compulsory mechanisms and people may observe them or not. But maxims such as those requiring sincerity or consistency (or specifying what sincerity or consistency should consist of with respect to a certain kind of illocutionary act) do not exist because of some arbitrary choice: people do not just agree on them. As Grice says (with reference to his maxims of conversational cooperation; see Grice 1989, 29), it is rational of speakers to follow them. Likewise, consistency is rationally mandated. Moreover, as we have seen, the violation of a maxim does not hinder the conventional effects of speech acts from being brought about.¹² Those speech act norms we have called objective requirements (Section 4) are also not conventional themselves, although they apply to performed speech acts (that is, executions of conventional procedures which achieve conventional effects). The point of a certain type of illocutionary act may be fixed by social agreement, but once it is fixed, the standard applying to it follows and the application of the standard must consider the actual performance in actual circumstances. It cannot therefore be a matter of convention, in any of the senses considered here, whether the function or point of a speech act of a certain kind is served well by a certain speech act token in certain circumstances.

9. Concluding Remarks The picture of speech act norms we have outlined comprises three main kinds: constitutive rules, upon which the performance of illocutionary acts depends; maxims, based on rational motivations, to be followed as requirements for perfection in order to perform a speech act optimally satisfactory in the perspective of the participants; and objective requirements for the overall correctness of the accomplished speech act with regard to the situation in the world to which it relates. These three kinds of norms engender three different sorts of penalty when ¹² [One might object that norms of politeness, which have been assimilated to maxims (see, e.g., Grice 1989, 28; Lakoff 1973), must be conventional because they change across cultures. I think that this is wrong. Not all politeness norms change with cultures (think of facework-related norms: Brown and Levinson 1987; Goffman 1967, 1971) and those that do change either are not speech act norms (and so our considerations about the sense in which a speech act norm may be conventional do not apply to them) or are indeed constitutive of some kind of a speech act. Think of those illocutionary acts whose point is to keep social relationships fair and smooth (such as greeting, thanking, apologizing, and congratulating) or whose linguistic procedures come in various forms, concerned to varying extents with avoiding threat or damage to face. See on this Sbisà 2021b.]

    

253

not complied with. The penalty for the violation of constitutive rules is infelicity, which may lead to failure in performing the act and therefore in bringing about its illocutionary effect. The penalty for failing to observe maxims never leads to the annulment of the performed act, but consists of blame and disrepute. The penalty for not meeting the objective requirements has both an interactional and an objective side. The former consists of the negative assessment of the speech act as not being the right thing to say and do in the circumstances, and of the speaker as being responsible for performing a “wrong” speech act. The latter consists of the mere fact that the speech act is unfit to contribute to the achievement of the goals of the speaker or possibly of other participants in that situation. Of the three kinds of norms, only constitutive rules can be said to be conventional, since they establish procedures that are repeatable and recognizable from one occasion to another and whose function (bringing about changes in the deontic roles of the participants) is only exercised against a background of social agreement. That our distinction is three-fold, however, has a consequence for the received picture of the relation between speech acts and conventions. In place of the received image of a group of “conventional” speech acts distinct from a group of “non-conventional” ones, one may conceive of all speech acts as conventional for certain aspects and non-conventional for others. Clearly, the examples presented in Sections 5 and 6 of how our three-fold distinction may apply to some speech act types are a very limited test of its potential as a descriptive framework and organizer of intuitions and doubts. I am confident, however, that it can provide fruitful and principled ways of organizing discussion about types of illocutionary act as well as about the illocutionary forces of speech act tokens, towards a fuller understanding of the dynamics of illocution.¹³

¹³ An early version of this paper was read at the 9th International Pragmatics Conference (held in Goteborg, July 8–13, 2007) at a workshop on “Speech Act Norms” coordinated by Mitchell Green. I am grateful to the participants for discussion and comments. Since I resumed the project, I have greatly profited from comments and objections by Maciej Witek, whom I would also like to thank wholeheartedly. I thank also Rae Langton for discussion about Lewis on accommodation, and two anonymous referees for their critical comments and suggestions.

15 Ways to Be Concerned with Gender in Philosophy 1. Introduction In this paper, I illustrate some (non-exhaustive) ways in which philosophy can tackle gender. My aim is not to provide an overview of philosophical work on gender and gender-related issues in the past decades, but to single out some kinds of approach to gender, whether actual or possible, and examine the difficulties or limitations of each. So, I will not comment upon feminist philosophical work done in the various areas of philosophy (such as that presented and discussed in Fricker and Hornsby 2000 and Saul 2003). I believe, however, that distinguishing between kinds of approach to gender as I will be trying to do is indirectly relevant also to that kind of philosophical work. Starting from mainly negative critical stances and proceeding towards more positive approaches, I distinguish: (i) (ii) (iii) (iv)

the critique of gender through discourse analysis; the critique of gender essentialism; research on how gender concepts work; reflection upon how gender issues relate to issues of intersubjective recognition.

Each of the four main sections of the paper is devoted to outlining one of these kinds of approach. Since there are close connections between the first and the second, and between the second and the third, these three can be considered as developing one and the same line of thought. The third way to tackle gender draws on recent research concerning the linguistic expression of categorizations, to which gender categories belong. This research, although not specifically about gender, offers a way in to understanding the normativity of gender models as well as gender-based discrimination. As to the fourth way, I propose considering gendered subjectivity in the context of the dynamics of intersubjective relations, framing it in a view of intersubjective recognition as a basic process in human life. Originally published in Phenomenology & Mind 15, “Methods of Philosophy,” ed. Stefano Bacin and Francesca Boccuni: 132–45. Turin: Rosenberg & Sellier, 2018.

Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0016

       

255

Hopefully, this kind of approach might shed some new light on the debates on gender and gender-related issues.

2. Critique of Gender through Discourse Analysis Because of the complexity of gender-related issues in contemporary society, and the fact that assumptions and expectations about gender are usually tacit, it is not easy to spell out what gender is, or even simply what gender (and genders) are commonly held to be. One instrument at hand for obtaining some clarity about how we think about the matter is the examination of how gender is dealt with in discourse and other social practices. Although one may object that such a task does not pertain to philosophy, I maintain that it does (in part at least), for two reasons. Firstly, let us not forget that at a certain time in the twentieth century philosophy was described as an activity of clarification (under the influence of Wittgenstein: see, e.g., 1922, §4.112; 1953, §133) and even though this characterization is now taken to be outdated, I believe that it might still be worth consideration as non-exhaustive, alongside more ambitious ones (for which see, e.g., Williamson 2006, 2007). The idea of philosophy as an activity of clarification is traditionally connected to an anti-metaphysical attitude, so that what is expected to be clarified is philosophical discourse itself, insofar as it expresses unsolvable “philosophical perplexities.” In my opinion, however, the idea can quite sensibly be extended to include all those issues we find difficult to tackle because of underlying conceptual confusion or lack of explicitness. In this sense, though seemingly a matter of applied linguistics, or somewhere between linguistics and sociology, discourse analysis (see Fairclough 1995; Wodak and Chilton eds. 2005) can also be practiced in a philosophical spirit, not only because it sometimes profits from the use of instruments coming from the philosophy of language such as speech acts, presupposition, and implicature but because it throws light upon the contents and organization of our ideas. Aspects of the use of language that are revealing of gender models are studied in the pioneering work of Robin Tolmach Lakoff Language and Woman’s Place (1975). For a use of speech act theory and of presupposition applied to gender-related issues, see for example Hornsby (1994); Hornsby and Langton (1998); and Langton and West (1999). I have used the instruments of pragmatics to make explicit various assumptions about our attitudes towards the feminine gender in my research on Italian women magazines (1976) and on books on pregnancy and birth addressed at women (1984c). Secondly, once our tacit assumptions about some complex and controversial issues and their implications are made explicit, the question remains open as to whether such assumptions are acceptable or even correct. It seems to me that the critical assessment of commonly held assumptions about gender and genders is to

256

       

a large extent a philosophical task, connected to the understanding that human beings have of the human condition, and one to be pursued by argument. Even if it goes beyond mere clarification, this task is at least contiguous to clarification activities, because it is made possible by the explicitness that follows from them. We cannot reject an assumption about gender (or criticize it or argue against it) unless we are aware of it, and we might not even realize it needs to be criticized unless we are aware of the precise role it plays in our discursive and social practices.

3. Critique of Gender Essentialism Many of the assumptions about gender revealed by analyzing discourse and observing social practices have the controversial feature of being “essentialist,” that is, they present themselves as concerning “essential” properties of a certain gender-related class of human beings. Now on the one hand, were this true, alleged members of the class who do not possess such properties should not be counted as real members; on the other, accepting people who lack those properties as members of the class amounts to falsifying the claim that the properties are essential. But this is not what actually happens. In fact, essentialist assumptions bear normative implications, that is, the members of the class of human beings at issue, provided they want to be (or cannot avoid being) members of that class, are accepted as members, expected to have or develop or acquire the alleged essential properties, and negatively evaluated as not “normal” or not “good” members if they do not. Claims about the alleged “essence” of a gender thereby become models to which people who are assigned that gender have to conform. For example, it is often held to be part of the essence of a gender that people belonging to it have certain abilities and typical dispositions, which make them suited to certain roles in society, and this can create in them a normative orientation towards actually assuming those roles. A further essentialist assumption is that belonging to a gender is a fact determined by nature or even a fact of human nature (and thus part of the essence of human beings). These latter assumptions add to the normativity of gender models the compulsion to follow one of them: if belonging to a gender is an essential part of being human, then no human being can escape gender altogether, which means that every one of us will be subject to one or other of the gender models available. Assumptions about gender and genders can be criticized as empirically false generalizations, and they can sometimes be proven to be such. But irrespective of their empirical truth or falsity (namely, whether the majority of the members of a gender actually display the properties supposed to be “essential” to it), they can (and should) be criticized as essentialist, which involves rejecting the idea that they describe the nature or essence of a gender, thereby limiting or suspending their normative import.

       

257

In their attempt to look at women’s difference afresh, and therefore not merely as the lack of allegedly masculine virtues but as rich in positive contents, some feminist thinkers have raised claims about the feminine gender that appeared to have essentialist implications (Gilligan 1982; Irigaray 1977).¹ Many other feminist thinkers, on the other hand, as well as many researchers in linguistics and sociology, have chosen to criticize and reject gender essentialism. Indeed, gender models are quite obviously cultural constructions² and the rhetoric of naturalness is, therefore, both false and politically detrimental. But this anti-essentialist stance is not without problems. To cope with the fact that most human beings have unambiguously sexed sets of chromosomes and bodies, anti-essentialists have used two strategies, one of which involves introducing the sex/gender distinction (see Lorber and Farrell eds. 1991), while the other radicalizes anti-essentialism, considering sexes too as cultural constructions (see, most famously, Butler 1999). As we know, the sex/ gender distinction assumes that in the opposition between male and female, what is natural can be distinguished from what is cultural (and therefore, in principle, subjected to choice). Unfortunately, though, it is empirically difficult, not to say impossible, to tell the difference between what people do because of their nature and what they do because of education, habit, tradition, and the like. Indeed, what appears to be cultural may be analyzed in a “naturalized” way, while what appears to be natural can almost always be modified by training or, when appropriate, by the use of technology. The sex/gender distinction, then, although at first sight clear and almost commonsense, actually turns out to be weak, and this weakness leads anti-essentialists to favor radical solutions, claiming that sex is as cultural as gender after all and that the sex/gender distinction does not hold. That all difference between human males and females is a matter of performance, however, thus belonging to the realm of culture, is not thoroughly convincing. One may wonder, for example, whether a woman’s failure to display certain properties belonging to the feminine gender (for example, a girl ignoring dolls and playing with cars and trains instead) and a woman undergoing surgery to change her sex are to be reasonably considered as graded manifestations of the same kind of process. The toughest point in the debate about essentialism is, however, reproduction. What about reproductive roles in human beings? Firstly, what about the physiological roles that male and female play in reproduction? From the biological point of view, human reproduction resembles that of ¹ Speculum (Irigaray 1974) is not an essentialist book, but in later works (among which Irigaray 1977) this author seems to derive the properties of the feminine subject directly from features of female sex. However, see Braidotti’s comments (1991, 248–63). Although Gilligan is often treated as the prototype of essentialist feminism (see for example Antony 2012), the interpretation of her work is controversial (see Saul 2003, 216–20). ² This assumption works quite well as a framework for the investigation of gender identities and relations in social and communicative interaction. See for example Baron and Kotthoff (eds) 2001; Kotthoff and Wodak (eds) 1997.

258

       

other mammals: male and female reproductive roles are not interchangeable nor symmetric, and (for the time being at least) the functions played in the reproductive processes by male and female individuals can only be partly replaced by technological aids. Secondly, does evolutionary adaptation to male and female reproductive roles determine other properties of male and female individuals? Perhaps not at all, or perhaps to some extent only. Granted, technological progress can turn biological functions into cultural choices. This is apparent nowadays in human reproduction, particularly with respect to motherhood: thanks to technology, women now have the means to decide whether the sexual intercourse they might want to have will be potentially fertile or not, so that conception (and therefore motherhood) can no longer be considered as matters of biological fatality. Other more complex technological interventions make it possible to replace intercourse with artificial insemination, or let a woman host an embryo of which she is not the genetic mother; moreover, after a certain minimum gestation time, fetuses can now survive and develop outside the mother’s womb. All these new techniques open the way to choices that were not “naturally” available, such as having a child from a dead husband, or getting pregnant after the end of one’s childbearing age. In these cases, the obvious contrast between what is done by means of technology and what would happen according to “nature” suggests that on each occasion we face a choice between a technologyguided and a nature-guided line of conduct. But these two kinds of behavior are not “chosen” by people in the same way and to the same extent. In most cases, it is reasonable to assume that what happens without technological intervention has a default status: the issue of choice only arises when something (a precise goal, for example) makes the use of technology pertinent, and so the two situations are not symmetrical. So, while whatever technology does rather obviously pertain to culture, it does not follow that what happens without the use of technology should be counted as cultural as well. After all, it seems only fair to concede that bodies enjoy a sort of givenness which philosophy should not sweep under the carpet but seriously thematize.³ It is quite understandable that people seeking to free themselves from gender models should attempt to shift the biology of reproduction from nature to culture: taking it all to be a cultural matter removes from gender models any possible biological legitimization. There might be other ways of rejecting gender essentialism, however, which do not need such a move. For example, it should be possible to see the nature–culture distinction as a continuous line between two poles, and recognize, in contrast to the way the distinction is construed when turned into a dichotomy, that culture has its determinism and nature its flexibility. As to our ³ More attention to this theme has been paid by phenomenology: a very interesting example is Heinämaa 2014. In my own work (Sbisà 1996), I have put forward a conception of feminine subjectivity as developing in the elaboration of female-sex related experiences.

       

259

bodies, it should be possible to take them as starting points (be these resources or constraints), namely as things that do not speak for themselves but which we elaborate on and assign significance to. Of course, it is no easy task to construct a defense from essentialism along these lines, especially due to the widespread expectation that any theory that is not anti-essentialist must be essentialist, and vice versa.

4. How Do Gender Concepts Work? Given the difficulties in dealing with essentialist assumptions about gender, one would expect that looking at language again would be useful, no longer to specify assumptions about gender making them available to criticism, but to investigate how it is that these assumptions are so pervasive of our experience and activity and tend to be perceived as normative, making one’s gender appear as one’s inescapable fate. Another way to concern ourselves with gender issues in philosophy might therefore be to discuss not so much the content of gender concepts as their nature or structure, and the way they work in cognition and communication. Let us first examine which ways of conceptualizing gender are available in principle. To begin with, gender can be (and often is) conceptualized as dichotomously exhaustive: there are two genders and each individual has to belong to one of them, or even, every property that an individual may have must be part of one or other of the two gender concepts. In this case, individuals have to belong to a gender, and if the latter assumption is also taken to hold, this gender will determine all their properties. This is the most restrictive way of conceptualizing gender. Emblematic of this is the astonishing claim I happened to come across recently on a poster promoting protest against the alleged influence of so-called “gender theory” in Italian educational programs: “boys are male and girls are female.”⁴ To lessen the compulsoriness and pervasiveness of gender, one may conceive of gender concepts as comprising only gender-relevant properties. This makes it possible, at least in theory, that an individual lacking those properties might fail to belong to a gender. Moreover, properties not relevant to gender could be possessed by individuals of any gender and therefore, for example, shared by individuals belonging to different genders. It seems to me that our usual way of conceiving of gender matches this model at least in this respect. Compulsoriness of gender would be also mitigated if a society and its language were to allow for an open number of genders: this would be made possible if the contents of existing gender concepts were conceived of as non-exhaustive of the possible sets of gender relevant properties. However, this way of conceiving genders is not without risks,

⁴ I saw this poster in Rome in September 2018.

260

       

since it fosters the individual’s identification with a gender, thereby indirectly confirming that gender belonging is compulsory after all. As a further alternative, gender belonging may be conceived of as gradable. In this perspective, one individual can be deemed more masculine than another, or even more masculine than feminine (or the other way around), depending on the number of properties he/she possesses belonging either to the masculine gender concept or to the feminine gender concept. Those properties are still either feminine or masculine, and it is their proportion that determines gender belonging. Adopting this perspective enables us to talk of more (or less) feminine women and more (or less) masculine men, and although we may be accepting actual people as “hybrid,” the fixed borders and received contents of gender concepts themselves are confirmed. This brief exploration of possible ways of conceptualizing gender is perhaps enough to show, or at least to strongly suggest, that none of these ways is quite free from the pervasiveness and compulsoriness that make gender concepts such a source of trouble in real life. Clearly, the solution to these troubles cannot be found in the structure of gender concepts considered in isolation, and so it is worth considering whether problems with gender concepts may be rooted in general facts about language and cognition. Gender labels are usually employed to make generic statements such as “Men love soccer” or “Women are emotional.” Recent research on generics, by SarahJane Leslie in particular (Leslie 2007, 2008, 2017),⁵ maintains that such statements are not equivalent to quantified statements, but are held to be true when the property assigned to the category is prevalent in the category or else when, even if not prevalent, the property is characteristic of the category or striking for some other reason (e.g., members of the category that possess it are dangerous, as in “Mosquitoes carry the West Nile virus”). Elaborating upon various experimental findings (showing, for example, that children understand generics as generalizations about kinds or categories before they fully master quantified statements, and that people tend to recall universally quantified statements as generics),⁶ Leslie claims that generics are the most basic way in which cognitive generalizations are expressed in language. So, generics support the fundamental cognitive activity of categorization both in the area of natural sciences (notably biology) and in social cognition, but at a price: the use of generics appears to regularly activate essentialist beliefs,⁷ that is, the category is taken to be grounded in an essence, which means that members of the category are taken to share something that causally grounds their manifestation of certain properties (Leslie 2017). It should ⁵ For an overview, see Leslie and Lerner 2016. ⁶ About children’s understanding of generics, see Gelman 2010; Gelman, Goetz, Sarnecka, and Flukes 2008; Gelman and Raman 2003; Hollander, Gelman, and Star 2002; Leslie 2012. About recalling quantified statements as generics, see Gelman, Sánchez Tapia, and Leslie 2016; Leslie and Gelman 2012. ⁷ Gelman, Ware, and Kleinberg 2010; Rhodes, Leslie, and Tworek 2012.

       

261

be noted that this use of generics does not suggest that the category they are about has fuzzy borders: rather, it makes the classic view of a category as having a perfectly delimited extension compatible with the possibility (and actual existence) of exceptions. Accepting a generic disposes the receiver to infer that any arbitrarily chosen member of that category is endowed with the property expressed by the predicate of the generic. It seems to me that the essentialist beliefs observed by Leslie in association with the use of bare plurals as subjects of generics could also be described as existential presuppositions (and accommodations thereof): the category is presupposed to “exist” as a category, and therefore, to have an essence, and the receiver is implicitly invited to share such a background assumption. It is insofar as a certain category is assumed to exist and be causally grounded (so that its members at least share the disposition to manifest certain properties), that it (or its members, qua its members) may become the target of derogatory assessment and discrimination. All this may explain some of the troubles we incur as speakers or receivers of generic statements about genders. What is explained, though, is not avoided or overcome. If Leslie is right about the way in which generics function, and therefore about how gender concepts work, what should philosophers do with generalizations about social kinds, gender in particular? Should we avoid using gender generics (for example, assertions the subject of which is expressed by bare plurals like “men” and “women”)? Leslie (2017) recommends the use of adjectives in place of nouns, since the former do not activate essentialist beliefs to the same extent as the latter: but this would hardly work with gender, and might even be a waste of energy, distracting us from tackling other sources of gender discrimination (as Saul 2017 warns). Should we, then, examine and reject the gender generics which fail to meet the conditions for a generic statement to be true (for example, the properties they predicate are not appropriate to generalization—not really “striking,” not really functional to the primary social function of the category)? Or would it be enough to argue that the normative understanding of gender generics is ill-founded? Establishing this last point would cancel part of the harmful implications of the use of gender generics, but would still not be enough to stop them being used to discriminate against a whole gender category. At this point, there is a side issue that needs to be mentioned: I have assumed throughout that philosophy has a professional deontology. But does it actually have a deontology and should it have one? And if so, what kind? Some colleagues might say that philosophers should only seek truth and give reasons for those claims that they make in so doing. In the twentieth century, numerous thinkers added to this, or else preferred to say that philosophers should refrain from making claims that are not suitable for being assessed as true or false, and should help non-philosophers distinguish claims that are so suitable from claims that are not. But apart from this direct or indirect relationship to truth, and the appeal to reasons, do philosophers have any further responsibilities as regards the society to

262

       

which they belong? My reply to this question is simple: everybody is responsible for what her speech brings about (illocutions) and to some extent also for their consequences on psychological attitudes and behavior (perlocutions); but even more responsible is whoever knows more or sees farther than others, as should be true of philosophers, or is in a position that can de facto have influence on others, as sometimes happens to them. I therefore think that philosophers facing choices about how to deal with matters that have moral or political implications should care as much about what is right as about what is true. Incidentally, these aims may well not be in competition. On the negative side, a philosopher should avoid licensing implications that may be expected to be harmful. On the positive side, a number of morally or politically relevant goals, such as freedom of individual realization, self-determination (within reasonable limits), smoothness in interpersonal relations, and transparency and reversibility of relational asymmetries, are such that philosophy can and should help in the pursuit of. Returning in the light of these reflections to Leslie’s worries and recommendations about the harmful assumptions associated with social kind and gender generics, I think that they are basically legitimate and worth consideration. Changing people’s linguistic habits by decree is obviously not a viable option, and after all, it is doubtful whether there is any simple way in which it would be worth changing them: but research on how certain harmful assumptions about genders are communicated and how their communication could be avoided or their content questioned can certainly be of some help on our way towards dealing with gender concepts in correct and appropriate ways.

5. Gender Issues and Intersubjective Recognition It seems that we have come to an impasse. We can make socially shared assumptions about gender explicit and criticize these assumptions. We can criticize not just their content but also their essentialist implications. We are aware that certain quite common ways of speaking (the use of gender generics), along with their important cognitive function, may also carry harmful presuppositions about gender categories, play a normative function, and pave the way to discrimination. But all this does not seem to give firm grounds to the rejection of gender discrimination, the refutation of essentialist beliefs about the genders, and the neutralization of the normative import of gender concepts. To these aims, we might profit from another (albeit converging) line of reasoning: I would suggest that philosophy tackle gender issues by considering them as a test of people’s capability and willingness to recognize others as subjects or, if you prefer, as human persons.⁸ ⁸ In what follows I will speak of “subjects,” in order to leave issues about the definition of “person” aside.

       

263

By “intersubjective recognition,” I am referring to the process by which two human individuals recognize each other as subjects. Many hints at this process and its indispensable role can be found in analytic philosophy. Recognition of mental functions, capacities, and dispositions is taken into consideration and even given a central place in accounts provided by various philosophers of the process of understanding another subject’s language and meaning. At the very least, I would like to mention Davidson’s Charity Principle (1973), Dennett’s “intentional stance” (1987), and Grice’s Cooperative Principle (1975).⁹ Intersubjective recognition is not a unilateral process. An attitude attributing subjectivity to an individual is not a private fact about the recognizer but, through her behavior and speech, contributes to establishing the intersubjectively shared space of the encounter and to characterizing its participants as subjects endowed with certain attitudes, dispositions, and other action-related properties. It cannot in general be entertained for long if there is no response or feedback on the part of the recognition’s target. The recognition of the other’s subjectivity, therefore, necessarily involves a framework of intersubjectivity. Being incapable of such recognition (which involves, for example, not realizing that beliefs, intentions, desires, and other attitudes vary across individuals and thus being unable to correctly attribute attitudes to others) is not merely being insensitive, but is actually a symptom of certain kinds of psychiatric disorders (see Baron Cohen 1997; Frith 2003). Local, but systematic, refusal to recognize certain others as subjects, possibly associated with the refusal of granting them human rights—as happens in reduction to slavery, or in genocide—is a unanimously condemned stance. Indeed, it should be pointed out that those who (for some reason of their own) do not want to condemn certain other people for taking such a stance make no attempt to defend them directly, but adopt a negationist strategy instead, attempting to deny that the events manifesting that stance have taken place or that they were an actual manifestation of that stance. There is no doubt that a responsible philosophy, in analyzing subject–other relations, should foster practices of intersubjective recognition, rather than hinder them, and the first thing to do is to pay attention to how they work. Recognizing someone as a subject involves assigning her various kinds of competence: attitudes, dispositions, but also rights, obligations, and other deontic properties (amounting to what may be called, adapting a notion from narrative semiotics, the subject’s “deontic competence”)¹⁰ that sometimes correspond to

⁹ I have touched upon this theme on various occasions (most recently in Sbisà 2013b). The idea of intersubjective recognition would also profit from a comparison with the theme of intersubjectivity in phenomenology, but the task is so huge that I am in no position to even begin hinting at it. ¹⁰ In narrative semiotics (Greimas 1970, 1983; Greimas and Courtés 1979), the dispositions and the deontic properties that prompt the subject’s action or enable her to act constitute her “modal competence” for acting. Focusing upon an agent’s deontic properties, which in my perspective differ from attitudes or dispositions because they are granted and withdrawn on the basis of social or

264

       

some kind of specific social status. It is tempting to think of this recognition as an all-in-one attribution, but while that may be so by default, in cases in which there are hurdles or problems either on the part of the recognizing subject or on the part of the subject to be recognized, in order to understand what is going on it is necessary to analyze various levels of attribution that may not all be active at the same time. Recognition of the other as a subject, then, consists at the very least of recognizing her as a center of perspective, while further steps consist of recognizing her as capable of (or actually exercising) agency, as a speaker mastering a language and expressing meaning by it, and, finally, as a speaker-agent of illocutionary acts, endowed with the deontic competence required by the procedure invoked, and affecting, by her acts, the deontic competence of addressees or other target participants. Recognition of another human being as a subject is completely achieved when it involves all these levels of subjectivity and, moreover, is granted unconditionally. Indeed, it can only be flawed when given under a condition: for example, if I take you to be an agent only insofar as you do certain things I want to be done, or to be a speaker insofar as that is indispensable to the exercise of a certain subordinate social role, or to be a center of perspective only when you perform your gender role in a way functional to my needs. When recognition is conditional in these ways, it can be suddenly withdrawn in ways which show that it was never actual and effective. It becomes apparent that I never really expected you to have your own perspective on your relationship to me, never really thought you were capable of autonomous decision-making, and never would grant that you be in a position to perform authority-presupposing illocutionary acts. When the other is recognized as a subject because he or she is similar to us, we are faced with an ambiguous kind of recognition. A recognition based upon the assumption that the subject recognized is similar to the recognizer would indeed be a conditional one and would therefore be flawed. But what if the cognitive bases of recognition, consisting in an ability to attribute attitudes and agency to other people, should involve some kind of simulation of the others’ stance and perspective (as in simulation theory, see Goldman 2006), on the basis of our own? Would this amount to making every recognition dependent on similarities between the subject to be recognized and the recognizing one? In many cases, perhaps, it will be enough if the interlocutors’ capacities to entertain attitudes and to act are considered similar, or even just parallel. Moreover, if it is true, as research on the role of mirror neurons in social cognition suggests (see Rizzolatti and Craighero 2004; Rizzolatti and Sinigaglia 2007), that what we do is simply use the same action schemes both to project our own action and to understand what another agent is doing, this clearly does not require any previous assessment of similarities interpersonal agreement, I use the label “deontic competence.” In my work on illocutionary acts since the 1980s (1984 [Essay 1], 1989a), I have connected changes in the deontic competence of participants to illocutionary acts as their conventional effects.

       

265

between recognizer and recognized. Rather, the feeling of a common belonging to humankind (or to the animal kingdom, or to life in the universe) will come later. But if there is a category of which both the recognizing and the recognized subject are held to be members, and this membership (involving participation in the category’s “essence”) works as a condition for the recognition, that recognition is flawed. I would say that women have long suffered from being recognized as subjects both conditionally and partially (namely, not at all levels). This may happen to men, too, in certain circumstances and for specific motivations; with racism, by the way, it happens to both genders at once; but for women, it is a chronic condition. The “silencing” of women, so much discussed in the philosophy of language and in political philosophy (Langton 1993 and Hornsby and Langton 1998 are key contributions to this debate),¹¹ can be seen as a manifestation of the failure to recognize the other’s subjectivity and deontic competence unconditionally: a typical and often dramatic case of “silencing” occurs when a woman refuses a man’s sexual advances, but the force of her utterance and the attitudes and the deontic competence it involves, both as regards the preconditions and the result of her speech act, are not recognized. It is no coincidence that the women’s movement of the 1970s based feminine autonomy upon unconditional intersubjective recognition occurring among women: the words used for describing this process were less abstract and more emotional but, insofar as it worked, that was what was going on. The issue of intersubjective recognition could be further investigated by considering it in the narrative dimension. I am not referring here to narrative theories of the subject in the philosophy of mind, but to the analytical techniques and terminology of narrative semiotics (Greimas 1970, 1983; Greimas and Courtés 1979). There, the Subject is first of all a narrative “actant,” that is, a narrative element defined for its functions in the narrative syntax: receiving a task, carrying it out, and being rewarded. The Subject is then enriched by the attribution of dispositions and deontic properties, to be turned into a more specific “actantial role,” and is manifested in discourse by an agent (an “actor”) endowed with descriptive characteristics. Whatever the details of the analytic terminology employed, in narrative semiotics the role of Subject is not one that can be played in isolation: another function in the narrative syntax is required to be at work in order for the Subject to receive from it its task, as well as its reward. That function, called the “Destinator,” is therefore indispensable to there being a Subject at all. It is the Destinator who assigns the Subject his or her task and after the performance recognizes the Subject as the performer. In the terms used here, the Subject of a narrative exists as such insofar as its Destinator recognizes it. This confirms that ¹¹ Contributions to the debate include: Bird 2002; Caponetto 2017; Hornsby 2011; Jacobson 1995; Maitra 2009; McGowan 2009, 2017; McGowan et al. 2011; Mikkola 2011; Sbisà 2009b [Essay 12].

266

       

our understanding of what it is to be a Subject (insofar as the structure of narratives reflects it) involves recognition of subjectivity as an indispensable part of the picture. A problem is raised, though, by the fact that the standard Destinator–Subject relationship appears to be mono-directional and hierarchical and therefore not suited to account for the intersubjective character of recognition of subjectivity. Not only are the functions of Destinator and Subject such that it is the Subject that depends on the Destinator, not the converse, but also this functional asymmetry is typically reflected in the descriptive characteristics that Destinator and Subject are endowed with: the Destinator, which is also the source of the values of the narrative, does not belong to the same sphere of activity as the Subject, but is presented as “transcendent” with respect to the Subject and its world. But intersubjective recognition, as I have presented it here, is realized completely only if it is bi-directional. The kind of narrative underlying actual intersubjective recognition should therefore envisage a Destinator whose manipulation and assessment activities are carried out in the same environment as the Subject’s performance, so that in principle agents expressing Destinator and Subject can switch their roles. Not by chance, it is especially in narratives in which the role of Subject is played by a female agent that this non-transcendence of the Destinator manifests itself: while, on the one hand, this confirms that the feminine gender has no access to transcendent sources of values, on the other, it opens up the relationship of Subject and Destinator to reversibility (on this, see Sbisà 2017b, 41–2) and to the bi-directionality of intersubjective recognition. To sum up, my suggestion is that further reflection on the ways in which intersubjective recognition is achieved and made manifest (or on the ways in which it fails and is revealed as flawed) can throw light on the dynamics of human relations, both positive and negative, and among these, on gender relations.¹² This reflection could also elicit an increased awareness of why gender discrimination is ill-founded and should be stopped, why essentialist beliefs about the genders should be abandoned, and why gender concepts should not be granted normativity. In fact, discrimination is a process by which a subject is denied full recognition because of his or her belonging, or not belonging, to a certain group. Essentialist beliefs pave the way to discrimination or at any rate, make recognition conditional on the subject having a certain essence. Lastly, the normative function of gender concepts might turn out to arise from the pressure put on the subject seeking recognition by the conditions on which the alleged recognition would be granted.¹³ ¹² Notwithstanding the great differences in the terms in which the two proposals are expressed, there may be similarities or even some complementarity between tackling gender issues in terms of recognition of subjectivity or failure thereof, as I have been suggesting to do in this section, and Mikkola’s humanist feminism as outlined in her work on dehumanization (2016). ¹³ [This paper was written as a contribution to the conference “Methods of Philosophy,” organized by the University Vita-Salute S. Raffaele (Milan) in 2017. I thank Elisabetta Sacchi, Sarah Songhorian, and Stefano Bacin for inviting me, the audience at the conference for the discussion, and Laura Caponetto for further discussion about the “silencing” of women.]

16 Assertion among the Speech Acts 1. Assertion as Illocution How is assertion collocated among the other speech acts? To reply to this question in speech-act theoretical terms may involve considering various ways in which speech acts can be described and classified. First of all, there is the locutionary, illocutionary, perlocutionary distinction. I do not assume that “speech act” and “illocutionary act” are interchangeable terms (as argued by Searle 1969, 22–3):¹ “speech act” is broader than “illocutionary act,” since it does not specify under what respect the speaker’s utterance is considered or what effect of it is highlighted. However, in this paper I will not discuss how assertion is collocated with respect to the locutionary– illocutionary–perlocutionary distinction, but I will take it that, as has been unanimously maintained in speech act theory since Austin (1975), assertion is an illocutionary act. I will also assume that the various views of illocution elaborated within speech act theory are the more insightful as to the phenomenon that Austin intended to bring to light, the more they prove able to describe performances of illocutionary acts as actions that bring about changes in the speech situation (cf. Sbisà 2002a [Essay 7], 2007a [Essay 10], 2019 [Essay 14]). Once we take assertion to be an illocutionary act, the question arises of how it relates to other illocutionary acts. This is the main issue that will be tackled in this paper, and it is two-fold. We will examine how assertion relates to illocutionary acts that are in some way similar to it, at least as to their involving the utterance of plain declarative sentences, and how assertion and its cognates should best be collocated within the whole gamut of illocutionary acts. The former exploration will rely upon a largely intuitive grasp of the “family” of assertive illocutionary acts; the latter will involve both a fuller characterization of assertion and reconsideration of illocutionary act classification.

Originally published in The Oxford Handbook of Assertion, ed. Sanford Goldberg, 159–78. Oxford: Oxford University Press, 2020. ¹ The fact that any performance of an illocutionary act is also a performance of a locutionary act, since the illocutionary act must have a content and this is expressed by a locutionary act, is not a final reason to identify the two act types.

Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0017

268

    

In conclusion, I will turn to the question of the role or rank of assertion among illocutionary acts: whether it is “just” one among them, or there are reasons for granting it some primacy or some special function.

2. Assertion within the Assertive Family Assertion as an illocutionary act is performed in uttering a plain declarative sentence. This means that the assertion is not the uttering of such a sentence, but another kind of act, in some way embodied in that uttering. Most versions of speech act theory would agree that assertion presents a proposition as true (see in particular Alston 2000, 4; but also Vanderveken 1990, 22) and that it commits the speaker to that proposition’s being true (Searle 1979, 2). All versions of speech act theory agree that assertion expresses belief (this feature is particularly central in the perspective of Bach and Harnish 1979). Given this sketchy picture, which we will enrich as we go on, we already know enough to identify a group of speech acts that either are varieties of assertion or are similar to assertion, which may be referred to as “the assertive family.”² Such a group includes speech acts whose characteristic linguistic form is the plain declarative sentence, such as describing or reporting, but also expressing an opinion and making a guess; the borders of the group are somewhat fuzzy, due to borderline cases. There are differences among these speech acts, which have been described in various ways that we will shortly reconsider here.

2.1 Members of the Assertive Family and Their Differences Let us begin by considering what illocutionary acts have been taken to be similar to assertion or even a species thereof. Searle (1969, 66) takes affirming and stating to have the very same felicity conditions as asserting. Searle (1979, 12–13) mentions as members of his class of Assertives the acts of stating, insisting, swearing (that p), hypothesizing, suggesting, concluding, deducing, boasting, and complaining: all these, he notices, can be literally characterized as true or false. While, on the one hand, they do not reduce to the “simple” act of making an assertion, on the other hand, there are many more illocutionary acts that may be held to share their main characteristics, particularly declarative sentence form and truth-aptness. So, Searle’s list of cognates of assertion is expanded by Vanderveken (1990, 169) in the form of a list of assertive performative verbs. Alston’s list of

² This label is used by Green (2013) with a precise definition inspired by Searle’s class of Assertives. Here I am using the word “family” in a sense closer to Wittgenstein’s (1953, §67), because I want to avoid assuming that we have to do with a perfectly delimited category of illocutionary acts.

    

269

Assertives (2000, 125–9), instead, is derived from Austin’s list of Expositives (illocutionary acts amounting to acts of exposition and discourse organization) (1975, 161–3). Not all of Austin’s Expositives involve the uttering of declarative sentences and therefore the presentation of a proposition as true, but of course only those that enjoy these properties can be assimilated to assertion. There is some obvious overlap between the designations of Assertives collected by Vanderveken and the selection of illocutionary acts similar to assertion made by Alston, but coincidence is far from being complete. Both lists appear to be heterogeneous and contain borderline cases not only with respect to the assertive family but to illocutionary acts in general. The same can be said of Assertives in the classification elaborated by Bach and Harnish, where Assertives are a subclass of the broader class of Constatives (1979, 42–6). Is disclosing an illocutionary act (Alston 2000, 126), or rather is it an act that we perform by de facto revealing something but may announce before performing it (Austin’s “suiting the action to the word”: 1975, 81–2)? Is warning (Alston 2000, 126; Vanderveken 1990, 169) really a cognate of assertion or is it, even if it may involve the use of a declarative sentence, a quite different illocutionary act? Are alarming (Vanderveken 1990, 169) or indicating (Bach and Harnish 1979, 42) illocutionary acts at all? The former is cited by Austin as a perlocutionary act (1975, 118), and the latter appears to belong to implicit communication rather than to illocution. Limiting our consideration to clear cases of illocution, we turn now to paying attention to the various kinds of differences that can be noticed within the assertive family. One kind of such differences consists in genus–species relations. Attention for this kind of differences inspires the description of the speech acts belonging to the assertive family (and surroundings) by Bach and Harnish (1979) and Alston (2000). For Bach and Harnish, all Constatives (a genus) share the characteristic of expressing both a belief and an intention that the hearer form, or continue to hold, a like belief (42): differences between kinds of Constatives (their species) are accounted for in terms of variations in the contents and the contextual relevance of the belief and the intention expressed. Assertives are one subclass of Constatives (a species); other subclasses include Predictives, Descriptives, Ascriptives, Informatives, Concessives, Suggestives, and Suppositives. Members of the same subclass all designate varieties of one and the same illocutionary act; so all members of the assertive subclass are assertions, while members of the other subclasses are not, properly speaking, “assertions,” but their illocutionary force is defined by the specific kinds of belief and intention that characterize their subclass. Alston (2000) uses a similar model in a different way: for him, assertion is a genus, comprising all Assertives, and other Assertives are its species, since their differences depend on additional conditions for the satisfaction of which the speaker may take responsibility when making an assertion. It is not clear

270

    

whether assertion, beyond being the genus, is also a species (with zero additional conditions): it seems to me that Alston thinks so. Also some remarks about assertions made by Grice and by Austin trace differences of the genus–species kinds. Grice (1989, 120–2) describes conventional implicatures as connected to “non-central” speech acts, that is, speech acts that share with “central” ones such as assertion one and the same kind of core intention (e.g., for assertion, the intention to make the hearer believe something— cf. 1989, 220—or in a later, amended formulation, to make the hearer believe that the speaker believes something, cf. 1989, 123), but the indicators for those acts express also other features of the function that the utterance is meant to play in the circumstances. Thus, “therefore” introduces a conclusion from premises that have already been asserted or are assumed to be true, so that its use activates the conventional implicature that the alleged conclusion follows from the premises (cf. 1989, 25–6). That is, in the case of “He is an Englishman, therefore he is brave” (Grice 1989, 25), “He is brave because he is an Englishman” is conventionally implicated just insofar as “therefore” indicates the non-central speech act of drawing a conclusion. Grice prefers to describe these phenomena in terms of implicating as opposed to saying, but to put it in speech-act theoretical terms, he is dealing with Austinian Expositives, and we can consider the “non-central” speech acts connected to his conventional implicatures as species of the genus assertion. As to Austin, his paper “How to Talk” (1953)³ deals with differences internal to the assertive family. There, Austin outlines an analysis of what we do in “simple” cases of predication (which are also cases of assertion: the copula “is a . . .” is called “assertive link”) such as saying that item 1222 is an F. Basically, Austin distinguishes cases in which we identify a referent and look for a type that can be applied to it from cases in which we are interested in finding a referent that belongs to a type; both actions can be embodied in the utterance of sentences such as “1222 is an F” but are different nevertheless. Austin also examines the link between predicates and types of items in the world (speaking of “names” whose “sense” matches a type), distinguishing cases in which we assign a name to an item (or the other way around), taking the sense of the name for granted and assuming that that sense matches the actual type of the item, from cases in which we assign a name to an item (or the other way around), taking the type of the item as given and assuming that the sense of the name used matches that type. In these terms, he distinguishes between stating, instancing, and identifying (of the last, there are two different varieties), and in contexts in which the simplifying presumption that there are as many senses of our names as types of items in the world is dropped,

³ Austin must have written this paper immediately after his first draft of How to Do Things with Words (1951–2). There are no references to the topics it deals with in the manuscript notes for How to Do Things with Words. An insightful analysis of it is provided by Fiengo (2017); see also Fiengo 2020 [and Sbisà in preparation].

    

271

also calling, exemplifying, describing, and classing. The most famous assertive speech act, statement, is therefore defined as a speech act to perform which we have to assign a type to a given item by finding a name whose sense matches that type (Austin 1953, 142–3, 145). As to assertion, in Austin’s account there is a sense in which all of the eight speech acts he defines are species of assertion, which should be therefore counted as a genus. This has the further consequence that “assert” cannot designate a full-fledged illocutionary act: the kind of judgment actually involved will always be one or other among those that the structure of language and the way speakers organize the speech situation make available. Perhaps it is not by chance that “assert” is not present in Austin’s classification of illocutionary acts and that “I assert that” is hardly ever cited by him as a performative formula. Searle’s Assertives include illocutionary acts that for various reasons one would not ordinarily call “assertions” (or deal with as with assertions), such as suggesting, hypothesizing, complaining, and deducing (which are indeed classified by Bach and Harnish as constituting subclasses of their Constatives other than that of Assertives). The description of differences between illocutionary acts belonging to the same illocutionary class, including differences among Assertives, is tackled by Searle and Vanderveken, relying mainly upon what they call the “degree of strength” of illocutionary force (1985, 98–9). In Searle and Vanderveken’s model of the illocutionary act, degree of strength affects illocutionary points, but also sincerity conditions; the notion can, moreover, be used beyond their model, to cope with the intuition that various elements of the procedure which is constitutive of an illocutionary act, such as the initial qualifications of the speaker or the illocutionary act’s result upon the speaker–hearer relation, can be graded (as I have argued in Sbisà 2001a [Essay 6]). Searle and Vanderveken claim that connections between illocutionary acts with the same illocutionary point but different degree of strength are regulated by “illocutionary entailment” (1985, 129–37); however, in their use of this notion, they oscillate between the implausible claim that a speaker who performs an illocutionary act is also performing all the other illocutionary acts which have the same illocutionary point, but a lower degree of strength, and an analogous claim, much weaker, related to sentences or even illocutionary verbs (Vanderveken 1990, 12, 50–3). In the former interpretation, a speaker who orders someone to do p also requests him to do p, and a speaker who swears that p also asserts that p. But it is not explained what performing these two acts in one and the same utterance amounts to; indeed, it cannot be said that there are two performances, or even merely two effects. In the latter interpretation, illocutionary entailment appears to be a semantic relation, such that if someone has sworn that p, we are entitled to describe her performance also in weaker, or more general, terms as an assertion that p (which amounts anyway, according to Searle and Vanderveken’s definition, to assigning to that speaker a commitment to the truth of p). In Searle and Vanderveken’s framework, not all Assertives are

272

    

assertions, but only those which have a degree of strength equal or higher than the neutral one. In addition, their theory appears to permit inferences from the attribution of a non-mitigated assertive force to that of a mitigated one. But saying that the speaker who has stated that p is also responsible for suggesting that p is like saying on a hot summer day that it is warm outside: not all semantic scales function like the scale; “cold,” “warm,” and “hot” are mutually exclusive as are shades of color, and so should be taken to be attributions of illocutionary force. Finally, in Searle and Vanderveken (1985, 110–12) we find also the rather vague and flexible notion of “mode of achievement” of the illocutionary point of the speech act. According to Searle and Vanderveken, there are Assertives that differ from assertion for the more specific character of their mode of achievement: these too are connected to assertion by illocutionary entailment, so that whoever performs one of these illocutionary acts also makes an assertion. Thus, we are sent back to the genus–species model once more. But I would like to notice that mode of achievement is the only element in the model of Searle and Vanderveken in which one finds a trace of Austin’s idea that the performance of an illocutionary act presupposes there being a procedure, which the speaker has to execute correctly and completely enough in order to bring about the effect that the illocutionary act is designed to produce (1975, 14–15; cf. Sbisà 2019 [Essay 14]). Variations in the procedure may correspond to major changes in the overall significance of a speech act, making aspects that do not pertain to the assertive family as salient as or even more salient than those that pertain to it. So, for example, postulating (taken to be an assertive by Vanderveken 1990, 172), considered not merely as the utterance of a sentence in a particular context but as an act performed by executing a procedure which has an effect, reveals stipulative and impositive aspects which make it quite different from the making of an assertion. Other examples might be announcing, defining, and even reporting and testifying. None of these is an act of judgment (as clearly are the acts considered as species of assertion in Austin 1953, 134–53) and the former two, albeit their execution usually involves the uttering of a declarative sentence, are too far from the procedure of making an assertion even to be taken as members of the assertive family.

2.2 The Assertive Family: Limits and Structure Drawing some morals from the explorations just made, I would now like to formulate some suggestions about the limits and internal structure of the “assertive family” as well as the position that assertion has in it. As to the limits, I would not admit as members of the assertive family speech acts that are not clearly illocutionary. Moreover, I would limit membership to illocutionary acts,

    

273

performed in uttering declarative sentences, whose effects are concerned with discourse or verbal interaction. I would also follow tradition in excluding value judgments. I do not think that value judgments are acts of approval or disapproval, recommendations, or prescriptions in disguise (see, e.g., Hare 1952; Stevenson 1944). Still, their function in verbal interaction is not the same as that of judgments of fact, as can be seen from the contents of the commitments they produce (which may involve lines of conduct) or from the different kinds of disagreement that arise in relation to them (which often involve criteria of judgment as opposed to their application). As to internal structure, the assertive family appears to comprise illocutionary acts that are assertions (because they are species or varieties of assertion) and illocutionary acts that cannot be properly called assertions but bear various kinds of relations to the making of an assertion. Members of the assertive family that can be properly called assertions include species of the genus assertion which consist of acts of judgment that have different internal articulation, taking different aspects of the speech situation for granted or inviting different kinds of challenge. They also include illocutionary acts that alongside being recognizable as assertions play various specific roles in the organization of discourse, for example as premises, conclusions, objections, or clarifications. Whatever one may say of the effects of assertions (we will discuss this later), it seems to me particularly doubtful whether illocutionary acts that do not involve judgment should be considered as species of assertion. It is, however, a fact that there are illocutionary acts such as reporting or testifying, whose most salient function is not that of issuing a judgment, which are intuitively members of the assertive family and can even be assimilated to assertion, possibly in consideration of other aspects of the procedure they consist in. Their relationship to assertion should be further analyzed.⁴ Members of the assertive family in which the assertive force is mitigated are not assertions. They are related to assertion through the downgrading of illocutionary force achieved by means of linguistic indicators of mitigation, which may affect in each case, in a salient way, different aspects of the procedure in the execution of which the illocutionary act is performed, including the effect achieved. For example, they may downgrade the speaker’s initial epistemic position (“It seems to me . . .”) or her resulting commitment (“Perhaps . . .”); these two elements of the procedure of asserting are connected, obviously, but it makes some difference which one the speaker chooses to mitigate on a given occasion. As to reinforced Assertives, they certainly are committal even more than simple assertions, but when the speaker is taken to perform one of them, very often what is emphasized

⁴ Hinchman (2020) addresses this in connection to testimony.

274

    

is not some core feature of the assertion, such as the speaker’s epistemic position or her issuing of a judgment, but some additional aspect related, for example, to authority (insisting) or commitment (swearing). Contrary to Searle and Vanderveken (1985), I think therefore that also these members of the assertive family are not species of assertion.

3. Assertion within Illocutionary Act Classifications We will now consider assertion’s place among illocutionary acts more generally. This amounts to considering into which class of illocutionary acts assertion falls, which, in turn, depends both on the definition we give to assertion and on the criteria for illocutionary act classification that contribute to determining illocutionary classes.

3.1 Definitions of Assertion and Illocutionary Classes We have anticipated some characteristics of assertion in the previous section. But assertion has received various more detailed definitions, often serving at the same time as the defining characteristics of an illocutionary class. Searle (1969, 66) gives a first definition of assertion in terms of conditions for successful performance (in particular, in terms of those necessary conditions that distinguish assertion from other illocutionary acts): the sentence uttered by the speaker must have a propositional content, which may feature any proposition; the speaker must have evidence or reasons for the truth of the proposition expressed; it must not be obvious to both speaker and audience that the audience knows that things are so; the speaker has to believe that things are so (or at any rate, must be responsible for so believing); and, last but not least, the speaker must intend that her utterance will count as an undertaking to the effect that p represents an actual state of affairs.⁵ In his work (1979), when he characterizes the class of Assertives, Searle builds upon this definition, modifying it slightly. Here the essential feature becomes, as is known, the illocutionary point, which (for Assertives) is specified as that of committing the speaker “to something’s being the case, to the truth of the expressed proposition” (1979, 12). This illocutionary point is accompanied by a “direction of fit” from words to world (which means that it is the Assertive that must fit the world and not the other way around), and by belief as the psychological state expressed. These characteristics hold for all Assertives and, therefore, for assertion too. ⁵ I have rephrased the shortened formula used to characterize the essential condition of assertion in Searle (1969, 66) according to the structure Searle himself assigns to essential conditions (1969, 60).

    

275

The definition of the point of Assertives in terms of commitment to the truth of the expressed proposition is widely but not universally accepted. For example, Bach and Harnish (1979), defining illocutionary classes on the basis of the attitudes expressed by the speakers in their utterances, the recognition of which by the audience would fulfill the speaker’s illocutionary intention, in their definition of Assertives (all of which are cases of assertion) focus only upon the speaker’s expression of a belief and of the intention that the audience form a like belief.⁶ Vanderveken, elaborating upon Searle’s theory, in his definition of Assertives seems to look back to Searle (1969), whose definition focused on the idea of representation: the utterance of an Assertive “serves to represent how things are” (1990, 16), or its point “consists of representing a state of affairs as actual,” “is to represent how things are in the world,” or even “consists of expressing a proposition P with the illocutionary point of representing as being actual in the world the state of affairs that P represents” (1990, 12, 22, 106). It is the word “actual” that makes most of the job in these definitions, distinguishing assertion from mere representation (in conformity to the Fregean heritage); commitment is not explicitly mentioned. Alston (2000) departs only slightly from Searle’s view by casting it in his own terms as the speaker’s “taking responsibility” for its being the case that p (where p is the proposition expressed in the sentence uttered). Since, in his view, all illocutionary acts can be described as requiring the speaker’s “taking responsibility” for something’s being the case (and, therefore, for the truth of propositions), Alston specifies that in the case of assertion what the speaker has to take responsibility for is not the obtaining of the states of affairs required by the rules of the illocutionary act, but the obtaining of the state of affairs that the asserted proposition says to hold, which must be presented explicitly in the utterance (with a waiver for cases of ellipsis). All these definitions (indirectly, also that of Bach and Harnish, through its appeal to belief) appear to rely upon truth-aptness. As Searle notices, “The simplest test of an assertive is this: can you literally characterize it (inter alia) as true or false” (1979, 13). Such a “simplest test” is, however, problematic. What does “literally” mean in this connection? How many kinds of speech acts may be characterized as true or false in some not strictly “literal” sense, and where does the border lie? Or perhaps “true in L” (as used in assigning truth conditions to sentences in the well-known biconditional “ ‘S’ is true in L if and only if p”) is literal whatever sentence of declarative syntactic form is substituted for “S”? If so, only utterances of imperative sentences, questions, and certain exclamations will fall short of being assertive speech acts. But to distinguish the various syntactic

⁶ MacFarlane (2011) discusses and criticizes this conception of assertion as an attitude-based account. The other speech-act theoretical conceptions of assertion referred to in this paper combine in various ways two of the other approaches he discusses, namely, the rule-based and the commitmentbased account.

276

    

forms of sentences, we do not need speech act theory. If it makes sense to speak of “assertion” as a speech act, it is because assertion is taken to be something other than the utterance of a declarative sentence with truth conditions. Be that as it may, Searle’s “simplest test” is clearly connected to the word-to-world “direction of fit,” which makes Assertives truth-apt. So, the role of truth-aptness in identifying Assertives would suggest that, after all, word-to-world direction of fit is their core property. However, Searle and Vanderveken stick to illocutionary point as the central property of classes of illocutionary acts: it is illocutionary point that determines from which direction success of fit must be achieved in the case of satisfaction of the utterance (Vanderveken 1990, 107). So, the burden of defining Assertives falls again upon the specification of their illocutionary point. But what is exactly that illocutionary point? What exactly do successful Assertives bring about? We have seen that the illocutionary point of Assertives has received a number of slightly different formulations: assuming they are equivalent would equate “representing the situation that p as actual” with “committing oneself to its being the case that p.” But the former expression, if it describes the aim of an act at all, deals with that act as an act of judgment. The latter expression, instead, refers to the undertaking of an obligation (the content of which is very unclear, as has been correctly remarked by MacFarlane 2011, 91).⁷ Granting that both things are part of what a successful assertive illocutionary act “does,” how exactly should they be connected? This is not explained by any of the speech act theorists we have referred to. The Grice-inspired solution adopted by Bach and Harnish is certainly clearer: but it falls short of collocating assertion (and communicative illocutionary acts in general) in the dimension of action, since they all are described merely in terms of the expression and intended recognition of attitudes. But if one takes illocutionary acts to be actions, it is reasonable to seek to identify their effects or outcomes. Moreover, it should be expected of the effects characteristic of the various classes of illocutionary acts (including Assertives) that they be species of one general kind of states of affairs. This condition is not met, for example, by Alston’s illocutionary classes: three of them, Directives, Commissives, and Exercitives, have “conventional effects,” while Expressives and Assertives lack them (Alston 2000, 133). Also Searle’s illocutionary points do not meet the condition: the core effect aimed at by an illocutionary act varies from bringing about a state of affairs to attempting to direct the hearer’s behavior, from expressing one’s attitude to committing oneself to some course of action or ⁷ Some clarification is provided by Green (2017) by distinguishing three dimensions of assertive commitment: liability (being made liable to be correct or incorrect depending on how things are), frankness (being subject to a norm requiring a psychological state of belief), and fidelity (being committed to responding appropriately to legitimate conversational challenges). The first and third dimensions in particular seem to me to articulate nicely two facets of the deontic state of a speaker who has issued a judgment as to how things are.

    

277

to the truth of the expressed proposition. Heterogeneity is apparent. Commitment is clearly a deontic state of an agent: when an agent is committed to something, she is in a state characterized by the possession of a deontic modal predicate of the “ought” kind. Also the states of affairs brought about by Declarations are characterized, albeit in a more complex way, by the attribution of deontic modal predicates of the “ought” and “can” kinds to the relevant participants in the social situation.⁸ But expressing an attitude can hardly be traced back to a deontic state of an agent, and were an analysis of this kind at all possible, it would certainly be impossible to extend it to the illocutionary point of Directives, which is attempting to direct someone’s behavior. So it appears that the effects, to which according to Searle illocutionary acts essentially aim, consist of the bringing about of states of affairs of different natures. The heterogeneity of the main criterion for classifying illocutionary acts is therefore a weakness of his classification.

3.2 The Place of Assertion in Austin’s Illocutionary Classes Austin, instead, consistently claims that all illocutionary acts have effects of the same kind, namely, “conventional” effects (1975, 14, 107, 117–19). By this, I understand that they are effects that depend on interpersonal or social agreement as opposed to straightforward causality and are cancellable if circumstances are discovered that make the illocutionary act misfire.⁹ Let us now examine how assertion is collocated within Austin’s general classification of illocutionary acts. Austin’s classification does not envisage a class of Assertives. He sketches a typology of whole procedures, grouping together procedures that are comparable as to their main steps such as the speaker’s initial position or the effect brought about, and characterized by the salience of the same element (for example, speaker’s authority as opposed to speaker’s commitment or speaker’s sincerity). Direction of fit (which, as we have seen, is associated with syntactic structure and therefore with aspects of the locution) is not decisive for his classes of illocutionary acts. Rather, the utterance of a declarative sentence, having truthconditions and therefore truth-apt, may contribute to different illocutionary procedures. Various illocutionary acts that we have considered as members of the assertive family, including some of the acts that Austin (1953) takes to be species of assertion, appear among his Expositives (illocutionary acts amounting to acts of ⁸ Searle (1995, 109–10) appears to dismiss the hypothesis that the effects of Declarations should be traced back to deontic states. But if he were right, the heterogeneity of his illocutionary points would be even greater. ⁹ I argue for this reading of Austin’s use of “conventional” in Sbisà 2007a [Essay 10], 2009a [Essay 11]. [See also Sbisà in preparation].

278

    

exposition and discourse organization, cf. 1975, 161–3) or his Verdictives (acts consisting “in the delivering of a finding, official or unofficial, upon evidence or reasons as to value or fact,” that “have obvious connexions with truth and falsity, soundness and unsoundness and fairness and unfairness”: 1975, 153), or as members of both classes (since Expositives are orthogonal to other classes, so that they may be at the same time classified in terms of one or other of the other four Austinian classes). “Assert” is not cited as a member of any one of the classes. This may depend on the fact that, as we have already seen, Austin takes “assertion” to be a generic term for a group of illocutionary acts that have more specific definitions of their own, so that, when an assertion is made, in fact one or other of these acts is performed. However, characterizations of assertion (or more specifically of statement) can be found where Austin compares performative utterances (which in this context means any utterances considered as performing illocutionary acts) with the so-called constative utterances, that is, utterances that are held to merely “say” something true or false (Austin 1975, 47–52, 136–40). In addition, various remarks about assertion and knowledge can be found in Austin’s paper “Other Minds” (1946, 77–86).¹⁰ The picture of assertion that emerges from Austin’s sparse suggestions gives a prominent role to the speaker’s being in a position to make the assertion. If it is apparent that she cannot claim the required epistemic reliability about the subject matter of the assertion, she will not be taken to make an assertion, but a conjecture or a guess (1975, 138). Moreover, an assertion may also be made “void” by referential failure (1975, 51, 137). It has to “secure uptake” like other illocutionary acts: we may doubt “whether I stated something if it was not heard or understood” (that is, the assertion, or the statement, must be recognizable as such: 1975, 139); it also “takes effect”: after it is made, says Austin, the speaker is committed to other assertions, or more precisely, other assertions that the speaker will make “will be in order or out of order,” and also the assertions made by the interlocutors will either contradict or not contradict the speaker (1975, 139). Finally, the making of an assertion is appropriate only if the speaker believes that things are as the assertion says, so that assertion implies belief (1975, 50, 136).

3.3 Do Assertions Have Conventional Effects? In what exactly the effect of the illocutionary act consists for each of Austin’s classes is not made completely explicit by Austin himself. However, one may consider that the deontic-modal properties of the agents involved, clearly affected by illocutionary acts, provide suitable material for a description (Sbisà 1984a

¹⁰ In this same paper, Austin first put forward the notion of performative utterance.

    

279

[Essay 1], 2001a [Essay 6], 2002a [Essay 7]). An agent’s possessing a certain deontic property can be considered as a “conventional” fact in the sense required: what an agent is allowed/not allowed, obliged/not obliged to do depends on social agreement or (in more informal situations) on agreement (explicit or tacit) among the participants in the specific interactional event; moreover, the deontic property vanishes if its attribution turns out not to be valid. However, while describing the illocutionary effects of Exercitives and Commissives only involves attributions of deontic properties of kinds associated with “can” and “ought,” Austin’s classes of Verdictives and Behabitives (the latter being speech acts by which the speaker reacts to events or to the behavior of others: 1975, 160)¹¹ appear to require the introduction of the epistemic dimension. But how precisely it should be introduced is not anything simple. A first problem is whether the epistemic modality to be introduced is knowledge or belief. As I shall argue, we should focus on knowledge. There is indeed a sharp difference between knowledge and belief as to their relationship to illocution in general and assertion in particular.¹² Whether a speaker is successful in producing a certain belief in the hearer depends (among other things) on whether the hearer actually comes to entertain that belief. But whether a speaker is successful in transmitting knowledge is not only a matter of producing a certain attitude in the hearer’s mind. Moreover, the fact that a speaker entertains a certain belief may explain why she sets out to perform a certain Verdictive or, more specifically, make a certain assertion. But it does not entitle her to that performance. Rather, what entitles her to perform a Verdictive or, more specifically, make a certain assertion is the possession of some relevant knowledge. We can also notice that the fact of being told about something does not entail believing that things are so, while in default conditions (when the competence of the speaker is not in question) it does entail that the hearer has been informed about how things are and now knows that things are so. One might even argue that the fact that the hearer has been reliably informed is enough to attribute to her the knowledge that things are so, whether she believes what she has been told or not. A second problem is whether knowledge is suitable for constituting the effect of an illocutionary act. If we accept, with Austin, that illocutionary effects are (in some sense of this word) “conventional” ones, we should not use “knowledge” to characterize an alleged illocutionary effect unless we identify at least some element of “conventionality” in the conception of knowledge, particularly as regards the ¹¹ Austin’s Behabitives are almost completely coextensive to Searle’s Expressives, but their description is different. ¹² Austin affirms a sharp difference between knowledge and belief both in his (1946) and in his essay on Plato’s Myth of the Cave (1979, 288–304), but he does not connect this theme to the analysis of illocutionary acts in any systematic way. For Austin on assertion, knowledge, and belief, see also Fiengo 2020 and Sbisà in preparation.

280

    

way in which an agent achieves knowledge of some state of affairs. As many have noticed (since Strawson 1964), Austin is not explicit as to the use he makes of the label “conventional.” But, for the current aims, it is relevant to point out that illocutionary effects obtain in situations in which there is social agreement (or at least agreement among the participants in the interactional event) that they obtain, and that their obtaining is defeasible, that is, holds by default unless certain sorts of infelicity are suspected or discovered, and is suspended or cancelled if they are actually discovered (see Section 3.2). If we consider the interactional dynamics of knowledge attributions, we see that something similar happens with them too, in certain cases at least. In order for a speaker to transmit knowledge to her audience by means of a Verdictive, she must be entitled to the act of judgment in which the Verdictive consists. If she is recognized as so entitled, any person in the audience is then held to have acquired the knowledge that has been linguistically formulated in her utterance: there is social or at least interpersonal agreement that the speaker was entitled to issue that verdictive and the audience came to know something in virtue of this. Obviously, many things may go wrong. The speaker may not really be in a position to issue that judgment, or may be lying, or make some mistake in issuing her judgment. In these cases, what the audience gets it not real knowledge. But it is difficult for an audience to explicitly consider and exclude these or similar flaws in real time during the interactional event (and sometimes even afterward). Therefore, the dynamics of the illocutionary effect of Verdictives proceeds by default, that is, unless some specific flaw in the speaker’s performance is revealed, any person who has received the Verdictive is taken to have acquired knowledge and (which is most in point here) can actually be dealt with as knowledgeable as regards the subject matter of the Verdictive. For example, in the case of assertion, she is dealt with as entitled to issue assertions with same or strictly related content in her own turn and to make decisions on the basis of the assertion received. All this is not far away from the dynamics of illocutionary effects of other classes. One might object that what is transmitted in this way is not knowledge, because whether someone knows something is a matter of fact (albeit one sometimes difficult to be certain about) and not anything “conventional,” not even in the sense of being dependent upon social or interpersonal agreement. But an agent’s being taken to possess knowledge appears to involve a judgment or decision or tacit agreement that the cognitive content she entertains has been arrived at in a way which is publicly recognized as reliable; and this is precisely what is meant by speaking of knowledge in describing the effect of verdictive illocutionary acts. We might therefore want to distinguish between knowledge in the full sense of the word, as an agent’s cognitive reliability enabling her to make assertions that are true, and knowledge in a weaker sense, as a (defeasible) status an agent is endowed with thanks to social or interpersonal agreement, entitling her to make assertions

    

281

and give the audience her guarantee for their truth.¹³ If one wants to preserve one’s conception of knowledge from any contamination with deontic and therefore “conventional” aspects, one might attempt to describe the dynamics of Verdictives using again only deontic properties of the “can” and “ought” kinds. But then the onus of accounting for the connections of Verdictives with the epistemic sphere is deferred to the sole content of those deontic properties. The main risk in doing so is to reduce Verdictives, assertion included, to the exercise of some right or authority, losing sight of the specific concerns and responsibilities involved in acts of judgment, which in the case of judgments of fact have to do with truth. A third point to be considered is whether the same kind of attribution of knowledge should figure in the description of the illocutionary effect of a verdictive and of the initial state of its speaker (as a felicity condition of the preparatory kind). Verdictives, including assertion, are acts of judgment and judgment can only be performed on the basis of a competence to judge whose components may vary across the different kinds of acts belonging to the Verdictive class. In the case of assertion, past experiences, opportunities to approach the assertion’s subject matter, possession of cognitive abilities and criteria of judgment, and capacity to exert one’s cognitive abilities accurately in the current situation (cf. Austin 1946, 79–81) all concur to putting the speaker in the position required. Since previous experiences generate knowledge, access to the assertion’s subject matter yields acquaintance with it, and criteria of judgment must be known to the speaker (implicitly at least), we may say that the competence to assert too consists of knowledge. Knowledge as speaker’s competence, insofar as it plays a role in the dynamics of illocution, shares with knowledge as illocutionary effect the nature of a (defeasible) status assigned to agents thanks to social or interpersonal agreement: it is in this quality that it can act as a requirement for the illocutionary successfulness of the assertion and (at least according to Austin 1975, 138), when not satisfied, make the assertion misfire. However, it should be pointed out that for any given assertion, this knowledge is not the same as the knowledge resulting from the assertion, which has the assertion’s content as its object. In this perspective (which was not worked out by Austin in the terms used here but is compatible with much of what he writes about knowledge, illocution, and assertion: see Section 3.2, and Sbisà in preparation), knowledge connects to Verdictives, and to assertion in particular, in a two-fold way: the speaker draws her entitlement to make an assertion from possessing the knowledge in which her competence to judge consists, while knowledge as the effect of the illocutionary act of asserting comes into being thanks to the exercise of that competence in an act of judgment and is transmitted to any participant in the interactional event who ¹³ The latter sense of “knowledge” could be usefully compared to the socially oriented conception of knowledge argued for by Kusch (2002).

282

    

receives the assertion. It is in virtue of the epistemic position of the speaker, recognized as amounting to knowledge, that her utterance can operate a change in the knowledge status of whoever receives it. It should be pointed out that the speech-act theoretical account of assertion outlined here, in the framework of Austin’s illocutionary classes, is compatible with Williamson’s account (2000, 238–69) to the extent to which it too foregrounds knowledge (rather than belief). But Williamson’s norm of assertion, the so-called Knowledge Rule (according to which a speaker asserting that p must know that p), is concerned with knowledge in its full factual sense, not with its “conventional” counterpart. Moreover, the knowledge that Williamson’s norm of assertion requires of the speaker to possess is knowledge of the very same content of the assertion and should therefore not be mistaken for the competence that entitles the speaker to assert. Even if Williamson presents his norm of assertion as a “constitutive” rule for assertion, it is not “constitutive” in the sense in which this word is used in speech act theory: if the Knowledge Rule is violated, this does not make the assertion misfire. I take it, therefore, that Williamson’s Knowledge Rule (albeit normatively characterizing assertion) is not part of the procedure for making an assertion. Rather, it constitutes an objective requirement for the assertion’s overall correctness (Sbisà 2019 [Essay 14]). The role of the speaker’s competence to judge is relevant to discriminate between Verdictives and Behabitives. If the latter may be said to provide their audience with knowledge, it is insofar as they make manifest an attitude or stance of the speaker (Sbisà 1984a, 106 [Essay 1, Section 3.4]): in them, the audience gets knowledge about the speaker, without this knowledge being grounded in an act of judgment of the speaker’s and behind that, in the speaker’s competence to judge. This difference is made apparent in a particularly vivid way if one considers the speaker’s commitments stemming from the performed illocutionary act. A Verdictive commits the speaker to supporting the result of her act of judgment not only by speaking and behaving subsequently in ways consistent with it but also by accepting to share with her audience, insofar as possible, the knowledge in which her competence consists. This is often done by providing justifications for the judgment issued. Also in the case of a Behabitive, the knowledge granted to the audience becomes a source of expectations for them and of commitments for the speaker, but those expectations and commitments concern the speaker’s future attitudes and behavior, rather than the logical consistency of her speech or her willingness to provide justifications. Summing up, in the view I have just outlined, assertion is identified not by reference to truth-aptness (although this is one of its characteristics) but as a procedure having a conventional effect, which may be described as production and transfer of knowledge and be analyzed in detail as entitling the hearer to further assertions and committing the speaker, not merely to consistency, but also to the defense of the assertion by evidence or argument. The place of assertion

    

283

among the speech acts is at the crossroad between two orthogonal illocutionary classes, Expositives and Verdictives, as an act that affects discursive and conversational relations, but at the same time involves judgment and allows for the transfer of knowledge. As to the relations of assertion so conceived with other illocutionary classes, it should be noted that borders between classes, in most illocutionary act classifications, are rigid. It is admitted sometimes that an illocutionary verb may be ambiguous between two different meanings and therefore designate two distinct illocutionary acts belonging to different classes (see, e.g., Vanderveken 1990, 174). But this is all. The classes here derived from Austin’s classification (and the original Austinian classes as well) are not so rigid. If illocutionary acts involve complex procedures, context may make salient one or other of their elements. Variations in the salience of elements of the procedure may be described as shifts from one kind of force to another.¹⁴ Reconsider, in this connection, the differences between making a statement and reminding, reporting, or testifying. Remaining always within Expositives, we have a shift from the basically Verdictive procedure of the statement to procedures in which either some kind of speaker’s authority becomes salient (reminding), or the speaker presents herself as just reacting to what is happening by mirroring it in discourse (reporting), or what is highlighted is the speaker’s commitment to truthfulness as regards matters of which she had direct experience (testifying).

4. Is Assertion “Special”? Up to this point, I have explored the relationship of assertion to illocutionary acts similar to it and its collocation within classes of illocutionary acts. It is now time to consider what can be concluded from those explorations as regards the role (or rank) of assertion among the other illocutionary acts. Is assertion just one among them, or does it have some exceptional property that sets it apart and gives it a special role? On the face of it, speech act theorists of any orientation take it that each kind of illocutionary act has its own peculiarities and that Assertives are just one among the other kinds. In most cases (Alston is an exception), the characteristic features of assertion are taken to be parallel to those of other illocutionary acts.¹⁵ But that analyses display parallel structure is not enough to guarantee that

¹⁴ I attempted to describe some kinds of such shifts (Sbisà 1984a, 108–12 [Essay 1, Section 4]). I still agree with the principle, but I am not sure I would endorse the same descriptions. ¹⁵ Alston’s requirements for an illocutionary act to be an assertion feature the explicit expression of the asserted proposition, in complete asymmetry to his requirements for other illocutionary acts (2000, 114–20).

284

    

assertion is really on a par with the other illocutionary acts. The issue is worth closer examination. Thus, let us consider in more detail what reasons there could be for treating assertion as a “special” speech act. There are at least three.

4.1 Assertion as Central to Inferential Semantics In Robert Brandom’s normative pragmatics (a trend of philosophical research external to speech act theory, but not far from it at least insofar as certain pragmatic themes are concerned), the practice of making assertions is given a central role. Brandom takes it to be the basic practice of our speech, enshrining the foundations of linguistic practices in general (1994, 167–8, 172–5). This is because, in positing inferential relations such as commitments and entitlements of speaker and hearer, it is held to generate content, according to the inferencebased view of semantics which is, for Brandom, strictly associated with normative pragmatics (1994, 83–4, 132–4). It should be clear that speech act theory does not endorse this kind of centrality. The speech-act theoretical conception of assertion (even considered in its varieties) does not endorse inferential semantics and cannot adhere therefore to “normative pragmatics” in Brandom’s sense. Similarities are accidental. If content does not depend on commitment, there is no primacy of illocutionary over locutionary acts. A fortiori, assertion as an illocutionary act cannot provide us with the foundations of language. Among speech act philosophers, Alston holds a form of use theory of meaning and could well be sympathetic with Brandom insofar as his theory too traces meaning back to (some form of) use. But the notion of illocutionary act potential upon which Alston builds his theory of meaning is equally concerned with all illocutionary acts and not primarily with assertion. Moreover, as Alston himself notices, Brandom’s theory does not seem to have room for the distinction between the speaker’s normative stance associated with assertion and her taking responsibility for certain states of affairs which, in Alston’s analysis, is connected to other illocutionary acts (2000, 116). The Searlean tradition is farther away from Brandom’s theory, since, in it, a basically Fregean notion of proposition is held to account for sentence meaning quite independently of illocutionary force (including assertive force). In the Gricean perspective, what is said (amounting to the truth-conditional meaning of the uttered sentence) and its truth-value are completely independent from any consideration of appropriateness or illocutionary successfulness. For Bach and Harnish (1979), the inferential steps by which the receiver understands the speaker’s locutionary act are prior to the inferential steps leading her to the understanding of the speaker’s illocutionary intention. As for Austin, his perspective entertains a problematic relationship to Brandom’s. There is in it the recognition of a normative dimension in the use of

    

285

language since, as we have seen, Austin’s conventional effects of illocutionary acts can be described in terms of deontic properties that are assigned to the participants and therefore (one may say) of their normative statuses.¹⁶ Moreover, with respect to assertion, we have seen above (Section 3.2) that Austin himself takes it to establish a normative order in discourse, which comes very close to Brandom’s view of assertion and its effects. But there are differences between the roles that the two philosophers assign to their remarks on this subject. Austin intends to show that every (and any) saying is also a doing, and that our assessments of utterances should not be forgetful of the actions performed in issuing them. But it is not his aim to explain meaning in terms of force; rather, he intends to keep these two notions distinct. Brandom intends to use his normative pragmatics to support his inferential semantics (1994, 83–4, 132–4): a theory of language in which the semantic content of a sentence is determined by the inferential network in which the sentence, when used, is collocated (Brandom 1994, 188–9). The Austinian variety of speech act theory shows that the recognition of a normative dimension in the use of language—through the description of illocutionary effects in terms of deontic states of the participants—is not necessarily connected to semantic inferentialism. It might even turn out that the need to pay tribute to inferential semantics constrains the descriptive possibilities of a normative pragmatics (as is argued by Labinaz 2018).

4.2 Assertion and the Truth/Falsity Assessment One might be tempted to take assertion as the central linguistic practice, or as being to some extent a “special” speech act, also by considering that only assertion is true/false, and therefore represents the most typical case of “saying.” It belongs to the classification policy of most speech act philosophers that all and only those speech acts that are assessed as true/false are gathered in one class of illocutionary acts (Assertives, Constatives). We have already cited Searle on this: “The simplest test of an assertive is this: can you literally characterize it (inter alia) as true or false?” (1979, 13). But Searle’s test is so simple just because we implicitly consider as truth-apt all utterances whose direction of fit is word-to-world. Since that direction of fit is easily recognizable from the use of declarative sentence form, any utterance of a declarative sentence turns out to be prima facie an assertion (or a variety of Assertive or Constative). At the same time, if one assumes (as most speech act philosophers do) that meaning is to be defined in terms of truth¹⁶ The relationship between deontic properties and normativity is not trivial. If an agent is assigned certain deontic properties, she becomes also endowed with a normative status, since her behavior will, or will not, conform to those deontic properties. While I prefer to describe the conventional effects of illocutionary acts in deontic terms, Witek (2015b) insists on the normative aspect of Austinian illocutionary effects.

286

    

conditions, and that what an utterance says amounts to its truth-conditions, the notion of “saying” most straightforwardly applies to truth-apt utterances, that is, to utterances of declarative sentences with word-to-world direction of fit and therefore assertive force. This comes close to conflating “saying” and “asserting.” Granting this kind of asymmetric position among the speech acts to assertion may turn out to be selfdefeating for the speech-act theoretical perspective, in two ways. First, the conflation of “saying” and “asserting” can be endorsed (and exploited) to deny that assertion is an illocutionary act at all (see, e.g., Cappelen 2011). Second, while sentence form contributes to making illocutionary force recognizable, it does not exhaust the field of linguistic force indicators, which moreover exert their indicating function only in relation to the actual context of utterance. To forget about this is to abandon the characteristic speech-act theoretical distinction between sentence form and illocutionary act. To admit of truth-conditional meaning in the context of a theory of speech acts in which assertion is illocution (even more so if we take it to be illocution along the lines that have been preferred in this paper), one might have to distinguish between the compositional calculus of truth-conditions and truth-values and the actual assessment of an assertion as an accomplished speech act in the truth– falsity dimension. This would also be an opportunity for realizing that there are two distinct ways to extend the consideration of truth/falsity to speech acts that are not Assertives (nor, in Austin’s classification, Verdictives). We may say that an uttered sentence’s truth-conditions are satisfied whenever there is a fit between words and world, independently of its direction: indeed, the proposition expressed in the utterance of a Directive establishes the conditions under which the Directive is complied with, exactly in the same way in which the proposition expressed in the utterance of an Assertive establishes the conditions under which the Assertive is true. But we may also consider, as Austin suggests (1975, 140–2, 145), that successfully performed speech acts of any kind, complete with meaning and force, may undergo an overall assessment in the light of the actual circumstances: were they the right thing to do? The assessment in terms of truth or falsity is the form that such an overall assessment takes in the case of assertion and its closest cognates.

4.3 Assertion and Knowledge Finally, one might want to grant assertion a special place among the speech acts because of its unique role in the epistemic domain. As we have seen, assertion presupposes a robust epistemic position of the speaker. In the characterization of assertion I have extracted from Austin, in order to make an assertion the speaker must be in a position to do so, and this position is to be specified, for

    

287

example, in terms of previously acquired competence, current opportunity, possession of data, and ability to apply criteria of judgment, which are all steps or elements concurring to achieving knowledge. In the characterization of the illocutionary effect of assertion that I have outlined earlier, assertion produces knowledge, that is, presents its content as the result of an epistemically reliable judgment of the speaker and therefore as object of knowledge for the receivers as well. Knowledge established and offered to the audience by an informal factrelated Verdictive is different from knowledge made available to the audience by an Expressive (or in Austin’s terms a Behabitive) for its role in the structure of conversation, and it is also different in subtler ways (mainly for its object and its relationship to evidence) from knowledge established by a value-related Verdictive. It is, therefore, specifically this kind of knowledge that plays a major role in the construction of conversational common ground as well as of the body of cultural and scientific knowledge. Certainly, assertion is worth particular attention in philosophy for these reasons. But if producing and transferring knowledge on the basis of the exercise of the speaker’s competence is assertion’s way of being an illocutionary act (the execution of a procedure designed to have conventional effects), recognizing its unique role in this third way is not in contrast with the classic speech-act theoretical attitude that considers all kinds of illocutionary acts to be on a par. The only warning to be added is that an Austinian speech-act theoretical perspective comes with some recommendation of modesty as to the cognitive capacities and performances of humankind. Assertion is no immediate or automatic “mirror of nature”; it is an action of which a speaker-agent is responsible, which involves choices at least as to the focus of attention and to the use of words, and which, to be performed successfully and correctly, needs the satisfaction of various requirements and the meeting of standards. Moreover, it is an action performed within a context, so that the assessments that apply to it should be expected to be context sensitive, be it the assessment of the initial position of the speaker (whose epistemic standard may vary with context, as is argued by Goldberg 2015, 255–9), or the assessment of the complete assertion in the dimension of truth and falsity (as has been proposed by Austin himself: 1975, 142–5).¹⁷

¹⁷ I would like to thank Paolo Labinaz for the many insightful discussions on topics connected with the theme of this paper, and Sanford Goldberg, editor of the Oxford Handbook of Assertion, for his helpful comments.

17 Illocution and Power Imbalance 1. Introduction Power imbalance is ubiquitous in social interaction, arising from various causes and having various effects. Some of its effects are wholly acceptable and can be seen as part of the physiology of social interaction, while others are unfair and even harmful, and of course there is a whole series of degrees in-between. It is a fact that a strong adult can lift heavy luggage, while an elderly person or a child cannot. The strong adult, thanks to their greater physical strength, can attack a weaker individual and harm them, while it is practically impossible for the opposite to happen. But if the weaker individual owns a gun and knows how to shoot, the person who has greater physical strength is in danger in turn. The boss can assign tasks to the staff and decide what has to be done, while ordinary staff members cannot: they are only expected to carry things out. This holds also when the boss is elderly or frail, and the subordinates younger and stronger. Suppose a group of friends is unsure about where to spend the evening and is hesitating, and one of them makes a clear proposal and all the others follow it: to that extent, that person is a leader. Power imbalance may begin de facto and then become de iure, or the other way around. There are circumstances which legitimize a power that did not exist before their being the case, and circumstances by which an authority that was already recognized as such is delegitimized. There are de iure powers which are hard to exert—a young teacher in a classroom of rebellious teenagers—until the corresponding de facto powers have been acquired too. In all that, speech has a role to play. Examining this role is the task (or one of the tasks) of a theory of speech acts, and particularly of the study of the illocutionary dimension of speech. This has always been an underlying conviction in my reflections and research on speech acts.¹ As appears already from Austin’s treatment of illocutionary acts in his (1975), an illocutionary act may affect what participants can or cannot do, in certain circumstances at least (think of acts such as naming a ship, issuing a permission or Originally published as “Illocuzione e dislivelli di potere,” in Rae Langton, Linguaggio d’odio e autorità. Lezioni milanesi per la Cattedra Rotelli, in dialogo con Raffaele Ariano, Claudia Bianchi, Laura Caponetto, Bianca Cepollaro, Roberto Mordacci, Marina Sbisà, ed. Claudia Bianchi and Laura Caponetto, 63–86. Milan-Udine: Mimesis, 2020. English translation by the author, revised by Judy Moss. ¹ This is part of a more general insight concerning the potential of speech act theory for social criticism. In my opinion, that potential was implicit from the very beginnings, but has surfaced rather slowly (see my remarks on this topic in Sbisà 1978, 42–3).

Essays on Speech Acts and Other Topics in Pragmatics. Marina Sbisà, Oxford University Press. © Marina Sbisà 2023. DOI: 10.1093/oso/9780192844125.003.0018

   

289

prohibition, betting, or challenging to a duel), thereby establishing a specific and often temporary power imbalance. Moreover, illocutionary acts must, in certain cases at least, be grounded in the possession of some form of power on the part of the speaker and therefore, in some form of power imbalance between the speaker and their interlocutors, albeit limited in scope and often to some extent provisional (think also, apart from cases such as naming a ship or issuing a permission or prohibition, of ordering, appointing, bequeathing, giving advice, or recommending). Speech act theory was hindered from focusing on power relations, with longlasting effects, by John Searle’s (1979, 1–7, 12) choosing to consider the status or position of the speaker with regard to authority—especially relevant, according to Austin, to the illocutionary class of Exercitives (1975, 151, 155–7)—as a merely supplementary parameter of the illocutionary act, whose value does not contribute to illocutionary act classification. Exercitives (as Austin conceived of them) disappeared not only from Searle’s own classification but also from those put forward subsequently. The interest in Exercitives then had a new start thanks to the work of Rae Langton and Mary Kate McGowan, who approached the relations between speech and power within the debate, initiated by Catharine MacKinnon, about the dynamics through which pornography “subordinates” and “silences” women (Langton 1993, 2009; MacKinnon 1987, 1993; McGowan 2003; see also Sbisà 2009b [Essay 12]). In this paper, I will illustrate the notion of exercitive illocutionary act in the framework of my elaboration of Austin’s idea of illocution (Section 2). After briefly considering some ways in which we bring about effects on what participants can or ought to do, other than the performance of Exercitives (Section 3), I will discuss a number of cases in which one can detect the exercitive force of an utterance (or stretch of discourse) occurring in an informal context, and thereby uncover the effects of that utterance on the degree of power or (more specifically) authority of the participants (Sections 4 and 5). In Section 6, then, I will discuss a case in which an utterance, insofar as it amounts to an exercise of authority, performs this function in a “backdoor” fashion (Langton 2018a). I shall return to the theme of power imbalance in my concluding remarks (Section 7).

2. Exercitives Austin’s Exercitives form a heterogeneous group, whose members can be assimilated to one another insofar as they involve the attribution of some form of authority to the speaker and the bringing about of an effect based on that authority on what the participants can or ought to do.² ² Austin defines Exercitives as “exercise of powers, rights or influence” (1975, 151). The notions of power and right merge with one another in the notion of authority (and are so understood in the subsequent literature on speech acts; see, e.g., Searle 1979, 10). As regards “influence,” I agree with Rae Langton who (in her lecture “Reimagining Free Speech,” Workshop “Speech, Harm, and Social

290

   

They comprise formal and informal varieties: some of them are ritually or legally regimented, while others (at first sight at least) seem to require the mere uttering of sentences of certain forms, for example unmitigated imperative sentences; some of them have the manifest goal of getting the addressee to do something, while others aim to create a normative framework in which further activity of the participants can be carried out and assessed. Among the various differences that can be noticed within Exercitives, none appears to me so clear-cut as to recommend replacing that class with two distinct classes (such as Exercitives and Directives in the classification proposed by Alston 2000). In fact, it is precisely in its capacity to reveal interrelations between the various sub-groups of Exercitives (especially formal and informal ones), that the theoretical interest of having that class lies. Indeed, only if we see continuity in that gamut of variations can we notice the presence in everyday life of authorities—not only formal and official but informal and ordinary—and get some idea of the extent to which our everyday social interactions are imbued with matters of power. Moreover, freeing the notion of authority from its association with institutional contexts is indispensable to becoming aware of the connections that exist between de facto and de iure powers. In these connections, an important role is played by a feature of the dynamics of illocution which (adopting the terminology of Lewis 1979) we can describe as the “accommodation” of the initial conditions or pre-conditions of the illocutionary act. When these pre-conditions envisage certain deontic powers on the part of the speaker, particularly their enjoying a certain authority (broadly intended), but the speaker is not endowed with any appropriate authoritative role, the required deontic powers can be assigned to them retrospectively merely by taking their illocutionary act as having been successfully performed (see Langton 2018a; Sbisà 2019 [Essay 14]). In my perspective, the class of Exercitives includes various different but interrelated groups of illocutionary acts, including: (i) Acts that bring about formal or institutional changes of status, such as appointing (to a job or role) or sentencing (to a punishment): in these cases, the linguistic form of the sentence that the speaker is required to utter and the circumstances of the utterance are usually explicitly regulated. (ii) Acts that establish rules (promulgating a law, naming): the circumstances of the linguistic utterance and the form of the sentence to be Norms,” Università Vita-Salute San Raffaele, Milan, October 3, 2019) uses a broad notion of authority, including informal cases which might be described in terms of influence. However, “influence” is ambiguous, because it lies somewhere between having a recognized (albeit informal) status and exerting a psychological effect (which would pertain to perlocution). For this reason, when referring to the kind of informally recognized power that makes an agent authoritative, I prefer to speak of “prestige.”

   

(iii)

(iv)

(v)

(vi)

(vii)

291

uttered are often explicitly regulated in these cases too, especially when the rules established belong to an institution. Directive Exercitives (ordering, commanding, urging, advising): these are acts whose characteristic perlocutionary goal is to direct the addressee’s behavior; they may or may not require formal circumstances and presuppose a fixed institutional hierarchy or merely an occasional and temporary one; they vary in degree of formality depending on their possible institutional function. For example, both a second-year student and a tutor can advise a first-year student about what courses to take: in the latter case, there is formal authority (since giving advice about courses is the tutor’s job), in the former, the speaker is informally accorded some prestige on the basis of their personal experience. In my framework, not all Directives are necessarily Exercitives, since we can use an utterance to attempt to direct the addressee’s behavior even without relying on a oneup position. Moreover, directivity is a matter of the perlocutionary goal associated with the illocutionary force of the utterance (its “perlocutionary object,” see Austin 1975, 118), not of its illocutionary effect. Evaluations and recommendations: apart from particular contexts, these are informal acts, requiring the recognition of the speaker as “morally” authoritative or enjoying prestige as regards decisions involving values of some sort. Warnings: here, the speaker is authoritative inasmuch as they have some reliable knowledge of interest to the addressee, the lack of which knowledge might cause the addressee damage. Their illocutionary effect is not merely a matter of assigning an obligation to the addressee (that is, the obligation to pay attention to the knowledge the speaker is providing), but also includes the redefinition of the respective responsibilities of the two interlocutors as to what may happen to the addressee relative to the situation which is the subject matter of the warning. Announcements and explanations: as with warnings, the speaker is epistemically authoritative, inasmuch as they possess the news or information which is the subject matter of the speech act, while the addressee does not. The truth of such news or information is taken for granted and the addressee is expected to receive it without questioning it. Exercitive Expositives: these are expositive illocutionary acts (Austin 1975, 152, 161–3)³ that control the distribution of rights and obligations (or other deontic properties) within discourse or conversation. The speaker performing them occupies a one-up position with respect to

³ Austin admitted of the possibility for Expositives (acts that organize discourse or conversation) to be also judgments, undertakings of commitment, display of attitudes, or exercise of authority (1975, 152, 161).

292

   

the interlocutor at least temporarily and relative to the development of the currently occurring stretch of discourse or conversational event. This is the case of questions, illocutionary acts that require the speaker to be in a position to ask the addressee about the pertinent subject matter and assign to them the obligation to answer. (viii) Exercitive assertions: this is the name I give to those illocutionary acts that are executed in uttering sentences of declarative linguistic form, but present their content as true beyond discussion on the basis of the authority or prestige of the speaker, therefore assigning to the audience the obligation to believe what the sentences uttered say. From the linguistic point of view, these utterances are indistinguishable from other assertions, but in discursive or conversational contexts, while judgments prompt responses expressing motivated agreement or disagreement and commit the speaker to account for their criteria, exercitive assertions appear to formulate stances for which no epistemic justification is either offered or expected.

3. Not Only Exercitives It should be pointed out that Exercitives are not an isolated case among illocutionary acts. First of all, the other kinds of illocutionary act also affect the distribution among the participants of deontic properties of the “can” or “ought to” kinds and may therefore contribute to creating, confirming, or modifying power imbalance. As I have argued elsewhere (Sbisà 1984a [Essay 1], 1989a), Verdictives (acts of judgment) and Commissives (acts of undertaking commitments) also require, for their successful performance, that the speaker be endowed with a deontic power of a certain kind, in the forms of competence (mastery of criteria, access to data, and possibly role of judge or evaluator officially or informally recognized) or of capacity (control over the satisfaction of the commitments undertaken). Only Behabitives (consisting of responses to events or behavior, often involving the display of attitudes: for example, greeting, apologizing, thanking) do not depend on what the speaker “can” do but rather on the position they are in, which has to be a position in which they are somewhat “indebted” with respect to an interpersonal situation and therefore obliged or expected to offer a ritualized move or at least take a stance. Behabitives too, however, involve deontic properties of the “can” and “ought to” kinds. On the one hand, those who perform them satisfy a state of indebtedness or obligation and thanks to that very satisfaction, are allowed to initiate new lines of conduct (for example: after greeting someone, thereby recognizing them, one is in a position to initiate a conversation); on the other hand, one commits oneself to a line of conduct consistent with the attitude displayed (for example: someone who apologizes for doing something

   

293

ought not to do the same thing again immediately afterwards). The difference of all these other kinds of act from Exercitives lies in the fact that in the case of the latter, the assignments of deontic properties of the “can” and “ought to” kind depend on the speaker’s possession of a specific kind of deontic power, namely, authority (broadly understood). Secondly, complex effects on the deontic properties of the participants can be brought about cumulatively, thanks to a sequence of discursive or conversational moves (whether by the same participant or distributed among the participants according to turn-taking rules). Although it may be difficult to single out the occurrence of a definite illocutionary act in these cases, the effects brought about may still be exercitive in character, that is, depend upon the authority (broadly understood) of some personal or collective agent. Thirdly, we should also leave it open as to whether it is possible for participants in an interactional event to acquire deontic properties by means other than illocutionary acts. Deontic properties can be affected by the use of presupposition-triggering linguistic expressions, since linguistically indicated presuppositions assign to the audience the obligation (conditional upon acceptance of the speech act as felicitous or appropriate) to take the truth of their contents for granted (Sbisà 1999a [Essay 3]). Cases in which presuppositions amounting to pre-conditions of an illocutionary act are accommodated are particularly relevant to the issue of power imbalance and we have already hinted at them in Section 2. For example, a command issued by a speaker who is not yet endowed with the necessary authority can still elicit the recognition of the speaker as authoritative. What happens in such cases can be traced back to the dynamics of illocution, since the status assigned to the speaker is part of the procedure that has to be executed in order to perform the illocutionary act of commanding (cf. Sbisà 2019 [Essay 14]). But there are also cases in which our linguistic utterances assign deontic properties of the “can” and “ought to” kinds to some target agent, that are more remote from the procedures for the execution of illocutionary acts and do not seem to require the recognition that the speaker be endowed with some special status. According to McGowan (2004, 2019), acts such as making the topic of a conversation shift should be considered “Conversational Exercitives,” a kind of act that produces normative effects in a basically automatic way. When the mere mention of some person or object suffices to make it legitimate as an object of reference for the subsequent conversational turns, and therefore available as the topic of further contributions, it seems that it is the exercise of a de facto power (the mere ability to speak) which produces a change in the norms of the current conversation. It might be observed that making the topic of a conversation shift toward the subject matter that one wishes to be focused upon is usually easier for participants who are somewhat authoritative. But a similar and even more problematic move from fact to norm appears to take place when insulting or derogatory words are used, which gain “permissibility” insofar as their utterance fails to

294

   

elicit any censorship or confrontation. This is what happens when an impolite or aggressive non-verbal behavior becomes a habit and is therefore taken as normal, but once taken as normal, appears to be legitimate too. Finally, the redistribution of deontic properties among the participants in an interactional event can also be prompted by conversational contributions conveying presupposed or not-at-issue contents not amounting to pre-conditions of illocutionary acts. I call “not-at-issue” those contents that are presented as marginal or accessory with respect to the main topic of an utterance, by formulating them in syntactically subordinate sentences (relative, temporal, or causal clauses), appositives, or parentheticals.⁴ When this is done, one may speak with Rae Langton of “back-door speech acts” (Langton 2018a), whose illocutionary force is not merely meant to be retrieved by inference (as is that of those “off-record” speech acts which are such for politeness reasons: cf. Brown and Levinson 1987), but takes effect covertly, through the back door, as it were, of the interactional event. “Back-door speech acts” thus have an effect upon the deontic properties of the participants even when they go unnoticed. In Sections 4 and 5, I will discuss Exercitives and their relation to authority, starting from some fairly clear cases of utterances having exercitive illocutionary force. I will not deal with the complex and hybrid phenomena we have hinted at in this section, but an awareness of them is an indispensable component of the background of our considerations. In Section 6, I shall return to back-door Exercitives in particular.

4. Everyday-Life Authorities Let us now go back to Exercitives, which I take to be the most central field in which to investigate power imbalance connected to speech. I wish to discuss some examples drawn from speech situations that are quite ordinary in character (albeit turned into performance by the media), considering them as representative of that exercise of authority which we meet and come at grips with in our everyday life, but which also lies at the basis of more formal and institutionalized cases. The first of these examples (from a TV show dated 2009; the speaker, a woman, is a right-wing Italian politician, while her addressee is the journalist hosting the TV show) contains the uttering of an imperative sentence with manifest directive goals.

⁴ Potts (2005) has argued for an analysis of what I call here “not-at-issue content” in terms of “conventional implicature.” I do not follow his approach, because I do not think that not-at-issue contents are retrieved by the audience by means of inferences, which is a hallmark of implicature (according to Grice 1975 at least). Indeed, they are represented in the text’s linguistic surface, and this should suffice to make them more akin to presuppositions, which can be so represented without losing their status (think, e.g., of the presuppositions of factive verbs or cleft constructions). [See Essay 5 and Sbisà 2021a.]

    (1)

295

Lei è un mantenuto – e io faccio l’imprenditore – è un cretino – questo è quanto ( . . . ) allora Lei sia rispettoso – per le donne come me – che lavorano, che lavorano.⁵ “You are a kept man—and I am an entrepreneur—you are a moron, that’s it . . . therefore behave respectfully to women like me who work, who work.”

Here, the speaker is urging the addressee to adopt a certain line of conduct (characterized as “respectful”) on the grounds of her one-up position, presented as a position of (informal) authority or prestige. The words “allora Lei sia rispettoso – per le donne come me” (“therefore behave respectfully to women like me”) clearly have a directive goal, but the illocutionary act is at the same time an Exercitive: the speaker presents herself as legitimately imposing herself on the host of the TV show, since she is “an entrepreneur” while according to her, the host is “a kept man” (possibly because he earns a salary) and moreover “a moron.” Besides, it is conveyed (as not-at-issue content) that women like the speaker, who ought to be shown respect, are women “who work, who work” and this contributes to making the directive Exercitive “sia rispettoso” (“behave respectfully”) appear justified on the grounds of the value of what the speaker considers to be work.⁶ The insult “you are a moron” emphasizes the contrast between speaker and addressee, making manifest the contempt already implicit in describing the addressee as a “kept man.” It could be disputed whether that insult is in turn, in its own way, an exercise of some kind of authority on the part of the speaker: in that case, it would be an authority to discriminate, accommodated in absence of objections on the part of the audience (in fact, the audience present at the TV show seems to appreciate what the speaker says, including the insult). My second example (also from a TV talk show, 2008; the speaker is an Italian moderate-left politician, who is addressing an Italian right-wing politician, a woman) is a case of exercitive assertion: the utterance of a syntactically simple declarative sentence, with a nominal predicate whose copula is in the present tense, in which the speaker affirms a value-laden principle in a quite dogmatic way: (2)

L’extracomunitario e l’omosessuale – è fratello e sorella //a me e a te – a me e a te.⁷ “The non-EU citizen and the homosexual is brother and sister to me and to you, to me and to you.”

⁵ The dash in the transcripts “–” stands for a short pause (shorter than 2 seconds). The dots in brackets indicate that some words have been omitted. Words that are commented upon (for their forceindicating function) are underlined in both the original and its translation. ⁶ It should be pointed out that the speaker is thinking of self-employed, entrepreneurial work. Working for someone else, as the TV host does, would seem not to have any value for her. ⁷ The double bar indicates the point at which the interlocutor starts speaking, in overlap with the speaker of (2).

296

   

Utterance (2) affirms the value of brotherhood/sisterhood and therefore non-discrimination and could be considered as a mere assertion (to the effect that the value of brotherhood/sisterhood applies to relations with any human being) or the undertaking of a commitment (for example, to foster brotherhood/ sisterhood with respect to people of the two discriminated categories that are mentioned). In the former case, it would be a verdictive illocutionary act (Sbisà 2020a [Essay 16]) and in the latter, a Commissive. However, it is no judgment, either as to fact or as to value. It does not provide any new finding, description, or evaluation. The metaphor of human beings being brothers and sisters is already available in the cultural background of the participants, with the well-known implication that no human being should be ever discriminated. There are no traces or hints of the reasons why the speaker approves of the value of brotherhood/sisterhood. As to the possible commissive reading, the speaker is certainly committed to defending the value he affirms; the point of the utterance does not seem, though, to assign him obligations about his future conduct or give any guarantee to the audience about it. The value of brotherhood/sisterhood is simply affirmed, proclaimed, in a neat and unhedged way, thus attempting to elicit its adoption by the audience. This suggests that (2) should be seen as an exercitive assertion: an assertion made from a position of authority or prestige, whose illocutionary effect consists in assigning to the audience the obligation to believe what the speaker has asserted (and, perhaps, behave in a consequential way). This analysis is confirmed by the insistence with which the speaker refers deictically to himself and his interlocutor, repeating this part of his turn twice (“a me e a te—a me e a te,” “to me and to you, to me and to you”): he is attempting to get his interlocutor involved in the adoption of the affirmed value. The interlocutor’s reply to (2) is consistent with such an understanding: (3)

//ma allora come sei ecumenico ( . . . ).⁸ “But then how ecumenical you are . . .”

With her words, the interlocutor characterizes the speaker of (2) ironically, implicitly comparing him to a religious preacher and thus underscoring the dogmatic and hortatory character of (2). At the same time, although (3) obviously does not grant the speaker the authority of a priest, it does recognize that he is implicitly staking a claim to some sort of moral prestige. So, although it expresses disagreement with the speaker of (2) in a way that throws on (2) the doubt of a possible infelicity, (3) recognizes (2) as a (purported) exercitive assertion. We shall now move on to an example that can be described as a case of authoritarian (rather than merely authoritative) discourse. I am speaking here of

⁸ The turn starts in overlap with (2) as indicated by the double bar.

   

297

discourse and not of a single utterance or turn, because the effect that the text we are about to examine has (or attempts to have) on the participants’ deontic properties of the “can” and “ought to” kinds depends cumulatively on various aspects of it: not only is there the indication of exercitive features, but also the communication of presupposed or not-at-issue contents which are unlikely to belong to the common ground of all the receivers. Our example (4) is drawn from a social network (Facebook, June 29, 2019); the page is that of Matteo Salvini, who at the time was Minister for Internal Affairs of the Italian Republic. The author of the text is therefore endowed with institutional authority, but the example can nevertheless be assimilated to the more informal ones we have already discussed, since a post on a social network does not have the characteristics of formal communication. Moreover, the illocutionary acts performed in uttering that post, albeit presupposing authority, do not depend on the specific formal authority possessed by the text’s author. (4)

Stanotte comportamento CRIMINALE della comandante (tedesca) della nave pirata (olandese) che ha tentato di schiacciare contro la banchina del porto di Lampedusa una motovedetta della Guardia di Finanza, con l’equipaggio a bordo, mettendo a rischio la vita degli agenti. DELINQUENTI! ⁹ “Last night CRIMINAL behavior of the (German) commander of the (Dutch) pirate ship which tried to crush a police motor launch against the seawall of the port of Lampedusa, with the crew on board, putting those officers’ lives at risk. DELINQUENTS!”

This post does not state that the commander of a certain ship held a criminal line of conduct (which would be a judgment), but announces that she did so. Indeed, the description of the event, suitable for presenting it as a case of “criminal behavior,” is taken for granted and therefore not subject to discussion: “la comandante . . . della nave pirata . . . che ha tentato . . .” (“the commander of the pirate ship which tried . . .”) is a definite noun phrase including a relative clause, and all its content is conveyed as presupposed. Accepting that presupposition, we take for granted among other things that there exists one and only one pirate ship which is salient in the context, and therefore, that the ship to which the post is referring is (whether literally or metaphorically) a “pirate” one: so, not only the description of the event, but its evaluation too, is at least in part presupposed. Also the fact that the commander of the ship jeopardized the life of the crew of the police motor launch is not stated, but communicated as not-at-issue by the gerund clause “mettendo a rischio la vita degli agenti” (“putting those officers’ lives at risk”). This is what announcements are: given a situation whose occurrence is not ⁹ https://www.facebook.com/salviniofficial/posts/stanotte-comportamento-criminale-della-comandantetedesca-della-nave-pirata-olan/10156726797208155/ (last accessed December 3, 2019).

298

   

in question and whose description (insofar as any is provided) is presupposed, a speaker who is in a position to do so makes the audience knowledgeable about it. This is what characterizes the announcement as an exercitive illocutionary act. It is often held that those who receive an announcement, get information about something and therefore some knowledge: that is quite true if the speaker is not wrong or deceptive. But the procedure of announcing is not one of formulating knowledge in order to share it and possibly check on it (as judgments do), nor does it necessarily require that the speaker have knowledge of the object of the announcement (since an announcement may concern a future event which has merely been planned, but has not occurred yet and therefore cannot be properly “known”). In our example (4), the fact that a certain description of the event referred to is presupposed invites the audience to believe (by accommodation) that things have actually gone that way, which in turn supports the assignment to the audience of the obligation to believe that the behavior of the commander of the ship was criminal. So the announcement “Comportamento CRIMINALE de [ . . . ]” (“criminal behavior of . . .”) also obliges the audience to endorse an assessment of that behavior as criminal. As a conclusion of the post, we find the exclamation “DELINQUENTI!” (“Delinquents!”), emphasized by the use of the capitals, which presents itself as a reaction to what has been presupposed and announced and therefore should most naturally amount to a behabitive illocutionary act. But given the exercitive character of the preceding part of the post, that exclamation too takes an exercitive shade: the reaction of an utterer endowed with authority, such as the author of this post, typically becomes a normative model for the reactions of their audience, since their reaction is seen as an exemplary one.

5. Established, Accommodated, or Contested Authorities Like formal Exercitives, everyday-life Exercitives too are grounded in a one-up position of the speaker as to specific kinds of power (authority or prestige). Such deontic properties undoubtedly differ from the authority required by formal Exercitives: nevertheless, it is reasonable and convenient (as argued for by Langton 2018b) to consider all of them as varieties of one and the same kind of social fact. The difference between formal authority and informal authority or prestige of the everyday-life and conversational kind lies above all in the means and the manner by which they are established. A formal authority is established explicitly and can be exercised only after being established. There exist procedures for assigning formal authority to people, such as elections, nominations, or appointments. In some cases, the assignment of formal authority over a certain domain is integral to the assignment of a role that has other contents too (for example, a teaching position or a career promotion). So, formal authority has

   

299

nothing basic: to perform Exercitives requiring it, a speaker must wait for others to assign it to them by means of other Exercitives. But it is unclear how the chain gets started. Semi-formal or informal authority and prestige are by far freer and can arise by themselves, springing into existence out of nowhere. Moreover, while being endowed with a role that comprises the possession of some formal authority is obviously of help in performing Exercitives, it is neither a sufficient condition for the exercise of informal authority or prestige nor a necessary one. On the one hand, the young teacher in the rebellious teenagers’ classroom has got the formal deontic properties of her role (provided she was duly appointed), but she will have to make a good deal of effort to get recognized as authoritative, that is, as possessing that basically informal, personal prestige without which it is impossible to control a classroom. On the other hand, in a group of peers without any formal structure, in which decisions have to be made, a leader may emerge in a wholly informal way, whose authority is merely context-bound and temporary, but may also constitute a first step toward a position of long-lasting prestige and perhaps some kind of formalization. Let us re-examine examples (1) and (2) in this light. It is no trivial matter for the speaker of (1) to gain a one-up position with respect to her interlocutor, since, due to the structure of TV talk shows (as asymmetric conversations which have a “director,” Orletti 2000), it is the host who has the right to the one-up position with respect to all other participants. A TV host’s authority is basically organizational and conversational, but is often exploited to support the opinions and stances that the host intends to highlight or promote. It is perhaps for this reason that the speaker of (1) makes explicit on what grounds she claims to be in a position to address a directive Exercitive at the host. The one-up position she stakes a claim to can be described as one of social prestige, to be gained or lost on the spot. So, whether she is or is not in a position to perform a felicitous directive Exercitive addressed at the host essentially depends on the acceptance by the audience of the setting that she herself outlines: the acceptance of that setting amounts to the accommodation of her informal, prestige-based authority. It is not merely a matter of believing that she has such an authority, but of granting it to her (at least temporarily, within the bounds of the current interactional situation). Likewise, it is not clear whether the speaker of (2) really has the moral authority he implicitly claims to have. His affirmation of the value of brotherhood/sisterhood can be effective as an exercitive assertion if the audience is willing to grant him the moral standing required for proclaiming values and urging others to endorse them or, at the very least, if the audience recognizes him as the spokesperson of some already accepted authority. If this is not the case, his turn remains the void imitation of a homily, as his interlocutor and antagonist ironically insinuates, thus distancing herself from the obligation to believe in the value of brotherhood/sisterhood. Also the derogatory exclamation “Delinquents!” in example (4) takes on the value of exemplary reaction to the news that has just

300

   

been announced only if its utterer is recognized as authoritative, which can happen not so much on the grounds of his formal, institutional role (which does not give him license to judge who is and who is not a “delinquent”), as at the personal level of his prestige as a political leader. When a speaker with officially established authority issues an Exercitive whose effect runs counter to the sensibility or the principles of the addressee, the latter faces a problem: obligations coming from different sources clash in their deontic state, so that it proves impossible to satisfy them all and some of them have to be disattended. However, a law of the State, or the command issued by a hierarchically higher authority do not cease to be such if some individual addressee decides to disobey them. But when the speaker’s authority is informal and negotiated on the spot, or obtained thanks to the by-default accommodation of the preconditions of the exercitive force of some of their utterances, there is a way out of any possible dilemma: it can be prevented from arising by “blocking” the recognition of the speaker as authoritative (as is argued by Langton 2018a) and with it, the success of the exercitive act. Refusing to accommodate the authority of the utterer of (4) does not amount to denying that at the time in which he published that post he was Minister of Internal Affairs of the Italian Republic. But it does amount to denying him the kind of informal authority or prestige that is required in order to succeed in imposing on one’s audience the obligation to believe that a certain event took place in a certain way, or that the behavior of certain people was criminal. It also involves refusing to grant any exemplary value to his reaction to the event.

6. A Back-Door Exercitive and How to Block It Is it possible to block the accommodation of the speaker’s (informal) authority, whenever we disapprove of the exercitive distribution of deontic properties of the “can” and “ought to” kinds that the speaker is purporting to bring about? And if it is possible, how? One of the hurdles to overcome is our by-default inclination to accommodate that which is presupposed. Therefore, it becomes easier to criticize and reject a presupposition, and thus also a pre-condition of an illocutionary act, once they are expressed explicitly (as I argued in various works, highlighting the potential for criticism that the capacity of explicitation possesses: see Sbisà 1999a [Essay 3] and 2007c).¹⁰ However, as is remarked by Langton (2018a), explicitation by itself does not suffice to ensure the blocking of undesirable accommodations.

¹⁰ [See also Sbisà 2021a.]

   

301

Indeed, making a content explicit does not necessarily amount to its rejection. What strategies can block presupposition accommodation, and thus also the accommodation of the speaker’s prestige or authority, is still largely to be explored. A contribution may come from the analysis of the messages that, in social networks, respond to utterances or texts displaying exercitive features. At this point, it will be useful to consider an exchange drawn from a social network. Our example (5) is a declaration attributed by the media to the president of an Italian football club (in which he defends the racist behavior of the supporters of the club’s team). We will then consider some of the comments responding to it on Twitter. This is the president’s declaration, as can be read on a tweet which reproduces the headline and the first lines of a newspaper article: (5)

I buuu li fanno anche a chi ha pelle normale. Non sempre i buuu sono un atto discriminatorio o razzista.¹¹ “They boo those with normal skin as well. Booing is not always a discriminating or racist act.”

The utterance of the two sentences in (5) is an exercitive act, performed on the basis of the apical role the speaker is endowed with: in it, the president of the football club defends the behavior of its supporters and even absolves them from the charge of racism. Since booing is not always addressed at colored players, it is not necessarily motivated by racism and is therefore permissible. But the speaker’s underrating the risk of racism among football supporters, while at the same time using the expression “normal skin,” suggests a sort of complicity. The use of the phrase “chi ha pelle normale” (“those with normal skin”) presupposes that there are people with normal skin and this assumption in turn suggests that there is a kind of skin that is normal (easily identifiable, in the speech situation, with so-called “white” skin). But introducing a distinction between what is normal and what is not normal as regards to skin is discriminating: in (5) an act of discrimination is therefore performed, targeting whoever is not “white.” This act is not a full-fledged illocutionary act, because it is not the main act performed in uttering the first sentence of (5) “I buuu li fanno anche a chi ha pelle normale” (“They boo those with normal skin as well”), which is instead the act of providing a justification for the subsequent assertion “Non sempre i buuu sono un atto discriminatorio o razzista” (“Booing is not always a discriminating or racist act”). That act of discrimination is a back-door speech act (to use Langton’s metaphor) and, as an act that discriminates against whoever has a skin that is not white, performed by a speaker endowed with a

¹¹ https://twitter.com/gadlernf thertweet/status/1179245939241955328 (last accessed December 3, 2019).

302

   

specific authority in exerting functions connected with the latter (such as defending the supporters of the club’s team), it has the characteristics of an Exercitive. Letting it pass without contestation amounts to validating it. If the audience is not willing to allow for it to be validated, they must react in such a way as to “block” the act of discrimination, that is, keep it from functioning. The replies elicited by (5) can be read on the web. Most of them are designed to reject the act of discrimination implicitly performed by its utterer. In attempting to do so, they resort to two main strategies, the former of which consists in trying to block the accommodation of the presupposition that there is a kind of skin that is “normal” by explicitly focusing on its content and ironically commenting on it: (6)

Non sapevo esistesse la pelle normale “I did not know normal skin existed.”

(7)

La mia è un po’ grassa. Dice che rischio, presidente? Eh? “Mine is a bit greasy. Would you say that I am running some risks, president? Eh?”

Neither of these replies is a case of full-fledged explicitation, but their effects are analogous to the effect which would be brought about by the explicitation of the discriminatory content of (5), since they open the way to dissent. In (6), that discriminatory content is expressed in the text’s surface by the complement clause “(che) esistesse la pelle normale” (“that normal skin existed”). Although this clause is subordinated to a factive verb, the commentator’s dissent from the content it expresses is revealed by the fact that the factive predicate is negated: “Non sapevo” (“I did not know”), while prima facie expressing astonishment, implicitly suggests skepticism. In (7), the commentator refers to his own skin and deals with kinds of skin as a matter of dermatology, rather than as markers of ethnic differences. So, the comment equivocates, with ironic effects, on the sentence “Normal skin exists” which would make the discriminatory presupposition of (5) explicit. Moreover, it expresses the worry that it might be easier to be ridiculed if one’s skin is not “normal” (in whatever sense), therefore highlighting the fact that being more exposed to ridicule is integral to the subordinate status which the back-door act of discrimination in (5) assigns to colored people. Since the identification of the presupposed content makes disagreement possible, but does not mandate it, we find also comments that approve of what is expressed by “Normal skin exists.” In the following comment, for example, there is not only an (allegedly) charitable reading of what is expressed by “Normal skin exists” but also a manifest endorsement of the way of speaking employed in (5), which implies a permissive attitude with respect to whatever act of discrimination that way of speaking might contribute to:

    (8)

303

Per noi europei, storicamente, la pelle normale ha un colore tra il bianco e il rosa. Per gli africani la pelle normale è più scura. Per gli asiatici è giallognola, tendente al marrone verso sud. Ripeto, storicamente. Perciò, non credo ci sia nulla di sbagliato. “For us Europeans, historically, normal skin has a color between white and pink. For Africans normal skin is darker. For Asian people it is yellowish, close to brown to the South. I repeat, historically. Therefore, I do not think there is anything wrong.”

The second strategy aimed at blocking the act of discrimination which is implicit in (5) targets the exercitive character of that act, and in general of (5), by challenging the speaker’s status as to authority or prestige: (9)

Ma è un genio! “But he is a genius!”

(10)

No, è il classico “burino” che ha fatto i soldi e per questo viene considerato uno in grado di rilasciare dichiarazioni. “No, he is the typical person coming from the countryside who has made money and therefore is taken to be in a position to make declarations.”

Example (9) is a manifestly ironic comment, which deals with the racist defense of racist football team supporters as revealing lack of wit and thereby undermines, not so much the formal authority of the speaker as the president of a club, but the informal aura of prestige which comes with it. Example (10), responding to (9), is even more explicit and provides an account of why the utterer of (5) is considered as authoritative, implicating at the same time that he is not worth such consideration. It should be pointed out, sadly, that the comment’s author himself uses a discriminatory term, “burino” (roughly: a person coming to Rome from the countryside at the border of what in the nineteenth century was the Papal State). He will apologize for using that term a few lines later. But it is cause for concern that, in order to block the accommodation of authority and therefore the exercitive force of a discriminatory utterance, discrimination should once again be resorted to. In fact, this shows to what extent our ordinary speech is permeated by power imbalance, or rather, by the widespread longing for one-up positions, to be reached by putting others one-down. Going from one extreme to the other appears to be easier than finding a fair balance.

7. Concluding Remarks By examining informal exercitive acts and the kinds of authority on which they are grounded or which they establish by accommodation, we were able to describe

304

   

some of the ways in which speech creates or modifies the rights and obligations of the participants in a social interaction or, more generally speaking, their deontic properties of the “can” and the “ought to” kinds. In the cases we have examined in detail, the deontic powers of the participants are never equal (as standardly occurs with Exercitives, including informal ones) and never irrevocably fixed. Attempts are made by one of the participants to become more authoritative than the interlocutors by gaining either some kind of practical authority or prestige. There are members of the audience who consent to this, sometimes enthusiastically, but there may also be conflicts between different sources of authority or prestige, and attempts to resist the power imbalance that the accommodation of authority and the exercise of that authority on the part of the speaker would create. Sometimes, it is the use to which a speaker appears to be willing to put the authority or prestige they claims to possess (for example, the fact that it would be used to perform an act of discrimination) that alerts some members of the audience and prompts them to deny the speaker that authority or prestige (insofar as possible). Broadening our view beyond Exercitives to illocutionary acts in general, we should admit that in one form or other, some power imbalance among the participants in a social interaction exists at least in the presupposed initial situations. But there are compensations for the imbalance, which may counterbalance it in the last resort. The addressee of a Verdictive acknowledges the speaker’s power to issue it (that is, their epistemic authority or competence), but is thereby authorized to rely upon that Verdictive and even on the criteria according to which it was issued (if the speaker was transparent enough about them) in their subsequent speech acts and practical decisions. The addressee of a Commissive too, while recognizing the speaker’s power in the form of the capacity to sustain a commitment, acquires a right to expect of the speaker a certain action or line of conduct, which is a form of deontic power (albeit not amounting to authority). The sphere of deontic powers is complex and articulated and their assignments to agents are subject to change with time and across occasions of interaction. Even in the case of those Exercitives that assign only obligations to the addressee (or limit their rights, as in the case of prohibition), if the authority exercised is local, has no claims to extend to other contexts, and serves aims that can be shared, its exercise does no harm. Indeed, in accepting such an Exercitive and possibly complying with the obligation it imposes, the addressee is not subject to some life-long penalty or assigned an irreversibly subordinated role, and in my opinion, this is where the border lies between the physiology of power imbalance and its pathology. Attempts to enhance one’s authority by transferring it from one context to another, as in the case of (4) and in part of (5), attempts to make redemption or compensation impossible as in (1) (“è un cretino – e questo è quanto”), and dogmatism as in (2) and many other cases of Exercitive assertion, tend to stabilize power imbalance and keep it mono-directional. It is not power imbalance in itself, but its stability and mono-directionality that must be avoided. Uniformity of

   

305

deontic properties across participants in a social interaction is no sensible goal. But, as it is desirable that participants recognize each other as human persons and that the actions they perform in and by speaking be transparent, it is also desirable (especially when power imbalance is at issue) that assignments of deontic properties be local, mobile, and reversible.

References Alston, William P. 2000. Illocutionary Acts and Sentence Meaning. Ithaca, NY: Cornell University Press. Antony, Louise. 2012. “Different Voices or Perfect Storm: Why Are There So Few Women in Philosophy?” Journal of Social Philosophy 43 (3): 227–55. Aristotle. 2009. The Nichomachean Ethics. English translation by D. Ross, ed. L. Brown. Oxford: Oxford University Press. Austin, John L. 1946. “Other Minds.” Proceedings of the Aristotelian Society Supplementary Volume 20. Reprinted in Austin 1979, 76–116. Austin, John L. 1953. “How to Talk. Some Simple Ways.” Proceedings of the Aristotelian Society 53. Reprinted in Austin 1979, 134–53. Austin, John L. 1956. “A Plea for Excuses.” Proceedings of the Aristotelian Society 57. Reprinted in Austin 1979, 175–204. Austin, John L. 1957. “Pretending.” Proceedings of the Aristotelian Society Supplementary Volume 32. Reprinted in Austin 1979, 253–71. Austin, John L. 1962. Sense and Sensibilia. Edited by G. J. Warnock. Oxford: Oxford University Press. Austin, John L. 1966. “Three Ways of Spilling Ink.” Philosophical Review 75. Reprinted in Austin 1979, 272–87. Austin, John L. 1975. How to Do Things with Words. 2nd ed. by J. O. Urmson and M. Sbisà. Oxford: Oxford University Press. (1st ed. by J. O. Urmson, 1962.) Austin, John L. 1979. Philosophical Papers. 3rd ed. by J. O. Urmson and G. J. Warnock. Oxford: Oxford University Press. (1st ed. 1961.) Avramides, Anita. 1989. Meaning and Mind. Cambridge, MA: MIT Press. Avramides, Anita. 2001. Other Minds. London: Routledge. Bach, Kent. 1994. “Conversational Impliciture.” Mind and Language 9: 124–62. Bach, Kent and R. M. Harnish. 1979. Linguistic Communication and Speech Acts. Cambridge, MA: MIT Press. Baker, Judith. 1991. “Introduction.” In Paul Grice, The Conception of Value, 1–21. Oxford: Oxford University Press. Ball, Brian. 2019. “Commitment and Obligation in Speech Act Theory.” In Normativity and Variety of Speech Actions (Poznań Studies in the Philosophy of the Sciences and the Humanities 112), ed. M. Witek and I. Witczak-Plisiecka, 51–65. Leiden: Brill. Ballmer, Thomas T. 1978. Logical Grammar: With Special Consideration of Topics in Context Change. Amsterdam: North-Holland. Ballmer, Thomas T. 1979. “Probleme der Klassifikation von Sprechakten.” In Sprechakttheorie und Semantik, ed. G. Grewendorf, 247–74. Frankfurt a/M.: Suhrkamp. Ballmer, Thomas T. 1981. “Context Change and Its Consequences for a Theory of Natural Language.” In Possibilities and Limitations of Pragmatics, ed. H. Parret. M. Sbisà, and J. Verschueren, 17–55. Amsterdam: John Benjamins. Baron, Bettina and Helga Kotthoff (eds.). 2001. Gender in Interaction: Perspectives on Femininity and Masculinity in Ethnography and Discourse. Amsterdam: John Benjamins.

308



Baron-Cohen, Simon. 1997. Mindblindness. An Essay on Autism and Theory of Mind. Cambridge, MA: MIT Press. Bazzanella, Carla. 1994. Le facce del parlare. Florence: La Nuova Italia. Bazzanella, Carla, Claudia Caffi, and Marina Sbisà. 1991. “Scalar Dimensions of Illocutionary Force. In Speech Acts: Fiction or Reality?,” ed. I. Ž. Žagar, 63–76. Ljubljana: International Pragmatics Association. Beaver, David I. 1997. “Presupposition.” In Handbook of Logic and Language, ed. J. Van Benthem and A. ter Meulen, 939–1008. Amsterdam: Elsevier. Bermejo-Luque, Lilian. 2011. Giving Reasons: A Linguistic-Pragmatic Approach to Argumentation Theory. Berlin: Springer. Bianchi, Claudia. 1999. “Three Forms of Contextual Dependence.” In Modeling and Using Context. Second International and Interdisciplinary Conference, CONTEXT ’99, Trento, Italy, September 1999. Proceedings, ed. P. Bouquet, L. Serafini, P. Brézillon, M. Benerecetti, and F. Castellani, 67–76. Berlin: Springer. Bird, Alexander. 2002. “Illocutionary Silencing.” Pacific Philosophical Quarterly 83: 1–15. Black, Max. 1963. “Austin on Performatives.” Philosophy 38: 217–63. Blum-Kulka, Shoshana. 1985. “Modifiers as Indicating Devices: The Case of Requests.” Theoretical Linguistics 12: 213–29. Blum-Kulka, Shoshana, Juliane House, and Gabriele Kasper (eds.). 1989. Cross-Cultural Pragmatics: Requests and Apologies. Norwood, NJ: Ablex. Braidotti, Rosi. 1991. Patterns of Dissonance. Cambridge: Polity Press. Brandom, Robert B. 1994. Making It Explicit: Reasoning, Representing, and Discursive Commitment. Cambridge, MA: Harvard University Press. Brown, Penelope and Stephen C. Levinson. 1987. Politeness. Some Universals in Language Usage. Cambridge: Cambridge University Press. Butler, Judith. 1999. Gender Trouble. Feminism and the Subversion of Identity. New York: Routledge. (1st ed. 1990.) Caffi, Claudia. 1992. “Il concetto di coinvolgimento nella linguistica pragmatica.” In La Linguistica Pragmatica, ed. G. Gobber, 267–97. Rome: Bulzoni. Caffi, Claudia, 1999. “On Mitigation.” Journal of Pragmatics 31: 881–909. Caponetto, Laura. 2017. “On Silencing, Authority, and the Act of Refusal.” Rivista di estetica 64: 35–52. Caponetto, Laura and Paolo Labinaz (eds.). Forthcoming. Sbisà on Speech as Action. Cham, Switzerland: Palgrave Macmillan. Cappelen, Herman. 2011. “Against Assertion.” In Assertion: New Philosophical Essays, ed. J. Brown and H. Cappelen, 21–48. Oxford: Oxford University Press. Carston, Robyn. 2002. Thoughts and Utterances. The Pragmatics of Explicit Communication. Oxford: Blackwell. Chierchia, Gennaro and Sally McConnell-Ginet. 1990. Meaning and Grammar. Cambridge, MA: MIT Press. Collini, Stefan (ed.). 1992. Interpretation and Overinterpretation. Cambridge: Cambridge University Press. Crary, Alice. 2002. “The Happy Truth: J. L. Austin’s How to Do Things with Words.” Inquiry 45: 59–80. Davidson, Donald. 1973. “Radical Interpretation.” Dialectica 27. Reprinted in D. Davidson, Inquiries into Truth and Interpretation, 125–39. Oxford: Oxford University Press, 1984. Davidson, Donald. 1980. Essays on Actions and Events. Oxford: Oxford University Press. Davidson, Donald. 1984. “Communication and Convention.” In D. Davidson, Inquiries into Truth and Interpretation, 265–80. Oxford: Oxford University Press.



309

Davis, Steven. 1981. “Perlocutions.” In Speech Act Theory and Pragmatics, ed. J. R. Searle, F. Kiefer, and M. Bierwisch, 37–55. Dordrecht: Reidel. Davis, Wayne. 1998. Implicature. Intention, Convention and Principle in the Failure of Gricean Theory. Cambridge: Cambridge University Press. Dennett, Daniel C. 1987. The Intentional Stance. Cambridge, MA: MIT Press. Derrida, Jacques. 1972. “Signature événement contexte.” In J. Derrida, Marges de la philosophie, 365–93. Paris: Minuit. (English translation Glyph. John Hopkins Textual Studies 1, 1977: 172–97.) Diamond, Julie. 1996. Status and Power in Verbal Interaction. Amsterdam: John Benjamins. Doerge, Friedrich C. 2006. “Re-definition and Alston’s ‘Illocutionary Acts’.” Grazer Philosophische Studien 73: 97–111. Dressler, Wolfgang and Lavinia Merlini Barbaresi. 1994. Morphopragmatics. Berlin: de Gruyter. Ducrot, Oswald. 1972. Dire et ne pas dire. Paris: Hermann. Duranti, Alessandro. 1988. “Intention, Language, and Social Action in a Samoan Context.” Journal of Pragmatics 12: 13–33. Duranti, Alessandro. 1992. Etnografia del parlare quotidiano. Rome: Carocci. Duranti, Alessandro and Charles Goodwin. 1992. Rethinking Context. Language as an Interactive Phenomenon. Cambridge: Cambridge University Press. Eco, Umberto. 1990. I limiti dell’interpretazione. Milan: Bompiani. Fabbri, Paolo. 1973. “Le comunicazioni di massa in Italia: sguardo semiotico e malocchio della sociologia.” Versus 5: 57–109. Fairclough, Norman. 1995. Critical Discourse Analysis. The Critical Study of Language. London: Longman. Fiengo, Robert. 2017. “Austin’s Cube.” In Act-Based Conceptions of Propositional Content, ed. F. Moltmann and M. Textor, 209–34. Oxford: Oxford University Press. Fiengo, Robert. 2020. “Austin on Asserting and Knowing.” In The Oxford Handbook of Assertion, ed. S. Goldberg, 643–60. Oxford: Oxford University Press.I Fillmore, Charles J. 1971. “Verbs of Judging: An Exercise in Semantic Description.” In Studies in Linguistic Semantics, ed. C. J. Fillmore and D. T. Langendoen, 273–90. New York: Holt, Rinehart and Winston. Forguson, L. W. 1966. “In Pursuit of Performatives.” Philosophy 41: 341–7. Fraser, Bruce, 1980. “Conversational Mitigation.” Joumal of Pragmatics 4: 341–50. Frege, Gottlob. 1960. “On Sense and Reference” (English translation of “Über Sinn und Bedeutung,” 1892). In Translations from the Philosophical Writings of Gottlob Frege, ed. P. Geach and M. Black, 56–78. Oxford: Blackwell. Frege, Gottlob. 1972. “Conceptual Notation” (English translation of Begriffschrift, 1879). In Conceptual Notation and Related Articles, translated and ed. T. W. Bynum. Oxford: Oxford University Press. Frege, Gottlob. 1997. “Thought” (English translation of “Der Gedanke. Eine logische Untersuchung,” 1918). In The Frege Reader, ed. M. Beaney, 325–45. Oxford: Blackwell. Fricker, Miranda. 2007. Epistemic Injustice. Oxford: Oxford University Press. Fricker, Miranda and Jennifer Hornsby (eds.). 2000. The Cambridge Companion to Feminism in Philosophy. Cambridge: Cambridge University Press. Frith, Uta. 2003. Autism: Explaining the Enigma. Oxford: Blackwell. (1st ed. 1991.) Gauker, Christopher. 1994. Thinking Out Loud. An Essay on the Relation between Language and Thought. Princeton, NJ: Princeton University Press. Gauker, Christopher. 1997. “Universal Instantiation: A Study of the Role of the Context in Logic.” Erkenntnis 46: 185–214. Gauker, Christopher. 1998. “What Is a Context of Utterance?” Philosophical Studies 91: 149–72.

310



Gauker, Christopher. 2003. Words without Meaning. Cambridge, MA: MIT Press. Gavioli, Laura and Gillian Mansfield (eds.) 1990. The PIXI Corpora. Bookshop Encounters in English and ltalian. Bologna: Clueb. Gazdar, Gerald. 1979. Pragmatics. Implicature, Presupposition and Logical Form. New York: Academic Press. Gazdar, Gerald. 1981. “Speech Act Assignment.” In Elements of Discourse Understanding, ed. A. K. Joshi, B. L. Webber, and I. A. Sag, 64–83. Cambridge: Cambridge University Press. Geis, Michael L. 1995. Speech Acts and Conversational Interaction. Cambridge: Cambridge University Press. van Gelder, Tim. 1998. “Monism, Dualism, Pluralism.” Mind & Language 13: 76–97. Gelman, Susan A. (2010). “Generics as a Window onto Young Children’s Concepts.” In Kinds, Things, and Stuff: The Cognitive Side of Generics and Mass Terms (New Directions in Cognitive Science v. 12), ed. F. J. Pelletier, 100–22. New York: Oxford University Press. Gelman, Susan A., P. J. Goetz, B. S. Sarnecka, and J. Flukes. 2008. “Generic Language in Parent–Child Conversations.” Language Learning and Development 4 (1): 1–31. Gelman, Susan A. and L. Raman. 2003. “Preschool Children Use Linguistic Form Class and Pragmatic Cues to Interpret Generics.” Child Development 74 (1): 308–25. Gelman, Susan A., I. Sánchez Tapia, and S. J. Leslie. 2016. “Memory for Generic and Quantified Sentences in Spanish-Speaking Children and Adults.” Journal of Child Language 43 (6): 1231–44. Gelman, Susan A., E. A. Ware, and F. Kleinberg. 2010. “Effects of Generic Language on Category Content and Structure.” Cognitive Psychology 61 (3): 273–301. Gilligan, Carol. 1982. In a Different Voice. Cambridge, MA: Harvard University Press. Goffman, Erving. 1967. Interaction Ritual: Essays on Face to Face Behavior. Garden City, NY: Doubleday. Goffman, Erving. 1971. Relations in Public. New York: Basic Books. Goffman, Erving, 1974. Frame Analysis. New York: Harper and Row. Goffman, Erving. 1981. “Footing.” In E. Goffman, Forms of Talk, 124–57. Philadelphia: University of Pennsylvania Press. Goldberg, Sanford C. 2015. Assertion: On the Philosophical Significance of Assertoric Speech. Oxford: Oxford University Press. Goldberg, Sanford C. (ed.). 2020. The Oxford Handbook of Assertion. Oxford: Oxford University Press. Goldman, Alvin. 1970. A Theory of Human Action. Englewood Cliffs, NJ: Prentice-Hall. Goldman, Alvin. 2006. Simulating Minds: The Philosophy, Psychology, and Neuroscience of Mindreading. Oxford: Oxford University Press. Goodwin, Charles. 1995. “The Negotiation of Coherence within Conversation.” In Coherence in Spontaneous Text, ed. M. A. Gernsbacher and T. Givón, 117–37. Amsterdam: John Benjamins. Green, Mitchell S. 2013. “Assertions.” In Pragmatics of Speech Actions, ed. M. Sbisà and K. Turner, 387–410. Berlin: de Gruyter. Green, Mitchell S. 2017. “Assertion.” Oxford Handbooks Online. DOI: 10.1093/oxfordhb/ 9780199935314.013.8. Greimas, Algirdas J. 1970. Du sens. Paris: Seuil. Greimas, Algirdas J. 1976. “Pour une théorie des modalités.” Langages 43: 90–107. Reprinted in Greimas (1983). Greimas, Algirdas J. 1983. Du sens II. Paris: Seuil.



311

Greimas, Algirdas J. and Joseph Courtés. 1979. Sémiotique: dictionnaire raisonné de la théorie du langage. Paris: Hachette. (English translation: Semiotics and Language. An Analytical Dictionary. Bloomington, IN: Indiana University Press, 1982.) Grice, Paul. 1957. “Meaning.” Philosophical Review 64. Reprinted in Grice 1989, 213–23. Grice, Paul. 1969. “Utterer’s Meaning and Intentions.” Philosophical Review 78. Reprinted in Grice 1989, 86–116. Grice, Paul. 1975. “Logic and Conversation.” In Syntax and Semantics, Vol. 3: Speech Acts, ed. P. Cole and J. L. Morgan. New York: Academic Press. Reprinted in Grice 1989, 22–40. Grice, Paul. 1982. “Meaning Revisited.” In Mutual Knowledge, ed. N. V. Smith. New York: Academic Press. Reprinted in Grice 1989, 283–303. Grice, Paul. 1986. “Reply to Richards.” In Philosophical Grounds of Rationality, ed. R. E. Grandy and R. Warner, 45–106. Oxford: Oxford University Press. Grice, Paul. 1989. Studies in the Way of Words. Cambridge, MA: Harvard University Press. Grice, Paul. 1991. The Conception of Value. Oxford: Oxford University Press. Grice, Paul. 2001. Aspects of Reason. Oxford: Oxford University Press. Gu, Yueguo. 1993. “The Impasse of Perlocution.” Journal of Pragmatics 20: 405–32. Hare, Richard M. 1952. The Language of Morals. Oxford: Oxford University Press. Harnish, Robert M. 1994. “Mood, Meaning and Speech Acts.” In Foundations of Speech Act Theory, ed. S. Tsohatzidis, 407–59. London: Routledge. Harnish, Robert M. 2009. “Internalism and Externalism in Speech Act Theory.” Lodz Papers in Pragmatics 5 (1): 9–31. Hart, Herbert L. A. 1949. “The Ascription of Responsibility and Rights.” Proceedings of the Aristotelian Society 49: 171–94. Heim, Irene. 1983. “On the Projection Problem for Presuppositions.” In Proceedings of the Second West Coast Conference on Formal Linguistics, ed. D. Flickinger et al. Stanford, CA: Stanford University Press. Reprinted in Pragmatics. A Reader, ed. S. Davis, 397–405. Oxford: Oxford University Press, 1991. Heinämaa, S. 2014. “ ‘An Equivocal Couple Overwhelmed with Life’: A Phenomenological Analysis of Pregnancy.” Philosophia 4 (1): 12–49. Hinchman, Edward S. 2020. Assertion and Testimony. In The Oxford Handbook of Assertion, ed. S. Goldberg, 555–79. Oxford: Oxford University Press. Hjelmslev, Louis. 1961. Prolegomena to a Theory of Language (English translation of Omkring sprogteoriens grundlaeggelse, 1943). Madison, WI: University of Wisconsin. Hollander, Michelle A., S. A. Gelman, and J. Star. 2002. “Children’s Interpretation of Generic Noun Phrases.” Developmental Psychology 38 (6): 883–94. Holmes, Janet. 1984. “Modifying Illocutionary Force.” Journal of Pragmatics 8: 345–65. Hornsby, Jennifer. 1988. “Things Done with Words.” In Human Agency. Language, Duty and Value. Philosophical Essays in Honor of J. O. Urmson, ed. Jonathan Dancy, J. M. E. Moravcsik, and C. C. W. Taylor, 27–46. Stanford, CA: Stanford University Press. Hornsby, Jennifer. 1994. “Illocution and Its Significance.” In Foundations of Speech Act Theory, ed. S. Tsohatzidis, 187–207. London: Routledge. Hornsby, Jennifer. 2004. “Agency and Actions.” Royal Institute of Philosophy Supplement 55: 1–23. Hornsby, Jennifer. 2010. “The Standard Story of Action: An Exchange,” Part 2. In Causing Human Actions: New Perspectives on the Causal Theory of Action, ed. J. H. Aguilar and A. A. Buckareff, 57–66. Cambridge, MA: MIT Press. Hornsby, Jennifer. 2011. “Subordination, Silencing, and Two Ideas of Illocution.” Jurisprudence 2: 379–85.

312



Hornsby, Jennifer and Rae Langton. 1998. “Free Speech and Illocution.” Legal Theory 4: 21–37. Hugues, George E. and M. J. Cresswell. 1996. A New Introduction to Modal Logic. London: Routledge. Irigaray, Luce. 1974. Speculum. De l’autre femme. Paris: Minuit. Irigaray, Luce. 1977. Ce sexe qui n’en est pas un. Paris: Minuit. Isard, Stephen. 1975. “Changing the Context.” In Formal Semantics of Natural Language, ed. E. Keenan, 287–96. Cambridge: Cambridge University Press. Jacobson, Daniel. 1995. “Freedom of Speech Acts? A Response to Langton.” Philosophy and Public Affairs 24: 64–79. Johnson, Lawrence E. 1992. Focusing on Truth. London: Routledge. Kant, Immanuel. 1998. Critique of Pure Reason (English translation of Kritik der reine Vernunft (1781/1787), ed. P. Guyer and A.W. Wood). Cambridge: Cambridge University Press. Kaplan, David. 1989. “Afterthoughts.” In Themes from Kaplan, ed. J. Almog, J. Perry, and H. Wettstein, 565–614. Oxford: Oxford University Press. Karttunen, Lauri. 1974. “Presupposition and Linguistic Context.” Theoretical Linguistics 1: 181–94. Karttunen, Lauri and Stanley Peters. 1979. “Conventional Implicature.” In Syntax and Semantics, Vol. 11: Presupposition, ed. Ch.-K. Oh and D. Dinneen, 1–56. New York: Academic Press. Katriel, Tamar and Marcelo Dascal. 1989. “Speaker’s Commitment and Involvement in Discourse.” In From Sign to Text, ed. Y. Tobin, 275–95. Amsterdam: John Benjamins. Katz, Jerrold J. 1977. Propositional Structure and Illocutionary Force. New York: Crowell. Kiparsky, Paul and Carol Kiparsky. 1971. “Fact.” In Semantics, ed. D. Steinberg and L. Jakobovits, 345–69. Cambridge: Cambridge University Press. Kissine, Michail. 2013. “Speech Act Classifications.” In Pragmatics of Speech Actions, ed. M. Sbisà and K. Turner, 173–201. Berlin: de Gruyter. Kotthoff, Helga and Ruth Wodak (eds.). 1997. Communicating Gender in Context. Amsterdam: John Benjamins. Kusch, Martin. 2002. Knowledge on Agreement. Oxford: Oxford University Press. Labinaz, P. 2018. “Brandom’s Deontic Scorekeeping Model and the Assertive Family.” Journal of Pragmatics 128: 53–66. Labinaz, Paolo and Marina Sbisà. 2020. “Informazione su argomenti di medicina: un’analisi dell’attendibilità e della fruibilità di testi online secondo una prospettiva pragmatica.” In Politiche del linguaggio in medicina, ed. S. Bigi, C. Caporale, and R. M. Zagarella, 41–74. Pisa: ETS. Labinaz, Paolo and Marina Sbisà. 2021a. “The Problem of Knowledge Dissemination in Social Network Discussions.” Journal of Pragmatics 175: 67–80. Labinaz, Paolo and Marina Sbisà. 2021b. “Speech Acts and the Dissemination of Knowledge.” In Approaches to Internet Pragmatics: Theory and Practice, ed. C. Xie, F. Yus, and H. Haberland, 145–72. Amsterdam: John Benjamins. Labov, William and David Fanshel. 1977. Therapeutic Discourse. New York: Academic Press. Lakoff, Robin T. 1973. “The Logic of Politeness: Or Minding Your P’s and Q’s.” In Papers from the 9th Regional Meeting of the Chicago Linguistic Society, ed. C. Corum, T. Cedric Smith-Stark, and A. Weiser, 292–305. Chicago: University of Chicago. Lakoff, Robin T. 1975. Language and Woman’s Place. New York: Harper and Row. Landowski, Eric, 1989. La société réfléchie. Paris: Seuil.



313

Langton, Rae. 1993. “Speech Acts and Unspeakable Acts.” Philosophy & Public Affairs 22 (4): 292–330. Langton, Rae. 1998. “Subordination, Silence, and Pornography’s Authority.” In Censorship and Silencing: Practices of Cultural Regulation, 261–83. Los Angeles: Getty Research Institute. Langton, Rae. 2009. Sexual Solipsism. Oxford: Oxford University Press. Langton, Rae. 2018a. “Blocking as Counter-speech.” In New Work on Speech Acts, ed. D. Fogal, D. Harris, and M. Moss, 144–64. Oxford: Oxford University Press. Langton, Rae. 2018b. “The Authority of Hate Speech.” In Oxford Studies in the Philosophy of Law, vol. 3, ed. J. Gardner, L. Green, and B. Leiter, 123–52. Oxford: Oxford University Press. Langton, Rae and Caroline West. 1999. “Scorekeeping in a Pornographic Language Game.” Australasian Journal of Philosophy 77 (3): 303–19. Leech, Geoffrey. 1983. Principles of Pragmatics. London: Longmans. Leonardi, Paolo and Marina Sbisà. 1977. “Atti linguistici e presupposizioni.” In Aspetti sociolinguistici dell’Italia contemporanea, ed. R. Simone and G. Ruggiero, 535–47. Rome: Bulzoni. Leonardi, Paolo and Marina Sbisà. 1978. “Presupposizioni.” Studi di grammatica italiana 7: 245–52. Leonardi, Paolo and Maurizio Viaro. 1990. Conversazione e terapia. Milan: Cortina. Leslie, Sarah Jane. 2007. “Generics and the Structure of the Mind.” Philosophical Perspectives 21 (1): 375–403. Leslie, Sarah Jane. 2008. “Generics: Cognition and Acquisition.” Philosophical Review 117 (1): 1–47. Leslie, Sarah Jane. 2012. “Generics Articulate Default Generalizations.” Revue Linguistique de Vincennes 41: 25–44. Leslie, Sarah Jane. 2017. “The Original Sin of Cognition.” Journal of Philosophy 114 (8): 393–421. Leslie, Sarah Jane and Susan A. Gelman. 2012. “Quantified Statements Are Recalled as Generics: Evidence from Preschool Children and Adults.” Cognitive Psychology 64 (3): 186–214. Leslie, Sarah Jane and Adam Lerner. 2016. “Generic Generalizations.” In The Stanford Encyclopedia of Philosophy (Winter 2016 Edition), ed. E. N. Zalta, URL = https://plato. stanford.edu/archives/win2016/entries/generics/. Levinson, Stephen C. 1983. Pragmatics. Cambridge: Cambridge University Press. Levinson, Stephen C. 2000. Presumptive Meanings. Cambridge, MA: MIT Press. Lewis, David. 1979. “Scorekeeping in a Language Game.” Journal of Philosophical Logic 8: 339–59. Lorber, Judith and Susan A. Farrell (eds.). 1991. The Social Construction of Gender. Newbury: Sage. Lyons, John. 1977. Semantics. Cambridge: Cambridge University Press. MacFarlane, John. 2011. “What Is Assertion?” In Assertion: New Philosophical Essays, ed. J. Brown and H. Cappelen, 79–96. Oxford: Oxford University Press. MacKinnon, Catharine. 1987. Feminism Unmodified. Cambridge, MA: Harvard University Press. MacKinnon, Catharine. 1993. Only Words. Cambridge, MA: Harvard University Press. Maitra, Ishani. 2009. “Silencing Speech.” Canadian Journal of Philosophy 39 (2): 309–38. Maitra, Ishani. 2011. “Assertion, Norms and Games.” In Assertion. New Philosophical Essays, ed. J. Brown and H. Cappelen, 277–96. Oxford: Oxford University Press.

314



McDonald, Lucy. 2020. “Your Word against Mine: The Power of Uptake.” Synthese 199 (1–2): 3505–26. McGowan, Mary Kate. 2003. “Conversational Exercitives and the Force of Pornography.” Philosophy & Public Affairs 31: 155–89. McGowan, Mary Kate. 2004. “Conversational Exercitives: Something Else We Do with Our Words.” Linguistics & Philosophy 27: 93–111. McGowan, Mary Kate. 2009. “On Silencing and Sexual Refusal.” Journal of Political Philosophy 17 (4): 487–94. McGowan, Mary Kate. 2017. “On Multiple Types of Silencing.” In Beyond Speech. Pornography and Analytic Feminist Philosophy, ed. M. Mikkola, 39–57. Oxford: Oxford University Press. McGowan, Mary Kate. 2019. Just Words: On Speech and Hidden Harm. Oxford: Oxford University Press. McGowan, Mary Kate, S. Helmers, J. Stolzenberg, and A. Adelman. 2011. “A Partial Defense of Illocutionary Silencing.” Hypathia 26 (1): 132–49. Mikkola, Mari. 2011. “Illocution, Silencing and the Act of Refusal.” Pacific Philosophical Quarterly 92: 415–37. Mikkola, Mari. 2016. The Wrong of Injustice: Dehumanization and Its Role in Feminist Philosophy. Oxford: Oxford University Press. Millikan, Ruth G. 1998a. “Language Conventions Made Simple.” Journal of Philosophy 95: 161–80. Millikan, Ruth G. 1998b. “Proper Function and Convention in Speech Acts.” In The Philosophy of P. F. Strawson, ed. L. E. Kahn, 25–43. Chicago: Open Court. Millikan, Ruth G. 2005. Language: A Biological Model. Oxford: Oxford University Press. Neale, Stephen. 1992. “Paul Grice and the Philosophy of Language.” Linguistics and Philosophy 15: 509–59. Ochs, Elinor. 1976. “The Universality of Conversational Postulates.” Language in Society 5: 67–80. Orletti, Franca. 2000. La conversazione diseguale. Rome: Carocci. Pagin, Peter. 2015. “Assertion.” The Stanford Encyclopedia of Philosophy (Spring 2015 Edition), ed. E. N. Zalta, URL = http://plato.stanford.edu/archives/spr2015/entries/asser tion/. Penco, Carlo. 1999. “Objective and Cognitive Context.” In Modeling and Using Context. Second International and Interdisciplinary Conference, CONTEXT ’99, Trento, Italy, September 1999. Proceedings, ed. P. Bouquet, L. Serafini, P. Brézillon, M. Benerecetti, and F. Castellani, 270–83. Berlin: Springer. Perrault, C. Raymond. 1990. “An Application of Default Logic to Speech Act Theory.” In Intentions in Communication, ed. Ph. R. Cohen, J. Morgan, and M. E. Pollack, 161–85. Cambridge, MA: MIT Press. Petrus, Klaus. 2005. “Illokution und Konvention, oder auch: was steckt nun wirklich hinter Austins ‘Securing of Uptake’?” Grazer Philosophische Studien 70: 101–26. Pitcher, George (ed.). 1964. Truth. Englewood Cliffs, NJ: Prentice-Hall. Potts, Christopher. 2005. The Logic of Conventional Implicatures. Oxford: Oxford University Press. Pozzato, Maria Pia. 1982. Il romanzo rosa. Milan: Espresso Strumenti. Pratt, Marie-Louise. 1977. Toward a Speech Act Theory of Literary Discourse. Bloomington, IN: Indiana University Press. Rajagopalan, Kanavillil. 2000. “On Searle (on Austin) on Language.” Language & Communication 20: 347–91.



315

Recanati, François. 2004. Literal Meaning. Oxford: Oxford University Press. Rizzolatti, Giacomo and Laila Craighero. 2004. “The Mirror-Neuron System.” Annual Review of Neuroscience 27: 169–92. Rizzolatti, Giacomo and Corrado Sinigaglia. 2007. “Mirror Neurons and Motor Intentionality.” Functional Neurology 22 (4): 205–10. Rhodes, Marjorie, S. J. Leslie, and C. Tworek. 2012. “Cultural Transmission of Social Essentialism.” Proceedings of the National Academy of Sciences 59 (34): 13526–31. Rossi-Landi, Ferruccio. 1961. Significato, comunicazione e parlare commune. Padua: Marsilio. Rossi-Landi, Ferruccio. 1978. Ideologia. Milan: ISEDI. Sadock, Jerrold. 1994. “Toward a Grammatically Realistic Typology of Speech Acts.” In Foundations of Speech Act Theory, ed. S. Tsohatzidis, 393–406. London: Routledge. Sadock, Jerrold and A. M. Zwicky, 1985. “Speech Act Distinctions in Syntax.” In Language Typology and Syntactic Description, Vol. 1: Clause Structure, ed. T. Shopen, 151–95. Cambridge: Cambridge University Press. van der Sandt, Rob A. 1988. Context and Presupposition. London: Croom Helm. van der Sandt, Rob A. 1989. “Presupposition and Discourse Structure.” In Semantics and Contextual Expressions, ed. R. Bartsch, J. van Benthem, and P. van Emde Boas, 267–94. Dordrecht: Foris. van der Sandt, Rob A. 1992. “Presupposition Projection as Anaphora Resolution.” Journal of Semantics 9: 333–77. Saul, Jennifer M. 2002a. “Speaker Meaning, What Is Said, and What Is Implicated.” Nous 36: 228–48. Saul, Jennifer M. 2002b. “What Is Said and Psychological Reality: Grice’s Project and Relevance Theorists’ Criticisms.” Linguistics and Philosophy 25: 347–72. Saul, Jennifer M. 2003. Feminism. Issues and Arguments. Oxford: Oxford University Press. Saul, Jennifer M. 2006. “Pornography, Speech Acts and Context.” Proceedings of the Aristotelian Society 106: 61–80. Saul, Jennifer M. 2017. “Are Generics Especially Pernicious?” Inquiry. An Interdisciplinary Journal of Philosophy. Published online: 13 Feb 2017. DOI: 10.1080/0020174X.2017.1285995. von Savigny, Eike. 1988. The Social Foundations of Meaning. Berlin: Springer. Sbisà, Marina. 1972. “Il problema della classificazione degli atti illocutori.” In Ricerche di filosofia linguistica, ed. Renzo Piovesan, 3–41. Florence: Sansoni. Sbisà, Marina. 1975. Wittgenstein. Rome: Astrolabio. Sbisà, Marina. 1976. “Speech acts e femminilità.” Problemi 47: 260–83. Sbisà, Marina. 1978. “Introduzione.” In Gli atti linguistici. Aspetti e problemi di filosofia del linguaggio, ed. M. Sbisà, 11–43. Milan: Feltrinelli. Sbisà, Marina. 1979. “Perlocuzione e presupposizioni.” In Retorica e scienze del linguaggio. Atti del X Congresso Internazionale di Studi della Società di Linguistica Italiana, ed. F. Albano Leoni and M. R. Pigliasco, 37–60. Rome: Bulzoni. Sbisà, Marina. 1980. “Per una metodologia della soggettività: domande post-wittgensteiniane.” Scienze umane 6: 153–79. Sbisà, Marina. 1981. “Wittgenstein e la crisi del soggetto.” In Uomini senza qualità. Atti delle Giornate di Studio, ed. F. Castellani, 21–49. Trento: U.C.T. Sbisà, Marina. 1983a. “Actes de langage et (acte d’)énonciation.” Langages 60: 99–106. Sbisà, Marina. 1983b. “Come non essere wittgensteiniani.” In Ludwig Wittgenstein e la cultura contemporanea, ed. A. Gargani, 109–15. Ravenna: Longo. Sbisà, Marina. 1984a. “On Illocutionary Types.” Journal of Pragmatics 8: 93–112. Reprinted in this volume as Essay 1.

316



Sbisà, Marina. 1984b. “Directivity and Sequencing.” In Discourse Analysis and Natural Rhetorics, ed. V. D’Urso and P. Leonardi, 143–51. Padua: Cleup. Sbisà, Marina. 1984c. La mamma di carta. Per una critica dello stereotipo materno. Milan: Emme. Sbisà, Marina. 1984d. “Frazer e Wittgenstein. Pathos delle spiegazioni causali e mistica della descrizione fisionomica.” La ricerca folklorica 10: 33–40. Sbisà, Marina. 1986a. “Manipulation et sanction dans la dynamique des actes de langage.” In Exigences et perspectives de la sémiotique. Recueil d’hommages pour A. J. Greimas, ed. H. Parret and H. G. Ruprecht, vol. 1, 529–38. Amsterdam: John Benjamins. Sbisà, Marina. 1986b. “Atti linguistici in un romanzo popolare.” In Dame, droga e galline, ed. A. Arslan Veronese, 291–305. Milan: Unicopli. Sbisà, Marina. 1987a. “Speech Acts and Context Change.” In Process Linguistics, ed. Th. T. Ballmer and W. Wildgen, 252–79. Tübingen: Niemeyer. Sbisà, Marina. 1987b. “Acts of Explanation: A Speech Act Analysis.” In Proceedings of the First International Conference on Argumentation, ed. F. H. van Eemeren et al., vol. 2, “19887–17. Dordrecht: Foris. Sbisà, Marina. 1987c. “Atti linguistici e dimensione timica.” Versus 47/48: 191–201. Sbisà, Marina. 1988. “Fra interpretazione e iniziativa.” In Le donne e i segni, ed. P. Magli, 38–49. Ancona–Bologna: Transeuropa. (1st ed. in Luoghi Comuni, Urbino: Centro Internazionale di Semiotica e Linguistica, 1985.) Sbisà, Marina. 1989a. Linguaggio, ragione, interazione. Per una teoria pragmatica degli atti linguistici. Bologna: Il Mulino. (2nd ed. as an e-book, 2009, Trieste: EUT.) Sbisà, Marina. 1989b. “Atti linguistici di spiegazione.” In La spiegazione nell’interazione sociale: aspetti cognitivi ed evolutivi, ed. M. S. Barbieri, 71–103. Turin: Loescher. Sbisà, Marina. 1990. “Speech Acts and the Expression of Affect.” Grazer Linguistische Studien 33/4: 279–95. Sbisà, Marina. 1991. “Sperimentazione di strategie pragmatiche nella comprensione e produzione di testi scritti.” In Scrivere nella scuola dell’obbligo, ed. M. Cortelazzo, 171–91. Florence: La Nuova Italia. Sbisà, Marina. 1992a. “Speech Acts, Effects and Responses.” In J. R. Searle et al., (On) Searle on Conversation, 101–11. Amsterdam: John Benjamins. Reprinted in this volume as Essay 2. Sbisà, Marina. 1992b. “Affetto e diritto come dimensioni dell’interazione verbale.” In La conversazione. Prospettive sull’interazione psico-sociale, ed. C. Galimberti, 185–203. Milan: Guerini. Sbisà, Marina. 1992c. “Atti linguistici ed espressione di affetto.” In La linguistica pragmatica, ed. G. Gobber, 353–78. Rome: Bulzoni. Sbisà, Marina. 1992d. “Parlando di parto. Vissuti e saperi nei discorsi delle donne.” In Come sapere il parto. Storia, scenari, linguaggi, ed. M. Sbisà, 185–237. Turin: Rosenberg & Sellier. Sbisà, Marina. 1994. “Language and Dialogue in the Framework of the Analytical Philosophy of Ordinary Language.” In Concepts of Dialogue Considered from the Perspective of Different Disciplines, ed. E. Weigand, 159–70. Tübingen: Niemeyer. Sbisà, Marina. 1995a. “Speech Act Theory.” In Handbook of Pragmatics, ed. J. Verschueren et al., 496–506. Amsterdam: John Benjamins. Reprinted with updates in Key Notions for Pragmatics, ed. J. Verschueren and J.-O. Östman, 229–44. Amsterdam. John Benjamins, 2009. Sbisà, Marina. 1995b. “The Relationship between Argumentation and Explanation in the Framework of a Speech-Act Analysis of Verbal Interaction.” In Proceedings of the III



317

ISSA Conference on Argumentation, Vol. 3: Reconstruction and Application, ed. F. H. van Eemeren et al., 3–15. Amsterdam: Sic Sat. Sbisà, Marina. 1996. “The Feminine Subject and Female Body in Discourse about Childbirth.” European Journal of Women’s Studies 3: 363–76. Sbisà, Marina. 1998. “Azione linguistica e status dei partecipanti.” In Di Pietro e il Giudice. L’interrogatorio al Tribunale di Brescia, ed. R. Galatolo and G. Pallotti, 61–7. Bologna: Pitagora. Sbisà, Marina. 1999a. “Ideology and the Persuasive Use of Presupposition.” In Language and Ideology. Selected Papers from the 6th International Pragmatics Conference, ed. J. Verschueren, 492–509. Antwerp: International Pragmatics Association. Reprinted in this volume as Essay 3. Sbisà, Marina. 1999b. “Presupposition, Implicature and Context in Text Understanding.” In Modeling and Using Context. Second International and Interdisciplinary Conference, CONTEXT ’99, Trento, Italy, September 1999. Proceedings, ed. P. Bouquet et al., 324–38. Berlin: Springer. Reprinted in this volume as Essay 5. Sbisà, Marina. 1999c. “La pragmatica nell’educazione linguistica e nella didattica. Parafrasi e acquisizione di informazioni.” In Insegnare italiano nella scuola del 2000, ed. U. Cardinale, 183–98. Padua: Unipress. Sbisà, Marina. 1999d. “Soggetto femminile, corpo e punto di vista.” In Reinventare la natura. Ripensare il femminile, ed. P. Cordin et al., 107–27. Trento: Edizioni Università degli Studi di Trento, Dipartimento di Scienze Filologiche e Storiche. Sbisà, Marina. 2001a. “Illocutionary Force and Degrees of Strength in Language Use.” Journal of Pragmatics 33: 1791–1814. Reprinted in this volume as Essay 6. Sbisà, Marina. 2001b. “Intentions from the Other Side.” In Paul Grice’s Heritage, ed. G. Cosenza, 185–206. Turnhout: Brepols. Reprinted in this volume as Essay 4. Sbisà, Marina. 2002a. “Speech Acts in Context.” Language & Communication 22: 421–36. Reprinted in this volume as Essay 7. Sbisà, Marina. 2002b. “Cognition and Narrativity in Speech Act Sequences.” In Rethinking Sequentiality, ed. A. Fetzer and C. Meierkord, 71–98. Amsterdam: John Benjamins. Reprinted in this volume as Essay 8. Sbisà, Marina. 2003. “Subject and Gender in H.D.’s Trilogy.” In H.D.’s Poetry: The Meanings That Words Hide (Ams Studies in Modern Literature, No 24), ed. M. Camboni, 91–113. New York: AMS Press. Sbisà, Marina. 2006a. “Speech Acts without Propositions?” Grazer Philosophische Studien 72: 155–78. Sbisà, Marina. 2006b. “Two Conceptions of Rationality in Grice’s Theory of Implicature.” In Rationality of Belief and Action. Proceedings of the International Philosophical Conference Held in Rijeka, May 27–28, 2004, ed. Elvio Baccarini and Snježana PrijićSamaržija, 233–47. Rijeka: University of Rjieka, Faculty of Arts and Sciences and Croatian Society for Analytic Philosophy. Reprinted in this volume as Essay 9. Sbisà, Marina. 2006c. “Communicating Citizenship in Verbal Interaction: Principles of a Speech-act Oriented Discourse Analysis.” In Analysing Citizenship Talk, ed. H. Hausendorf and A. Bora, 151–80. Amsterdam: John Benjamins. Sbisà, Marina. 2007a. “How to Read Austin.” Pragmatics 17 (3): 461–73. Reprinted in this volume as Essay 10. Sbisà, Marina. 2007b. “Pathways to Explicitness.” Lingue e Linguaggio 6: 101–20. Sbisà, Marina. 2007c. Detto non detto. Le forme della comunicazione implicita. Rome–Bari: Laterza. (2nd ed. 2010.)

318



Sbisà, Marina. 2009a. “Uptake and Conventionality in Illocution.” Lodz Papers in Pragmatics 5 (1): 33–52. Reprinted in this volume as Essay 11. Sbisà, Marina. 2009b. “Illocution and Silencing.” In Language in Life, and a Life in Language: Jacob Mey—A Festschrift, ed. B. Fraser and K. Turner, 351–7. Bingley: Emerald. Reprinted in this volume as Essay 12. Sbisà, Marina. 2011. “J. L. Austin’s Linguistic Method: Towards a Reassessment.” In John L. Austin et la philosophie du langage ordinaire, ed. S. Laugier and C. Al-Saleh, 109–35. Hildesheim: Olms. Sbisà, Marina. 2012a. “Austin on Meaning and Use.” Lodz Papers in Pragmatics 8 (1): 5–16. Sbisà, Marina. 2012b. “Classifying Illocutionary Acts, or, a Tale of Theory and Praxis.” In Pragmaticizing Understanding: Studies for Jef Verschueren, ed. M. Meeuwis and J.-O. Östman, 39–52. Amsterdam: John Benjamins. Sbisà, Marina. 2013a. “Some Remarks about Speech Act Pluralism.” In Perspectives on Pragmatics and Philosophy, ed. A. Capone et al., 227–43. Berlin: Springer. Sbisà, Marina. 2013b. “Soggetto e riconoscimento.” In Semiotica delle soggettività. Per Omar, ed. M. Leone and I. Pezzini, 169–92. Rome: Aracne. Sbisà, Marina. 2014a. “Evidentiality and Illocution.” Intercultural Pragmatics 11 (3): 463–83. Sbisà, Marina. 2014b. “The Austinian Conception of Illocution and Its Implications for Value Judgement and Social Ontology.” Ethics & Politics 16 (2): 619–31. Reprinted in Moral Realism and Political Decisions. Practical Rationality in Contemporary Public Contexts, ed. G. De Anna and R. Martinelli, 135–50. Bamberg: University of Bamberg Press, 2015. Sbisà, Marina. 2014c. “Austin on Language and Action.” In J. L. Austin on Language, ed. B. Garvey, 13–31. London: Palgrave Macmillan. Sbisà, Marina. 2015. “C’è del pragmatismo in J. L. Austin? Una rilettura delle proposte austiniane sul tema della verità.” Esercizi Filosofici 10 (2): 230–45. Sbisà, Marina. 2017a. “Implicitness in Normative Texts.” In Pragmatics and Law. Practical and Theoretical Perspectives (Series: Perspectives in Pragmatics, Philosophy and Psychology), ed. F. Poggi and A. Capone, 23–42. Berlin: Springer. Sbisà, Marina. 2017b. “Il soggetto femminile: dimensioni di analisi.” In S/Oggetti Immaginari. Letterature comparate al femminile, ed. L. Borghi and R. Svandrlik, 36–42. Ebook@Women. (1st ed. Urbino: Quattroventi, 1996.) Sbisà, Marina. 2018. “Ways to Be Concerned with Gender in Philosophy.” Phenomenology and Mind 15: 132–45. Reprinted in this volume as Essay 15. Sbisà, Marina. 2019. “Varieties of Speech Act Norms.” In Normativity and Variety of Speech Actions (Poznań Studies in the Philosophy of the Sciences and the Humanities 112), ed. M. Witek and I. Witczak-Plisiecka, 23–50. Leiden: Brill. Reprinted in this volume as Essay 14. Sbisà, Marina. 2020a. “Assertion among the Speech Acts.” In The Oxford Handbook of Assertion, ed. S. Goldberg, 159–78. Oxford: Oxford University Press. Reprinted in this volume as Essay 16. Sbisà, Marina. 2020b. “Illocuzione e dislivelli di potere.” In Rae Langton et al., Linguaggio d’odio e autorità. Lezioni milanesi per la Cattedra Rotelli, ed. C. Bianchi and L. Caponetto, 63–86. Milan-Udine: Mimesis. English translation in this volume as Essay 17. Sbisà, Marina. 2021a. “Presupposition and Implicature: Varieties of Implicit Meaning in Explicitation Practices.” Journal of Pragmatics 182: 176–88. Sbisà, Marina. 2021b. “(Im)politeness and the Human Subject.” In The Philosophy of (Im) politeness, ed. C. Xie, 157–77. Cham: Springer.



319

Sbisà, Marina. In preparation. Austinian Themes. Illocution, Action, Knowledge, Truth, and Philosophy. Oxford: Oxford University Press. Sbisà, Marina and Paolo Fabbri. 1980. “Models (?) for a Pragmatic Analysis.” Journal of Pragmatics 4: 301–19. Sbisà, Marina and Simona Regina. 2003. “Chi lo capisce questo libro? Questioni di comprensibilità nei libri di testo di fisica.” La fisica nella scuola 36: 91–102. Sbisà, Marina and Paola Rodari. 1989. “Il ruolo dell’implicito nella comprensione del testo.” In I luoghi della parola. L’educazione linguistica di base, ed. A. M. Mitri and A. De Rossi, 277–90. Florence: La Nuova Italia. Sbisà, Marina and Patrizia Vascotto. 2007. “How to Conceive of the Other’s Point of View: Considerations from a Case Study in Trieste.” In The Discourse of Europe. Talk and Text in Everyday Life, ed. S. Millar and J. Wilson, 153–71. Amsterdam: John Benjamins. Schegloff, Emanuel. 1988. “Presequences and Indirection: Applying Speech Act Theory to Ordinary Conversation.” Journal of Pragmatics 12: 55–62. Schegloff, Emanuel. 1992. “Repair after Next Turn: The Last Structurally Provided Defense of Intersubjectivity in Conversation.” American Journal of Sociology 97: 1295–1345. Schiffer, Stephen R. 1972. Meaning. Oxford: Oxford University Press. Searle. John R. 1965. “What Is a Speech Act?” In Philosophy in America, ed. M. Black, 221–39. London: Allen and Unwin. Searle, John R. 1969. Speech Acts: An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. Searle, John R. 1977. “Reiterating the Differences: A Reply to Derrida.” Glyph. John Hopkins Textual Studies 1: 198–208. Searle, John R. 1979. Expression and Meaning. Cambridge: Cambridge University Press. Searle, John R. 1989. “How Performatives Work.” Linguistics and Philosophy 12: 535–58. Searle, John R. 1992. “Conversation.” In (On) Searle on Conversation, ed. Searle et al., 7–29. Amsterdam: John Benjamins. Searle, John R. 1995. The Construction of Social Reality. New York: The Free Press. Searle, John R. and Daniel Vanderveken. 1985. Foundations of Illocutionary Logic. Cambridge: Cambridge University Press. Searle, John R. et al. 1992. (On) Searle on Conversation. Amsterdam: John Benjamins. Sinclair, John McH. and R. M. Coulthard. 1975. Towards an Analysis of Discourse: The English Used by Teachers and Pupils. Oxford: Oxford University Press. Soames, Scott. 1982. “How Presuppositions Are Inherited. A Solution to the Projection Problem.” Linguistic Inquiry 13: 483–545. Soames, Scott. 1989. “Presupposition.” In Handbook of Philosophical Logic, ed. D. Gabbay and F. Guenthner, vol. 4, 553–616. Dordrecht: Reidel. Sperber, Dan and Deirdre Wilson. 1995. Relevance. 2nd ed. Oxford: Blackwell. (1st ed. 1986.) Stalnaker, Robert. 1973. “Presuppositions.” Journal of Philosophical Logic 2: 447–57. Stalnaker, Robert. 1974. “Pragmatic Presupposition.” In Semantics and Philosophy, ed. M. Munitz and P. Unger. New York: New York University Press. Reprinted in Stalnaker 1999, 47–62. Stalnaker, Robert. 1978. “Assertion.” In Syntax and Semantics, Vol. 9: Pragmatics, ed. P. Cole. New York: Academic Press. Reprinted in Stalnaker 1999, 78–95. Stalnaker, Robert. 1998. “On the Representation of Context.” Journal of Logic, Language and Information 7. Reprinted in Stalnaker 1999, 96–113. Stalnaker, Robert, 1999. Context and Content. Oxford: Oxford University Press. Stevenson, Charles L. 1944. Ethics and Language. New Haven: Yale University Press.

320



Strawson, Peter F. 1950a. “On Referring.” Mind 59. Reprinted in P. F. Strawson, LogicoLinguistic Papers, 1–27. London: Methuen, 1971. Strawson, Peter F. 1950b. “Truth.” Proceedings of the Aristotelian Society, Suppl. Vol. 24. Reprinted in P. F. Strawson, Logico-Linguistic Papers, 190–213. London: Methuen, 1971. Strawson, Peter F. 1964. “Intentions and Conventions in Speech Acts.” Philosophical Review 73. Reprinted in P. F. Strawson, Logico-Linguistic Papers, 149–69. London: Methuen, 1971. Strawson, Peter F. 1971. Logico-linguistic Papers. London: Methuen. Streeck, Jürgen. 1980. “Speech Acts in Interaction: A Critique of Searle.” Discourse Processes 3: 133–54. Stubbs, Michael. 1996. Text and Corpus Analysis. Oxford: Blackwell. Sweetser, Eve. 1990. From Etymology to Pragmatics. Cambridge: Cambridge University Press. Thomason, Richmond H. 1990. “Accommodation, Meaning and Implicature: Interdisciplinary Foundations for Pragmatics.” In Intentions in Communication, ed. Ph. R. Cohen, J. Morgan, and M. E. Pollack, 325–65. Cambridge, MA: MIT Press. Travis, Charles. 2000. Unshadowed Thought: Representation in Thought and Language. Cambridge, MA: Harvard University Press. Travis, Charles. 2008. Occasion Sensitivity. Selected Essays. Oxford: Oxford University Press. Tsohatzidis, Savas (ed.). 1994. Foundations of Speech Act Theory. London: Routledge. Urmson, James O. 1977. “Performative Utterances.” Midwest Studies in Philosophy 2: 120–7. Vanderveken, Daniel. 1990. Meaning and Speech Acts, Vol 1: Principles of Language Use. Cambridge: Cambridge University Press. Vendler, Zeno. 1972. Res Cogitans. Ithaca, NY: Cornell University Press. Vision, Gerald. 2004. Veritas. The Correspondence Theory and Its Critics. Cambridge, MA: MIT Press. Walker, Tracy. 2013. “Requests.” In Pragmatics of Speech Actions, ed. M. Sbisà and K. Turner, 445–66. Berlin: de Gruyter. Weigand, Edda. 1989. Sprache als Dialog. Sprechakttaxonomie und kommunikative Grammatik. Tübingen: Niemeyer. Weigand, Edda. 1994. “Discourse, Conversation, Dialogue.” In Concepts of Dialogue Considered from the Perspective of Different Disciplines, ed. E. Weigand, 49–75. Tübingen: Niemeyer. Williamson, Timothy. 2000. Knowledge and Its Limits. Oxford: Oxford University Press. Williamson, Timothy. 2006. “Past the Linguistic Turn?” In The Future for Philosophy, ed. Brian Leiter, 106–28. Oxford: Oxford University Press. Williamson, Timothy. 2007. The Philosophy of Philosophy. Oxford: Blackwell. Witek, Maciej. 2015a. “Mechanisms of Illocutionary Games.” Language & Communication 42: 11–22. Witek, Maciej. 2015b. “An Interactional Account of Illocutionary Practice.” Language Sciences 47: 43–55. Witek, Maciej. 2019. “Coordination and Norms in Illocutionary Interaction.” In Normativity and Variety of Speech Actions (Poznań Studies in the Philosophy of the Sciences and the Humanities 112), ed. Maciej Witek and I. Witczak-Plisiecka, 66–97. Leiden: Brill. Wittgenstein, Ludwig. 1922. Tractatus Logico-Philosophicus. With English translation. London: Routledge and Kegan Paul.



321

Wittgenstein, Ludwig. 1953. Philosophische Untersuchungen/Philosophical Investigations. With English translation. Edited by Elizabeth Anscombe and Rush Rhees. Oxford: Blackwell. Von Wright, Georg H. 1972. An Essay in Deontic Logic and the General Theory of Action. Amsterdam: North Holland. Wodak, Ruth and Paul Chilton. 2005. A New Agenda in (Critical) Discourse Analysis. Theory, Methodology and Interdisciplinarity. Amsterdam: John Benjamins. Wunderlich, Dieter. 1976. Studien zur Sprechakttheorie. Frankfurt a/M.: Suhrkamp. Wunderlich, Dieter. 1977. “Assertions, Conditional Speech Acts, and Practical Inferences.” Journal of Pragmatics 1: 13–46. Yeager, Daniel. 2006. J. L. Austin and the Law. Lewisburg: Bucknell University Press.

Index For the benefit of digital users, indexed terms that span two pages (e.g., 52–53) may, on occasion, appear on only one of those pages. accommodation 3–4, 18–20, 29–30, 53–4, 62–4, 66, 98, 103, 133–4, 228, 242–4, 246–50, 260–1, 290, 293–5, 297–301, 303–4 actant 2, 4–5, 28–31, 35, 38, 155–6, 265–6 action 14, 17–19, 21–2, 129–30, 141, 144, 154–5, 166, 188, 193, 203–4, 212, 225, 227–8 ascription of (attribution of ) 21–2, 249, 264–5 Action (in narrative semiotics) 144, 156–7, 160, 166 adjacency pair 49–50 advice 39, 49, 116–19, 133, 240, 242–3 Alston, William 268–70, 275–6, 283–4, 290 agreement 138–9 by-default 19–20, 147–8, 150–1, 165–6, 210, 222–3, 229 on conventional states of affairs 139–40, 146–7, 160–1, 163, 227–30, 249–51 on illocutionary act performance (illocutionary effects) see also negotiation, interactional 1–3, 19–20, 24, 26–7, 34, 110–12, 142–3, 148–9, 152–4, 185–7, 192–3, 208–11, 221–3, 277–81 tacit 19–20, 185–6, 193, 210, 222–3 announcement 291, 297–8 apology 36–7, 136–7, 153–4, 224, 292–3 appointment 33, 290, 298–9 assertive family, the 237, 267–74, 277–8 assertion 27, 40, 46–9, 65, 71, 94, 141–3, 226, 239, 244–6, 267–87 norm of 234–5, 245–6, 282 Assertives 26, 35–6, 268–77, 283–6 Austin, John L. 1–3, 12–15, 17, 129, 191–4, 284–5, 287 How to Do Things with Words 13–14, 181–91 on action 21–2, 188, 193 on aetiolation 189–91, 214 on assertion 270–1, 277–8 on constative utterances 135–6, 182–3, 203–4 on correspondence with facts 25n.2, 238–41, 286 on felicity conditions 29–30, 65, 129–35, 185, 207, 232–8

on the hearer’s uptake 43, 146–7, 184, 187, 195–6, 208–9, 219–20 on the illocutionary act 18–19, 23–4, 26–7, 109–10, 146–8, 184–7, 196–7, 204–8, 219–21 on illocutionary act classification 24–6, 277–8 on infelicities 130–1, 185–7, 207, 233–4 on the locutionary act 106, 239 on making undone 185–7, 222 on performative utterances 129–30, 182–3 on the perlocutionary act 44–5, 110, 124, 187–9, 268–9 authority 5, 30, 33, 35–7, 39, 115–16, 118, 130–1, 139–40, 216–17, 227–8, 242–3, 249–50, 263–4, 273–4, 280–1, 283, 288–300, 303–5 Avramides, Anita 21, 75 Bach, Kent 16–17, 43, 94–5, 109, 130–1, 200–3, 220, 268–71, 275–6, 284 Bazzanella, Carla 15, 108n.2, 113n.9, 114, 124–5, 160n.8 Behabitives 3–5, 24–6, 31, 34, 36–42, 49, 123, 224, 240, 278–9, 282, 286–7, 292–3, 298 bet 39, 45–7 blocking 152, 300–3 Blum-Kulka, Shoshana 107, 127 Brandom, Robert 284–5 Brown, Penelope 127–8, 235, 294 bushes 114, 121n.14 Caffi, Claudia 15, 108n.2, 113n.9, 114, 114n.10, 121n.14, 123n.16, 160n.8, 191n.9 challenge 34, 36–7, 41 childbirth, discourse on 14 command 33–4, 44–5, 136–7, 141–2, 227–8, 291, 293–4 Commissives 3–5, 24–31, 34–5, 37–42, 116, 224, 276–7, 292–3, 304–5 communication 53–4, 65–7, 81–2, 84, 98–9, 111n.6, 130–2, 145–6, 165–6, 169–70, 175–7, 179–80, 203–4, 212–13, 215–18, 296–7

324



compliment 45–6, 107 congratulations 243–4 Constatives 268–72, 277–8, 285–6 constitutive rules (norms) 50–2, 139n.3, 231–8, 240–6, 249–53 context 16–17, 40, 42, 67, 69–70, 90–2, 129–31, 143, 169–70, 210–11, 239–41, 287 change 92–4, 129, 141–3, 246–8 cognitive 130–2, 136–9 construction of 133, 139–40 delimitation of 134–6, 140–1 givenness of 132–4 objective 63–7, 92–4, 96–100, 102, 136–43, 200–1 of a speech act 129–32, 136–41, 283 representation of (take on) 67, 69–70, 91–7, 99–100, 102–4 conventionality of illocution 1–2, 14–15, 17–19, 26–7, 110–12, 126, 142–3, 145–8, 185–7, 191–2, 195, 199–211, 221–3, 250–2, 253 of speech act norms 250–3 of states of affairs 4–5, 93, 110, 139–40, 143, 148, 193, 223–5 conversation 15, 43–4, 50–2, 65, 98–9, 113, 134–5, 137, 192–3, 236–7, 293–4 conversational cooperativity 3, 51–2, 65–6, 80–1, 83–5, 89–91, 111, 142, 167–9, 174, 216–17, 236, 252 maxims 20, 82–5, 167–70, 231, 236–8 sequence 43–4, 51–2, 149–52 Conversational Exercitives 212, 293–4 Davidson, Donald 21–2, 214 Declarations 32, 36, 41, 202–3, 223n.4, 277 default assumptions (acceptance, recognition) 3, 64–6, 94, 98–9, 103, 130–1, 133–4, 138–40, 168–9, 229, 233–4, 249, 300–1 effectiveness see also agreement, by-default 94, 210, 279–80, 300 situation (conditions) 147–8, 168–9, 257–8, 279 defeasibility 84, 92, 133, 222, 233–4, 281 of illocutionary effects 2, 185–8, 192, 208, 222–3, 230, 279–81 degrees of strength (see also mitigation, mitigation/reinforcement, reinforcement) of speech acts 105–9, 111–12, 127–8 of illocutionary effects 115–21, 125–6 of expressed attitudes 106–7, 122 of perlocutionary goals 123–5

deontic modalities 27, 111–12, 126–8, 223 objects 228 power 292–3, 303–5 properties (roles, states) 2–3, 113–15, 117–20, 125–6, 263–6, 276–81, 292–4 Derrida, Jacques 189–90, 214 Directives 26, 28, 32, 34, 41, 107, 109, 118, 276–7, 286, 290–1 discourse, analysis of 3, 13–14, 17, 37, 53, 72, 83–5, 91–2, 107, 114–15, 127, 148, 166, 179–80, 223–5, 255–6 discrimination 254–5, 260–2, 266, 301–3 Ducrot, Oswald 65 Eco, Umberto 14, 72 epistemic injustice 215–16 evaluation see also value judgment 34, 40, 87–8, 115, 291 exercitive assertion 40, 292, 295–6 Exercitives 3, 24–8, 30–4, 37–42, 118, 216–17, 224, 240, 276–9, 289–305 explicitation 4, 20, 71, 300–2 explicit performative utterance (formula) 124, 129–30, 146–7, 192, 199, 207–9 Expositives 3, 20, 24–5, 35–6, 50–1, 224, 240, 268–70, 277–8, 282–3, 291–2 Expressives 26, 36, 41, 49, 107, 117, 122–3, 126–7, 276–7, 286–7 Fabbri, Paolo 14 face 4, 41, 105n.1, 107, 127–8, 235–6, 238, 252n.12 fact/value dichotomy 13, 225–7 feminism 14 Fricker, Miranda 215–17 Gauker, Christopher 16–17, 63–7, 92, 98, 111n.6, 136–8, 140, 143 Gazdar, Gerald 43, 66, 92–3, 141 gender concepts 254, 259–62 equal opportunities 194 essentialism vs anti-essentialism 256–9 generics 260–1 models 13, 70, 254–9 prejudice 215–18, 255–6 Generalized Conversational Implicature Theory 170, 175–6 Gilligan, Carol 257 Goffman, Erving 14–15, 41–2, 127–8, 133, 190–1 Green, Mitchell 268n.2, 276n.7 greeting 49–50, 292–3

 Greimas, Algirdas J. 2, 14, 27n.4, 111n.7, 144, 155, 263n.10, 265–6 Grice, Paul 2–3, 15–16, 72–4, 87–9, 130, 179 on conventional implicature 20, 95–6, 169, 177, 270 on conversational implicature 82–3, 95–6, 167–70, 177 on non-natural meaning (speaker meaning) 73–5, 78–9, 88, 197–8 on rationality 81–2, 86–9, 175–9 on the Cooperative Principle (conversational maxims) 80–3, 88–9, 167–8, 171–2, 236–8, 252 on the person 85–7, 176–7 on truth vs appropriateness 130–1, 284 on value 86–8, 176–7 Harnish, Robert M. 16–17, 20–1, 43, 109, 130–1, 200–3, 220, 268–71, 275–6, 284 Hart, Herbert L. A. 222 hedges 37, 49, 114 Hornsby, Jennifer 21–2, 212–13, 215–17, 255, 265 human rights 194, 229–30, 263 ideology 54, 67–8, 70 illocutionary act 1–3, 14–15, 18–21, 23–4, 43–4, 93, 126–7, 152–4, 156–7, 165, 181–7, 192, 195, 207–13, 215–19, 224–5, 228, 232–3, 235, 237–8, 267 act, conditional 45–7 act classification 3, 13, 23–37, 199–204, 224, 274–7, 283 effect 1–3, 26–31, 46–7, 50–1, 108–15, 122, 124–6, 141, 145–9, 184–7, 204–8, 219–25, 231, 238 entailment 271–2 force 44, 48, 106–12, 126, 141, 151–3, 184, 209 force indicators 18–19, 40, 42, 107, 127, 209, 224–5, 235–6, 286 procedure 18–20, 192, 207–9, 233–6, 238, 250–2, 272, 277, 282–3, 287 implicature 3–4, 95–7, 101–2, 167, 169 conventional 20, 91, 95–6, 100–1, 270, 294n.4 conversational 3, 20, 80–5, 90–1, 95–7, 102, 167–70, 172–4, 178–80 calculability of 72, 171–9 vs presupposition 4, 20, 91, 99–104 infelicity 3, 26–7, 48, 133, 147–8, 186–7, 230, 233n.2, 249–51, 279–80 abuse 130–1, 214, 216, 234–5 misfire 130–1, 200–1, 215–17

325

intention, ascription of (attribution of ) 20–1, 73–80, 82–5, 89, 173, 215–16, 249 interpretation of symptoms 83–5 intersubjective recognition 88, 254–5, 262–6 insisting 42, 124, 160–5 insult 293, 295 Irigaray, Luce 14, 257 Jacobson, Daniel 212–13, 215 justification 45–6, 176, 216, 282, 292 Kant, Immanuel 236 knowledge 2, 5, 17, 27, 31, 35–7, 39–40, 90–2, 94, 102–4, 115, 170, 215–17, 226–7, 229, 234–5, 240–1, 245–6, 278–83, 286–7, 291, 297–8 Lakoff, Robin Tolmach 255 Langton, Rae 212–18, 265, 289, 294, 298–301 Leonardi, Paolo 14 Leslie, Sarah-Jane 260–1 Levinson, Steve 127–8, 141, 170, 175–6, 235, 294 Lewis, David 2, 53–4, 63–4, 66, 133–4, 228, 246–8 MacKinnon, Catharine 212, 289 making undone 1–2, 26–7, 185–7, 222, 234, 251–2 Manipulation 37–9, 41–2, 155–6, 160, 166 McGowan, Mary-Kate 212, 289, 293–4 meaning 14, 62, 80–1, 94–5, 111, 134–6, 182, 191, 226–7, 236, 239, 263–4, 284–6 natural 78–9 non-natural see also speaker meaning 72–80, 83–4, 88–9, 170, 177, 179–80, 196 speaker 167, 184, 197–8, 200, 220 metaphysics 13–14, 17, 87–9, 193, 255 Millikan, Ruth G. 209n.8, 221n.2, 250–1 Mikkola, Mari 266n.12 mitigation see also mitigation/ reinforcement 15, 37, 105–8, 112–3, 126–8, 190–1, 271–4 examples of 116–21, 123, 125–6 mitigation/reinforcement indicators see also bushes, hedges 107, 127 of speaker commitment 120 of speaker entitlement 115 of addressee obligations 118–19 of expressed attitudes 122 of perlocutionary goals 123–4 naming of a ship, the 110, 146–7, 184–5, 205–6, 209–10, 220–1, 288–9 narrative schema 144, 155–66

326



narrative (Continued) semiotics 2, 14, 24, 37–8, 144, 155, 263–6 sequence 37–8, 41–2, 144, 155–7, 166 negotiation, interactional 18–20, 51–2, 106–7, 111, 143, 148–53, 166, 192–3, 300 not-at-issue content 20, 294, 297–8 objective requirements see also speech act correctness/incorrectness 231, 238–46, 252–3, 282 offer 45–7, 150–1, 158–9 order 32–3, 130–1, 133, 139–40, 147–9, 249–52, 288–9, 291 ordinary language philosophy 12–13, 73–4, 80–1 Penco, Carlo 16–17, 136 perlocutionary act 44–5, 110, 141–2, 187–9, 191, 268–9 effect 26–7, 42, 44–5, 145–6, 187–8, 192–3, 220, 224–5, 242, 251–2, 289n.2 goal 34–5, 45, 49–50, 117, 123–4, 198 object (response) 32, 34, 45, 124, 213–14, 291 sequel 124–5 permission 33, 39, 118–19, 130–1, 288–9 Petrus, Klaus 204, 211 politeness 112–13, 124–5, 127–8, 152 rules (norms, maxims) of 83–5, 231, 235–6, 238, 252n.12 pornography 212–15, 217–18 postulating 272 Potts, Christopher 294n.4 power see also deontic power 19–20, 32–4, 37, 41, 106–7, 111–12, 114–15, 117, 119, 125–6, 293–4, 298–9 relations 2, 204, 212, 288–90, 292–3, 303–5 prestige 289n.2, 291–2, 298–300, 303 presupposition 3–4, 53, 66–7 appropriateness-related 20 cancellation 102–3 force-related 20 inducers (triggers), examples of 55–6, 58–61, 69, 260–1, 297–8 informative 15–16, 53–6, 64, 133–4, 137–8 linguistically indicated 53–4, 65–7, 70, 91, 98–9, 137, 293–4 persuasive 4, 54, 56–62, 64–7, 70, 217–18, 260–1 pragmatic 62–4, 90–1, 98 semantic 97–8 speaker see pragmatic presupposition vs implicature distinction 4, 20, 91, 99–104 promise 27–8, 34, 45–7, 115, 118–19, 130–1, 200–1, 209–10, 224, 234–5, 237–8, 241–2, 251–2

refusal 213–18 reinforcement see also mitigation/ reinforcement 15, 105–8, 112–3, 126–8, 273–4 examples of 115–16, 119–20, 122–4 Relevance Theory 90–1, 130–1, 135, 169–70, 173–6 response 19–20, 24, 43–50, 110, 113–14, 117, 124, 146, 148–51, 152n.4, 159, 164–5, 205–6, 220, 263, 292 responsibility 2, 21–2, 141, 188–9, 193, 222, 225, 252–3, 287 Rossi-Landi, Ferruccio 12–13, 70 Sanction 37–42, 155–8, 160, 164–6 Saul, Jennifer M. 15–16, 168–9, 173, 179, 212–13 scorekeeping 2, 246–8 Searle, John R. 14, 181, 223n.4 on assertion see also Assertives 268, 274 on background 43n.1, 51–2, 130–1, 134–5 on felicity conditions 25–6, 130–5, 200–1, 232–5, 237–8 on the illocutionary act 106–9, 184, 197–8, 201–2, 267 on illocutionary point 26, 41, 47, 109, 131–2, 201–2, 276–7 on illocutionary act classification 26, 28, 32, 35–6, 41, 202–3, 268–9, 289 on non-serious speech acts 189–91, 214 silencing 212–18, 265, 289 situatedness 134–6, 140, 143 Soames, Scott 64, 66, 98, 137–8 social contract, the 193, 225 social ontology 193, 225, 227–9 speech act 1, 18–19, 129–32, 144, 247–8, 253, 267, 288–9 back-door 294, 301–2 correctness/incorrectness 226–7, 229–30, 240–1, 286 initiative vs reactive 156–7 maxims 231, 236–8, 242–6, 251–3 non-central 20, 95–6, 169, 270 non-serious 189–91, 214, 216–17 norms see also constitutive rules, speech act maxims, objective requirements 231–2, 247–53 sequence see also conversational sequence, narrative sequence 144–6, 148–55, 159, 165–6 Stalnaker, Robert 53–4, 63–4, 93–4, 130–3, 135, 137–8, 141–3 Strawson, Peter F. 2–3, 16–17, 65–6, 77–8, 97, 110n.4, 141, 146–7, 184, 195–206, 208–9, 219–21, 239

 subject (subjectivity) 14, 20–1, 28, 73–4, 87–9, 123n.16, 174, 179–80, 190–1, 214, 216–17, 236, 249, 254–5, 257n.1, 262–6 Subject (in narrative semiotics) 38, 265–6 subordinating 212–13, 217–18, 289 text 4, 13, 91–2, 126–7 text–context relation 90–3 truth 17, 26, 39–40, 70, 84–5, 92, 134–5, 197, 239–41, 245–6, 277–8, 280–1, 286–7 aptness 268–9, 275–6, 282–3, 285–6 commitment to 20, 45–6, 48, 94, 97, 125–6, 224, 234–5, 271–2, 274–5 conditions 14, 73, 167, 202–3, 236, 275–6, 284–6 value 65, 95, 97, 130–1, 169, 284 uptake, hearer’s see also agreement on illocutionary act performance; response 1–3, 15, 19–20, 24, 43–50, 109–11, 114–15, 146–8, 150–2, 164, 181,

327

183–8, 192–3, 195–200, 205, 207–11, 215, 219–23, 248 Urmson, James O. 13–14, 202–4 value judgment 70, 86–7, 89, 225–7, 229, 272–3, 286–7 van der Sandt, Rob 64, 66 Vanderveken, Daniel 109, 268–9, 271–6 Vendler, Zeno 32 Verdictives 3, 24–6, 31, 34–42, 86, 116, 150–1, 157, 216, 224, 226–7, 229–30, 240, 277–83, 286–7, 292–3, 304–5 warning 39, 49, 196, 205–6, 209–11, 224, 268–9, 291 Weigand, Edda 156–7 West, Caroline 217–18, 255 Williamson, Timothy 234–5, 245–6, 251–2, 255, 282 Witek, Maciej 221n.2, 223, 232n.1, 247–51 Wittgenstein, Ludwig 12–15, 231, 255, 268n.2