Sbisà on Speech as Action (Philosophers in Depth) 3031225279, 9783031225277

Citation preview

PHILOSOPHERS IN DEPTH

Sbisà on Speech as Action Edited by Laura Caponetto · Paolo Labinaz

Philosophers in Depth

Series Editor Constantine Sandis Department of Philosophy University of Hertfordshire Hatfield, UK

Philosophers in Depth is a series of themed edited collections focusing on particular aspects of the thought of major figures from the history of philosophy. The volumes showcase a combination of newly commissioned and previously published work with the aim of deepening our understanding of the topics covered. Each book stands alone, but taken together the series will amount to a vast collection of critical essays covering the history of philosophy, exploring issues that are central to the ideas of individual philosophers. This project was launched with the financial support of the Institute for Historical and Cultural Research at Oxford Brookes University, for which we are very grateful. Constantine Sandis

Laura Caponetto  •  Paolo Labinaz Editors

Sbisà on Speech as Action

Editors Laura Caponetto Newnham College University of Cambridge Cambridge, UK

Paolo Labinaz Department of Humanities University of Trieste Trieste, Italy

ISSN 2947-552X     ISSN 2947-5538 (electronic) Philosophers in Depth ISBN 978-3-031-22527-7    ISBN 978-3-031-22528-4 (eBook) https://doi.org/10.1007/978-3-031-22528-4 © The Editor(s) (if applicable) and The Author(s), under exclusive licence to Springer Nature Switzerland AG 2023 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Palgrave Macmillan imprint is published by the registered company Springer Nature Switzerland AG. The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface

Marina Sbisà is one of the world’s pre-eminent scholars in philosophical pragmatics, publishing landmark papers in speech act theory, implicit communication, and presupposition. In 2018, she retired from the University of Trieste after having worked there for over forty years. During this period, she also held prestigious visiting appointments at the University of Oxford, Johns Hopkins University, and Rutgers University. Professor Sbisà began her work in Austinian philosophy of language at a time when most publications in the field were conducted with little care for their practical import. Against the grain, she developed her ideas with a keen awareness for both their historical context and practical implications, extending far beyond philosophical debate. Sbisà’s socially engaged approach to the philosophy of language would anticipate the current pragmatic turn in philosophy by some decades. Her “who is doing what to whom” approach to discourse is deeply influential beyond the narrow confines of linguistic philosophy, for example, in contemporary discussions of feminism, citizenship, hate speech, silencing, and pornography. Her work on J. L. Austin remains equally unmatched. She has authored dozens of seminal articles on his philosophy and with J. O. Urmson co-edited the definitive second English edition of How to Do Things with Words (1975) as well as its Italian translation with Carlo Penco (1987). Sbisà is moreover a distinguished scholar of Wittgenstein and Grice, siding with the former against the latter in pioneering a v

vi Preface

conventionalist view of speech actions that is partly governed by illocutionary norms. The volume you hold in your hands (or are viewing on your screens) focuses on Sbisà’s understanding of speech as a kind of action. According to this Austinian approach to the relation of language to behaviour, we would do well to see the philosophy of language as a subset of the philosophy of action (albeit a rather large and sprawling one). This is because, to paraphrase Wittgenstein, words have no meaning outside of their uses in our behavioral practices. This insight, lost to most philosophers of language from Grice onwards, is a guiding thread throughout Sbisà’s work. Without a sound philosophy of action, the philosophy of speech acts is ungrounded. Indeed, on her account, the distinction between acts and actions is crucial to understanding “what notion of action Austin intended to apply to locutionary, illocutionary and perlocutionary acts.” Not all of the contributors in this volume will agree with the above way of putting things. But its world-class line up of philosophers of language, linguists, social epistemologists, action theorists, and communication scholars is unified in celebrating and exploring Sbisà’s ongoing legacy in the area. Sbisà on Speech as Action brings together essays on illocutionary force, speech act norms, discourse analysis, uptake, ideology, persuasion, interactional achievement, silencing, sexual negotiation, propaganda, deception, and more, with replies by Marina herself to each of the contributions. I am deeply grateful to Laura Caponetto and Paolo Labinaz for enthusiastically accepting my invitation to edit a volume on Sbisà’s work, not to mention bringing together such a fine cast of distinguished scholars to discuss its themes. I hope that you enjoy this tribute to Professor Sbisà as much as I have. London January 2022

Constantine Sandis

Contents

 arina Sbisà’s Deontic Approach to Speech Actions  1 M Laura Caponetto and Paolo Labinaz  hould Speech Act Theory Eschew Propositions? 27 S Mitchell Green  n the Conventional Nature of Illocutionary Acts: Uptake, O Conventions, and Illocutionary Effects 49 Bruno Ambroise  arieties of Uptake 75 V Claudia Bianchi Interactional Negotiation 97 Maciej Witek  ome Varieties of Illocutionary Pluralism121 S Casey Rebecca Johnson  peech Acts and Ventriloquation: The Contribution of Marina S Sbisà to a General Theory of Action and Performativity143 François Cooren vii

viii Contents

 owards a Unified Theory of Illocutionary Normativity165 T Neri Marsili l locutionary Force, Speech Act Norms, and the Coordination and Mutuality of Conversational Expectations195 Sanford C. Goldberg Speech in Non-ideal Conditions: On Silence and Being Silenced221 Alessandra Tanesini  Speech-Act Theoretic Analysis of White (Prosocial) Lies245 A Marina Terkourafi  resupposition and Propaganda: A Socially Extended Analysis275 P Michael Randall Barnes  eplies to Contributors299 R Marina Sbisà

Notes on Contributors

Bruno Ambroise  is Junior Researcher in Philosophy of Language at the French National Centre for Scientific Research (CNRS) in Paris. He is working on the epistemology and history of speech acts theories and pragmatics and studying the relations between linguistic theories and social sciences. He is the author of many papers on Austin’s philosophy, speech acts theories and pragmatics and a book entitled Qu’est-ce qu’un acte de parole? (Paris: Vrin, 2008) and has edited the volume De l’action du discours: Le concept de speech act au prisme de ses histoires (London: ISTE Editions, 2018). Michael Randall Barnes  is a Research Fellow with the Humanising Machine Intelligence project, at the Australian National University. He works in the social philosophy of language and in applied ethics. He works mainly on issues relating to how speech harms and is particularly interested in the harms made possible by online speech. He holds a PhD in Philosophy from Georgetown University. Claudia Bianchi  (PhD at École Polytechnique, Paris) is Full Professor of Philosophy of Language at the Philosophy Faculty of Vita-Salute San Raffaele University, Milan. Her main research interests are in the fields of philosophy of language, pragmatics and feminist philosophy of language. ix

x 

Notes of Contributors

In particular, she works on hate speech, slurs and discursive injustice. At Vita-Salute San Raffaele University, she is the Coordinator of the PhD programme in Philosophy and the Director of the research centre Gender (Interfaculty centre for gender studies) and of the Master in Communication of Science and Health. She has published four books, edited a further one and co-edited three more. Her latest authored work is Hate speech. Il lato oscuro del linguaggio (Laterza, 2021). Laura Caponetto  is the Sarah Smithson Research Fellow in Philosophy at Newnham College, University of Cambridge. Before joining Newnham, she was a Postdoctoral Researcher in Philosophy at Vita-Salute San Raffaele University (Milan) and an Adjunct Lecturer in Pragmatics at the University of Pavia. Her research is primarily in applied philosophy of language. She is especially interested in how speech act theory, and pragmatics more broadly, can be brought to bear on issues of moral, political and societal concern. Her articles have been published in journals such as Analysis, Synthese, Topoi and Ethical Theory and Moral Practice. She is among the founding members of the Italian Society for Women in Philosophy (SWIP Italia). She holds a PhD in Philosophy from VitaSalute San Raffaele University (2019). François Cooren  PhD (Université de Montréal, 1996), is a Professor at the Université de Montréal, Canada, where he is also Chair of the Department of Communication. His research focuses on organizational communication, language and social interaction, and communication theory. He is past president of the International Communication Association (ICA, 2010–2011) and of the International Association for Dialogue Analysis (IADA, 2012–2021) and former Editor-in-Chief of Communication Theory (2005–2008). He was elected ICA Fellow in 2013 and NCA distinguished scholar in 2017. He has published 15 volumes (4 as an author or co-author and 11 as an editor or co-editor) and authored more than 80 peer-reviewed articles and close to 60 chapters. He is one of the founding members of what is today known as the Montreal School of organizational communication, one of the main branches of the Communication as Communicative of Organization (CCO) approach.

  Notes on Contributors 

xi

Sanford C. Goldberg  is Chester D. Tripp Professor in the Humanities and Professor of Philosophy at Northwestern University, and Professorial Fellow in the Arché Research Centre at the University of St Andrews. He is the author, most recently, of Foundations and Applications of Social Epistemology (2021), Conversational Pressure: Normativity in Speech Exchanges (2020), To the Best of Our Knowledge: Social Expectations and Epistemic Normativity (2018), and Assertion: On the Philosophical Significance of Assertoric Speech (2015). Mitchell Green is Professor of Philosophy at the University of Connecticut, as well as Editor-in-Chief of the journal Philosophia. His research interests include the evolutionary biology of communication, speech acts and their role in conversation, empathy, self-expression, selfunderstanding and the epistemic value of fiction. He is author of Engaging Philosophy: A Brief Introduction (2006), Self-Expression (2007), Moore’s Paradox (co-ed. with J. Williams; 2007), Know Thyself: The Value and Limits of Self-Knowledge (2017) and The Philosophy of Language (2021), as well as over threescore articles in edited volumes and in such journals as Linguistics and Philosophy, Oxford Studies in the Philosophy of Language, Noûs, Mind, Topoi, British Journal of Aesthetics, Theoretical Linguistics, Minds and Machines, and Philosophical Studies. Special issues of the journals  Grazer Philosophische Studien (vol. 96 (2019)) and Organon Filozofia (vol. 29 (2021)) contain articles by other philosophers focusing on Green’s research contributions over the last quarter-century. Casey Rebecca Johnson  is an Assistant Professor at the University of Idaho, USA. Her research focuses on the effects of social position and power on knowers’ ability to do what they want with their words and their knowledge. She is the author of Epistemic Care: Vulnerability, Inquiry, and Social Epistemology (2023). She edited the volume Voicing Dissent (2017). She has also published papers on illocutionary force in Feminist Philosophy Quarterly and Synthese. Paolo Labinaz  is an Associate Professor at the Department of Humanities of the University of Trieste. His primary research areas include the philosophy of language, argumentation theory and the philosophy of cogni-

xii 

Notes on Contributors

tive science. He explores topics related to speech act theory and its applications, as well as the relationship between reasoning and argumentation. He is the author of the books La razionalità (Carocci, 2013) and L’asserzione come azione linguistica: aspetti epistemici, cognitivi e sociali (EUT, 2019) and has published numerous articles in national and international journals and edited volumes. Neri Marsili  (PhD 2018, Sheffield) is a “Talent Attraction” Research Fellow at the National Distance Education University (UNED), in Madrid, where he is leading a five-year interdisciplinary project on online disinformation. His primary areas of research include philosophy of language, epistemology, aesthetics and experimental philosophy. His work has been published in journals such as Cognition, The Journal of Philosophy, Philosophical Quarterly, Mind & Language, The Journal of Pragmatics, Philosophical Studies, Synthese and Analysis. Together with Peter Pagin, he manages the “Assertion” entry of the Stanford Encyclopedia of Philosophy. Marina Sbisà  is a Senior Scholar at the University of Trieste, where she was Professor of Philosophy of Language until retiring in 2018. She has held visiting positions at Fuji Women’s University, the University of Amiens and CURAPP-CNRS, the University of Szczecin, and Magdalen College and New College, University of Oxford. Her research interests include speech act theory, presupposition and implicature, and ordinary language philosophy (Wittgenstein, Austin, Grice), as well as the fields of discourse analysis and gender studies. She collaborated with J.O. Urmson on the second revised edition of J.L. Austin, How to Do Things with Words (1975). She is the author of Essays on Speech Acts and Other Topics in Pragmatics (2023), Detto non detto. Le forme della comunicazione implicita (Laterza, 2007) and Linguaggio ragione interazione. Per una pragmatica degli atti linguistici (Il Mulino, 1989; 2nd ed. EUT 2009). She is a member of the Consultation Board of the International Pragmatics Association and President of the Italian Society for Women in Philosophy (SWIP Italia). Alessandra Tanesini  is Professor of Philosophy at Cardiff University. Her latest book is The Mismeasure of the Self: A Study in Vice Epistemology (2021). Her work lies at the intersection of ethics, politics, the philoso-

  Notes on Contributors 

xiii

phy of language and epistemology with a focus on epistemic vice, silencing, prejudice and ignorance. Marina Terkourafi  studied linguistics at the University of Athens and the University of Cambridge, where she received her PhD in 2001 with a thesis entitled “Politeness in Cypriot Greek: The Frame-Based Approach”. She went on to be the first A.G. Leventis Postdoctoral Research Fellow at the British School at Athens (Greece) and then a Research Fellow at the Computer Lab of the University of Cambridge (UK), before taking up a tenure-track position in linguistics at the University of Illinois at UrbanaChampaign (USA). Since 2017, she is Professor and Chair of Sociolinguistics at the University of Leiden (The Netherlands). In that same year, she took up a five-year post as Co-Editor-in-Chief of the Journal of Pragmatics. In her work, she tries to bring into dialogue different areas of pragmatics, especially experimental, post-Gricean and crosscultural approaches, with a view to achieving a more realistic view of how users of different language varieties communicate, centring on the notions of conventionalisation and hearer’s meaning. Maciej Witek  is Professor of Philosophy at the University of Szczecin, where he is Head of the Institute of Philosophy and Cognitive Science and the coordinator of the Cognition & Communication Research Group. His main area of interest is philosophy of language, with particular emphasis on speech act theory and post-Gricean pragmatics. In his research, he develops the Austinian approach to speech acts and uses it to explain discursive phenomena such as accommodation, presumptions, irony, linguistic underdeterminacy, de se utterances, self-expression and meaning negotiation. His recent publications include “Illocution and Accommodation in the Functioning of Presumptions” (Synthese 198, 2021), “Irony as a Speech Action” (Journal of Pragmatics 190, 2022), and “An Austinian Alternative to the Gricean Perspective on Meaning and Communication” (Journal of Pragmatics 201, 2022). He is the co-editor, with Iwona Witczak-Plisiecka, of Normativity and Variety of Speech Actions (2019). Recently, he has co-edited, with Marcin Lewiński, Bianca Cepollaro and Steve Oswald, a special issue on Norms of Public Argument: A Speech Act Perspective (Topoi 42(2), 2023).

List of Figures

Towards a Unified Theory of Illocutionary Normativity Fig. 1

An account of illocutionary norms that distinguishes between norms for performance (upstream normativity) and rules for compliance (downstream normativity)

174

A Speech-Act Theoretic Analysis of White (Prosocial) Lies Fig. 1

The two types of intentions (top: Intention 1 = Gricean r-intention, bottom: Intention 2 = the lying intention) involved in malicious and non-malicious (gray and white) lies 254

xv

List of Tables

Should Speech Act Theory Eschew Propositions? Table 1

Three dimensions of commitment for four members of the assertive family, and for two non-members (sheer guesses and suppositions) (from Green 2016)

32

Towards a Unified Theory of Illocutionary Normativity Table 1 Table 2

The three categories of rules introduced by Sbisà (2019) A table visualising which family of rules falls under which category under the revised model. The triple line separates rules for performance (upstream normativity) from rules for compliance (downstream normativity)

170

179

A Speech-Act Theoretic Analysis of White (Prosocial) Lies Table 1

Bryant’s open-ended lie classification (from Bryant 2008, 32)

266

xvii

Marina Sbisà’s Deontic Approach to Speech Actions Laura Caponetto and Paolo Labinaz

1 Themes in Marina Sbisà’s Philosophy Marina Sbisà, Senior Scholar and former Professor of Philosophy of Language at the University of Trieste, is one of the leading speech act theorists in the contemporary scene. Sbisà became well known to speech act scholars as a co-editor, together with James O. Urmson, of the second, revised edition of John L. Austin’s How to Do Things with Words (1975, henceforth HTW).1 But her contribution to speech act theory goes far beyond this important editorship. Building upon Austin’s legacy, Sbisà  She is also a co-editor (together with Carlo Penco) of the Italian translation of HTW, titled Come fare cose con le parole (1987).

1

L. Caponetto (*) Newnham College, University of Cambridge, Cambridge, UK e-mail: [email protected] P. Labinaz Department of Humanities, University of Trieste, Trieste, Italy e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Caponetto, P. Labinaz (eds.), Sbisà on Speech as Action, Philosophers in Depth, https://doi.org/10.1007/978-3-031-22528-4_1

1

2 

L. Caponetto and P. Labinaz

has developed an influential conception of speech as action, according to which our utterances have conventional effects, that is, reshape the normative context they occur in by assigning or unassigning new deontic statuses to certain relevant parties. Sbisà’s conventionalist view, pivoted on the role of intersubjective agreement in deontic updating, challenges the main tenets of a Gricean-inspired intentionalist understanding of speech act performance—a long-dominant paradigm in speech act theory—and has paved the way for a theory of speech actions centered on the normatively transformative power of illocution. In Sbisà’s perspective, what we do with words is partly dependent on how our words are responded to by our audience. This explains why, according to Sbisà, examining speech acts in isolation is methodologically inappropriate, and scholars should shift the analytical focus from single utterances to discursive sequences. Sequentiality, Sbisà claims, is key to illocutionary force assignment. Sbisà’s interest in Austin’s philosophy is also mirrored in her attempt to disentangle Austin’s way of conceiving of a range of notions—from action to knowledge, truth, meaning, and context—and to illuminate how Austinian insights can be drawn upon to advance contemporary debates in analytic philosophy (e.g., Sbisà 2002a, 2006a, 2011, 2012, 2014c, 2015). Sbisà’s contribution to pragmatics is not confined to speech acts, but includes a thorough examination of the forms of implicit communication. In particular, she has developed a distinctive view on the role of presupposition and implicature in text understanding, according to which both have a normative character in that they can be retrieved from a text on the basis of some rule or reason, and thus must be accepted as contributing to that text’s overall meaning, even when its author denies that they meant to convey them (e.g., Sbisà 1999b, 2007b, 2007c). Sbisà’s theorizing has been strongly influenced by the systematic analysis of naturally produced speech, particularly conversation. Her empirically driven works have been concerned with either specific types of speech acts or aspects related to their performance (e.g., Sbisà 1987, 2001, 2006b, 2014a; Labinaz and Sbisà 2014), as well as with implicit communication (e.g., Sbisà 1999a, 2007c, 2017, 2021a), argumentative discourse (Labinaz and Sbisà 2018), feminism and gender stereotyping (particularly, pregnancy and motherhood stereotyping and the feminine

  Marina Sbisà’s Deontic Approach to Speech Actions 

3

subject; e.g., Sbisà 1984b, 1996, 2003), linguistic and cultural identities (esp., Sbisà and Vascotto 2007), and most recently the dissemination of knowledge (or presumed knowledge) in social networks (Labinaz and Sbisà 2021a, b). Last but not least, throughout her research, Sbisà has paid great attention to the notion of subjectivity, cashing it out as interactionally constructed (esp., Sbisà 2018, 2021b). The eleven chapters in this collection discuss Sbisà’s Austinian-inspired approach to speech acts in the context of some of the liveliest debates in current pragmatics. Some deal with foundational questions in the theory of speech acts (such as the place of propositions in speech act theorizing, the conventionalism-intentionalism dispute, uptake’s role in illocution, illocutionary pluralism, speech act norms and normativity), while others tackle issues in social and political philosophy of language (including the discursive status of lying, the dynamics of silencing, and the role of authority in hateful propaganda). This chapter aims to lay out the terrain by providing the theoretical background that is necessary to understand the main research questions addressed by the chapters in this collection. We identify and discuss the key claims grounding Sbisà’s deontic approach to speech actions, while locating her contribution within the broader literature on speech acts and their nature. In particular, we focus on her distinctive Austinian-inspired understanding of the conventionality of illocutionary acts (Sect. 2), their core effects (Sect. 3), and the role of presupposition accommodation in force assignment (Sect. 4). In closing, we outline the book’s structure and give summaries of each chapter in the collection (Sect. 5).

2 Illocution and Conventionality Sbisà’s work on Austin’s legacy has primarily focused on the notion of illocution. This section aims at presenting her approach on a key issue in speech act theory—namely, whether and how illocutionary acts are conventional. Even though Austin considers conventionality a distinguishing feature of illocutionary force (Austin 1975, 103, 107, and 119), he never came up with a rigorous definition of illocution, and only scattered remarks can be found in HTW on the conventionality of illocutionary acts. Partly because of this, there is no scholarly consensus on how to spell out Austin’s conventionalism.

4 

L. Caponetto and P. Labinaz

Austin’s commentators have tended to focus on certain passages of HTW, while backgrounding others that Sbisà (2009, 43–47) considers essential to understanding Austin’s way of shaping the link between conventionality and illocution. In particular, commentators have concentrated on 1. Austin’s formulation of the felicity condition (A.1), according to which, for an utterance to be performative (or differently put, for an utterance to have illocutionary force), there must be an accepted conventional procedure—and uttering the words the speaker utters must amount to, or be part of, its execution (Austin 1975, 14 and 26–28); and on Austin’s remarks that 2. illocutionary acts can be said to be conventional “[…] in the sense that at least [they] could be made explicit by the performative formula” (Austin 1975, 103), and that 3. “there cannot be an illocutionary act unless the means employed are conventional” (Austin 1975, 118). Against this backdrop, some prominent commentators have concluded that, in Austin’s view, illocutionary acts are conventional in that they are performed by conventional means.2 Illocutionary acts could only be performed in accordance with some linguistic or extra-linguistic conventions—and speakers would be required to engage in conventionally established behaviors that make their acts recognizable to their audiences.

Convention and Intention in Illocution Sbisà has developed her approach on the conventionality of illocution in opposition to those of two prominent speech act theorists—namely, Peter Strawson and John Searle. Let us look at their views in turn.  See, among others, Strawson (1964); Searle (1968); Bach and Harnish (1979, 120 and 131–132); Green (2021, Section 4.1). 2

  Marina Sbisà’s Deontic Approach to Speech Actions 

5

Strawson (1964) has famously objected to the claim that illocutionary acts are essentially conventional by arguing that the successful performance of “ordinary” illocutionary acts—ordering, asking, warning, and so on—depends on the hearer’s uptake rather than on the conventional means deployed by the speaker. In Strawson’s Gricean-like perspective, uptake amounts to the hearer’s recognition of the speaker’s communicative intention. For the speaker’s intention to be recognized (and recognized as intended to be recognized), and uptake to be achieved, the speaker must make that intention “wholly overt” (Strawson 1964, 459). They can do so by relying on the conventional meaning of the words they utter, but this isn’t necessary. Illocutionary force, says Strawson (1964, 459), “[…] is essentially something that is intended to be understood”— and what is required for illocutionary performance is the hearer’s uptake, however secured. Alongside ordinary illocutions, there are illocutionary acts—such as marrying or condemning—invoking “convention-­ constituted procedures” (Strawson 1964, 459). If ordinary illocutions are felicitously performed only if uptake is secured, the felicity of “conventional” illocutionary acts depends on their compliance with the relevant conventions. So, Strawson acknowledges that some illocutions essentially deploy conventional means, but insists that they do not form the whole of illocutionary acts. In particular, he stresses the key role of speaker intention, and its fulfillment, in performing a vast array of everyday illocutions. Searle (1969, 1979), for his part, has developed a view in which intention and convention both contribute to illocutionary performance. Searle (1969, 22) takes language to be “a rule-governed form of behavior,” and, while he agrees that the hearer’s recognition of the speaker’s intention is essential to successful illocution, he adds that, for this recognition to occur, both the speaker and hearer must be acquainted with the semantic rules for the use of illocutionary force indicators (e.g., word order, punctuation, grammatical mood, etc.). Searle (1969, 54–62) derives these rules from a set of necessary and sufficient conditions for the successful and non-defective performance of illocutionary acts.3 Such rules, he  As is well known, Searle analyzes the set of conditions for fully successful promises. He then extracts from these the set of rules for the use of the illocutionary force indicator for promising, and eventually generalizes his analysis to other types of illocutionary acts (Searle 1969, 57ff). 3

6 

L. Caponetto and P. Labinaz

claims (1969, 62–71), specify (1) what kind of proposition, if any, is to be expressed to perform a certain illocutionary act (propositional content rule); (2) what capacities and attitudes the speaker and hearer must have (preparatory rules); (3) what psychological state(s) the speaker has to entertain (sincerity rule); and (4) what kind of illocutionary act one’s utterance is to count as (essential rule).4,5 Complying with rules (1)–(4) is pivotal to making one’s audience understand the meaning and force of one’s utterance, and thus to bring about what Searle calls “the illocutionary effect.” This effect “consists simply in the hearer understanding the utterance of the speaker” (Searle 1969, 47)—and, in the performance of literal (or direct) illocutionary acts, it is brought about in virtue of the fact that the rules for using the expression the speaker uses associate it with a certain meaning and force.

Illocutionary Effects as Conventional Effects Both Strawson and Searle take the securing of uptake to be the (unique) characteristic effect of successful illocution. In contrast, and following Austin, Sbisà maintains that the securing of uptake, while being a necessary condition for illocution, does not exhaust the effects associated with its successful performance. Furthermore, in Sbisà’s perspective, the recognition of the speaker’s communicative intention does not play any central role in securing uptake. (We will get back to this below.) Sbisà (2009, 44–46) strongly claims that uptake is not the only illocutionary effect. Two further effects are connected with illocution: (1) the act “takes effect,” and (2) it may conventionally invite a response or sequel on the part of the audience (Austin 1975, 116–118). It is worth bearing in mind that, for Austin (1975, 14), any conventionally accepted procedure is designed to produce a (characteristic) conventional effect. According to Sbisà (2007a, 44–46), to say that a felicitous illocutionary act “takes effect” (in Austin’s terminology) is to say that it brings about its  For example, the essential rule for assertion states that the utterance of a declarative sentence counts as the undertaking that its content represents an actual state of affairs (Searle 1969, 51). 5  The existence of these rules implies that merely resorting to certain linguistic devices is not enough to mark one’s utterance as a certain illocutionary act. 4

  Marina Sbisà’s Deontic Approach to Speech Actions 

7

conventional effect, and does so in a way that differs from natural causation (see Austin 1975, 117). More precisely, conventional effects are the result of “human elaborations and choices, tacit agreements, conventions, and rule-following behaviour” (Sbisà 2013b, 32). So, Sbisà concludes, “the word ‘conventional’ can be used, in an Austinian framework, to characterize the effect which illocutionary acts are said to ‘take’” (Sbisà 2013b, 32). Note, however, that the example Austin (1975, 117) provides to illustrate how illocutionary acts “take effect” is somewhat misleading. He resorts to the act of naming a ship, which takes effect in the sense that it gives the ship a certain name and obliges speakers to call it by that name. As Sbisà observes (2007a, 464), for an act of naming to take effect, the speaker is expected to follow a ceremonial procedure. It may thus seem that only ceremonial or institutional illocutions “take effect,” and that Austin’s insight does not generalize to ordinary illocutionary acts. Sbisà has devoted much effort to explaining how and in what sense any illocutionary act takes effect (Sbisà 1984a, 1989, 2001, 2006b, 2007a, 2009). In her framework, the successful performance of an illocutionary act consists in the bringing about of its conventional effect, which—as we will see in the next section—Sbisà describes in terms of a change in the deontic statuses of the conversation participants, that is, in their commitments, entitlements, obligations, and so on. Upon observing that Austin himself “[…] shows a tendency towards foregrounding the effect of the act rather than the means for performing it” (Sbisà (1984a, 94), Sbisà advocates a shift from means conventionality to effect conventionality: 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. (Sbisà 1984a, 94)

On this perspective, uptake amounts to the hearer’s recognition of the procedure that the speaker is executing—not of their communicative intention. The speaker has to carry out the procedure for performing the act in question correctly and completely enough to be understood as

8 

L. Caponetto and P. Labinaz

executing it.6 If the speaker and hearer disagree on what procedure has been invoked, the illocutionary act the speaker purports to perform will not achieve its conventional effect. This is because conventional effects come into being by virtue of intersubjective agreement, which may be explicit or (as is more often the case) tacit (Sbisà 1989, 64–66; 2007a, 465; 2009, 47–50). Intersubjective agreement is made possible by the securing of uptake. Sbisà’s way of construing the notion of uptake differs sharply from Strawson’s and Searle’s. While, in an intention-based framework, the speaker—and their intentions—take pride of place, Sbisà conceives of uptake as a bilateral process. The speaker and the hearer both  contribute to the illocutionary act’s taking effect by reaching an agreement on what has been done with words. In addition to the indispensability of intersubjective agreement, effect conventionality is characterized by what Sbisà (2007a, 465–466) calls “defeasibility.” The states of affairs that “natural” (non-conventional) effects consist of “[…] can be further modified, but not just annulled” (Sbisà 2007a, 465). In contrast, illocutionary acts bring about effects that can be annulled (or “defeated”), if some fatal infelicity is discovered.7 Suppose, for example, that a man wearing a police vest stops my car. “License and registration,” he utters. I take that as an order. Suppose also that the man is a fake police officer. His words will not impose any ultima facie obligation upon me to give him my license and registration. Upon discovering that he lacks the relevant authority, and that his illocution is thus fatally infelicitous, his order will be “made undone” (Sbisà 2007a,

 According to Sbisà (2019, 24ff), illocutionary procedures are established by constitutive rules. Such rules determine the preconditions that must be in place for a speaker to invoke a certain procedure and the steps they must carry out in order to be recognized as executing it. While Searle’s constitutive rules are cashed out in terms of “internalist requirements,” that is, of attitudes and mental states that the speaker and addressee must have, Sbisa’s constitutive rules amount to “externalist requirements,” that is, they specify the circumstances, persons, and modalities that are necessary to perform a given illocutionary procedure (Sbisà 2002a, 423–424; 2019, 27, fn. 4; see Harnish 2009 for a different reading of Searle’s constitutive rules). Furthermore and interestingly, Sbisà (in preparation) observes that, while Searle considers the constitutive rules of an illocutionary act as comprising the necessary and sufficient conditions for its successful and non-defective performance, Austin stressed that, even when a speaker executes the appropriate procedure entirely, there is no guarantee of illocutionary success. 7  See Hart (1949) for an interestingly similar use of the term ‘defeasibility’. 6

  Marina Sbisà’s Deontic Approach to Speech Actions 

9

465–466).8 Notice that making a speech act undone does not amount to annulling each and every effect it has. On the one hand, Sbisà (2007a, 465) suggests that “[…] we have ways of redescribing null acts in terms of the performance of different acts and we may say that it is those acts which really took place.” So, the fake policeman’s null order can be redescribed in terms of a (successful) request. On the other hand, any utterance produces causal (or perlocutionary) effects that cannot be simply annulled, being, as they are, part of the natural realm. Suppose I hand the fake policeman my license and registration, in the erroneous belief that he is a true police officer. Unlike the purported obligation his words imposed on me, which can be retroactively recognized as null or non-­ binding, the things I do as a causal consequence of his utterance cannot be simply annulled. Once I discover the truth, I can blame myself for not being more careful, but I can’t just make it the case that I never handed him my license and registration. In short, on Sbisà’s perspective, to say that the core effect of illocution is conventional is to say that it depends on intersubjective agreement as to what the speaker has done with words, and relatedly, that it can be annulled or defeated. This core effect is essential to any illocutionary act, be it ordinary or institutional in character. On this basis, Sbisà has developed a unified account of illocutionary forces in terms of their characteristic conventional effects that does away with Strawson’s distinction between intention-based and convention-based illocutionary acts (see Sbisà 1984a, 1989). Focusing on effect conventionality, as Sbisà does, is to shift perspective in speech act studies, and to adopt a view that is fully in keeping with Austin’s emphasis on the performative dimension of language (see Sbisà 1989, 28–33; 2006b, 152–154; 2007a; 2014c). The shift involves (re) conceptualizing speech as a social action, rather than a mere rule-governed activity. Searle’s idea that “speaking a language is engaging in a rule-governed form of behavior” (Searle 1969, 12) might sound similar to Austin’s claim that speakers must execute conventionally accepted procedures in order to illocute. But, as Sbisà underlines (2009, 46), “[…] it is typical of the procedures of which performative utterances are part,  On the different ways in which illocutionary acts can be undone, see Caponetto (2020).

8

10 

L. Caponetto and P. Labinaz

that they are not mere rule-governed sequences of words or gestures, but give rise to effects of a conventional kind.” Illocutionary acts are social actions and involve holding the speaker responsible for bringing about some conventional effect. “All speech should be considered as action,” says Sbisà (2007a, 462). All speech consists in an outcome-oriented process, requiring certain conditions to be satisfied and certain steps to be carried out for the outcome to be achieved. But what does this outcome amount to?

3 Illocutionary Effects and Deontic Modality Sbisà characterizes the conventional effects of illocutionary acts—that is, the outcome toward which they point—by resorting to the lexicon of deontic modality, and in particular to expressions such as “permission,” “obligation,” “commitment,” “entitlement,” “license,” “right,” and “waiver” (see, especially, Sbisà 1984a, 1989, 2006b).9 Sbisà here elaborates upon some ideas developed within narrative semiotics. In particular, she draws upon the idea of “modal competence,” which narrative semioticians introduce to account for the set of modal attributes that an agent must possess at each stage of a narrative in order to complete their narrative trajectory.  As the narrative proceeds, the modal attributes that are necessary to perform certain actions come to be presupposed as part of the agent’s modal competence, and can be altered by their performances or by the intervention of other agents (see Greimas 1987; Greimas and Courtés 1982). Sbisà's model of linguistic exchanges is similar to this. Her idea is that, at each stage of a linguistic exchange, each participant is recognized as possessing some modal attributes that entitle them to perform certain illocutionary acts but not others. The successful performance of an illocutionary act alters the participants’ modal attributes. In  According to Greimas and Courtés (1982, 207), the narrative trajectory of an agent (or, in semiotic terms, an actant) consists of a series of “modal” and “realization” programs that succeed one another in time and are organized in a logical chain in which each step is presupposed by the other. 9

  Marina Sbisà’s Deontic Approach to Speech Actions 

11

particular, it changes what they can do (have a right, are entitled or allowed to do) and what they are committed or obliged to do with respect to one another. The conventional effects of illocutionary acts are thus effects on the deontic dimension of the relationship between the speaker and hearer.10 Sbisà describes them in terms of the […] assignments to or cancellations from each one of the participants of modal predicates related to the necessity or possibility of actions with respect to norms. 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 to and from agents of deontic modal values is basically “conventional”, namely, dependent on social factors such as the relevant intersubjective agreement. (Sbisà 2001, 1978)

In analyzing the dynamics of assignment and/or cancelation of deontic modal attributes, Sbisà (2001, 1799–1800) distinguishes between: (1) the deontic modal attribute(s) that the speaker claims (or presupposes) they have in performing a certain illocutionary act; (2) the deontic modal attribute(s) that a successful illocutionary performance assigns to and/or cancels from the audience, and (3) the deontic modal attribute(s) that are correspondingly assigned to and/or canceled from the speaker. As to (1), in invoking a certain procedure, the speaker presupposes, and implicitly invites the audience to presuppose, that they have the deontic modal attribute(s) that are necessary for their invocation to be felicitous. For example, since an order is felicitous only if the speaker has practical authority over the hearer, when one utters a sentence of the form “Do X,” and unless there are specific reasons for doubting it, one will be taken as having that authority, and thus being in the right position to oblige the hearer to do X. In this and similar cases, the speaker’s possession of the relevant deontic modal attributes may be tacitly “accommodated” by the audience.11  Interpersonal relations involve cognitive and emotional dimensions, too. Since these dimensions pertain to the perlocutionary, as opposed to the illocutionary, domain, we leave them aside. See Sbisà (2014b, 623). 11  More will be said on accommodation in speech act performance in Sect. 4. 10

12 

L. Caponetto and P. Labinaz

Deontic modal notions can be employed to describe a whole range of communicative and relational activities: from the imposition of or release from obligations (be they legal or moral), to the assignment (or removal) of rights and entitlements, the exertion of influence, the undertaking of commitments, the dissemination of knowledge (or presumed knowledge), and the ratification or approbation of desires. The richness of the lexicon of deontic modality makes it a particularly powerful tool to account for the varieties of illocutionary act types, and to analyze the complexities of real-world illocutionary act tokens. Here, we will focus on Sbisà’s classification of illocutionary act types (Sbisà 1984a, 1989, 2006b), leaving aside the results of her empirically oriented studies (see, esp., Sbisà 2001, 2006b, 2014a; Labinaz and Sbisà 2014, 2021a, b). Following Austin (1975, 150–163), Sbisà divides illocutionary act types into four classes, namely, Verdictives, Exercitives, Commissives, and Behabitives. She sets Austinian Expositives aside, since they cut across the other classes (as both Austin and Sbisà have suggested, any candidate for the expositive class could also be placed in some other class; see Austin 1975, 152 and 161–162; Sbisà 1984a, 126–130). Sbisà draws upon Austin’s classification—rather than Searle’s or others’—because of its internal flexibility, which makes it well suited to analyze not only prototypical, clear-cut cases, but also marginal and hybrid cases. Austin’s classification does not aim to identify well-defined categories of illocutionary acts, but “[…] to describe fuzzy and partially overlapping sets of illocutionary act types, each of which has some prototypical cases at its core” (Labinaz and Sbisà 2021b, 154). Most importantly, illocutionary acts are not classified according to the means for performing them (e.g., the standard linguistic form of the speaker’s utterance), but are grouped together in virtue of their characteristic conventional effects. Sbisà characterizes the four classes as follows: –– Verdictives—such as assessing or valuing—consist in the making of a judgment on the basis of evidence or reasons. Verdictive judgments may be official or informal, final or provisional, and about facts or values. They presuppose that the speaker has access to the data and criteria that are necessary to make the relevant judgment. In terms of deontic modal attributes, a speaker performing a verdictive commits

  Marina Sbisà’s Deontic Approach to Speech Actions 

13

themself (1) to providing evidence and/or reasons in support of their judgment, if requested, and (2) to acting in ways that are consistent with the verdictive that they have performed. The audience acquires an entitlement to rely on the speaker’s judgment in their subsequent verbal and non-verbal behavior, for example, to use it as the content of further, related verdictives, or as a premise in reasoning, or as grounds for decisions (Sbisà 1984a, 104–105; 1989, 90–92; see also Sbisà 2020a, and Labinaz and Sbisà 2021a, b); –– Exercitives—such as ordering or advising—are exercises of authority or influence. They presuppose that the speaker has some kind or degree of authority or authoritativeness. In terms of deontic modal attributes, they assign or remove rights or obligations to or from the audience. In the prototypical case of ordering, the speaker acquires the right to punish or otherwise sanction the addressee if they fail to comply with the order (Sbisà 1984a, 102–104; 1989, 85–88); –– Commissives—such as promising or betting—consist in the undertaking of some commitment on the part of the speaker. They presuppose that the speaker is able to do—and that they are recognized as being able to do—what they commit themself to doing. The commitment that the speaker undertakes may take either a specific action or a general trend of behavior as its content. Commissives assign the addressee an entitlement to expect that the speaker will honor their newly acquired commitment(s) (Sbisà 1984a, 104; 1989, 89–90); –– Behabitives—such as thanking or condoling—consist in the production of a behavioral response, such that an obligation toward the audience, or otherwise a social expectation, is fulfilled. The speaker need not have any special status: to successfully perform a behabitive, it is typically enough that the speaker finds themself in the appropriate circumstances. Behabitives entitle the audience to make assertions concerning the speaker and their psychological states and attitudes (Sbisà 1984a, 106–107; 1989, 92–94). Sbisà’s project of providing a unified account of illocutionary forces in terms of their characteristic conventional effects has proven successful: all illocutionary act types seem to have effects that can be appropriately described in deontic terms. Furthermore, as Sbisà’s empirically driven

14 

L. Caponetto and P. Labinaz

studies suggest, this way of describing illocutionary effects can be fruitfully employed to account for the force of illocutionary act tokens in natural discourse (Sbisà 2001, 2006b; Labinaz and Sbisà 2021b). We have already said that illocutionary acts are felicitous—that is, succeed in bringing about their deontic outcome—only if the speaker and hearer agree as to what was done with words. Importantly, the hearer’s interpretation of the speaker’s act, and thus their (dis)agreement with the speaker as to its illocutionary profile, can be (and often is) manifested in their response. “The response (verbal or non-verbal) which follows the illocutionary act under examination [...] makes manifest how the hearer has taken the speaker’s illocutionary act,” says Sbisà (1992, 101). Suppose that a speaker utters a sentence of the form “Do X,” intending to order, and the hearer replies “Don’t bother me.” An order is only such if it imposes an obligation upon the hearer. In the case at hand, the hearer would seem to disagree that they have been obliged to do X. If the speaker does have authority over the hearer, they might consider initiating a “repair procedure,” that is, taking action against the hearer’s reaction, so as to clarify that their utterance was intended as, and indeed constituted, an order. If no repair procedure is initiated, the speaker will ipso facto accept the failure of their purported order. This example should make it clear why, in Sbisà’s model, “[…] sequentiality is a fundamental dimension of speech acts” (Sbisà 2002b, 94). Since intersubjective agreement is a necessary condition for illocutionary success, in order to ascertain whether a certain illocutionary act has been felicitously performed, one must look at the hearer’s reaction—and at the negotiation process, if there is one, that follows. Moving away from a tradition that takes speech acts in isolation to be the basic unit of analysis, Sbisà develops a view that has speech act sequences at its core.12

 Adopting Greimas and Courtés’ (1982) “manipulation-action-sanction” scheme, Sbisà (2002b, 82–87) takes speech act units to have a three-place sequential organization. Importantly, she points out that the scheme can be applied to the same speech act sequence in multiple ways, giving rise to different interpretations of the same speech event—an insight that preludes to her distinctive illocutionary pluralism (Sbisà 2013a). 12

  Marina Sbisà’s Deontic Approach to Speech Actions 

15

4 Accommodation and Felicity Conditions As mentioned in the previous section, when a speaker invokes the procedure for performing a certain illocutionary act, hearers tend to assume that its felicity conditions are satisfied, unless some contextual signal indicates otherwise. Consider our fake police officer example again. He stops my car and says, “License and registration.” If no contextual signal makes me suspicious, I will take it for granted that he is a police officer, and thus has the authority to order drivers to hand him their license and registration. Sbisà (2002a, 425) highlights that an analogous story can be told about presuppositions. I overhear you saying to a friend, “George has stopped smoking,” and by default assume that George, whom I barely know, used to smoke. Austin himself acknowledged, in a little discussed passage of HTW, that felicity conditions—especially (A.2) conditions—and presuppositions are similar beasts. What is to be said of the statement that “John’s children are all bald” if made when John has no children? [...] Here I shall say “the utterance is void”. Compare this with our infelicity when we say “I name …”, but some of the conditions (A.1) and (A.2) are not satisfied (specially A.2 perhaps […]). Here we might have used the “presuppose” formula: we might say that the formula “I do” presupposes lots of things: if these are not satisfied the formula is unhappy, void. (Austin 1975, 50–51)

The formula “I name …,” pronounced in the context of naming a ship, for example, is felicitous only if, among other things, the speaker is the person appointed to name it and the ship has not been named already. In pronouncing the words “I name …,” the speaker presupposes that those conditions obtain—while the hearer, at least insofar as no alarm signal alerts them, will typically play along. “Accommodation,” in David Lewis’ sense, can do its job because presuppositions, as Sbisà points out, are assumed to be satisfied by default. When an utterance (“George has stopped smoking”) requires a presupposition (“George used to smoke”) that some conversation participant does not already share, the presupposition in question will be

16 

L. Caponetto and P. Labinaz

straightforwardly “accommodated,” that is, included among the participants’ background assumptions, unless someone openly challenges it (Lewis 1979). Accommodation is a context-repairing process. In principle, for an utterance to be fully appropriate, its presuppositions must be shared by the conversation participants before the speaker speaks. In practice, however, when a speaker introduces some informative presupposition, accommodation may promptly repair their fault by rendering their utterance appropriate after all.13 One interesting question, which has been explicitly tackled by Rae Langton in recent works (2015, 2018a, b), and is implicitly addressed by Sbisà (2002a), is whether felicity conditions—which, as Austin remarks, are akin to presuppositions—may be subject to accommodation processes. Both Langton and Sbisà focus on the authority condition for ordering and come to an affirmative answer. “Authority,” Sbisà (2002a, 430) writes, “does not exist apart from its recognition by relevant social participants.” When it comes to formal or institutional authority, the relevant social participants and the authority-­ attributing procedures are well defined and largely independent from how the conversation participants react to the speaker’s words. In order to become a police officer, for example, one must graduate from a police academy, and no one without police academy training has the authority to act as a police officer. The fake policeman may have me fooled, but this doesn’t give him the authority to order me my license and registration. His illocutionary act is fatally infelicitous, even though it may pass as felicitous and give me a prima facie duty to obey. But what about informal authority? 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 default assumption of the satisfaction of felicity conditions, if not challenged, leads directly to the relevant agree-

 See, for example, Simons (2003).

13

  Marina Sbisà’s Deontic Approach to Speech Actions 

17

ment and thus to fixing a conventional framework, which counts as an element of the objective context of that very speech act. (Sbisà 2002a, 431)14

The would-be leader of a gang, for example, may acquire authority over the other gang members—and thus become the leader—by acting as if they already have it. They may boss the others around, using the imperative mood to address them. If nobody objects, the would-be leader may well succeed in redefining their role: they may acquire the authority they presupposed they had, thanks to their audience’s compliance. Orders, as already pointed out, are authoritative speech acts: they are felicitous only if the speaker has authority over the hearer. In principle, the authority condition for ordering must be satisfied before the speaker speaks. In practice, however, when someone gives a peer orders, thus falsely presupposing that they have informal authority over them, accommodation may promptly repair their fault by enabling them to acquire authority in the moment, and rendering their orders felicitous after all. While Sbisà does not explicitly frame the process of informal authority acquisition in terms of presupposition accommodation, in her 2002a paper she develops an insightful parallelism between felicity conditions and presuppositions that anticipates, and paves the way for, a debate on the role of accommodation in illocution that will get underway about fifteen years later and is among the most discussed topics in contemporary speech act studies.15

5 This Collection: Structure and Contents It should be clear at this point that speech acts pose questions of concern not only to philosophy of language, but also to epistemology and social philosophy, action theory, linguistics and sociolinguistics, semiotics,  See also Sbisà and Fabbri (1980, 315): “Presuppositions […], as long as they are retroactive in attaching to the agent/speaker a different status (so that e.g. he/she might gain authority by succeeding in having his/her orders accepted as such by people who previously were not subordinate to him/her), appear to be one of the main devices for self-modification.” 15  See, esp., Maitra (2012), Witek (2013, 2021), Adams (2019), Langton (2015, 2018a, b), Caponetto (2022). 14

18 

L. Caponetto and P. Labinaz

communication studies, and more. This collection brings scholars with expertise in various disciplines into a productive dialogue with one another. By focusing on different aspects of Sbisà’s philosophy, the eleven chapters in this collection provide the first comprehensive critical treatment of her outstanding contribution to speech act theory, and offer a snapshot of some of the liveliest debates in contemporary pragmatics. The volume is structured in such a way as to guide the reader from foundational questions in the theory of speech acts to issues in applied philosophy of language and social and political pragmatics. The collection closes with Sbisà’s “Replies to Contributors.” In “Should Speech Act Theory Eschew Propositions?,” Mitchell Green tackles the question as to whether propositions should be removed from speech act theory’s conceptual toolkit. Contra Sbisà (2006a), Green argues that a speech act theory invoking propositions does not necessarily devolve into a view according to which speech acts primarily transmit propositional attitudes. Further, a speech act view that retains propositions may be able to account for the normative (or deontic) statuses that our illocutions contribute to bring about. On this basis, Green concludes that, while Sbisà’s reasoning justifies denying propositional attitudes a central role in illocution, we do not have compelling grounds for dispensing with propositions in speech act theorizing. In “On the Conventional Nature of Illocutionary Acts: Uptake, Conventions, and Illocutionary Effects,” Bruno Ambroise offers a supportive elaboration of Sbisà’s claim that uptake is necessary for an illocutionary act to bring about its characteristic conventional effects. By bringing Sbisà’s framework into fruitful contact with Richard Moran’s (2018) social account of telling and Pierre Bourdieu’s (1991) analysis of performativity in terms of “symbolic power,” Ambroise argues against an intentionalist approach to speech acts and in favor of a full recognition of their place in the social world. In “Varieties of Uptake,” Claudia Bianchi argues that theories giving the hearer’s uptake a prominent role in fixing the force of a speech act are theoretically unsatisfactory, and have controversial practical implications. Against hearer-dependent views (e.g., Langton 1993; Kukla 2014; McDonald 2022) and drawing upon theoretical tools developed by Sbisà, Bianchi advances a speaker-dependent account of illocutionary force,

  Marina Sbisà’s Deontic Approach to Speech Actions 

19

which has the advantage of granting victims of silencing illocutionary agency. In “Interactional Negotiation,” Maciej Witek makes use of the scorekeeping model of illocutionary dynamics (Witek 2015, 2019) to account for the mechanisms of what Sbisà (2013a) calls “interactional negotiation,” that is, the idea that the force of an act can be negotiated by the conversing agents. Witek sets out to illuminate whether and how the hearer’s uptake and the speaker’s subsequent reaction to it contribute to fixing illocutionary force. In particular, he distinguishes between three types of uptake—explicit, implicit though active, and passive—and investigates the different roles they play in negotiation. In “Some Varieties of Illocutionary Pluralism,” Casey Rebecca Johnson elaborates upon Sbisà’s (2013a) remarks on speech act pluralism to argue in favor of a relativist approach to illocutionary force. After clarifying how relativism deals with both one-on-one conversations and multi-­ party conversations, Johnson discusses her approach vis-à-vis what Marcin Lewiński (2021) has recently called “horizontal pluralism.” In “Speech Acts and Ventriloquation: The Contribution of Marina Sbisà to a General Theory of Action and Performativity,” François Cooren relies on Sbisà’s conventionalist framework to investigate the “ventriloquial” dimension of communication. Other-than-humans do things with or without words: situations confirm or disconfirm what we take to be true (Peirce 1877), arrows indicate the direction to follow (Cooren 2006), and agreements commit signatories to certain terms and conditions (Ashcraft, Kuhn and Cooren 2009). Cooren submits that situations, arrows, and agreements are made to say or do things—which is the essence of ventriloquism. In “Towards a Unified Theory of Illocutionary Normativity,” Neri Marsili points out that speech act scholars have been using headings misleadingly: sometimes, multiple headings are used for the same type of illocutionary norm, and sometimes the same heading is used for different types of norms. Drawing upon Sbisà’s (2019) tripartition between constitutive rules, maxims, and objective requirements, Marsili sets out to develop a unified, principled framework for studying illocutionary normativity, which has the tools to draw all the relevant distinctions, while avoiding overlaps and terminological confusion.

20 

L. Caponetto and P. Labinaz

In “Illocutionary Force, Speech Act Norms, and the Coordination and Mutuality of Conversational Expectations,” Sanford C. Goldberg brings together his view on conversation as a site for the generation, modification, and satisfaction of normative expectations (Goldberg 2020), and Sbisà’s idea that illocutionary acts redefine conversation participants’ deontic statuses. After fruitfully combining this approach with a constitutive rules account of speech acts, Goldberg highlights that one important upshot of his reflections is that we should reject Austin’s thesis that illocutionary success requires the hearer’s uptake, as well as Sbisà’s claim that Austin’s thesis is needed if we are to make sense of the normative dimension of speech. In “Speech in Non-ideal Conditions: On Silence and Being Silenced,” Alessandra Tanesini objects to Goldberg’s (2020) claim that linguistic agents are defeasibly entitled to presume that, in conversation, silence indicates acceptance. Drawing inspiration from Sbisà (2020b), Tanesini argues that, far from being general norms of conversation, norms of no silent rejection are temporarily enacted by means of exercitive assertions, and promote silencing, often at the expense of underprivileged speakers. In “A Speech-Act Theoretic Analysis of White (Prosocial) Lies,” Marina Terkourafi focuses on (so-called) white lies, that is, harmless lies that are uttered out of concern for the addressee. Adopting Sbisà’s neo-Austinian approach, Terkourafi claims that malicious lying cannot be a type of speech act, but that white lying is—a claim supported by three types of evidence: the conventionality of the means by which white lies are told, their online processing, and their acquisition. In closing, Terkourafi points out that the existence of a continuum between white lies (which are a type of speech act) and malicious lies (which are not) suggests that whether an utterance counts as a speech act or not can be a matter of degree. In “Presupposition in Propaganda: A Socially Extended Analysis,” Michael Randall Barnes argues that hate speakers may gain informal, non-ideal, authority in ways that have so far been largely neglected. While Langton’s (2018a) and other prevailing models depict authority as a property of individual speakers, Barnes advances an approach that recognizes the power of the crowd. By revisiting Sbisà’s (1999a) view  on

  Marina Sbisà’s Deontic Approach to Speech Actions 

21

presupposition as a means for conveying ideology, Barnes submits that the mechanisms by which hateful presuppositions are accommodated are much more active and social than the literature has suggested.

References Adams, Nate P. 2019. Authority, Illocutionary Accommodation, and Social Accommodation. Australasian Journal of Philosophy 98 (3): 560–573. Ashcraft, Karen Lee, Timothy R.  Kuhn, and François Cooren. 2009. Constitutional Amendments: “Materializing” Organizational Communication. The Academy of Management Annals 3 (1): 1–64. Austin, John L. 1975. In How to Do Things with Words, ed. James O. Urmson and Marina Sbisà, 2nd ed. Cambridge, MA: Harvard University Press. Bach, Kent, and Robert M.  Harnish. 1979. Linguistic Communication and Speech Acts. Cambridge, MA: MIT Press. Bourdieu, Pierre. 1991. Language and Symbolic Power. Cambridge: Polity Press. Caponetto, Laura. 2020. Undoing Things with Words. Synthese 197: 2399–2414. ———. 2022. Accommodated Authority: Broadening the Picture. Analysis. https://doi.org/10.1093/analys/anac049 Cooren, François. 2006. The Organizational World as a Plenum of Agencies. In Communication as Organizing: Empirical and Theoretical Explorations in the Dynamic of Text and Conversation, ed. François Cooren, James R. Taylor, and Elizabeth J. Van Every, 81–100. Mahwah, NJ: Lawrence Erlbaum. Goldberg, Sanford C. 2020. Conversational Pressure: Normativity in Speech Exchanges. Oxford: Oxford University Press. Green, Mitchell. 2021. Speech Acts. In The Stanford Encyclopedia of Philosophy, edited by Edward N.  Zalta (ed.). https://plato.stanford.edu/archives/ fall2021/entries/speech-­acts/. Greimas, Algirdas Julien. 1987. On Meaning. Selected Writings in Semiotic Theory. Trans. Paul J. Perron, and Frank Collins. London: Frances Pinter. Greimas, Algirdas Julien, and Joseph Courtes. 1982. Semiotics and Language: An Analytical Dictionary. Trans. Larry Christ, Daniel Patte, James Lee, Edward McMahon II, Gary Phillips, and Michael Rengstorf. Bloomington: Indiana University Press. Harnish, Robert M. 2009. Internalism and Externalism in Speech Act Theory. Lodz Papers in Pragmatics 5: 9–31.

22 

L. Caponetto and P. Labinaz

Hart, H. L. A. 1949. The Ascription of Responsibility and Rights. Proceedings of the Aristotelian Society 49 (1): 171–194. Kukla, Quill R. 2014. Performative Force, Convention, and Discursive Injustice. Hypatia 29 (2): 440–457. Labinaz, Paolo, and Marina Sbisà. 2014. Certainty and Uncertainty in Assertive Speech Acts. In Communicating Certainty and Uncertainty in Medical, Supportive and Scientific Contexts, ed. Andrzej Zuczkowski, Carla Canestrari, Ramona Bongelli, and Ilaria Riccioni, 31–58. Amsterdam: John Benjamins. ———. 2018. Argumentation as a Dimension of Discourse. The Case of News Articles. Pragmatics and Cognition 25 (3): 602–630. ———. 2021a. The Problem of Knowledge Dissemination in Social Network Discussions. Journal of Pragmatics 175: 67–80. ———. 2021b. Speech Acts and the Dissemination of Knowledge in Social Networks. In Approaches to Internet Pragmatics: Theory and Practice, ed. Hartmut Haberland, Francisco Yus, and Chaoqun Xie, 145–172. Amsterdam: John Benjamins. Langton, Rae. 1993. Speech Acts and Unspeakable Acts. Philosophy and Public Affairs 22: 293–330. Reprinted in Rae Langton, Sexual Solipsism: Philosophical Essays on Pornography and Objectification (2009), 25–63. Oxford: Oxford University Press. ———. 2015. How to Get a Norm from a Speech Act. The Amherst Lecture in Philosophy 10: 1–33. ———. 2018a. The Authority of Hate Speech. In Oxford Studies in Philosophy of Law, ed. John Gardner, Leslie Green, and Brian Leiter, vol. 3, 123–152. Oxford: Oxford University Press. ———. 2018b. Blocking as Counter-Speech. In New Work on Speech Acts, ed. Daniel Fogal, Daniel W. Harris, and Matt Moss, 144–164. Oxford: Oxford University Press. Lewiński, Marcin. 2021. Illocutionary Pluralism. Synthese 199: 6687–6714. Lewis, David. 1979. Scorekeeping in a Language Game. Journal of Philosophical Logic 8 (3): 339–359. Maitra, Ishani. 2012. Subordinating Speech. In Speech and Harm. Controversies over Free Speech, ed. Ishani Maitra and Mary Kate McGowan, 94–120. Oxford: Oxford University Press. McDonald, Lucy. 2022. Reimagining Illocutionary Force. The Philosophical Quarterly 72(4): 919–939. Moran, Richard. 2018. The Exchange of Words. Oxford: Oxford University Press. Peirce, Charles S. 1877. The Fixation of Belief. Popular Science Monthly 12: 1–15.

  Marina Sbisà’s Deontic Approach to Speech Actions 

23

Sbisà, Marina. 1984a. On Illocutionary Types. Journal of Pragmatics 8: 93–112. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 1984b. La mamma di carta. Per una critica dello stereotipo materno. Milano: Emme. ———. 1987. Acts of Explanation: A Speech Act Analysis. In Proceedings of the First International Conference on Argumentation, ed. Frans H. van Eemeren et al., vol. 2, 7–17. Dordrecht: Foris. ———. 1989. Linguaggio, ragione, interazione. Per una teoria pragmatica degli atti linguistici [Language, Reason, Interaction. Towards a Pragmatic Speech Act Theory]. Bologna: Il Mulino. 2nd edition as an e-book (2009). Trieste: EUT. ———. 1992. Speech Acts, Effects, and Responses. In  (On) Searle on Conversation,  eds. Herman Parret and Jef Verschueren, 101–111. Amsterdam: John Benjamins. ———. 1996. The Feminine Subject and Female Body in Discourse about Childbirth. European Journal of Women’s Studies 3: 363–376. ———. 1999a. Ideology and the Persuasive Use of Presupposition. In Language and Ideology. Selected Papers from the 6th International Pragmatics Conference, edited by Jef Verschueren, 492–509. Antwerp: International Pragmatics Association. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 1999b. Presupposition, Implicature and Context in Text Understanding. In Modeling and Using Context. Second International and Interdisciplinary Conference, edited by Paolo Bouquet et al., pp. 324–338. Berlin: Springer. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2001. Illocutionary Force and Degrees of Strength in Language Use. Journal of Pragmatics 33: 1791–1814. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2002a. Speech Acts in Context. Language & Communication 22, 2002: 421–436. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2002b. Cognition and Narrativity in Speech Act Sequences. In Rethinking Sequentiality, edited by Anita Fetzer and Christiane Meierkord, 71–97. Amsterdam: John Benjamins. Reprinted in Marina Sbisà­ (forthcoming), Essays On Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press.

24 

L. Caponetto and P. Labinaz

———. 2003. Subject and Gender in H.D.’s Trilogy. In H.D.’s Poetry: The Meanings That Words Hide, ed. Marina Camboni, 91–113. New  York: AMS Press. ———. 2006a. Speech Acts Without Propositions? Grazer Philosophische Studien 72: 155–178. ———. 2006b. Communicating Citizenship in Verbal Interaction. Principles of a Speech Act Oriented Discourse Analysis. In Analysing Citizenship Talk, ed. Heiko Hausendorf and Alfons Bora, 151–180. Amsterdam: John Benjamins. ———. 2007a. How to Read Austin. Pragmatics 17 (3): 461–473. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2007b. Pathways to Explicitness. Lingue e Linguaggio 6: 101–120. ———. 2007c. Detto non detto. Le forme della comunicazione implicita. Roma– Bari: Laterza (2nd ed. 2010). ———. 2009. Uptake and Conventionality in Illocution. Lodz papers in Pragmatics 5: 33–52. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2011. J.L.  Austin’s Linguistic Method: Towards a Reassessment. In John L. Austin et la philosophie du langage ordinaire, ed. Sandra Laugier and Christophe Al-Saleh, 109–135. Hildesheim: Olms. ———. 2012. Austin on Meaning and Use. Lodz Papers in Pragmatics 8 (1): 5–16. ———. 2013a. Some Remarks About Speech Act Pluralism. In Perspectives on Pragmatics and Philosophy, ed. Alessandro Capone, Franco Lo Piparo, and Marco Carapezza, 227–244. Heidlberg: Springer. ———. 2013b. Locution, Illocution, Perlocution. In Pragmatics of Speech Actions, ed. Marina Sbisà and Ken Turner, 25–75. Berlin: De Gruyter. ———. 2014a. Evidentiality and Illocution. Intercultural Pragmatics 11 (3): 463–483. ———. 2014b. The Austinian Conception of Illocution and Its Implications for Value Judgement and Social Ontology. Ethics & Politics 16 (2): 619–631. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2014c. Austin on Language and Action. In J.L. Austin on Language, ed. Brian Garvey, 13–31. Basingstoke, UK: Palgrave Macmillan. ———. 2015. C’è del pragmatismo in J.L. Austin? Una rilettura delle proposte austiniane sul tema della verità. Esercizi Filosofici 10 (2): 230–245.

  Marina Sbisà’s Deontic Approach to Speech Actions 

25

———. 2017. Implicitness in Normative Texts. In Pragmatics and Law. Practical and Theoretical Perspectives, ed. Francesca Poggi and Alessandro Capone, 23–42. Berlin: Springer. ———. 2018. Ways to Be Concerned with Gender in Philosophy. Phenomenology and Mind 15: 132–145. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2019. Varieties of Speech Act Norms. In Normativity and Variety of Speech Actions, edited by Maciej Witek and Iwona Witczak-Plisiecka, 23–50. Leiden: Brill. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2020a. Assertion Among the Speech Acts. In Oxford Handbook of Assertion, ed. Sandford Goldberg, 159–178. Oxford: Oxford University Press. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2020b. Illocuzione e dislivelli di potere. In Linguaggio d’odio e autorità. Lezioni milanesi per la Cattedra Rotelli, ed. by Claudia Bianchi and Laura Caponetto, 63–86. Milano-Udine: Mimesis. Transl. as Illocution and Power Imbalance in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2021a. Presupposition and Implicature: Varieties of Implicit Meaning in Explicitation Practices. Journal of Pragmatics 182: 176–188. ———. 2021b. (Im)politeness and the Human Subject. In The Philosophy of (Im)politeness, ed. Chaoqun Xie, 157–177. Cham: Springer. ———. 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: 301–319. 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. Sharon Millar and John Wilson, 153–171. Amsterdam: John Benjamins. Searle, John R. 1968. Austin on Locutionary and Illocutionary Acts. Philosophical Review 77 (4): 405–424. ———. 1969. Speech Acts. An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. ———. 1979. Expression and Meaning. Cambridge: Cambridge University Press. Simons, Mandy. 2003. Presupposition and Accommodation: Understanding the Stalnakerian Picture. Philosophical Studies 112: 251–278.

26 

L. Caponetto and P. Labinaz

Strawson, Peter F. 1964. Intention and Convention in Speech Acts. Philosophical Review 73 (4): 439–460. Witek, Maciej. 2013. How to Establish Authority with Words: Imperative Utterances and Presupposition Accommodation. In Logic, Methodology and Philosophy of Science at Warsaw University, ed. Anna Magdalena Brożek, vol. 7, 145–157. Warsaw: Semper. ———. 2015. Mechanisms of Illocutionary Games. Language & Communication 42: 11–22. ———. 2019. Accommodation in Linguistic Interaction. On the So-Called Triggering Problem. In Philosophical Insights into Pragmatics, ed. Piotr Stalmaszczyk, 163–191. Berlin: De Gruyter. ———. 2021. Illocution and Accommodation in the Functioning of Presumptions. Synthese 198: 6207–6244.

Should Speech Act Theory Eschew Propositions? Mitchell Green

1 Introduction The theory of speech acts was introduced into the philosophical mainstream by Austin (1975) and then refined by Strawson (1964), Searle, Sbisà, Bach and Harnish, and Schiffer (1972), among others.1 The theory attempts to articulate our folk practices of ascribing significant utterances to one another, particularly our practices of using those utterances to achieve effects beyond the production of clauses bearing semantic contents. As such it aims to provide insight into the many respects in which saying is also a kind of doing. Marina Sbisà was instrumental in showing that the first edition of Austin’s How to Do Things with Words was inaccurate in important ways ­relative to the manuscript on which that work was based. For this reason  My thanks to the editors of this volume, Laura Caponetto and Paolo Labinaz, for their insightful comments on an earlier draft of this chapter. 1

M. Green (*) Department of Philosophy, University of Connecticut, Storrs, CT, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Caponetto, P. Labinaz (eds.), Sbisà on Speech as Action, Philosophers in Depth, https://doi.org/10.1007/978-3-031-22528-4_2

27

28 

M. Green

she collaborated with J. O. Urmson to edit and publish a second edition of that work which has become the standard version thereof. Since that time, Sbisà has made a memorable contribution to the theory of speech acts by revealing several respects in which saying is doing, and more specifically by highlighting the ways in which speech acts shape our normative statuses relative to one another (Sbisà 2013). In so doing, Sbisà has guided researchers to appreciate dimensions of language use that go well beyond the communication of thoughts. Although not initially developed as a contribution to the theory of speech acts, Grice’s discussion of meaning (Grice 1989) also spurred that theory’s development. In the years immediately following the appearance of Austin’s and Grice’s theories, their relation to one another was not entirely clear. However, Strawson (1964) made a strong case for the relevance of Grice’s account of non-natural meaning (now typically referred to as “speaker meaning”) to the general theory of speech acts. The case begins with the observation that Austin seems to have thought that all speech acts require for their performance an invocation of extra-semantic conventions.2 Whereas semantic conventions are those imbuing words with their literal meaning, extra-semantic conventions go beyond these to enable certain utterances to perform socially significant acts. Thus for instance, “Check!” is to be said by a chess player when she has put her opponent in check; and likewise for many other types of case. This fact goes beyond a plausible semantic characterization of the word “check.” Strawson argues against Austin that not all speech acts depend on extra-semantic conventions. Accepting, for instance, may be carried out by formalized means, but it strictly requires only the overt manifestation of the agent’s willingness to take something that another agent is offering. An agent can make such willingness overt without invoking extra-­ semantic conventions; indeed without invoking semantic conventions.3  Sbisà (2009) and Green (2020a, b) discuss Austin’s version of conventionalism about speech acts in greater detail. 3  Sbisà (2009) proposes a reading of Austin on which the conventional nature of speech acts is due to the conventionality of their effects, which are in turn characterized as coming into being by dint of an agreement between agents. (“A conventional effect is such insofar as it comes into being by being agreed upon by the relevant members of a social group” [Sbisà 2009, 49].) We may doubt that an agreement so described is sufficient for the institution of a convention, and I have raised such doubts in Green (2018b). See also Green (2021a) for further discussion of conventions. 2

  Should Speech Act Theory Eschew Propositions? 

29

Strawson proposes accordingly that in order to perform an act of accepting, a speaker need only “speaker-mean” that she is doing so. This speaker-­ meaning condition is commonly understood as requiring that the speaker perform an action with the intention to produce a psychological effect in an addressee, by means at least in part of that addressee’s recognition of the speaker’s intention. Green (2007) offers a refinement of this approach that begins with the observation that speaker meaning does not require intentions to produce effects on addressees. Instead it depends on a speaker’s intention to make some aspect of her commitments overt: she must intend to make those matters knowable (though not necessarily known) to appropriate others, as well as intend that this very intention be knowable to appropriate others. Further, speech acts often depend on speaker meaning thus construed; yet many speech acts, including those that do not depend essentially on extrasemantic conventions, do not require such complex intentions for their achievement. I will develop these suggestions in Sect. 4.

2 Sbisà’s Argument Against Propositions Sbisà (2006, 2020) argues that a prominent way of thinking about speech acts as inherited from the work of Strawson and, later, Bach and Harnish (1979), backgrounds the agentive aspects of speech acts in favor of speakers’ psychological states and their propositional contents. Such a methodology is central to the influential work of Bach and Harnish (1979), for whom speech acts consist essentially in various configurations of intentions to produce a psychological effect in an addressee by means at least in part of that addressee’s recognition of the speaker’s intention.4 In developing this argument, Sbisà describes what she terms the strong speech act view, according to which it is essential to speech acts that they have a role in helping to create a world that is cultural rather than merely natural, and in which agents relate to one another in terms of their deontic statuses (Sbisà 2006, 161). Sbisà (2007, 471) writes,  I here set aside the question whether Austin invoked propositions in his account of illocutionary acts; Sbisà makes a strong case that it’s doubtful that he did. 4

30 

M. Green

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 civilisation as opposed to “state of nature”.

Sbisà distinguishes the strong from a weak speech act view according to which speech acts primarily serve the purpose of transmitting thoughts, and more specifically propositional attitudes. Unless we accept that the transmission of propositional attitudes would under appropriate conditions suffice for the creation of a world that is cultural rather than natural, replete with deontic statuses and the like—that is, unless the weak speech act view can account for the data that the strong speech act view offers to explain—the weak view would seem to be unsuited to explain the role of speech acts in human society. We have then two questions: (A) Is the weak speech act view unsuited to explain the role of speech acts in society? (B) Does invocation of propositions in our theory of speech acts force us to accept the weak rather than the strong view of speech acts? If it does, then on the assumption that the answer to question (A) is in the affirmative, we will have reason to resist the introduction of propositions into speech act theory. The scope of the present chapter prevents me from being able to answer question (A) conclusively by means of a consideration of all theories of the weak kind. However, in the next section, I will provide grounds for concluding that, with reference to the speech act of assertion, one prominent version of the weak view is unsuited to account for the kind of normative structure that Sbisà emphasizes. This will provide some reason for an affirmative answer to (A). After that (Sect. 4), I will defend a negative answer to question (B) by showing that a distinct form of speech act theory, which invokes propositions in a different way, does not devolve into the weak view and is, further, able to account for the normative phenomena that speech acts help to generate. On this basis I conclude that we do not have compelling grounds for removing propositions from the speech act theorist’s toolkit.

  Should Speech Act Theory Eschew Propositions? 

31

3 Limitations of the “Attitudinal” Approach to Speech Act Theory Sbisà, as we have noted, argues that the introduction of propositions into speech act theory makes it difficult to account for the normative statuses that speech actions create, writing: The introduction of propositions in speech act theory created the conditions for turning the effect of the illocutionary act from a full-fledged effect (a change in states of affairs obtaining in the world, albeit of a peculiar kind) to a merely cognitive one. (Sbisà 2006, 162)

By speaking of merely cognitive effects of illocutionary acts, I take Sbisà to mean that on the weak speech act view, speech acts have at most perlocutionary powers by virtue of which they bring about psychological effects in our addressees; such effects may in turn create or modify the normative statuses of speakers and their addressees.5 This is in contrast to a view on which speech acts are able constitutively to modify normative statuses among speakers, addressees, and other pertinent parties. In assessing this claim, we cannot hope to determine whether all varieties of speech act theory formulated in terms of the communication of propositional attitudes are vulnerable to Sbisà’s criticisms. Instead, in this section, we focus on Bach and Harnish’s influential attitudinal approach with attention to the question whether it has the resources to account for the normative structure which Sbisà highlights. Our focus for this purpose will be the case of assertion. Searle (1969, 66) characterizes assertion as an “undertaking to the effect that P.” This characterization needs elucidation because while ­ordinary discourse speaks of undertaking courses of action, talk of undertaking propositions is not idiomatic. Much subsequent literature on assertion elucidates this notion of propositional undertaking as a commitment to a  The editors of this volume have suggested to me that by “cognitive effect,” Sbisà means not a perlocutionary effect, but instead merely the state of understanding what the speaker is attempting to say. If so, hers is a stronger claim than the one I shall be examining. It seems wiser to evaluate the weaker claim, since if it is found wanting, then the stronger claim will be as well. 5

32 

M. Green

proposition.6 Yet although such talk is by now second-nature within philosophy of language and epistemology, we might still wish for more clarity about what it is to commit oneself to a proposition. Green (2013, 2016) argues that commitment to a proposition will take different forms depending on whether it occurs in the course of an assertion, or instead in the course of a conjecture or educated guess. These different members of the “assertive family” may be characterized normatively along the three dimensions of Liability, Frankness, and Fidelity (LFF). For assertion, those dimensions may be characterized as follows: Liability: one who asserts that P is right or wrong on the issue of P depending on P’s truth value. Frankness: one who asserts that P is sincere only if she believes that P. Fidelity: one who asserts that P opens herself to the challenge, “How do you know?” By contrast, a conjecture has the same liability condition as assertion, but its frankness and fidelity conditions are different. The main points are summarized in Table 1. Table 1  Three dimensions of commitment for four members of the assertive family, and for two non-members (sheer guesses and suppositions) (from Green 2016)

 See the useful overview in Shapiro (2018).

6

  Should Speech Act Theory Eschew Propositions? 

33

While the Liability and Frankness dimensions of assertoric commitment are uncontroversial, there will be less consensus on the exact formulation of the Fidelity requirement. Fortunately, we need not agree on the exact characterization of the Fidelity dimension of assertoric commitment in order to ask whether the Bach and Harnish approach has the resources to handle it. Instead, we note that the preponderance of current accounts of assertion have in common their appeal to something like Fidelity.7 With this background, we may assess one form of a weak speech act view. Bach and Harnish (1979) base their approach to speech acts on the concept of expression, itself defined in terms of reflexive intentions. A reflexive intention (or R-intention) is an intention that is intended to be recognized as intended to be recognized (pp. xiv-v). On this basis these authors define expression as follows: Expressing: For S to express an attitude is for S to R-intend the hearer to take S’s utterance as reason to think S has that attitude. (Bach and Harnish 1979, 15)

For instance, if S makes an utterance with the R-intention that her addressee to take it as reason to think S believes that P, then she expresses the belief that P. Further, different speech act types are defined in terms of both expression and the psychological state the speaker is intending to produce in their addressee. For instance, assertion is characterized as follows: Assertives (simple): (affirm, allege, assert, aver, avow, claim, declare, deny (assert … not), indicate, maintain, propound, say, state, submit): In uttering e, S asserts that P if S expresses:

i. the belief that P, and ii. the intention that H believe that P. (Bach and Harnish 1979, 42) By contrast, promises express a mixture of beliefs and intentions: Promises: (promise, swear, vow): In uttering e, S promises H to A if S expresses:  See Marsili and Green (2021) for further discussion and references.

7

34 

M. Green

i. the belief that his utterance obligates him to A, ii. the intention to A, and iii. the intention that H believe that S’s utterance obligates S to A and that S intends to A. (Bach and Harnish 1979, 51) Observe that these conditions are presented as jointly sufficient but not as necessary for their analysanda. (There thus may be other sufficient conditions.) Nonetheless, it is doubtful that the first set of conditions secure a speaker’s undertaking of a commitment that is sufficient for assertion. For while the expression of belief condition (i) would seem to secure the Liability and Frankness conditions, matters are less clear for the Fidelity condition. To see why, imagine a speaker S producing an utterance in a way that meets the above conditions for assertives. She may do so without presenting herself as either knowing or having strong grounds for believing that P. Instead, she may merely be laying bare her state of mind with the further overt intention that her addressee come to share that state of mind. She may do that without presenting the content of her belief as something that she knows to be true, has strong grounds for believing, and so on. Indeed, for all Bach and Harnish’s sufficient conditions for assertion tell us, a speaker could meet those conditions and go on to say, “… but I don’t know that P,” with no impropriety. However, one necessary feature of assertion is that one who asserts that P presents herself as knowing, or at least as having strong grounds for believing that P.8 In the imagined case the speaker is not so presenting herself, and so is not asserting that P.9 Accordingly, a speaker may meet the conditions that Bach and Harnish present as sufficient for asserting without meeting the Fidelity condition that seems a crucial component of the commitment that this act involves. We cannot hope to assess the entire range of analyses that Bach and Harnish offer of speech acts generally. However, given the centrality of assertion to everyday communicative life, we may conclude that Bach  See van Elswyck (2021) for discussion and references.  We may also observe that Bach and Harnish’s putative sufficient condition for promising is not in fact sufficient. The reason is that it builds in no requirement of addressee uptake. Yet in the absence of uptake, speakers are not committed to the future courses of action that their words describe. 8 9

  Should Speech Act Theory Eschew Propositions? 

35

and Harnish’s approach to speech acts will be inadequate to capture an important aspect of everyday communication.10 We may also appreciate the limitations of the propositional attitude approach to speech acts by considering whether it has the capacity to account for the behavior of certain explicit performatives.11 For one may reasonably expect an approach of this kind to explain how speakers who describe themselves as undertaking certain commitments, manage thereby to undertake those commitments. For instance, normally one who says, “I claim that P,” manages to claim that P, and normally one who says, “I promise to meet you tomorrow at noon,” manages to issue a promise to her addressee. That is not so when we change person or tense, or when one of these sentences is used to report what the speaker generally does under certain circumstances. How do these processes work in the first-person, present indicative active cases? Bach and Harnish attempt to account for how an utterance of “I claim that P” can be a way of claiming that P, that is, of putting forth P as an assertion. Their strategy is to invoke the idea of an indirect speech act (ISA). Some theories of ISA’s hold that they consist of two speech acts, one that is direct, and another that is indirect. Thus consider: (1) Excuse me, but you’re standing on my foot. We might suppose that the speaker of (1) both asserts that the addressee is standing on her foot and requests that he remove himself from that foot. The assertion would be direct, while the request would be indirect. This “two speech act” approach to ISA’s seems reasonable for cases such as (1), but less so for examples such as: (2) Can you pass the olive oil? when spoken in a typical dining situation. Here, while we are inclined to say that the speaker is requesting that the addressee pass the olive oil, it does not seem mandatory to suppose that she has done so indirectly by means of performing the illocutionary act of asking a question. Instead,  The range of topics covered in Goldberg (2020) are one indication of assertion’s centrality to everyday communication. Also, Green (2013) raises other objections to Bach and Harnish’s account of assertion. 11  The argument that follows is a condensed version of that given in Green (2021b). 10

36 

M. Green

we might say of case (2) that only one speech act has been performed, namely, that of making a request, while the sentence tokened therein has only been used in an act of speech, or locuted. An analogous choice faces one who wishes to theorize, from an ISA point of view, about how a speaker of (3) I claim that P. may use it to claim that P. Are they asserting the entire sentence, and on that basis indirectly putting forth P as a claim? Or are they only locuting (3) and, somehow, thereby illocuting its complement P? Bach and Harnish (1979, 208) advocate a “two speech act” version of the ISA approach to the cases that concern us. On analogy with their account of the performative “I order you to leave,” we may reconstruct that account to apply to such cases as (3) as follows: 1. He is saying, “I claim that P.” 2. He is stating that he is claiming that P. 3. If his statement is true, then he is claiming that P. 4. If he is claiming that P, it must be his utterance that constitutes his claim. (What else could it be?) 5. Presumably, he is speaking the truth. 6. Therefore, in stating that he is claiming that P, he is claiming that P. Bach and Harnish’s earlier remarks (1979, 29–33) support interpreting “say” in the foregoing argument in the thin, locutionary sense. On this interpretation, as Jary (2007) rightly notes the move from premise (i) to premise (ii) is questionable: a speaker’s utterance of an indicative sentence, even if that sentence is true, need not be an assertion. (I assume that Bach and Harnish use “assertion” and “statement” to refer to the same illocutionary act type.) What, however, if we read “stating” in (ii) in the weaker, locutionary sense? Then (ii) will follow from (i), but the conditional in (iv) will be dubious: a mere locutionary act will not be enough to account for how a speaker manages to make a claim. On either interpretation of its premises, then, the above argument does not establish its conclusion.

  Should Speech Act Theory Eschew Propositions? 

37

4 A Middle Way: Contentful Commitments In accounting for utterances such as (3), is there a way of proceeding that moves from the premise that a locutionary act has been performed (premise (i) in Bach and Harnish’s argument of the last section) to a more modest next step not going as far as the positing of an illocutionary act as explanans and yet one that is strong enough to yield the result that the speaker is claiming that and thus (assertorically) committing herself to P? Doing so would require developing an account of a range of communicative acts that are in some sense “between” locutionary and illocutionary acts. This will in turn require identifying the relevant notion of betweenness. Green (2021b) offers an approach that does not assume that speech acts depend crucially on reflexive intentions. I instead invoke the notion of a verbal signal, in which a speaker uses conventionally meaningful language with the objective of conveying the information encoded by that language. A verbal signal is thus a richer notion than that of a locutionary act (since such acts need not be performed with the intention of conveying information), but does not require the complex communicative intentions associated with speaker meaning (including Green’s preferred notion of speaker meaning in which the notion of overtness plays a central role). If this approach is viable, then speech acts may be performed either by means of verbal signals or by appropriate configurations of communicative intentions sufficient for speaker meaning. The approach of Green (2021b) also explains how speakers may use verbal signals to undertake commitments. As background for this approach assume that adult, socially competent speakers generally have final authority on what they commit themselves to. (Exceptions are cases in which they fail fully to grasp the meaning of the words they attempt to use to undertake those commitments, or have other commitments or sources of information patently incompatible with those commitments they are purporting to undertake.) With this background, we may discern how a speaker may use “I claim that P” to verbally signal that she is undertaking commitment to P as an assertion rather than with some other illocutionary force, while her use of “I conjecture that P” verbally

38 

M. Green

signals that she is undertaking commitment to P as a conjecture rather than as a claim.12 Speakers may undertake their commitments with more complex communicative intentions than are required for verbal signals, but need not do so.13 Further, when a speaker indicates that her utterance of an indicative sentence is an assertion but does so without the help of a preface such as “I claim that …,” she is still signaling but not verbally signaling the force of her remark. This strategy supports a larger project of understanding many speech acts as ways in which speakers undertake commitments as underwritten by social institutions. Elsewhere (Green 2013, 2016, 2018a) I have advocated a view of many speech act types as institutions that result from a process of cultural evolution. For instance, the practice of assertion is an institution characterized by a particular configuration of LFF norms, while the practice of conjecturing is an institution characterized by a different configuration of such norms. While there is no guarantee that a viable linguistic community must have a practice of assertion, those communities that do so have an institution that, by virtue of its normative structure, helps safeguard the reliable transmission of information.14 By contrast, norms governing speech acts having a commissive role, such as promises, safeguard the function of such acts in achieving coordination among multiple agents.15  Here and in what follows we may remain neutral on the nature of propositions. For our purposes they may be sets of possibilia, structured entities, or abstractions from utterances. 13  Green (2021b) uses the notion of verbal signal, as well as the related notion of verbal index, to reconstruct a line of reasoning analogous to that given in lines (1)–(4) in the text, and to show it to be valid. This reconstruction makes good on our earlier claim that verbal signaling is sufficient under certain conditions for the performance of a speech act. 14  I have elsewhere (Green 2019b) imagined a linguistic community that only has a practice of “ur-­ assertion,” in which speakers undertake primitive forms of commitment to the contents that they express. (In a similar spirit, Witek (2021) discusses what he calls proto-assertion.) Also, the fact that a given society does adopt the practice of assertion although it might not have, does not make that practice conventional. One may see this by noting that a practice of, for example, cutting might, or might not, be adopted by a society. (Some societies might not cut their food or other items of interest, but instead tear or break them.) But if a society does adopt cutting, there need be nothing arbitrary about it. Accordingly, the premise that having a practice of assertion is not mandatory for the existence and continuation of a linguistic community, does not imply that that practice, or any of the effects it makes possible, is arbitrary in a way that is necessary for convention. See Green (2021a) for further discussion of convention. 15  In Green (2019a), I argue in response to Geurts (2019) that we do an injustice to the richness and variety of social life if we attempt to explain the variety of communicative phenomena in terms of coordination alone. 12

  Should Speech Act Theory Eschew Propositions? 

39

The cultural-evolutionary approach to speech acts does not require us to suppose that the norms governing a particular speech act type were ever deliberately laid down. We may also adopt this approach without assuming that such norms are always ideally suited to the job for which they are designed. Further, we need not assume that cultural evolution has produced institutions whose norms provide guidance for all conceivable contingencies: there may be scenarios for which the norms associated with a given speech act type S simply fail to provide guidance, whether it be on the question whether a certain speech act has been performed, or whether it has been carried out felicitously, or precisely what commitments its performance engenders. On this cultural-evolutionary approach to speech act types as social institutions, propositions have a job to do in speech act theory, but not because the main role of speech acts is to communicate our psychological states to one another. Instead, propositions are among various devices we use to manifest and track our commitments, usually but not exclusively in social contexts. (Commitments are sometimes intrapersonal instead, such as when I carry out a line of reasoning in private, or attempt to make a promise to myself.) Such commitments may support either the reliable transmission of information or the reliable coordination of our actions. We may appreciate the value of proposition-involving commitments in light of a number of points. First, given the LFF facts depicted in Table  1, we see that members of the assertion family involve commitments that enable us to keep track of a person’s reliability along multiple dimensions: someone who, for instance, has a record of making claims that she is unable to back up when challenged, will be less apt to have her contributions added to our conversational common ground. Yet such challenges crucially rely on the propositions she has expressed in the course of her assertion, conjecture, or educated guess. So too, we understand the requirement of sincerity as one pillar safeguarding the reliability of our speech acts for the purpose either of information transfer or of action coordination; but depending on the speech act in question, this sincerity condition may make crucial reference to a psychological state with a propositional content.

40 

M. Green

Another reason for retaining propositions in our theorizing about speech acts concerns argumentation. Assertion is not deductively closed: one who asserts that P, and asserts that if P, then Q, has not thereby asserted Q. However, she is committed to Q, and we often exploit this fact in deriving surprising implications from what we say, as well as in arguing against a speaker’s position by pointing out that her explicitly avowed views yield unpalatable commitments. Viewing assertions as exhibiting a force/content duality helps us to keep track of these facts. Similar remarks apply to the phenomenon of agreement and disagreement, as well as to the related notion of changing one’s position. One way of changing one’s mind about an issue would seem to depend upon undergoing a transition from believing some proposition P, to either believing P’s negation or refraining from taking a stand on P’s truth altogether. Contents may play a role in speech act theory without that fact entailing that the only contents that may play that role are propositional contents. Instead, for all we have said so far, the contents associated with imperative and interrogative sentences might not be propositional, and instead have a different abstract structure (Hornsby 1998; Green 2018a). What is more, we need not assume that all communicatively significant acts have contents that correspond to those of any sentences. One who expresses disdain toward an object is not doing something needing to be characterized in propositional terms. Instead, she may be expressing an attitude better construed as prepositional (Tormey 1971) than as propositional. On this way of thinking, an agent might express disappointment over the broken vase, surprise about the great news, or love for her son. Such acts may be understood without our invoking propositions for the purpose. After arguing against the inclusion of propositions into speech act theory, Sbisà suggests that speech acts might make a different kind of conversational contribution from those commonly imputed to them by philosophers of language: Speech might make a difference in much the same way in which your room changes when you hang a new poster on the wall. Faced with a given circumstance, we find our way through it thanks to the way we perceive it and

  Should Speech Act Theory Eschew Propositions? 

41

pre-verbally (or indeed also verbally) categorize it. When words are spoken in that circumstance, the performed speech act adds its own occurrence to the circumstance and the way we perceive and categorize it is likely to change accordingly. (Sbisà 2006, 176)

There is no doubt that the very performance of a speech act will likely make a difference to our conversation or other verbal exchange: as Jary (2007) stressed, the fact that a speaker has made a request, a prediction, or a concession will usually be among the facts that interlocutors take into account in deciding how to proceed. However, the poster analogy is only apt if we stress that some posters are representational rather than just abstract configurations of line and color, and their appreciation in such cases depends on our taking account of their representational content. Similarly, interlocutors will often also need to take account of what a speaker has requested, predicted, or conceded, and this is usually best captured by the semantic contents of their words. Because those contents often take the form of propositions (alongside other semantic objects), these theoretical entities have a role to play in our theorizing about speech acts. Yet, as I have argued, we may acknowledge this role while also highlighting the ways in which agents performing speech acts invoke social institutions in the course of altering their normative statuses.

5 Contentful Indicators of Force In the course of her argument for eschewing propositions from speech act theory, Sbisà (2006, 163–165) also suggests that we reconsider the widely accepted view that linguistic devices that indicate illocutionary force must be bereft of semantic content. While I have offered reasons for resisting Sbisà’s proposed jettisoning of propositions from speech act theory, I endorse her suggestion to rethink the alleged incompatibility of force indication and the possession of semantic content. Accordingly, in this section, I will challenge one influential line of thought supporting this incompatibility claim, and offer a perspective on which we can also make good sense of semantically contentful indicators of illocutionary force.

42 

M. Green

According to a long tradition in philosophy of language, any expression that serves to indicate illocutionary force must lack semantic content. In support of this conclusion, Stenius (1967) argues as follows. In a logically perspicuous language, each sentence that is used in the performance of a speech act will contain both an indicator of illocutionary force and other material indicating the semantic content expressed by the sentence. The latter pertains to what is said, while the former is one aspect of how it is said. But indicators of force must be themselves bereft of semantic content. The reason is that if they did have semantic content, they would contribute that content by familiar compositional means to the sentence in which they occur; in so doing leave undetermined the force of the entire content thus expressed. Thus for instance, Stenius would hold that “I claim that …” is not an indicator of force, since in any given speech act it contributes its semantic content to the sentence in which it occurs, thereby leaving unanswered the question what is the force with which that entire sentence is being tokened. Stenius evidently construes the idea of force indicator in such a way that an expression can indicate force only if it does so indefeasibly. On this way of thinking, a linguistic device can be a force indicator only if it is a strong ifid (Green 2000), that is, a linguistic device that indicates force in any utterance in which it occurs. As Hare (1989) points out, nothing in principle rules out the possibility of strong ifids being used in the performance of certain speech acts. (Perhaps affixing your signature on a check would approach this ideal, indicating nearly indefeasibly one’s promise to pay the amount listed on that check.) However, while there may be special, technical purposes for which such strong ifids would play a valuable role, everyday speakers manage to indicate the force of their utterances without relying on strong ifids. Instead, against the backdrop of conversational record and interlocutors’ conversational goals, a variety of factors such as grammatical mood, intonation, facial expression, and prosody enable speakers to indicate the force of their utterances with no need of strong ifids. In this more tolerant sense of “force indication,” the process of force indication is a context-involving and imperfect one, but effective for all that.

  Should Speech Act Theory Eschew Propositions? 

43

Once we are comfortable thinking of force indication in this more permissive sense, there need be nothing improper about describing semantically contentful expressions as force-indicators as well. In earlier work (Green 2000) I have advocated such a position in discussing certain parenthetical expressions. To explain this idea, let us first define the notion of illocutionary validity. Let S be an arbitrary speaker, and a sequence of force/content pairs; then: is illocutionarily valid iff if speaker S is committed to each Ai under mode Δi, then S is committed to B under mode Δ.

This notion will piggyback on whatever is one’s favored notion of validity at the semantic level. It also requires reference to a partial ordering of illocutionary forces in terms of strength. Assertion is, in this sense, stronger than a guess, and a command stronger than a suggestion. Illocutionary validity is closed only under the weakest force occurring in the inference. (This comports with our earlier observation that one who asserts P, and conjectures that if P, then Q, is committed to Q but only in a way appropriate to conjectures.)16 Next consider: (4) If, as is the case, snow is white, then grass is green. The conditional in which the as-parenthetical is embedded may be used in an assertion, conjecture, or any of a variety of other speech acts. However, if the sentence is used in a speech act, then the speaker performing that act is also undertaking assertoric commitment to the proposition that snow is white. We may thus call constructions such as the as-parenthetical in (4) weak ifids: let “Δ__” be, syntactically, a function from sentences into sentences, chosen from a set of connectives each element of which is in the domain of a function IF, whose range comprises illocutionary forces. This allows us to speak of “the force associated with connective ‘Δ__.’” Let f Δ denote that force. Then we may say that “Δ__” is a weak illocutionary force indicating device (hereafter weak ifid) just in case for all illocutionary forces f ’ and sentences A, the inference  Vanderveken (1990) provides tableaux for various families of forces that helps elucidate the sense in which one force may be stronger or weaker than another.

16

44 

M. Green

f ’ … Δ(A) … --------------fΔA

is illocutionarily valid. Nothing in the concept of a weak ifid precludes its having semantic content. Indeed, the as-parenthetical such as we find in (4) appears to be a candidate for being a weak ifid. If so, we will have a case of a force indicator that also possesses semantic content. Similarly, on a more permissive understanding of force indication, we may see “I claim that …” as used in sentences such as (3) above to verbally signal that the complement clause following it is being put forth as an assertion, as adumbrated in Sect. 4. In response to these suggestions, Stenius might ask, what is the force with which the resulting semantic content expressed by “I assert (conjecture) that P” is being put forth? To this we may properly reply: that semantic content need not be being put forth with any illocutionary force at all. Instead, it is enough that the preface, “I assert that …” is used as a verbal signal as that notion was characterized in Sect. 4. This point leaves untouched the semantic facts that “I assert (conjecture) that P” is true just in case the speaker of that sentence asserts (conjectures) that P. Accordingly, so-called performative prefixes such as “I claim that …” and “I assert that …” do not change their semantic profile from one context of utterance to another. Instead, in certain grammatical and discursive environments, they typically serve as what I have called verbal signals, and in that role are able to indicate the force of other semantic material such as the complement clause that they embed.17 I have argued that Sbisà’s criticism of the weak speech act view has merit, by showing that one prominent version of that view appears to be too weak to account for the normative statuses that speech acts create. However, I have resisted Sbisà’s contention that the, or at least a culprit in this failing is the employment of propositions in speech act theory. Instead, I have contended that alongside other semantic contents, propositions have a role to play in helping us to elucidate, challenge, and  The liberalization of the force/content distinction advocated here does not justify rejecting that distinction entirely. In Green (2018a) I have defended a liberalization of that distinction, while showing that recent attacks thereon are not cogent. 17

  Should Speech Act Theory Eschew Propositions? 

45

coordinate with our own commitments and those of others. At the same time I have endorsed and elaborated upon Sbisà’s call for a relaxing of the longstanding banishment of semantic content from force indicators. Whether it is in the course of challenging or elaborating Sbisà’s arguments, however, we have been able to highlight the powerful role that language plays in the enhancement of human agency, thereby confirming a theme that has animated much of her philosophical contribution.

References Austin, John L. 1975. In How to Do Things with Words, ed. James O. Urmson and Marina Sbisà, 2nd ed. Cambridge, MA: Harvard University Press. Bach, Kent, and Robert M. Harnish. 1979. Linguistic Communication and Speech Acts. Cambridge, MA: MIT Press. Geurts, Bart. 2019. Communication as Commitment Sharing. Theoretical Linguistics 45: 1–30. Goldberg, Sandford, ed. 2020. Oxford Handbook of Assertion. Oxford: Oxford University Press. Green, Mitchell. 2000. Illocutionary Force and Semantic Content. Linguistics & Philosophy 23: 435–473. ———. 2007. Self-Expression. Oxford: Oxford University Press. ———. 2013. Assertions. In Pragmatics of Speech Actions, ed. Marina Sbisà and Ken Turner, 387–410. Berlin: De Gruyter. ———. 2016. Assertion. In Oxford Handbooks Online, ed. D. Pritchard. Oxford: Oxford University Press. https://www.oxfordhandbooks.com/view/10.1093/ oxfordhb/9780199935314.001.0001/oxfordhb-­9780199935314-­e-­8. ———. 2018a. A Refinement and Defense of the Force/Content Distinction. In New Work on Speech Acts, ed. Daniel Fogal, Daniel W. Harris, and Matt Moss. Oxford: Oxford University Press, 99–122. Oxford: Oxford University Press. ———. 2018b. Showing, Expressing, and Figuratively Meaning. In Beyond Semantics and Pragmatics, ed. Gerhard Preyer, 157–173. Oxford: Oxford University Press. ———. 2019a. Think Twice before Paving Illocutionary Paradise. Theoretical Linguistics 45: 39–51. ———. 2019b. Assertion, Implicature and Speaker Meaning. Rivista Italiana di Filosofia del Linguaggio 13: 100–115.

46 

M. Green

———. 2020a. Assertion and Convention. In Oxford Handbook of Assertion, ed. Sandford Goldberg, 347–370. Oxford: Oxford University Press. ———. 2020b. Speech Acts. In The Stanford Encyclopedia of Philosophy (Fall 2021 Edition), edited by Edward N. Zalta. https://plato.stanford.edu/ archives/fall2021/entries/speech-­acts/. ———. 2021a. The Philosophy of Language. Oxford: Oxford University Press. ———. 2021b. Force, Content, and Translucent Self-Ascriptions. In Force, Content & the Unity of the Proposition, ed. Gabriele M. Mras and Michael Schmitz, 195–214. London: Routledge. Grice, Herbert Paul. 1989. Studies in the Way of Words. Cambridge, MA: Harvard University Press. Hare, Richard M. 1989. Some Subatomic Particles of Logic. Mind 98: 23–37. Hornsby, Jennifer. 1998. Things Done with Words. In Human Agency: Language, Duty, and Value: Philosophical Essays in Honor of J.O. Urmson, ed. Jonathan Dancy, Julius M.E. Moravcsik, and Christopher C.W. Taylor, 27–46. Stanford: Stanford University Press. Jary, Mark. 2007. Are Explicit Performatives Assertions? Linguistics and Philosophy 30: 207–234. Marsili, Neri, and Mitchell Green. 2021. Assertion: A (Partly) Social Speech Act. Journal of Pragmatics 181: 17–28. Sbisà, Marina. 2006. Speech Acts without Propositions? Grazer Philosophische Studien 96: 155–178. ———. 2007. How to Read Austin. Pragmatics 17: 461–473. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2009. Uptake and Conventionality in Illocution. Lodz Papers in Pragmatics 5: 33–52. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2013. Locution, Illocution, Perlocution. In Pragmatics of Speech Actions, ed. Marina Sbisà and Ken Turner, 25–75. Berlin: De Gruyter. ———. 2020. Assertion Among the Speech Acts. In Oxford Handbook of Assertion, edited by Sandford Goldberg, 159–178. Oxford: Oxford University Press. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. Schiffer, Stephen. 1972. Meaning. Oxford: Oxford University Press. Searle, John R. 1969. Speech Acts. An Essay in the Philosophy of Language. Cambridge: Cambridge University Press.

  Should Speech Act Theory Eschew Propositions? 

47

Shapiro, Lionel. 2018. Commitment Views of Assertion. In Oxford Handbook of Assertion, ed. Sandford Goldberg, 75–98. Oxford: Oxford University Press. Stenius, Erik. 1967. Mood and Language Game. Synthese 17: 254–274. Strawson, Peter F. 1964. Intention and Convention in Speech Acts. The Philosophical Review 73: 439–460. Tormey, Alan. 1971. The Concept of Expression: A Study in Philosophical Psychology and Aesthetics. Princeton, NJ: Princeton University Press. Van Elswyck, Peter. 2021. Representing Knowledge. The Philosophical Review 130: 97–143. Vanderveken, Daniel. 1990. Meaning and Speech Acts: Volume 1, Principles of Language Use. Cambridge: Cambridge University Press. Witek, Maciej. 2021. Self-Expression in Speech Acts. Organon Filozofia 28: 326–359.

On the Conventional Nature of Illocutionary Acts: Uptake, Conventions, and Illocutionary Effects Bruno Ambroise

1 Introduction Marina Sbisà is probably the best specialist of John L. Austin’s texts and, by far, the best interpreter of his philosophy; but she also is a great philosopher and, in defending Austin’s positions, she thereby supports singular and important philosophical theses. Such is the case with conventionalism in speech act theory: by reminding the philosophical community of Austin’s conventionalist conception of the efficiency of speech, she demonstrates that only a conventionalist position can provide a conceptually clear understanding of the fact that speech acts. To put it another way: it is not (conceptually) arbitrary that Austin’s wish to study the proper capacity of speech to do things, or to produce actions, led him to adopt a conventionalist conception of speech acts—and not an intentionalist conception, though this latter is dominant today. B. Ambroise (*) Institute of Legal and Philosophical Sciences of the Sorbonne (ISJPS: UMR 8103), Université Paris 1 - Panthéon Sorbonne & CNRS, Paris, France e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Caponetto, P. Labinaz (eds.), Sbisà on Speech as Action, Philosophers in Depth, https://doi.org/10.1007/978-3-031-22528-4_3

49

50 

B. Ambroise

Indeed we can contend that, following Peter F. Strawson’s first criticisms (Strawson 1954), and then the works of John R.  Searle (1969), H. Paul Grice (1989), and Kent Bach and Robert Harnish (1979), the now standard conception of speech acts is an intentionalist one, according to which acting through or with speech consists, in one way or another, in making intentions (of one or more orders) manifest to others and, therefore, in intervening on their beliefs. This conception, very widespread in philosophy of language and pragmatics, retains an important, but ambiguous, point of the Austinian theory: the idea that for a speech act—and more precisely its illocutionary part—to be accomplished, the famous “securing of uptake” must be obtained (Austin 1975, 117). For Austin, this necessary condition is to be added to the conditions for the happiness of a given speech act (Austin 1975, 117–118).1 For most speech act theorists today, this condition often becomes a necessary and sufficient condition for a speech act to be successful—but this “securing of uptake” is then reinterpreted as the “hearer’s recognition of the speaker’s intention to perform the illocution in question” (Horsnby and Langton 1998, 25).2 It is thus understood as the condition for a communicative act to be achieved, which only consists in (correctly) making a certain kind of the speaker’s intention(s), or some other propositional attitude, manifest to the receiver. For instance, for a promise to be achieved, it is necessary and, according to some, sufficient that the promisor uses a sentence such that the promisee, in understanding it, recognizes the promisor’s intention to make a promise (given an appropriate context and other ancillary conditions). This way of understanding the Austinian condition of uptake might have good reasons—at least within the standard framework of speech act theory, which departs from Austin on many decisive points. But it certainly misses an important aspect of Austin’s conceptual analysis: the idea that saying is acting—and, more precisely, that performing illocutionary acts produces real effects. Indeed, within the framework of the standard theory, the efficiency of language is reduced to the fact of producing the  More precisely, one can contend that it specifies the A conditions. See below.  I only give this quote for the sake of convenience because it offers a kind of synthesis of the positions of Searle, Grice, Bach and Harnish. Surely, this quote is taken from a text that is not solely devoted to this issue, but Hornsby develops her position further in Hornsby (1994). 1 2

  On the Conventional Nature of Illocutionary Acts: Uptake… 

51

correct grasp, by the hearer, of the speaker’s intention(s)3—that is to the fact of producing certain beliefs in the hearer. One can certainly admit that the production of such beliefs corresponds to effects of the speaker’s discourse—in that sense, the “uptake” would be the main effect of illocutionary acts. But that is certainly not what Austin had in mind when he was speaking of illocutionary acts, at least for two reasons: (i) this kind of “cognitive” effect is precisely what he called “perlocutionary”; (ii) wishing to escape the descriptivist illusion, he wanted to identify effects (produced by illocutionary acts) that Sbisà calls “full-fledged” (Sbisà 2006, 162), such as a promise, a marriage, or a christening. Austin was thus investigating the capacity of language to produce real changes “in states of affairs obtaining in the world” (Sbisà 2006, 162), and not only its very well-known capacity to transmit content and to affect the beliefs of interlocutors. What is important in Austin’s “discovery” of performative utterances and illocutionary acts is precisely the uncovering of a singular and specific power of language, which any purely cognitive understanding of the effects of language will miss. The intentionalist conception of speech acts, which forms the standard framework of analysis, precisely offers such an understanding and (at least) runs the risk of failing to grasp Austin’s important conceptual point. That is why it is not so easy to dispense with the Austinian conception of illocutionary acts, which takes a conventionalist form. If Austin insists on the conventional aspect of the illocutionary act, it is not because he focuses excessively on institutional speech acts (as it is often presupposed), but because he seeks to elucidate this capacity of speech to produce objective, and not merely cognitive, effects. His idea is that, to account for the capacity of speech to produce such effects, it is necessary to admit conventions which govern their productions—by (the joint participation of ) the speaker and the hearer(s). For, if these effects are objective, they have a (conceptual and probably ontological) peculiarity: they precisely are the results of the uptake, by the hearer(s), of the speech act intended by the speaker. Two consequences follow, well noted by Sbisà:

 Most often, in these approaches, intentional terms are nominalized and “intentions” are therefore understood as “mental states.” For a critique, see “A Plea for Excuses” in Austin (1979). 3

52 

B. Ambroise

(i). Illocutionary effects are conventional and belong to an order of reality that is not that of nature (in the physicalist sense of the term)— Sbisà says they are “normative” (Sbisà 2009, 45). (ii). These are effects which depend, for their achieving, on the collaboration of the speaker and the hearer, or on their joint action. In this chapter, I would like to return to Sbisà’s defense of the necessarily conventional character of the speech act, emphasizing the logic of her argument, which unearths often misunderstood points of Austin’s philosophy (especially his philosophy of action). I would also like to complete this by offering a clear account of the illocutionary act as a joint action that requires its uptake, by resorting to the work of Richard Moran. Finally, I would like to emphasize the socially situated dimension of uptake—of which Sbisà is fully aware but which she does not always take into account in her explanation. Thus I would like to insist on the social conditions of the efficiency of the illocutionary act, as they have been underlined by Pierre Bourdieu. In doing so, I hope to indicate (if not to demonstrate) that Austin’s insistence on institutionally supported speech acts was not gratuitous, nor mistaken, but well founded.

2 Sbisà’s Criticism of the Intentionalist Understanding of Speech Efficiency  he Securing of Uptake as the Backbone T of the Intentionalist Conception Sbisà devoted at least one full paper to the defense of a conventionalist approach of illocution against the intentionalist conception. In “Uptake and Conventionality in Illocution” (Sbisà 2009), she explicitly criticizes the intentionalist account of illocution that relies on the necessity, for the illocutionary act to be achieved, of the uptake of the intended act. According to this conception, the uptake, whose necessity had been noticed by Austin, must be understood in terms of the recognition of the speaker’s intention.

  On the Conventional Nature of Illocutionary Acts: Uptake… 

53

Let us first look at what Austin said: 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, 116–117)

As Sbisà reminds us, here Austin contends that the securing of uptake is “a necessary condition for the performance of an illocutionary act” (Sbisà 2009, 34). But Strawson, wishing to interpret the “illocutionary force” in Gricean terms and, more generally, to replace the Austinian account with a Gricean framework, suggests that –– 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; –– if this recognition is necessary for the speaker’s performance of the illocutionary act to be successful […], 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); –– the illocutionary act can therefore be traced back to (one kind of ) speaker meaning. (Sbisà 2009, 35–36) As a result, the illocutionary act is reduced to the understanding, by the hearer, of the speaker’s intention to produce such an act. To perform an order, or a promise, is to make one’s intention to order (or to promise) recognized by the hearer through the production of a certain utterance expressing (in a more or less explicit way) this intention. Therefore, the uptake is considered as “the only effect essentially connected with the performance of an illocutionary act” (Sbisà 2009, 36). Such a conception became highly influential and widely taken up, for example, by Searle or by Bach and Harnish, who proposed different

54 

B. Ambroise

versions of it. Searle, to avoid confusing illocutionary and perlocutionary effects, thus reformulates the role of uptake: 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)

Searle maintains that this “illocutionary effect” is the only effect produced at the illocutionary level. He also gives a (modified) Gricean intentionalist account of its production: The speaker S intends to produce an illocutionary effect IE in the hearer H by means of getting H to recognize S’s intention to produce IE. (Searle 1969, 47)

The recognition of the speaker’s intention thus becomes the only necessary condition to produce an illocutionary act. And this conception allowed speech acts theorists to reject Austin’s conventional account of speech acts. More precisely, they consider that a sharp distinction must be made between illocutionary effects, which are strictly a matter of communication (of intentions), and perlocutionary effects which, when they provoke changes in the interactional order and, more broadly, in the social order, are supported by conventions and institutions. As such, illocutionary effects do not need conventions to occur—except, Strawson admits, in a few remarkable cases (marriages, christenings, etc.). But even if Strawson makes this concession, he bases this bipartition on a very particular reading of the Austinian conventions: he considers that Austin, “in speaking of the conventionality of illocutionary acts, was concerned [only] with the conventional means by which such acts are performed” (Sbisà 2009, 38). This reading allows him to make the conventional aspect inessential to the understanding of uptake and “inaugurate[s] 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” (Sbisà 2009, 39).

  On the Conventional Nature of Illocutionary Acts: Uptake… 

55

As Sbisà correctly notices, even if Searle seems to have partly resisted Strawson’s analysis in underlining the conventional aspect of illocutions (Searle 1969), his subsequent focus on the condition of sincerity, understood in intentional terms, also leads him to explain the efficiency of speech in terms of uptake, by the hearer, of the speaker’s intentions. So it has become accepted in speech act theory in general that illocutionary efficiency, insofar as it is based on the correct grasp of the speaker’s intentions, is a communicative and not a conventional matter.

Sbisà’s Defense of the Conventionalist Position Sbisà raises (at least) two important objections to the tenability of the intentionalist approach,4 rightly claiming that “behind the divide between conventional and intention-based speech acts there lies the contrast between saying and doing [the very contrast that Austin wanted to combat], reborn within the very context of speech act theory” (Sbisà 2009, 42): (i). To reduce the uptake to the (correct) grasp of the speaker’s intention(s) is to consider the illocutionary effect as a cognitive effect on the interlocutor’s beliefs.5 It is therefore to make the illocutionary effect a subjective and cognitive effect, whose status is contrary to the proven and observed objectivity of the illocutionary effects of the most common speech acts. This conception of illocutionary effects also abolishes the conceptual distinction between the illocutionary and perlocutionary domains. (ii). The conventionality of illocutionary acts in Austin’s analysis is not only a conventionality of the means used to produce them, but also, and above all, a conventionality of illocutionary effects. This point is, in a sense, close to the previous one: the observed objectivity of illocutionary effects can only be produced because of conventions that make the agreement of the relevant parties as to what is produced possible.  In saying this, I do not claim that all the objections apply to all the authors previously mentioned. They apply to a now widespread framework of analysis. 5  This is debatable, but even in Searle’s model it is difficult to see the illocutionary effect as anything other than ultimately shared beliefs and intentions. 4

56 

B. Ambroise

Let’s take a closer look at these two important objections. Objection i. Sbisà sometimes makes the first point by saying that speech act theories moved from an objective conception of context to a cognitive conception (Sbisà 2002, 427–429; Sbisà 2009, 39). An important part of the objection consists in showing that a cognitive conception of context is unable to explain how the hearer can come to consider that a given speech act is correctly performed: whatever the speaker thinks, or whatever his intentions, they are not sufficient to understand what speech act is being performed, if the publicly available objective context does not help this understanding or, worse, runs counter to it. In fact, the available objective context takes precedence over the speaker’s beliefs and intentions. As Sbisà reminds us: […] speaker’s beliefs hardly suffice to get the interlocutor to take a speech act as successfully performed. […] 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. (Sbisà 2002, 428)

Here Sbisà’s almost materialist analysis6 stresses that beliefs are not enough to establish the success of an act and that much more concrete conditions, observable by all participants in the exchange, must be fulfilled. To put it bluntly, if the material, social, objective conditions— which correspond to what Austin called the A and B “felicity conditions”—of a given illocution are not fulfilled in the context of its accomplishment, then it cannot be accomplished. Of course, the argument is most convincing with clearly institutional examples—most of Austin’s original examples of performatives—but Sbisà shows that it is also valid in the case of constative utterances: “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” (Sbisà 2002, 429).  “Materialist” is my qualification: I am not sure that Sbisà would accept it. But I think it would be possible to defend this rapprochement in more detail. 6

  On the Conventional Nature of Illocutionary Acts: Uptake… 

57

Another part of the objection would be to point out,7 like Austin, the possibility, and, sometimes, the necessity that a judge be able to decide whether a certain illocutionary act was performed, regardless of the beliefs and intentions of the participants in the exchange. Of course, in some circumstances, beliefs and intentions do matter, especially in relation to the question of the speaker’s sincerity or commitment. As such they matter to determine whether the act is completely successful, that is whether it satisfies all the conditions identified by Austin, from A1 to Γ2, given its context. But as a matter of principle, all other things being equal, an external observer should be able to determine (and sometimes decide and impose) which speech act has been performed, given the context in which it was performed. This principal objectivity of the illocutionary act performed seems to prevent its reduction to the simple recognition, by the interlocutor, of the speaker’s beliefs and intentions: this is because the illocutionary effect obtained does not depend solely on the speaker and on what he wants to communicate. It is—and must be—objectively decidable, at least in a given context. Another way of putting the same point is to remark, as Sbisà does repeatedly, that to conceive of the illocutionary effect as a purely cognitive effect is to risk annulling the distinction between the perlocutionary and the illocutionary, which was precisely Austin’s conceptual issue. As Sbisà reminds us, analyzing Stalnaker’s proposals, “[…] producing belief is a typical perlocutionary act. So Stalnaker’s [cognitive] context change does not account for speech acts qua illocutionary acts” (Sbisà 2002, 433). Indeed, if we support the idea that producing an illocutionary act consists in making the hearer grasp one’s intention to perform a certain act, that is, consists in changing one’s beliefs (Stalnaker’s “context”), then the act is reduced to communicating (and making the listener recognize) this intention, the result of which corresponds precisely to what Austin called perlocutionary effect (or consequence): a certain belief resulting from what has been (intentionally) said. Now, a characteristic feature of the perlocutionary effect is precisely that it is contingent and depends centrally on the hearer’s own idiosyncrasy, in this case on how, given her  I use the conditional because, if Sbisà seems to be perfectly aware of this point (see Sbisà 2013, 58), she regularly decides not to insist on it. 7

58 

B. Ambroise

beliefs, she interprets the speaker’s intentions. To put it bluntly again: nothing can determine how the hearer’s beliefs will be affected by the communication of the speaker’s beliefs (or intentions). But this large flexibility of interpretation is not admissible in the case of illocutionary effects: if a promise, or an order, or even an assertion is to be done, then its accomplishment does not depend on the hearer’s idiosyncrasy. Moreover, the effects of the promise (or the order or the assertion) do not depend on the intentions of either the speaker or the hearer: the illocutionary effects take place according to a normativity that cannot depend solely on the (good) will of the participants in the verbal exchange.8 So, in this type of analysis, the illocutionary effect itself—that is, an effect which does not reduce to cognitive and subjective attitudes of the interlocutors—vanishes.9 Objection ii. If we admit the objectivity of the illocutionary effect, then we need to explain it. This is a relatively easy task once one understands that it is conventional in nature. This is a major conceptual point that Sbisà regularly emphasizes: “[…] 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” (Sbisà 2009, 43). This reminder amounts to taking seriously the idea, philosophically and strongly supported by Austin throughout his work, that speaking is acting in the full sense of the term. To take seriously the idea that speech acts is to consider that it produces real, full-­ fledged, effect(s). And, in the case of the illocutionary act, the effects cannot but be conventional by nature. As Sbisà regularly insists, “Austin distinguishes three effects that […] are connected with the illocutionary act, the first of which is the securing of uptake” (Sbisà 2009, 44). This first one is best seen as a condition to be fulfilled for the illocutionary act to be really accomplished.10 But a  On this precise point, inspired by Hume, see Ambroise (2013), Lance and Kukla (2009), and Moran (2018). Austin vigorously defended this idea in his “Other Minds” (see Austin 1979, 100–101). 9  Which, admittedly, would not be a problem for Grice. 10  For a discussion of this point, see Longworth (2019). 8

  On the Conventional Nature of Illocutionary Acts: Uptake… 

59

central effect to be obtained by an illocutionary act for it to take place— and which is now completely ignored by mainstream pragmatics—is the second one noticed by Austin: 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. Thus “I name this ship the Queen Elizabeth” has the effect of naming or christening the ship […]. (Austin 1975, 117)

Later on, Austin refers to this second effect as “taking effect.” I will skip most exegetical details given by Sbisà, as—I think—they are basically correct and based on a thorough knowledge of all Austin’s manuscripts. What is important is that this “taking effect,” as the example given by Austin shows, is intrinsically conventional. Austin claims that this conventional effect is shared by all illocutionary acts qua illocutionary acts. How can we understand this conventionality of the effect produced? Sbisà explains: 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. (Sbisà 2009, 45).

Two ideas are important here: the kind of reality produced by the act is not natural but conventional; a normative dimension is introduced in the discursive exchange, which is another aspect (or consequence) of the illocutionary effect’s objectivity noticed before. Concerning the first idea, Sbisà adds: […] to extend this idea to [all] other illocutionary acts, we should look at changes in states of affairs belonging to the same level of reality as norms. Such are, for examples, assignments or withdrawals of rights and licenses, assignments of obligations or waivers, undertakings of commitments.

60 

B. Ambroise

Indeed, a large number of illocutionary act types have effects that can be described in such terms […] and it is moreover possible to describe this way the effects of illocutionary acts tokens […] Even for a type of illocutionary act that is explicitly invoked by Strawson to show that not all illocutionary acts are conventional, that is, warning (Strawson 1954, p. 440), there is, quite intuitively, a state of affairs which is brought about and can be described in terms similar to those introduced above […]. (Sbisà 2009, 45)

This latter situation11 is what we can call the disclaimer of the speaker’s responsibility for what she warns about. This is a new norm which is established in the verbal and behavioral interaction. As Austin puts it in talking about promising, “a new plunge is taken” (Austin 1979, 99): once you have made a promise/an order/an assertion/a marriage/a warning, you can no longer ignore the commitment made there. In the most obvious case of the promise, if you have made a promise, then you are bound to keep it (whatever you think of it), because the illocutionary act of the promise creates precisely that commitment which does not belong to the natural order of things—but to another dimension. The same is true, in fact, of the assertion: once you have asserted something, you are bound to continue your speech in a manner consistent with what you have asserted and to behave accordingly. The assertion, as an illocutionary act, does introduce a norm.12 Austin never made this idea of norm completely clear, probably refusing to engage in ontological speculation. However, he admitted that the illocutionary effect was different from natural effects, and he repeatedly criticized the idea that reality was reducible to material reality.13 Moreover, when he sought to qualify illocutionary effects, he consistently spoke of created commitments and obligations—all things that fall within the  I am a bit reluctant to qualify the illocutionary effect in terms of “states of affairs,” as Sbisà does— because I think this qualification runs the risk of being understood in an overly substantial manner. 12  On this point, also see Brandom (1983) and Fricker (2007). 13  This is evident in his analyses of what it is to act, both in Austin (1975) and in “A Plea for Excuses” (Austin 1979, 175–204), and in his analysis of perception (Austin 1962). 11

  On the Conventional Nature of Illocutionary Acts: Uptake… 

61

domain of the norms and which do require an explanation in terms of conventions.14 Sbisà proposes the following interpretation: 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 states 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. (Sbisà 2009, 45–46)

It is therefore because the type of reality to which the illocutionary effect belongs falls under the normative domain that we must look for a necessarily conventional explanation for its establishment: without a convention making it possible to determine what normative effect follows from such and such a linguistic procedure—a speech act of a certain type—this would not produce an effect in the illocutionary sense.15 And it is precisely this need to appeal to conventions that explains the role played by the securing of uptake. Far from being based on the transmission of intentions,16 it consists in recognizing the validity of a socially established procedure, that is, a convention. Indeed, if a conventional effect is understood, as proposed by Sbisà, as “an effect [whose production is achieved] thanks to the agreement on the part of the relevant participants on which act it is that has been performed” (Sbisà 2009,

 We must add a characteristic peculiar to illocutionary effects, noted by Austin, which distinguishes them from “natural” effects: they are cancellable in a way distinct from physical effects, which cannot be undone but only modified. Sbisà calls this characteristic “defeasibility.” See Sbisà 2007, 465–466; see also Caponetto 2020. 15  Of course, it could produce many other kinds of effects (physical, rhetorical, perlocutionary, cognitive, etc.). 16  This is not to say that intention has no role to play in this process. See Sec. 3 and Moran’s arguments below. 14

62 

B. Ambroise

265), then the uptake plays a decisive role, as a necessary condition, in the implementation of the convention that is invoked. As Sbisà sums up: […] what is revealing of the conventionality of the illocutionary act (understood as the conventionality of its effect) is the need to secure uptake. A conventional effect is such insofar as it comes into 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 […] she is committed to a further certain course of action. […] Under this respect, there is no substantive difference between so-called “conventional”, institutional or ritual cases and informal, conversational ones. (Sbisà 2009, 49)

We can thus understand why any illocutionary act, insofar as it produces conventional effects, requires uptake: indeed, its effect (as token) is the product of an “intersubjective agreement” (Sbisà 2007, 470) that conditions the correct instantiation of the convention that establishes the type of (normative) effect of which it is a token. In this process, the intentions of the participants in the interaction, whether communicated or not, have no decisive role to play. They cannot create a conventional effect anyway. Rather, the causal relationship must be reversed if we are to understand Austin’s felicity conditions: it is because certain conventional procedures imply the speaker’s sincerity (in the case, e.g., of promise or assertion) that certain intentions on her part are required. But, as such, intentions do not explain the performance of illocutionary acts that produce conventional effects. And since every illocutionary act is defined by the production of this type of effect, we can understand why an intentionalist explanation is unable to account for it. This is one of the strong conclusions that must be drawn from reading the important texts of Sbisà. Another strong conclusion is that Austin’s appeal to the securing of uptake is intrinsically part of a conventional explanatory device: the securing of uptake is precisely the mark of a conventional understanding of the effectiveness of the illocutionary act, insofar as it depends on a conventionally determined recognition of the effect it is supposed to produce.

  On the Conventional Nature of Illocutionary Acts: Uptake… 

63

3 On the Conventional Nature of the Illocutionary Act as a Social Act and Its Social Conditions  he Speech Act of Telling as a Public Act T of Commitment The lesson we have drawn from Sbisà is that an illocutionary act is necessarily conventional since it produces specific normative and objective effects. Now I would like to concur with her and reinforce her hypotheses by resorting to a recent analysis of a specific kind of speech act put forward by Moran (2018).17 What is interesting about Moran’s analysis is that he focuses precisely on assertive speech acts that are usually used to support the intentionalist thesis. Yet, he offers an analysis of telling—the uttering of which consists in conveying some information to someone— as an act that produces commitments because it is a social act. In this respect, Moran joins Sbisà’s conventionalist analysis while focusing on the specific act of assertion that is telling. Moran’s strategy is to start from Reid’s idea that there are “social acts of the mind,” of which telling is one, while basing his analysis of telling on a characteristic already noted by Austin and Sbisà, but which he treats differently: that of the normative commitment made through a telling. Surprisingly enough, given his general position, Moran insists on the intentional nature of—and on the sincerity implied by—the act of telling as an act of telling something to someone who is, thereby, given reasons to believe what is told: […] what I am calling the ordinary situation of verbal communication, insofar as this centers on acts like that of one person telling something to another, is an intersubjective act in which the status of the utterance as a reason to believe what is said depends on the speaker and her audience being related in a way with a particular structure. (Moran 2018, xi)

 In what follows, I will only present schematically some ideas from this very rich and important book. 17

64 

B. Ambroise

The act of telling thus establishes a particular structured relationship between the speaker and the hearer, who occupy two distinct but related positions. In this type of relationship, it is important that the speaker has a sincere intention to convey some information to the hearer in speaking: for this reason, the hearer is justified in believing him. But the intentional character of the act should not be understood as depending on an intention seen as a mental state to be communicated, by the utterance, to the hearer. The intentional character of the act must rather be understood, in an Anscombian way, as a non-mentalistic property revealing the way the act is done: in this case, as an act engaging the authority of the speaker.18 Sincerity matters epistemically to the hearer not because it provides him with a window onto the speaker’s beliefs but because it tells him what the person as such is assuming responsibility for, as this relates to the hearer’s own belief. (Moran 2018, 95)

And this commitment cannot be derived from a simple transmission of information (be it speaker’s beliefs or intentions),19 but from the fact that the speaker, through his statement, presents himself as committing himself to the truth of what he says: he implies his responsibility by saying what he says as he says it, that is, by performing the speech act of telling (rather than another speech act): It is the status of her utterance as committed act rather than as phenomenon or found object that makes it a matter of belief or disbelief for another person, and this status is not for the speaker herself a matter of theoretical speculation. (Moran 2018, 27)

 It could be argued, in a way that exceeds the limits of this text, that Austin’s understanding of the intentional character of an act, as analyzed in his texts on action (see “A Plea for Excuses” and “Three Ways of Spilling Ink” in Austin 1979), is not so far from Anscombe’s analyses (Anscombe 1957). Moran implicitly draws on this proximity to propose an analysis of the act of telling as an intentional illocutionary act and to criticize a mentalist reading (encouraged by Grice himself ) of Grice’s proposals on communication. 19  Moran devotes a good part of his book to challenging a certain reading of Grice according to which telling can be reduced to a simple transmission of information that would function as a cause (and not only a reason) of the hearer’s beliefs. 18

  On the Conventional Nature of Illocutionary Acts: Uptake… 

65

How is this possible? Precisely because the act of telling is a speech act using a public procedure: it is a social act of the mind. What exactly does that mean? Moran insists on the fact that the transmission of information necessarily rests on a linguistic act because the type of information given by an act of telling can only be given in a public manner: it is because the speaker presents the information he gives as intentional through the public procedure of telling that others can grasp it as information he gives them—because it is what the speech act of telling is supposed to be used for. It is precisely at this level that Moran resorts to the illocutionary dimension of speech acts elaborated by Austin: the act of telling is a speech act in the full sense of the term, because the proper action of this act, if successful, would consist precisely in informing others of what I take to know20 by making them aware that I am presenting this information intentionally in making this (and not another) speech act (Moran 2018, 31–32). The illocutionary dimension would thus take over the dimension of address intrinsic to this informative assertion. Moran then takes up Austin’s idea that a successful illocutionary act requires “the securing of uptake” and, as Sbisà does but in another way, links this necessity to the social aspect of these acts of telling (and of speech acts in general): The simple act of telling involves an interlocking system of authority and dependence on, or deferral to, the role played by the other person. As Reid says, social acts of mind are distinctive in that they “can have no existence without the intervention of some other intelligent being”. (Moran 2018, 34)

Indeed, they must be expressed to another party who recognizes them as what they are: The speaker has not managed to perform the act of telling or promising if there is no recognition of what she is up to by the person she is addressing  As Laura Caponetto reminds me, one might be worried that the telling might not be completely sincere, either because the speaker was trying to deceive the interlocutor or because she does not know what she is saying. This is why Moran (2018, 91) insists that the speaker must be prepared to take responsibility for what she says, and present, in her telling, what she takes herself to know. 20

66 

B. Ambroise

[…]. The proper understanding of Austin’s “uptake conditions” for ordinary speech acts [will lead us to see] that the understanding of illocution generally requires an understanding of practical knowledge that involves two parties and not just one, what they know together rather than what they know about each other. (Moran 2018, 34)

Here, what the speaker and the hearer know together refers to the speech act accomplished, that is, to a verbal procedure. Later on, Moran adds: […] the illocutionary force of an utterance, which is the prior condition for its having the status of an assertion or an act of telling, has an internal relation to the recognition of its audience, for it is made possible by that very recognition. Insofar as the speaker’s audience sees that in this verbal act she is intending to warn, promise, or tell him something, then the speaker has in virtue of that fact succeeded in warning, promising, or telling him. (Moran 2018, 137)

Thus, uptake is what both enables and ensures that an illocutionary act is successful: it is a necessary condition of its “taking effect.” And Moran understands this internal necessity as being linked to what he calls the social aspect of these acts: The sufficiency of recognition for illocutions is also part of what is essentially overt or manifest in “social acts of mind”, as expressed in the fact that, for instance, a “false promise” is still a promise, a false testimony is still testimony. (Moran 2018, 137)

In other words, Moran links the social character of the illocutionary act to its necessary and sufficient recognition by the hearer. He even argues that this explains the committing dimension of the act outside any recourse to internal, private, or mental elements. In this sense, the reality of the illocutionary act resides in the discursive exchange, such that it must meet certain (social or conventional) rules to be, for example, an assertion. According to Moran, social normativity always results from the implementation of a certain linguistic procedure by two individuals who are free to do so and who, in an encounter, decide for one to take on a

  On the Conventional Nature of Illocutionary Acts: Uptake… 

67

commitment and for the other to ratify the commitment presented by the first. As such, his understanding of the sociality of the illocutionary act and its effects is minimal; but it is not null and even decisive as the following passage proves: […] the speaker’s “illocutionary authority” is other-dependent in that […] the content of what she knows (which illocution she is performing) must include the perspective of the other person on her performance. This dependence is connected with the general notion of authority, for the speaker’s illocutionary authority obtains insofar as it is recognized by others to obtain. This is perhaps most clearly seen when we consider that ordinary illocutions exist along a spectrum from the minimally institutional to the fully bureaucratic and judicial […]. The institutional context is present throughout the spectrum of illocutions; it is just that it embraces a more or less universal class of speakers and a more basic set of speech acts at the minimally institutional end. In speaking of the “illocutionary authority” of the speaker, and the normative power to constitute her utterance as having some illocutionary status, there should be no suggestion that this is somehow a kind of unilateral authority, or that the various illocutionary statuses in question are not, like other statuses generally, dependent for their very substance on being recognized by others. The speaker may be said to “confer” a certain illocutionary status on her utterance, but only against an institutional background that is not hers alone, and with the cooperation of the others in her speaking community. (Moran 2018, 142–143)

In this excerpt, Moran does seem to recognize the primacy of the social practices that condition the speech acts that a given speaker may perform, even though the speaker must ultimately remain in control of the speech act she performs, as she is supposed to be the only one who has authority over the real reasons for her action (and who, therefore, takes on a commitment through it). But, as he says, “an institutional background” is necessary to ensure that members of a linguistic community will understand and recognize what speech acts are performed and what their accomplishment normatively implies: a speaker can only take on a commitment through a speech act because there is an institutional background that supports the existence of the normative practice that this

68 

B. Ambroise

speech act is, and that guarantees that its execution produces a certain normative effect. It is therefore by relying on the specific characteristics of an assertive act—on its own rationale or logic—that Moran shows that telling is necessarily an illocutionary act that has normative effects insofar as it is a social act of the mind. In other words, he shows that this type of assertion is necessarily, given its own objective, regulated by social rules. I have the impression that this analysis of the act, which can be called “internalist,” in that it takes into account the speaker’s point of view and her relationship to what she says, is a useful complement to the more externalist analyses proposed by Sbisà. Of course, Moran’s solution, as attractive and rigorous as it may be, has to face two pitfalls. Moran should be careful 1. not to reduce the illocutionary act to the transmission and modification of beliefs—something that he is sometimes inadvertently guilty of in his analysis of telling21—otherwise he would ultimately fall back into offering a perlocutionary analysis of the act of telling and lose the objectivity of its illocutionary effects;22 2. not to endorse an overly individualistic conception of the social, otherwise he would reduce the social dimension to mere intersubjectivity. It is on this last point that I would like to return, developing remarks that could eventually concern Sbisà’s analysis.

 n Social Conditions and the Limits O of Context Plasticity Both Sbisà and Moran offer a conventionalist, or social, explanation of the efficacy of the illocutionary act, which gives a central role to the “securing of uptake” and allows us to understand that it produces, as an  This risk is also due to the fact that Moran (partially) borrows, in an uncritical manner, the analysis of uptake from Hornsby. 22  Rather, one must retain from Moran’s analysis the insistence that an illocutionary act is linked to a question of recognized objective status. 21

  On the Conventional Nature of Illocutionary Acts: Uptake… 

69

effect, norms. In this respect, they agree with the French sociologist and anthropologist Pierre Bourdieu, when he proposes to treat performative efficiency23 in terms of “symbolic power” (Bourdieu 1991). For him, symbolic power is precisely a power whose effectiveness lies in the (sometimes unconscious) recognition it obtains from those over whom it is exercised. And he proposes to analyze illocutionary efficiency as a specific modality of symbolic power, that of discourse, insofar as it is held by speakers who manage to have their hearers recognize the validity and efficiency of the speech acts they intend to produce. Now, according to him, this symbolic power of discourse is based on the authority, in a given context, of the speaker over his interlocutor—an authority which itself depends on the resources (material, institutional, educational, political, epistemic, and more generally symbolic24) available to the speaker at the time and in the context in which she uses her word. Therefore, symbolic power is not equally distributed and not all agents have it equally: the ability to produce illocutionary acts, if these are modalities of symbolic power, is not equally distributed among speakers. To put it another way: not all speakers have the same resources—what Bourdieu generically calls “capitals” —to succeed in performing illocutionary acts, that is, to ensure that the validity of their acts is recognized (to secure the uptake) and thus to perform effective speech acts. As Bourdieu puts it: The question of performative utterances becomes clearer if one sees it as a particular case of the effects of symbolic domination, which occurs in all linguistic exchanges. The linguistic relation of power is never defined solely by the relation between the linguistic competences present. And the weight of different agents depends on their symbolic capital, i.e., on the recognition, institutionalized or not, that they receive from a group. (Bourdieu 1991, 76)

 Bourdieu speaks of performative efficiency but he clearly has in mind any form of illocutionary efficiency. See De Fornel (1983) for a discussion. 24  According to Bourdieu, each “field” of activity has its own type of capital: money in the economic field, knowledge and integrity in the scientific field, artistic value in the artistic field, and so on. Each capital has its own value. Each capital holds its value according to the recognition it is given, given the functioning of each field. See, for more details, Bourdieu (2000). 23

70 

B. Ambroise

A performative utterance is destined to fail each time that it is not pronounced by a person who has the “power” to pronounce it, or, more generally, each time that “the particular persons and circumstances in a given case” are not “appropriate for the invocation of the particular procedure invoked”; in short each time that the speaker does not have the authority to emit the words he utters. […] the success of these operations of social magic […] is dependent on the combination of a systematic set of interdependent conditions which constitutes social rituals. (Bourdieu 1991, 111)

To sum up, there is no equality in the capacity to produce speech acts, because this capacity depends on conditions which are unevenly distributed in the social world.25 I will not elaborate further here.26 But I would like to take these analyses as a starting point to discuss an ambiguity in one of the hypotheses put forward by Sbisà, which seems to me to be too optimistic (from a moral and political point of view). Now, Sbisà knows and regularly uses sociological and anthropological work: her preferred author is rather Erving Goffman, who concentrates on micro-sociology, but I am inclined to think that she would not disagree profoundly with Bourdieu’s hypotheses in general. Indeed, she is also fully aware of power and domination issues in language,27 unlike most philosophers of language in the analytic tradition.28 But, at times, her understanding of the effectiveness of illocutionary acts as capable of constructing or modifying what she calls “the objective context”—or at least some of her formulations of it—seems to overlook certain objective dimensions of the social context, which allow or prevent these modifications from taking place—that is to say, which endow (or not) the speaker  The same type of remarks was sometimes developed in philosophy, notably in the important work of Rae Langton. See Langton (1993). 26  For more details, see Ambroise (2016). 27  I would even add that it is Sbisà’s particular understanding of the efficiency of language, and thus the particular conceptual framework she constructs, that allows her to be sensitive to these immediately practical aspects of language use. So there are direct moral and political implications of her work on language. 28  Indeed, for a long time, traditional analytical philosophy of language has tended to ignore issues of power and domination in language. There is now a growing trend in contemporary (analytic) philosophy of language that is sensitive to those issues. See, for instance, Langton (1993), Fricker (2007), Lance and Kukla (2009), McGowan (2019), and Stanley (2015). 25

  On the Conventional Nature of Illocutionary Acts: Uptake… 

71

with the capacity to succeed in his illocutionary acts, or to obtain the requested recognition. For example, Sbisà says, in a discussion of the given or constructed nature of the context of the speech act (based on well-known works by linguistic anthropologists such as Alessandro Duranti and Charles Goodwin): I would like to suggest that […] 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 interlocutor’s 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 […]. 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 redescribed […]. But if the satisfaction of felicity conditions and presuppositions need not be taken in 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. (Sbisà 2002, 425)

I must confess that I find it difficult to understand how that conclusion follows from what precedes it. Surely Sbisà’s aim here is to conceive the dynamics of the success of a speech act, since this success depends on the recognition (the uptake) that the hearer grants it. And she is certainly right to think that, by default, the participants in the exchange consider the conditions of felicity fulfilled, either by ignorance or by impossibility to verify that it is indeed the case. But it does not follow that “the context […] is set up by its participants” or, to put it otherwise, that the securing of uptake would be sufficient to ensure the success of the act, for at least two reasons:

72 

B. Ambroise

1. Firstly, as Sbisà herself points out, if the conditions of felicity are not fulfilled, no matter what the participants think (or don’t know) about them, then the speech act cannot be performed. In other words, and to use Sbisà’s words, if the appropriate context is not given, then the act fails. This conclusion seems to me to derive from the objectivity of the context, that is, from the fact that it is not (purely) cognitive and that it cannot be changed by a simple movement of the will29—except in rare and exceptional cases where the speaker has absolute power (if she is a God or a Dictator maybe). 2. Secondly, as Bourdieu reminds us, the recognition granted to the speaker and necessary for her to succeed in her speech act is itself socially conditioned by the context in which the act is performed: not everyone will obtain this recognition, because not everyone necessarily has, not only the competence, but also the social authority to succeed in such an act. In other words, the securing of uptake is itself conditioned by the contextual, and particularly social, conditions in which the act is performed.30 Therefore, it is not clear to me how the context could be set up by the participants of an interaction, at least if this interaction takes place in the real world.

4 Conclusion Sbisà’s work makes a very convincing case for the idea that illocutionary efficiency is conventional and can only be conventional. It also explains why the securing of uptake plays such an important role: since the illocutionary act aims at producing conventional effects, these can only occur if the participants in the discursive exchange recognize both that the  Or, to put it in the form of a criticism of idealism: it is not because we think that the conditions are fulfilled that they become so. However, see Langton (2015), for the opposite idea according to which speaker authority is a felicity condition that can spring into existence thanks to the audience’s acceptance that the speaker has authority. This important and controversial issue would deserve a full treatment, which I cannot offer in the space of this chapter. 30  Again, see Langton (1993). 29

  On the Conventional Nature of Illocutionary Acts: Uptake… 

73

utterance is correctly used as an act with these effects and that these effects must follow from this act and be imposed as normative consequences on all participants in the exchange. We have also seen how the very logic of assertive speech acts, which, according to Moran, involves a commitment from the speaker, requires analysis in conventional terms. In this respect, Moran’s work seems to converge with Sbisà’s. In so doing, these works re-inscribes speech acts in the social world and explain how they, and their efficiency, are derived from it. As such, and as Sbisà explains on several occasions, this approach, which seems to me to be conceptually correct, should be decisive for “applied pragmatics” and, more generally, the sociology of language and linguistic anthropology. Provided, however, that we never neglect the conditions of felicity, social or otherwise, of speech acts which determine their recognition and therefore their effectiveness.

References Ambroise, Bruno. 2013. Promising. In Pragmatics of Speech Actions, ed. Marina Sbisà and Ken Turner, 477–498. Berlin: De Gruyter. ———. 2016. Efficacité du discours et rapports de force. In Langage et politique, ed. Bruno Ambroise and Bertrand Geay, 195–212. Lille: Presses Universitaires du Septentrion. Anscombe, G. Elizabeth M. 1957. Intention. Oxford: Basil Blackwell. Austin, John L. 1962. Sense and Sensibilia. Oxford: Oxford University Press. ———. 1975. In How to Do Things with Words, ed. James O.  Urmson and Marina Sbisà, 2nd ed. Cambridge, Ma: Harvard University Press. ———. 1979. Philosophical Papers. 3rd ed. Oxford: Oxford University Press. Bach, Kent., and Robert M. Harnish. 1979. Linguistic communication and speech acts. Cambridge: M.I.T. Press. Bourdieu, Pierre. 1991. Language and Symbolic Power. Cambridge: Polity Press. ———. 2000. Pascalian Meditations. Stanford: Stanford University Press. Brandom, Robert. 1983. Asserting. Nous 17 (4): 637–650. Caponetto, Laura. 2020. Undoing Things with Words. Synthese 197: 2399–2414. De Fornel, Michel. 1983. Légitimité et actes de langage. Actes de la recherche en sciences sociales 46 (1): 31–38. Fricker, Miranda. 2007. Epistemic Injustice. Oxford: Oxford University Press.

74 

B. Ambroise

Goffman, Erving. 1971. Relations in Public. New York: Basic Books. Grice, Paul H. 1989. Studies in the Ways of Words. Cambridge: Harvard University Press. Hornsby, Jennifer. 1994. Illocution and Its Significance. In Foundations of Speech Act Theory: Philosophical and Linguistic Perspectives, ed. Savas Tsohatzidis, 187–207. London: Routledge. Hornsby, Jennifer, and Rae Langton. 1998. Free Speech and Illocution. Legal Theory 4: 21–37. Lance, Mark, and Rebecca Kukla. 2009. Yo and Lo! The Pragmatic Topography of the Space of Reasons. Cambridge: Harvard University Press. Langton, Rae. 1993. Speech Acts and Unspeakable Acts. Philosophy & Public Affairs 22 (4): 293–330. ———. 2015. How to Get a Norm from a Speech Act. The Amherst Lecture in Philosophy 10: 1–33. Longworth, Guy. 2019. Illocution and Understanding. Inquiry. https://doi. org/10.1080/0020174X.2019.1667869. McGowan, Mary Kate. 2019. Just Words. Oxford: Oxford University Press. Moran, Richard. 2018. The Exchange of Words. Oxford: Oxford University Press. Sbisà, Marina. 2002. Speech Acts in Context. Language and Communication 22: 421–436. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2006. Speech Acts without Propositions? Grazer Philosophische Studien 96: 155–178. ———. 2007. How to Read Austin. Pragmatics 17: 461–473. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2009. Uptake and Conventionality in Illocution. Lodz Papers in Pragmatics 5: 33–52. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2013. Locution, Illocution, Perlocution. In Pragmatics of Speech Actions, ed. Marina Sbisà and Ken Turner, 25–75. Berlin: De Gruyter. Searle, John R. 1969. Speech Acts. An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. Stanley, Jason. 2015. How Propaganda Works. Princeton: Princeton University Press. Strawson, Peter. 1954. Intention and Convention in Speech Acts. The Philosophical Review 73 (4): 439–460.

Varieties of Uptake Claudia Bianchi

1 Introduction The debate about the determination of the illocutionary force of a speech act revolves around the notion of uptake and, more specifically, around the role played by the audience: many scholars consider the hearer’s recognition of the meaning and the force of the locution a necessary condition for the successful performance of an illocutionary act. A variety of theories has been put forward. According to one theory, the hearer’s uptake determines whether a successful act has been performed (Langton 1993, 2018). According to an alternative account, the hearer’s uptake not only determines whether a successful act has been performed, but also constitutes the nature of the act performed (Kukla 2014). According to a third account, the illocutionary force of a speech act is the result of a process of negotiation between the hearer and the speaker (McDonald 2021, 2022). C. Bianchi (*) Faculty of Philosophy, Vita-Salute San Raffaele University, Milan, Italy e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Caponetto, P. Labinaz (eds.), Sbisà on Speech as Action, Philosophers in Depth, https://doi.org/10.1007/978-3-031-22528-4_4

75

76 

C. Bianchi

Drawing on the theoretical tools provided by Marina Sbisà’s work, in this chapter, I will show how the theories giving the audience a central role in fixing the illocutionary force of a speech act fall short. The topic proves relevant not only for theoretical reasons, but also for social and political ones. In particular, it has a bearing on debates about pragmatic phenomena of illocutionary distortion such as discursive injustice and silencing, where marginalized speakers have trouble performing particular speech acts they are entitled to perform. I will argue that focusing on the hearer’s role contributes to hinder, rather than improve, the performative potential of members of underprivileged groups, and to undermine their agency. The structure of this chapter is as follows: In Sect. 2, I introduce uptake and distinguish two readings of the notion: a speaker-dependent and a hearer-dependent reading. In Sect. 3, I focus on the hearer-dependent reading and examine three different accounts, put forward by Rae Langton, Quill Kukla, and Lucy McDonald. In Sect. 4, I underline the shortcomings of the theories favoring a hearer-dependent reading of the notion of uptake. In Sect. 5, I track down the misunderstanding these theories have in common: they tend to focus on the communicative and cognitive dimension of (ordinary) speech acts, and tend to overlook their performative and normative dimension. In Sect. 6, I conclude by briefly sketching an alternative account, based on a speaker-dependent reading of the notion of uptake, and endorsing a normative perspective—where uptake concerns the way in which the utterance should be taken, considering the procedure invoked by the speaker or the speaker’s intentions, when adequately made manifest and public.

2 What Is Uptake? As is well known, Austin claims that the performance of an illocutionary act “involves the securing of uptake,” and defines uptake as “the understanding of the meaning and of the force of the locution” (Austin 1975,

  Varieties of Uptake 

77

116–117). Here I will focus exclusively on the determination of the force of a locution, leaving the analysis of the determination of its content for another time. In traditional speech act theory, the understanding of the force of the locution is usually spelled out in two different ways, traced back, respectively, to Austin (1975) and Strawson (1964): (1) the recognition of the convention invoked by the speaker; (2) the recognition of the illocutionary intentions of the speaker. For the purposes of this chapter, I will remain neutral between a conventionalist and an intentionalist perspective: nothing in the following discussion hangs on this difference. I take the expression “securing of uptake” as having two readings: (a) A speaker-dependent reading: the securing of uptake is an effect depending on the speaker. It depends on the speaker’s effort to produce the hearer’s recognition of the force of her locution—may this be either by invoking a suitable procedure in suitable circumstances (conventionalism) or by making her communicative intentions public and available to the audience (intentionalism); (b) A hearer-dependent reading: the securing of uptake is an effect depending on the audience. It depends on the actual recognition by the hearer of the force of the speaker’s locution—may this be either the actual recognition of the procedure invoked by the speaker or the actual recognition of the speaker’s communicative intentions. Both Austin’s and Strawson’s theories are compatible with a speaker-­ dependent reading. In the Austinian framework, the securing of uptake is an effect brought about by the speaker in issuing the utterance: “unless a certain effect is achieved, the illocutionary act will not have been happily, successfully performed […] Generally the effect amounts to bringing about the understanding of the meaning and of the force of the locution” (Austin 1975, 116–117, emphasis mine). True, Austin is not completely explicit about whether what is necessary for the successful performance of a speech act is the speaker’s reasonable effort to produce the hearer’s recognition or the actual recognition by the hearer of the force of the act. Nonetheless, as Marina Sbisà (2013, 32) remarks,

78 

C. Bianchi

One position that may plausibly be attributed to Austin is that uptake is secured when the speaker manages to make it possible for the audience to understand. This indeed is already an achievement […] and therefore an effect brought about by the speaker in issuing the utterance.

A speaker-dependent reading is also in line with Strawson’s suggestions. In the Strawsonian framework, the hearer’s actual recognition of the speaker’s intentions is not a necessary condition for the successful performance of an illocutionary act. Indeed, Strawson suggests that it is the aim and not the achievement of securing uptake that is key for the performance of an illocutionary act.1 The speaker’s illocutionary intentions need not actually be recognized by the audience, but merely made explicit and public; they must be overt, manifest and “avowable” (Strawson 1964, 454). Elsewhere I have argued in favor of a speaker-directed reading and developed an account reconciling convention-based and intention-based theories (Bianchi 2021). Here I will focus on the hearer-dependent reading—a reading generally favored in the literature on silencing and discursive injustice.2 In the next section, I will present three different accounts giving the hearer a central role in determining the illocutionary force of a speech act. I will get back to the speaker-directed reading in Sect. 6.

 Cf. Strawson (1964, 449): “the aim, if not the achievement, of securing uptake is essentially a standard, if not an invariable, element in the performance of the illocutionary act.” 2  There are some exceptions, maintaining that an illocutionary act can be performed without the achievement of uptake in the hearer-dependent reading. See, for example, Bird (2002, 3): “uptake is not necessary for illocution in general, nor […] is it necessary for refusal in particular”; or McGowan (2017, 45): “I do not regard uptake as necessary for illocution, but it is necessary for communication”; Maitra (2009, 313, fn. 7), has a weaker position: “even if it is right, contra Austin, that an illocutionary act can be performed without uptake, there is clearly a sense in which that act is less than fully successful – or happy.” The hearer-dependent reading is usually attributed to Austin: see Bianchi (2021) for criticism of this attribution. 1

  Varieties of Uptake 

79

3 Securing of Uptake: Hearer-Dependent Reading Some authors (Rae Langton, Quill Kukla, and Lucy McDonald among the others) consider the hearer’s uptake—namely the actual recognition of the force of the locution by the hearer—a necessary condition for the successful performance of a speech act. Unless the hearer recognizes the procedure invoked by the speaker, or her communicative intention, a certain speech act is not successfully performed (therefore its conventional effects in terms of rights, duties, obligations, and commitments are not brought about).3 Three different accounts of the hearer’s role have been suggested:4 (A). Ratification Theory: the hearer’s uptake ratifies the speaker’s act: the audience has the power to determine whether the act is successfully performed or not. (B). Constitution Theory: the hearer’s uptake constitutes the speaker’s act: the audience has the power to determine not only whether the act is successfully performed, but also the nature of the illocution performed. (C). Collaboration Theory: the illocutionary force of an act is the result of a negotiation between the speaker and the hearer—where the force the speaker and the hearer agree upon may reflect neither the speaker’s illocutionary intention (or the procedure invoked by her) nor the hearer’s original recognition. Let us look at the three accounts more closely.

Ratification Theory Suppose that a speaker S attempts to perform speech act A. According to the Ratification Theory, there are two cases.  On this point, see Sbisà (2013).  I borrow McDonald’s labels: McDonald (2021, 3511) and (2022).

3 4

80 

C. Bianchi

(i). The hearer H recognizes the force of the S’s act (the procedure invoked by S, or her communicative intention), and thereby ratifies her act as an illocutionary act A: S performs successfully the illocutionary act A. (ii). The hearer H fails to recognize the force of the S’s act (the procedure invoked by S, or her communicative intention), and thereby fails to ratify her speech act: S fails to perform the illocutionary act A and, for that matter, fails to perform any illocutionary act at all. A version of the Ratification theory is held by Rae Langton. According to Langton (1993), there are contexts in which particular illocutionary acts performed by women are unsuccessful because the hearer fails to recognize their illocutionary force. This is the case, for instance, of acts of refusal of sexual advances: a man imbued with false beliefs about women’s sexuality may fail to recognize (hence fail to ratify) a woman’s refusal. Since there is no ratification, the woman’s act does not succeed and she is silenced. The H’s actual uptake makes it the case that the woman’s locution counts as no speech act at all.

Constitution Theory Suppose again that S attempts to perform speech act A. According to the Constitution Theory, there are two cases: (i). H may recognize the procedure invoked by S, or her communicative intention, and thereby contribute to constitute it as an act A.5 (ii). H may fail to recognize the procedure invoked by S, or her communicative intention, and recognize another procedure, or another communicative intention, thereby constituting S’s utterance as a different speech act, an act B.6  In Kukla’s words, the hearer may “help to finish” it as an act A (Kukla 2014, 444).  There is indeed a third option: (iii) H may fail to recognize the procedure invoked by S, or her communicative intention, and actually fail to recognize any procedure, or communicative intention, at all. In this case, S’s illocutionary act fails and counts as no speech act at all. I thank a reviewer for pointing out this third option. 5 6

  Varieties of Uptake 

81

The H’s uptake of the S’s utterance determines not only whether the S’s illocutionary act succeeds or not, but also the nature of the illocution performed by S; this determination takes place even when the hearer’s interpretation is at odds with the speaker’s intention (or with the procedure invoked by the speaker). A version of the Constitution Theory is held by Kukla (2014). Quill Kukla (writing as Rebecca Kukla) claims that hearers can contribute to the nature of illocutionary acts and determine not only whether an act is successfully performed, but also which act is performed. In Kukla’s example, Celia is a floor manager at a factory where most of the workers are males: she issues orders but because of her gender, her orders are taken by her male subordinates as requests. The hearer’s uptake of Celia’s utterance as a request constitutes it as a request: the audience’s actual uptake makes it the case that Celia’s locution counts as a different act from the one she intended to perform (see Kukla 2014, 446).

Collaboration Theory Suppose that S attempts to perform speech act A.  According to the Collaboration Theory, the force of an illocution results from an agreement reached by the speaker S and the hearer H. The nature of the speech act is determined by a negotiation engaged between S and H. S’s utterance has illocutionary force A iff both (a) and (b) occur: (a) H communicates to S that he interprets S’s utterance as having force A; (b) S then communicates to H that she accepts H’s interpretation of the utterance as having force A (see McDonald 2022, 924). Actually, S may simply continue the conversation, which signals that she accepts H’s interpretation. Alternatively, when H’s interpretation of S’s utterance does not match her original intentions, S may choose to correct or to veto H’s interpretation. A version of the Collaboration Theory is held by McDonald (2022). Take the Coffee Order scenario: Alex and Zoe are colleagues with equal

82 

C. Bianchi

rank, but Alex attempts to order Zoe to bring her coffee. Zoe says, “Sorry, I can’t right now!” communicating to Alex that she interprets her utterance as having the force of a request. Alex might decide to reply with something like “That’s okay,” signaling that Zoe’s interpretation is acceptable to her; or else she might correct or veto Zoe’s interpretation: “if Alex rejects Zoe’s interpretation of the utterance as a request, then it cannot be a request; this possibility gets ruled out. Therefore while the collaboration theory renders communication a more collaborative affair than on the ratification theory, speakers still retain a ‘veto’, in line with the intuition that speakers have more authority over the force of their utterances than hearers” (McDonald 2022, 927). In the next section, I will show how the theories giving the audience a central role in fixing the illocutionary force of a speech act fall short. Furthermore, I will argue that focusing on the hearer’s role contributes to hinder, rather than improve, the illocutionary potential of members of underprivileged groups, and to undermine their agency.

4 Challenges and Objections Scope A first group of objections concerns the scope of the models giving a hearer-dependent reading of the “securing of uptake”: the three accounts seem plausible only in more or less idealized conditions: 1. The necessity of H’s uptake for the performance of a speech act is more or less explicitly restricted to ordinary speech acts. Institutional acts (or “essentially conventional speech acts”: Strawson 1964, 456), like marrying, surrendering, or passing sentence, are generally ruled out.7 Alexander Bird raises this issue in Bird (2002): when a judge passes sentence, her words may well be addressed to the prisoner, but sentence is passed whether or not the prisoner realizes it. According to  See, for example, McDonald (2021, 3508, fn. 4), who acknowledges that, unlike the speech acts of ordinary conversation, institutional acts do not require uptake. 7

  Varieties of Uptake 

83

Bird, no one has to realize that sentence has been passed “so long as the judge performs [her] duty in accordance with the law and established procedures” (Bird 2002, 7). Someone has to realize that sentence has been passed for the sentence to be carried out, but carrying out a sentence amounts to a perlocutionary, not illocutionary, effect. Actual uptake of the actual hearer does not seem necessary for the successful performance of institutional illocutionary acts. 2. Another issue may be raised concerning ordinary speech acts, whenever the conversational setting involves more than one hearer. Different hearers may interpret the same utterance in different ways. Whose uptake matters for the successful performance of a speech act? Whose uptake should be secured? The three models are bound to postulate that multiple hearers interpret the speaker’s utterance in the same way.8 This limitation is seen as an innocent idealization, but is actually at odds with a vast array of conversational contexts: written utterances, radio, television, and video utterances that may be read or heard by millions of people and interpreted in a variety of different ways. On whose uptake does the felicity of these speech acts depend? On whose ratification or constitution? The Collaboration Theory provides no improvement here. According to the Collaboration Theory, the speaker has to negotiate the force of her act with the audience, gain their ratification and communicate that she accepts it; but it is unclear whom this negotiation should be engaged with. 3. The three models are bound to a further idealization of the conversational setting: intended addressee and actual hearer must coincide. However, there are contexts in which an individual can overhear an utterance addressed to someone else, again opening the possibility of different understandings of the same utterance, and different attributions of illocutionary force (see McDonald 2021, 3508, fn. 6). Moreover, suppose that S attempts to perform a speech act and no one is in the position to give an uptake, or an appropriate uptake, as when  This assumption is explicitly spelled out by McDonald (2021, 3508, fn. 6). See Casey Johnson’s essay in this volume for a pluralist perspective on illocutionary force (Johnson 2023); on illocutionary pluralism see also Lewiński (2021). 8

84 

C. Bianchi

someone promises something to a baby. Is this promise taking effect and changing S’s normative status, by putting an obligation on her? Also, imagine that someone overhears the promise. Is this in some way affecting the performance of S’s act? Similar remarks are in order when discussing other cases of lack of uptake or delayed uptake. Imagine that S and H are talking over the phone: S promises to come to H’s party, but the phone call ends abruptly—leaving no time for H to ratify S’s speech act, to constitute it, or to negotiate its force with her. Or imagine that S promises to come to H’s party by letter, but the letter is delivered after the party, or gets lost in the mail. What is at stake is the normative dimension of the speech act: from a normative point of view, I claim, the speaker assumes an obligation even if there is no uptake by the hearer, or no uptake at the intended or expected time. I will return to this point in Sect. 5.

The Power of the Hearer Another group of objections concerns the power of the hearer. Even when dealing with acts performed in idealized conditions (namely ordinary speech acts, involving one single hearer who is also the intended addressee), a hearer-dependent reading runs the risk of conferring an excessive power to the hearer. In fact, the three models end up turning many utterances that we intuitively regard as successful illocutions into failures, or distorted acts. Indeed, according to the Ratification Theory, if the hearer’s uptake is not achieved, or does not correspond to the speaker’s intentions (or procedures), the illocutionary act fails, and its conventional effects (rights, duties, obligations, and commitments) are not brought about. In Langton’s example, the uptake actually secured by the woman intending to refuse makes it the case that her locution counts as no speech act at all. There is no ratification, hence her speech act is unsuccessful. According to the Constitution Theory, if H’s uptake does not correspond to S’s intentions (or procedures), the act is constituted according to H’s recognition, as a different speech act from the one intended by the speaker. In

  Varieties of Uptake 

85

Kukla’s example, the uptake actually secured by Celia makes it the case that her locution counts as a request, and not as an order: its alternative uptake constitutes it as a request. Both the Ratification Theory and the Constitution Theory entail that some speakers, and especially marginalized speakers, are deprived of agency, autonomy, and authority in their conversational practices. Victims of silencing or discursive injustice fail to perform speech acts they are entitled to perform: they invoke the appropriate procedures, entertain the appropriate intentions, have the required authority, but their acts are not ratified or constituted as the acts they intended to perform—and this is only because their addressees are inattentive, incompetent, or imbued with prejudices. In Daniel Jacobson’s words, the two models have the unwelcome consequence of holding the performance of an illocutionary act “hostage to the perversity of one’s audience” (Jacobson 1995, 73–74).9 There is indeed an important difference between the Ratification and the Constitution Theory as long as sexual refusals are concerned: for the Constitution Theory, but not for the Ratification Theory, the audience’s actual uptake may be sufficient to turn refusal into consent.10 Yet, for the Ratification Theory H’s actual uptake may be sufficient to turn a refusal into nothing at all. The same kind of objections can be extended to the Collaboration Theory. According to the Collaboration Theory, if there is no agreement between S and H, the act fails. In the sexual refusal case, the disagreement between the man and the woman is sufficient to turn her refusal into nothing at all. The act fails also in cases of the speaker’s veto: speakers can veto interpretations of their utterances when these interpretations do not match their original intentions. This is in line with the intuition that speakers have more authority over the force of their utterances than hearers: however, S has the power only to block mistaken interpretations, not  Hornsby and Langton (1998) is a reply to Jacobson’s criticism.  Cf. Langton (2018, 151, fn. 44): “There are limits to what hearers can do, in my view. A hearer may weaken what would have been an order into a mere request, if an order requires a certain hearer-dependent felicity condition. But a hearer cannot e.g. twist sexual refusal into sexual consent, since consent requires a certain speaker-dependent felicity condition—the speaker’s decision or intention to consent.” On Langton’s proposal, see Bianchi (2021). 9

10

86 

C. Bianchi

to actually impose her interpretation, and perform the act she intends to perform. More generally, a negotiation as the one envisaged by McDonald tends to be governed by the more powerful. The same conversational dynamics leading to discursive injustice apply to the possibility to ratify, reject, or negotiate interpretations. Again, we can endorse a collaboration model only by further restricting it to conversational settings displaying no particular asymmetries between speakers and hearers. However, as McDonald herself acknowledges, “situations in which all interlocutors participate freely seem few and far between, due to widespread social power asymmetries along many axes” (McDonald 2022, 936): this has the implausible consequence of denying performativity to a vast amount of utterances, whenever speakers belonging to marginalized groups are involved.11 Another potential worry concerns not only sincere misunderstandings but also instrumental misinterpretations.

Deliberate Misinterpretation According to the three models, a speaker may fail to perform a speech act not only because her addressee is absent-minded, or lacking competence or else biased, but also because her addressee is instrumentally misinterpreting her utterance.12 Neither Langton nor Kukla addresses this case, which represents a potential amplification of the injustice suffered by individuals belonging to oppressed groups. Imagine that, in Kukla’s scenario, Celia’s subordinates recognize her intention to perform an order, yet instrumentally pretend to interpret it as a request and respond to it accordingly, thereby making it a request: their deliberate misunderstanding turns Celia’s attempted act A into a different act, an act B.  Or imagine that, in  At least unless “certain social supports are in place in the context of utterance”: cf. McDonald (2022, 936): “Quill Kukla has recently suggested, for example, that in situations of compromised autonomy, agents can still perform the illocutionary act of consenting, provided their ability to do so is ‘socially and interpersonally scaffolded’ (Kukla 2020). We could extend this argument to all illocutionary acts; it may be that agents who are not fully autonomous can still perform illocutionary acts, provided certain social supports are in place in the context of utterance.” 12  This worry is raised by McDonald herself (McDonald 2021, 3512). 11

  Varieties of Uptake 

87

Langton’s scenario, the man merely pretends to interpret the woman’s refusal as a form of teasing or seducing. The man’s instrumental misinterpretation cannot turn a refusal into a consent, but can still turn the woman’s act into nothing at all, just silence. This problem is further amplified by the Collaboration Theory, where instrumental misinterpretations may indeed concern not only H’s uptake, but also S’s ratification (or her veto) upon H’s interpretation. Indeed, speakers appearing to have performed harmful speech acts may simply reject the audience’s interpretation of their utterances as harmful—even when this interpretation is perfectly reasonable. In this way, hate speakers can unfairly reject responsibility for their objectionable speech acts. A mirror objection may be raised concerning bad faith agreements. Suppose that S makes a racist remark, and H unintentionally gives a non-­ racist interpretation, more charitable than the one intended by the speaker; and suppose that S pretends it was exactly what she meant, or simply continues the conversation, which signals that she accepts H’s interpretation. According to the Collaboration Theory, S would end up performing the more charitable speech act. A speaker may pretend to agree with H’s “ameliorative” interpretation of her utterance, even if this interpretation does not match her original intentions, nor the procedure she invoked—ending up performing a “better” speech act than her intended one. Instrumental misinterpretations (as far as the hearer’s uptake is concerned) may occur also as deliberate and disingenuous misinterpretations of S’s conversational move, taking the form of what Laura Caponetto and Bianca Cepollaro call “bending.” To “bend,” according to them, is to deliberately give a distorted uptake to a speaker’s illocutionary move— precisely, an ameliorative uptake, which turns that move into a contribution whose potential for harm is much lesser (Caponetto and Cepollaro 2022). Suppose that S makes a racist remark, and H intentionally gives a non-racist interpretation; and suppose that S does not veto H’s interpretation, but pretends it was exactly what she meant, or simply continues the conversation, choosing not to correct H’s interpretation. According to the Collaboration Theory, S ends up performing the more charitable speech act, even if this interpretation does not match her original intentions, nor the procedure she invoked. Note that, while illocutionary

88 

C. Bianchi

distortions (sincere or insincere that might be) undermine the speaker’s agency and autonomy, bending can be employed to counter hate speech and its oppressive power. Yet, are we prepared to maintain that S performed a non-racist speech act in the first place? Alternatively, and perhaps more precisely, we could argue that the speaker performed a racist speech act and then retracted her original speech act.13

5 Communicative Versus Normative Dimension In this section, I will track down what I take as the key misunderstanding the three accounts have in common. The models giving the hearer a central role in determining the illocutionary force of a speech act tend to focus on the communicative and cognitive dimension of speech acts, and to neglect their performative and normative dimension—actually the element of novelty in Austin’s work. The emphasis on illocutionary acts as forms of communication, rather than actions, and the divide between institutional and ordinary speech acts are symptoms of the tendency to revive the distinction between acts that do things and acts that say things—the very distinction Austin was trying to eradicate.14 According to Austin, illocutions are acts bringing about effects on social states of the world: they assign or withdraw rights, obligations, commitments, and may establish new social states of affairs.15 While this is  On retractions, see Caponetto (2020).  Cf. Sbisà (2009a, 42): “I surmise that behind the divide between conventional and intention-­ based speech acts there lies the contrast between saying and doing, reborn within the very context of speech act theory.” According to Sbisà, another point reviving the distinction between doing and saying is the introduction of propositions in speech act theory: “Assuming that there are propositions and that language expresses them independently of its expressing force(s) as well, a speech act may very naturally be described as an act of expressing a proposition with a certain communicative intention applying to it. Speech act successfulness may very naturally be identified with the recognition of the speaker’s communicative intention on the part of the hearer, which is a merely cognitive effect (in particular, a matter of acquiring a certain belief about the speaker). As a result, the communication that the speaker is performing an action is foregrounded at the expense of the very action performed” (Sbisà 2006, 162). 15  See Austin (1975, 117): “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.” 13 14

  Varieties of Uptake 

89

generally acknowledged as far as institutional acts are concerned, in a hearer-dependent perspective the effects of ordinary speech acts tend to be regarded as merely cognitive, as changes in the beliefs of the interlocutors: their main illocutionary effect is taken to be the production of the hearer’s uptake.16 Therefore, the recognition of the meaning and force of the speaker’s utterance on the part of the hearer is the condition of the success of the illocution.17 Take for instance how McDonald characterizes the speech act of promising: “It seems that part of what it means to promise is to successfully communicate. When we say that someone made a promise or an assertion, what we are tracking is the fact that the speaker, by transmitting information to us, made us believe that our normative statuses and their normative statuses had changed in a distinctive way” (McDonald 2022, 922).18 According to McDonald, the main effect of the illocutionary act of promising is a cognitive one, amounting to the recognition of the speaker’s communicative intention, along with the belief that the normative setting has been changed. However, there is an important difference between a change in our normative status and believing that there is a change in our normative status. When S utters, (1) I will fix your car in suitable circumstances, she puts herself in a state of obligation to fix H’s car; moreover, H gains an entitlement to S’s fixing his car, no matter what H himself believes (even if, e.g., H thinks that S is joking). The emphasis on the communicative dimension at the expenses of the normative one is especially harmful as far as cases of discursive injustice or silencing are concerned, such as refusals of sexual advances. By uttering,  See Strawson (1964) and Searle (1969), who defines the “illocutionary effect” as “the hearer understanding the utterance of the speaker” (Searle 1969, 47). For a critical assessment, see Sbisà (2007, Section 3). 17  This applies mostly to Langton’s and McDonald’s accounts, while Kukla has a theory explicitly conflating the illocutionary and the perlocutionary dimensions. 18  See also McDonald (2022, 921): “I take it most of us agree that illocutionary acts are a form of communication; when we make promises, assertions, and requests, we are communicating” (emphasis mine). 16

90 

C. Bianchi

(2) No in suitable circumstances, S puts the addressee in a state of obligation to stop any further advance: she assigns obligations of some kind, withholds permissions and the like—she changes the normative status of both S and H, no matter what H believes (and sometimes no matter what S herself believes).19 Of course, obligations and entitlements require sometimes a form of social recognition (especially if some disagreement occurs between S and H). However, this kind of recognition differs from H’s actual uptake in two respects. On the one hand, the participants involved in the recognition of the changes in the normative landscape may not include H—and, for that matter, may not include S; on the other hand, this kind of social uptake involves more than mere recognition of force of the speaker’s utterance. Let us go back to institutional acts. We said earlier that, at the end of a trial, sentence is passed whether or not the prisoner realizes it. Or take the act of surrendering on the battlefield: as long as the soldiers have thrown down their weapons, held up their hands and cried out “I surrender,” they have surrendered, even if the enemies do not recognize their act and shoot them.20 Actual uptake of the actual hearer does not seem necessary for the successful performance of institutional speech acts. However, someone in the relevant section of the social world has to realize that sentence has been passed for the sentence to be successfully performed. And someone in the relevant section of the social world has to recognize the procedure invoked by the soldiers as a procedure of surrender (say, a judge on a trial ruling on the war crime of shooting surrendering soldiers). In both cases, the relevant audience is not merely recognizing the act, but also validating and confirming it. Not only must they recognize that sentence has been passed, or that someone was surrendering on the battlefield, but also in some sense certify that sentence has been  As Sbisà (2009b, 355) correctly points out, refusals are not communications, and not even judgments (verdictives), but the exercising of powers and rights (exercitives). See also McGlynn (2019) for objections against Fricker’s epistemic account of silencing. 20  This example is Bird’s (2002). 19

  Varieties of Uptake 

91

correctly passed, or that someone was suitably surrendering. In other words, the relevant uptake is not the one provided by the hearer but a sort of social uptake—a form of ratification delivered by relevant participants and conferred more to the social practice as a whole than to the specific speech act. This point extends to ordinary speech acts and is related to Austin’s claim that illocutionary acts (both ordinary and institutional) are conventional. The conventionality of speech acts has traditionally been traced back to the conventionality of the means (or procedures) by which they are performed. Marina Sbisà has convincingly argued that their conventionality resides, primarily, in the conventional status of the effects they bring about: the effect of a speech act 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” (Sbisà 2007, 464). For sure, “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 … it is at least necessary that what the speaker does and says be socially accepted as having that effect” (Sbisà 2009a, 45). However, this social acceptance: (i). does not coincide with H’s uptake: rather, it involves the uptake of members of the relevant social group; (ii). does not amount to mere recognition: rather, it requires acceptance and validation. Not only a promise or a refusal must be recognized, but also in some sense certified: members of the relevant social group must validate them, namely establish that the promise was correctly performed, or that someone was suitably refusing; (iii). does not call for an agreement between the speaker and the hearer: rather, it could be at odds with what both the speaker and the hearer believe has happened. On this last point, take the case of a sexual refusal in a long-term relationship: it may well be that both the man and the woman do not recognize the “no” uttered by the woman as a binding refusal—because, say, both are imbued in sexist stereotypes such as “no act of sexual refusal is in

92 

C. Bianchi

order in long-term relationships.”21 The case may end up in court to be settled: the judges (the members of the relevant social group) would not only recognize the “no” as a speech act of refusal, but also validate it as a successful speech act of refusal, a binding refusal. Note that the institutional setting of this social validation doesn’t make the act of refusal issued by the woman an institutional speech act: the judges in court are deciding about the force and felicity of an ordinary speech act. The hearer-dependent reading assigns pride of place to actual uptake, which might be arbitrary or just wrong. We must, then, abandon it and endorse a speaker-dependent reading, and a normative perspective— where uptake is the way in which the utterance should be taken, considering the procedure invoked by the speaker or the speaker’s intentions, when adequately made manifest and public.

6 Conclusion: A Speaker-Dependent Reading In Sect. 2, we said that the securing of uptake has two readings—a speaker-dependent reading and a hearer-dependent reading—and that both Austin’s and Strawson’s theories are compatible with a speaker-­ dependent reading. In an Austinian, conventionalist, perspective, in order to secure uptake—and hence perform a particular speech act—the speaker has to invoke a suitable procedure in suitable circumstances, and resort to a complex array of discursive conventions. In a Strawsonian, intentionalist, perspective, in order to secure uptake—and hence perform a particular speech act—the speaker must make her communicative intentions public and available to the hearer. Following Austin’s and Strawson’s analyses, in Bianchi (2021) I have endorsed a speaker-dependent reading (that I have called weak intentionalism), where the illocutionary intentions of the speaker need not actually be recognized by the audience, but merely made available to the audience  About 24% of people in the UK think that sex without consent in long-term relationships isn’t rape (despite laws against rape in marriage being in place since 1991): see the End Violence Against Women report at https://yougov.co.uk/topics/resources/articles-reports/2018/12/01/publics-­ attitudes-­sexual-consent (last accessed 27/05/22). 21

  Varieties of Uptake 

93

by the speaker. To successfully perform a promise, a refusal or an order, the speaker must secure uptake in a weaker sense, meaning that she must put her audience in a position to recognize her illocutionary intention or the procedure she is invoking. To this end, the speaker can exploit any feature of the context of performance, the content of the locution, syntactic and prosodic devices, gestures, tone of voice and so on. In this sense, securing uptake is a necessary condition for the felicitous performance of an illocutionary act. My weak intentionalist account involves a normative component: an intention—in order to be illocutionary relevant—must be something that a (normal) addressee in normal circumstances is able to work out using conventional means and contextual information—ultimately reconciling convention-based and intention-based theories. This is in line with Sbisà’s notion of “reasonably secured” uptake: “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” (Sbisà 2009a, 48).22 S succeeds in performing a certain speech act if she has arranged things in such a way that competent, attentive, and unprejudiced hearers can reasonably be expected to recognize her illocutionary intention or the procedure she is invoking. Even if a “perverse” hearer (to say it with Jacobson) fails to recognize the force of S’s locution, or takes it to have a different force than the one S intended, or else fails to agree with S on a certain force, S still counts as having performed the speech act she  Cf. Sbisà (2009a 48–49): “I am particularly interested in the hypothesis that the understanding of illocutionary force may be based on pattern recognition … Patterns may be cognitively processed in different ways … 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.” 22

94 

C. Bianchi

intended and was entitled to perform. A crucial outcome of speaker-­ dependent proposals, allowing to grant individuals belonging to marginalized groups their illocutionary potential, hence their agency and autonomy.23

References Austin, John L. 1975. In How to Do Things with Words, ed. James O. Urmson and Marina Sbisà, 2nd ed. Cambridge, Ma: Harvard University Press. Bianchi, Claudia. 2021. Discursive Injustice: The Role of Uptake. Topoi 40: 181–190. Bird, Alexander. 2002. Illocutionary Silencing. Pacific Philosophical Quarterly 83: 1–15. Caponetto, Laura. 2020. Undoing Things with Words. Synthese 197 (6): 2399–2414. Caponetto, Laura, and Bianca Cepollaro. 2022. Bending as Counterspeech. Ethical Theory and Moral Practice.  https://doi.org/10.1007/ s10677-022-10334-4. Hornsby, Jennifer, and Rae Langton. 1998. Free Speech and Illocution. Legal Theory 4: 21–37. Johnson, Casey. 2023. Some Varieties of Illocutionary Pluralism. In Sbisà on Speech as Action, ed. Laura Caponetto, Paolo Labinaz. London: Palgrave Macmillan. Jacobson, Daniel. 1995. Freedom of Speech Acts? A Response to Langton. Philosophy and Public Affairs 24: 65–79. Kukla, Quill R. 2014. Performative Force, Convention, and Discursive Injustice. Hypatia 29 (2): 440–457. ———. 2020. A Nonideal Theory of Sexual Consent. Ethics 131 (2): 270–292. Langton, Rae. 1993. Speech Acts and Unspeakable Acts. Philosophy and Public Affairs 22: 293-330. Reprinted in Sexual Solipsism: Philosophical Essays on Pornography and Objectification (2009a), 25–63. Oxford: Oxford University Press.

 This chapter has particularly benefited from conversations (and disagreement) with Laura Caponetto, Bianca Cepollaro, Rae Langton, Mary Kate McGowan, and Marina Sbisà. Thanks also to two anonymous reviewers for their valuable feedback. 23

  Varieties of Uptake 

95

———. 2018. Blocking as Counter-Speech. In New Work on Speech Acts, ed. Daniel Fogal, Daniel W. Harris, and Matt Moss, 144–164. Oxford: Oxford University Press. Lewiński, Marcin. 2021. Illocutionary Pluralism. Synthese 199: 6687–6714. Maitra, Ishani. 2009. Silencing Speech. Canadian Journal of Philosophy 39 (2): 309–338. McDonald, Lucy. 2021. Your Word against Mine: The Power of Uptake. Synthese 199: 3505–3526. ———. 2022. Reimagining Illocutionary Force. The Philosophical Quarterly 72(4): 919–939. McGlynn, Aidan. 2019. Testimonial Injustice, Pornography, and Silencing. Analytic Philosophy 60 (4): 405–417. McGowan, Mary Kate. 2017. On Multiple Types of Silencing. In Beyond Speech. Pornography and Analytic Feminist Philosophy, ed. Mari Mikkola, 39–58. New York: Oxford University Press. Sbisà, Marina. 2006. Speech Acts without Propositions? Grazer Philosophische Studien 96: 155–178. ———. 2007. How to Read Austin. Pragmatics 17: 461–473. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2009a. Uptake and Conventionality in Illocution. Lodz Papers in Pragmatics 5: 33–52. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2009b. Illocution and Silencing. In Language in Life, and a Life in Language: Jacob Mey – A Festschrift, edited by Bruce Fraser and Ken Turner, 351–357. Bradford: Emerald. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2013. Locution, Illocution, Perlocution. In Pragmatics of Speech Actions, ed. Marina Sbisà and Ken Turner, 25–75. Berlin: De Gruyter. Searle, John R. 1969. Speech Acts. An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. Strawson, Peter F. 1964. Intention and Convention in Speech Acts. The Philosophical Review 73: 439–460.

Interactional Negotiation Maciej Witek

1 Introduction In her numerous works, Marina Sbisà (1984, 1992, 2002a, b, 2006, 2007, 2009, 2013a, b, 2018) has developed an Austin-inspired theory of speech acts which constitutes an alternative to the Gricean account of illocutionary interaction. According to John L. Austin, “the illocutionary act is a conventional act: an act done as conforming to a convention” (Austin 1975, 105); the force of a speech act, then, is determined not by the speaker’s intention, but by the conventional procedures she invokes in making her utterance (Austin 1975, 128). By contrast, the Griceans (Strawson 1964; Bach and Harnish 1979) claim that “a central class of illocutionary acts is communicative in nature” (Harnish 2005, 15) and that the force of a communicative speech act—for example, that of a statement, a request, a warning, a promise, an apology, and so on—is

M. Witek (*) Institute of Philosophy and Cognitive Science, University of Szczecin, Szczecin, Poland e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Caponetto, P. Labinaz (eds.), Sbisà on Speech as Action, Philosophers in Depth, https://doi.org/10.1007/978-3-031-22528-4_5

97

98 

M. Witek

determined by the audience-directed and reflexive intention with which it is made (Strawson 1964; Bach and Harnish 1979). They also argue that only some illocutionary acts—such as betting, dubbing, giving a verdict, naming a ship, and other formal or ritualized conversational moves that Austin (1975) discussed as paradigmatic cases of illocutions—are conventional in that their performance requires certain extralinguistic conventions and normally involves the use of certain conventional means. Sbisà adopts Austin’s central idea according to which to perform an illocutionary act is to invoke “an accepted conventional procedure having a certain conventional effect,” which includes “the uttering of certain words by certain persons in certain circumstances” (Austin 1975, 14). She argues, however, that illocutionary acts are conventional not due to the conventional means with which they are or could be performed,1 but because of the conventional nature of the effects they produce; the effects are conventional in that they come into being and exist in virtue of “an interpersonal or social agreement” (Sbisà 2009, 50; cf. Labinaz and Sbisà 2021, 69) and, as a result, are defeasible in the sense of being “liable to annulment” (Sbisà 2009, 47; cf. Sbisà 2007, 465). Normally, the agreement is tacit and by default: communication runs smoothly and the interacting agents take their conversational contributions at their face value (Sbisà 2002b, 78). In some cases, however, the agreement is achieved through more or less open negotiation. The speaker’s utterance may have linguistic or contextual properties which “may be taken to indicate [more than] one illocutionary force” (Sbisà 2013b, 236) and it “is up to interactional negotiation […] to make the utterance count as one of the [possible] illocutionary acts it is compatible with” (Sbisà 2013b, 236); there may also be situations in which there is a divergence between the speaker’s intended force and the way her utterance is taken by the hearer (Corredor 2018; Bentley 2020; McDonald 2022). In such cases, the communicating agents may produce a sequence of conversational moves whose  As Austin (1975, 103) put it, the use of language for performing illocutionary acts “may […] be said to be conventional, in the sense that at least it could be made explicit by the performative formula.” According to Peter F. Strawson (1964, 450), however, “the general suitability of an illocutionary act for performance with the help of the explicitly performative formula for that act” can be explained along the Gricean lines, that is, by reference to the speaker’s motive to get the hearer to recognize her communicative intention. For a discussion of this issue, see Sbisà (2009, 38 and 47–48). 1

  Interactional Negotiation 

99

purpose is to negotiate the function of the speaker’s act and thereby achieve the agreement in virtue of which its conventional effect exists. In short, illocutionary underdeterminacies and divergent force assignments invite interactional negotiation. My aim in this chapter is to use Sbisà’s idea of interactional negotiation to consider what it is for conversing agents to follow language conventions or to act “as conforming to a convention.” In Sect. 2, I use the Austinian notions of uptake and response as well as the concept of accommodation (Lewis 1979; Langton 2015, 2018; Witek 2021) to discuss a few examples of force negotiation and develop a model of its underlying mechanisms. In Sect. 3, I suggest that interactional negotiation plays a key role in the functioning of all language conventions—phatic, rhetic, illocutionary, and so on—construed as families or lineages of linguistic precedents (Millikan 1998, 2005; cf. Witek 2015a, b, 2018). My contention is that the ideas presented in Sects. 2 and 3 are compatible with Sbisà’s theory of speech acts and can be regarded as its possible extension.

2 Uptake, Responses, and Accommodation in Force Negotiation Stephen C. Levinson (2012, 104) argues that what he calls “action ascription”—that is, the assignment of a certain more or less definite action to a conversational turn—“is revealed by the response of the next speaker, which, if uncorrected in the following turn(s), becomes in some sense a joint ‘good enough’ understanding” or, in other words, takes effect as a jointly accepted interpretation of the initiating turn. According to Sbisà (2006, 2013b), the above-mentioned ascription-centered view of action is an integral element of Austin’s theory of speech acts and, by extension, of her Austin-inspired model of illocution. She claims: [T]he hearer’s responses manifesting how the hearer has received the speaker’s illocutionary act must be taken into account when what has been done is at issue. An analyst can tell what has been done (at the illocutionary level) from the consideration of illocutionary force indicating devices (what I would like to call the speech act’s illocutionary physiognomy) and from

100 

M. Witek

the hearer’s response, provided the speaker does not further challenge the latter. (Sbisà 2006, 153)

Viewed from the perspective of Sbisà’s theory of speech acts, then, the force of an utterance is jointly co-determined by the speaker, the hearer, and possibly other relevant social agents. In other words, what Gerald Gazdar (1981) calls “speech act assignment” is fixed through interactional negotiation: a discursive process which involves (1) the speaker’s initiating move, (2) the hearer’s response, and (3) the speaker’s validation or repair. Quite often, turn (3) manifests or signals the speaker’s default and tacit acceptance of the hearer’s understanding of turn (1). In some cases, however, the speaker can make more or less explicit attempts at correcting the hearer’s uptake manifested in turn (2), thereby initiating another round of negotiation. The above-mentioned initiation-response-evaluation schema plays a key role in a conversation-oriented theory of speech acts. Even though Emanuel A. Schegloff (2006) identifies adjacency pairs as minimal conversational units, other researchers studying the sequential organization of discourse focus on expanded (Jefferson and Schenkein 1978) or three-­ part exchanges (Tsui 1989), for example, question-answer-comment triads (McHoul 1979, 191) or repair ternary sequences (McHoul 1990, 350), which determine or negotiate the meaning of their initiating turns (Tsui 1989, 550).2 Likewise, some scholars working within the speech act-­ theoretic tradition (Corredor 2018; McDonald 2022) assume, following Sbisà’s (1992, 2002b, 2006) works on sequencing in illocutionary interaction, that the initiation-response-evaluation schema represents the structure of the discursive mechanism of force negotiation. Let us then look at the ternary sequence under discussion through the lens of Sbisà’s Austininspired theory of speech acts. While producing her initiating turn, the speaker uses various devices— lexical, grammatical, prosodic (Witek et al. 2022), kinesic, contextual, and so on—whose function is to indicate the procedure she invokes and, by the same token, show what she takes to be the conventional effect of her words. The hearer’s response manifests his uptake, that is, shows how  For an extensive discussion of sequencing in speech acts and meaning negotiation, see Fetzer (2018).

2

  Interactional Negotiation 

101

he understands the force and meaning of the speaker’s utterance. Normally, the evaluating turn takes the form of a default or implicit validation of the force ascription revealed by the hearer’s response; in some cases, however, it involves a series of corrections and repairs which are subject to further evaluation. In the remaining part of this section, I use the above-presented framework to discuss a number of examples of interactional negotiation. Before I get into the details, however, let me consider the relation between the hearer’s response and his uptake. The former is said to manifest (Sbisà 2006, 2007, 2009, 2013b), reveal (Levinson 2012), or signal (McDonald 2021) the latter.3 One may ask, however, what uptake is and what it is for the hearer’s response to manifest it. According to Austin (1975, 117), to secure uptake on the part of the hearer is to bring “about the understanding of the meaning and of the force of the locution”; in other words, it is to get the hearer to recognize the force and meaning of the speaker’s act. Herbert H. Clark (1996, 199) claims, in turn, that the hearer’s response displays or gives evidence that he understands or recognizes what the speaker means; in short, the response “is evidence of understanding”4 (Clark 1996, 200, italics in the original). It is natural to assume, then, that the hearer’s response gives evidence of his uptake construed as his understanding or recognition of the force and meaning of the speaker’s act.  Marcin Matczak (2019) argues, however, that uptake construed along the Hartian lines— for example, as an act of recognition which confirms that a certain act of enacting a piece of legislation is valid—is to be identified with a certain reaction to a change in the law. At first sight, such a proposal seems to depart from Austin’s original perspective, since Austin (1975, 116–117) explicitly distinguished the securing of uptake and the inviting of a response as two distinct effects of the successful illocutionary act. Nevertheless, Matczak’s account of the response-uptake relation seems to allow for the peculiarity of legal communication, which involves (1) collective agents—such as a legislature and a community of lawyers—rather than individual speakers and hearers and (2) producing complex and structured texts rather than individual utterances constituting conversational turns. 4  In fact, Clark (1996, 200) uses the sentence “Uptake is evidence of understanding.” It is instructive to stress, however, that he uses the term “uptake” to refer to the hearer’s response, for example, the words “I take you on” uttered in response to the speaker’s “I bet you sixpence” (see Clark 1996, 139); at the same time, he claims that the uptake so construed displays the hearer’s recognition and argues that “[w]hat is needed to complete an illocutionary act, however, is not its uptake, but its recognition” (Clark 1996, 139). In this paper, however, I use the term “uptake” along the Austinian lines, that is, as standing for what is manifested by or embodied in the hearer’s response rather than for the response itself. 3

102 

M. Witek

In my view, the above presented picture is correct as far as it goes. However, I would like to go further and make the following four points. First, as Levinson (2012, 104; cf. Sbisà 2013b, 242) notes, describing uptake in terms of recognition “is potentially misleading, because it presupposes that actions have a correct identity, when actually the process of attributing an action to a turn is a fallible, negotiated, and even potentially ineffable process.” Therefore, it is better to describe uptake in terms of action and meaning ascription rather than in terms of recognition. Second, uptake construed as action ascription may, but does not have to, take the form of the hearer’s occurrent mental state whose content represents the force and meaning of a conversational contribution.5 Normally, uptake is practical and implicit. In other words, it is embodied in the hearer’s response to the speaker’s act. When John follows Ann’s order, his behavior manifests that he takes her words to constitute a binding and felicitous directive act. Nevertheless, it is difficult to think of his uptake as a mental state that is formed prior to his response and plays a causal role in its production. Rather, John’s mental uptake, if there is any, is formed for the sake and in the course of acting and in this respect is akin to “discourse-constituted thoughts”: mental states that have no independent existence but are formed in the course of a progressing discourse (Jaszczolt and Witek 2018; Witek 2020). In sum, in most cases uptake does not take the form of an occurrent thought but is embodied in the hearer’s response. It remains to be considered, however, what it is for the hearer’s response to embody his uptake. Third, the relation of embodying that holds between the hearer’s response and his uptake is primarily ontological and only derivatively evidential. In my view, the hearer’s uptake is in a sense a structural component of his response; more specifically, the former coincides in content with a precondition of the latter. Following Mandy Simons (2013, 345), I take precondition to be an ontological notion which holds “[…] of events: a precondition on an event E is a condition which must be satisfied by the world in order for the event E to take place.” By extension, we  Elsewhere I have called this view “uptake externalism”; for a discussion, see Witek (2015c, 21).

5

  Interactional Negotiation 

103

can take the notion of precondition to hold of actions: a precondition on action A is a condition which must be satisfied by the context of a stretch of behavior in order for this stretch to count as the performance of action A; in other words, the felicity of action A presupposes that its preconditions are satisfied.6 Ann’s utterance of an imperative sentence is a felicitous order provided she stands in an appropriate authority relation to John; in other words, the felicity of her directive act presupposes that she has the required authority. By analogy, John’s practical response to Ann’s utterance can be adequately described as an act of compliance—or, alternatively, as his failure or refusal to comply—only if her utterance is a binding directive act; in other words, the ascription of the action of compliance to John’s behavior presupposes that Ann’s words constitute a felicitous order. The response-uptake relation, then, is akin to presupposing: in producing his response, the hearer presupposes that the speaker’s utterance to which he reacts has a certain force.7 What is more, uptake construed as force ascription seems to “project”: if John explicitly refuses to execute Ann’s order, his refusal, like compliance, manifests that he takes her words to constitute a felicitous directive act. John may also defuse Ann’s purported order—and thereby undo it (Caponetto 2020) or make it misfire (Austin 1975)—by saying that she does not have the required authority; that is to say, he can block the tacit acceptance or accommodation8 of what the felicity of her utterance qua order presupposes (Langton 2018; Witek 2021).

 A similar view is proposed by Thomason, Stone, and De Vault (2006). For a discussion of the idea of preconditions and the role it plays in an action-based approach to presupposing, see Witek (2019). 7  The idea that the response-uptake relation can be described in terms of presupposing comes from Sbisà (1992, 102–103); see also McDonald (2022, 933–934). 8  Roughly speaking, accommodation is a context-adjusting (Stalnaker 2002, 2014) or context-­ repairing (Lewis 1979) process which is guided by a default assumption that what the speaker says constitutes an appropriate conversational move of a certain type. For a discussion of accommodation as it occurs in illocutionary interaction—as well as of the phenomena of blocking the accommodation of some presuppositions—see Langton (2015, 2018), Sbisà (2018), and Witek (2013, 2015c, 2021). 6

104 

M. Witek

By analogy, let us consider a situation in which Ann utters sentence (1) and John responds by uttering one of the sentences listed in (2): (1)   Is Carl a philosopher? (2)  a.  Yes, he is.     b.  Ask Peter.    c.  I don’t know.    d.  Why do you ask?    e.  It depends on what you mean by “philosopher.” In my view, each of his possible responses listed in (2) manifests that John takes Ann’s utterance of (1) to be a genuine question9 rather than, say, an ironic comment on Carl’s bizarre behavior. In uttering any variant of (2) John presupposes that Ann’s utterance of (1) is a felicitous question. That is to say, his uptake “projects” in that it is embodied in different conversational moves that are equally appropriate responses to Ann’s initiating turn.10 By analogy, in uttering any member of the following family of sentences (3)  a.  I have to visit my sister.    b.  I do not have to visit my sister.    c. If I have to visit my sister, I will not be able to come to your lecture.    d.  I have to visit my sister or call my father. the speaker presupposes that she has a sister. What enables the hearer’s response to display or give evidence of his uptake, then, is the fact that the latter coincides in content with a certain precondition of the former. The relation of being a precondition of is  Viewed from the perspective of Segmented Discourse Representation Theory (Asher and Lascarides 2001, 2003), speech acts made in uttering different variants of (2) have relational illocutionary forces; for instance, John’s utterance of (2a) stands to Ann’s utterance of (1) in the Question-­ Answer-­Pair relation (Asher and Lascarides 2003, fn. 314), whereas his utterances of (2b) and (2c) have relational illocutionary forces of Request-Elaboration (Asher and Lascarides 2003, 326) and Not-Enough-Information (Asher and Lascarides 2003, 319), respectively. 10  They are appropriate because they make up a family of responses that Ann’s initiating turn invites by convention (Austin 1975, 117) as a question and as such they contribute to the development of a rhetorically coherent discourse (Asher and Lascarides 2001, 2003); for a discussion of this topic, see Witek (2015a) and (2018). 9

  Interactional Negotiation 

105

ontological. When it holds of actions, it is determined by the procedures for their performance. What is more, it is akin to presupposing. In uttering any variant of (3) the speaker presupposes that he has a sister and thereby displays his belief to the same effect. By analogy, John’s utterance of any variant of (2)—or, more precisely, the speech act he makes in this utterance—manifests his uptake and thereby gives evidence that he takes Ann’s utterance to be a felicitous act of asking a question. Fourth, the third turn in an initiation-response-evaluation sequence can be described either as licensing or as blocking the accommodation of what the second turn presupposes. In other words, it can either validate or repair the hearer’s uptake construed as force ascription. By way of illustration, let us consider the following interaction: (4)  Ann:  a.  Do you have a pencil?    John:  b.  Yes, I do.        c.  Here you are.        [John hands Ann a pencil]    Ann:  d.  Thanks!       [Ann takes the pencil.] Ann’s utterance of (4a) constitutes the performance of two speech acts or, in other words, is ascribed two different illocutionary forces. In uttering (4b), John presupposes that Ann’s initiating turn is a question, whereas in uttering (4c) and handing her a pencil he shows that he takes it to be a request. The former ascription is validated or agreed on by default: it goes through unchallenged and its accommodation is licensed simply because it is not blocked. The latter is explicitly endorsed by Ann’s evaluative turn (4d). By way of comparison, let us consider interaction (5): (5)  Ann:  a.  Do you have a pencil?    John:  b.  Yes, I do.      c.  Here you are.       [John hands Ann a pencil]    Ann:  d.  No, thanks.         e. I don’t need it.       f.   I just wanted to know.

106 

M. Witek

This time Ann explicitly blocks the accommodation of John’s presupposition—which is manifested in his response (5c)—that her initiating turn constituted a request. In uttering (5d) she refuses to accept the pencil, whereas in uttering (5e) she explicitly states that one of the felicity conditions for making the request under discussion is not fulfilled. What is more, she utters (5f ) to license the accommodation of John’s first ascription to the effect that her utterance of (5a) has been a question. Unlike in the previous case, she validates this ascription explicitly rather than implicitly and thereby makes sense of her initiating turn which, as evidenced by John’s utterance of (5c), has been misunderstood. It is instructive to stress that the interactional negotiation of the force of turn (5a) can be accounted for along the Gricean lines. That is to say, it can be described in terms of miscommunication and repairing the hearer’s faulty action ascription. One may say, then, that the force of Ann’s utterance of (5a) is determined by Ann’s communicative intention (Strawson 1964; Bach and Harnish 1979) rather than by John’s uptake. Nevertheless, there are cases in which the negotiated force of an utterance depends on the audience’s response—or, more specifically, on the action ascription it manifests—rather than on what the speaker has in mind. By way of illustration, let us consider the following three examples which, it seems, require an Austinian explanation. Consider dialogue (6), which is part of an attested example discussed by Sbisà (2002b, 77) in one of her papers on speech act sequences. A and B are sitting in A’s kitchen. A is the hostess, whereas B is a guest who wants to record an interview of A. The following conversation takes place:11 (6)  A:  a.  If you’d wanted we could have gone into the living room.      b.  We’d have been much more comfortable there.    B:  c.  It is all the same.      d.  Let’s bring the stuff through.    A:  e.  Yes, OK. I will carry this for you.

 The conversation of which example (6) forms a part was in Italian; the English translation comes from Sbisà (2002b). 11

  Interactional Negotiation 

107

As Sbisà notes, A’s utterance of (6a) can be read either as an offer or as a proposal. In short, it is illocutionarily underdetermined and as such invites interactional negotiation: Here it is possible to observe how sequencing selects one reading for an illocutionarily ambiguous conversational move. [(6a)] 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 [(6c)], denies having any desire or interest in moving […], so that the reading of [(6a)] as an offer does not become effective. […] Utterance [(6a)], however, manages to become effective as a proposal. […] B’s turn [(6d)] is quite appropriate as a reply to a proposal to move. So turn [(6cd)] can be viewed as operating a selection among the forces that [(6a)] 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 [(6e)] […], A ratifies what has been done up to that point. (Sbisà 2002b, 78–79)

In other words, B’s utterance of (6c) prevents turn (6a) from being read as an offer; that is to say, if A intended his words to take effect as an offer, B’s utterance of (6c) would defuse or undo this purported act by explicitly stating that one of its preconditions—that is, that B has “any desire or interest in moving”—is not satisfied. In uttering (6d), B presupposes that A’s initiating turn constitutes a proposal; as Sbisà notes, turn (6d) “is quite appropriate as a reply to a proposal to move.” Next, this presupposition is accommodated by A’s utterance of (6e). In sum, the force of turn (6a) is jointly co-determined by A and B. The next example comes from “The Queen’s Gambit,” a miniseries written and directed by Scott Frank. Beth and her friend Harry are talking about her anger that arises when she plays chess: (7)  Beth:  a.  Anger clears my head.    Harry:  b.  Anger is a potent spice.         A pinch wakes you up, too much dulls your senses.    Beth:  c.  Where’d you get that from, a fortune cookie?    Harry:  d.  Mrs. Grecco, my second grade teacher.     Beth: e.   Ah.    Harry:  f.   Mm.

108 

M. Witek

In uttering (7a), Beth asserts that anger has a positive effect on her thinking. Harry’s utterance of (7b) is a complex metaphor which presents an alternative and more nuanced view of anger and its impact on us. It is natural to assume that the point behind his remark is to draw Beth’s attention to the potential negative effects of strong anger and warn her against them. Nevertheless, rather than saying “Thanks for the warning, but I keep my emotions under control,” Beth attacks Harry and his statement. More specifically, she utters (7c) with an intention to ridicule Harry and his statement and thereby convey that she does not intend to take his words seriously. To achieve this effect, she uses a rhetorical technique that consists in asking a question of the form “Where’d you get that from?” in response to the speaker’s advice or opinion. To defuse her attack, Harry utters (7d). More specifically, his response manifests or gives evidence that he takes Beth’s utterance of (7c) to be a standard question that invites the response of giving an answer; in other words, his words can be naturally taken to take effect as an indirect answer to Beth’s question.12 In sum, the process of negotiating the effective function of Beth’s utterance of (7c) involves the interaction of two conflicting factors: her intention to ridicule the statement made in turn (7b) and Harry’s uptake embodied in his response; the exchange of turns (7e) and (7f ), which is a characteristic conclusion of information-exchanging dialogues, shows that the latter prevails over the former. As a result, Beth’s attempt to ridicule Harry’s opinion does not come off: her purported attack is a misfire and her utterance of (7c) takes effect as a standard question. Harry recognizes Beth’s intention to ridicule what he said, but effectively blocks its fulfillment. It is instructive to stress that in speculating about the intention behind Beth’s utterance of (7c) I do not adopt the Gricean perspective on speech acts (for a non-Gricean discussion of this example, see Witek 2022a, 71). According to the Griceans, what determines the force or function of an act is the communicative intention with which it is made, that is, an intention whose fulfillment requires (Strawson 1964) or even consists in (Bach and Harnish 1979) its recognition. In my view, however, Beth’s intention  As Asher and Lascarides (2003, fn. 314) would put it, Harry’s utterance of (7d) is linked to Beth’s utterance of (7c) by the rhetorical relation of Indirect-Question-Answer-Pair (IQAP). 12

  Interactional Negotiation 

109

to ridicule Harry and his words is not communicative; that is, it is not intended to be fulfilled by means of its being recognized. What is more, I am inclined to say that it is not perlocutionary either. The perlocutionary intention behind Beth’s utterance of (7c), if there is any, is to embarrass Harry, that is, to get him to feel awkward and unsettled. The central point behind ridiculing, by contrast, is not to “produce certain consequential effects upon the feelings, thoughts, or actions of the audience, or of the speaker, or of other persons” (Austin 1975, 101) but, rather, to express one’s negative or mockery attitude toward a certain target—a person, a speech act, an idea, or an opinion—and thereby publically stigmatize it as unreliable and untrustworthy. My hypothesis, then, is that ridiculing is not a perlocutionary act but, rather, a complex speech action which can be likened to ironizing construed as a speech action whose point is to present its target—that is, a certain contextually available thought—in an unfavorable light (Witek 2022b). A detailed discussion of this issue, however, goes beyond the scope of this chapter. For the present purposes it suffices to assume that Beth’s intention to ridicule, which can be described as a Searlean intention-in-action (Searle 1983), is neither Gricean nor perlocutionary.13 One might also argue that Harry’s utterance of (7d) is designed to get Beth to believe that he takes her words as a mere question, when in fact he perfectly got that her utterance of (7c) was uttered with an intention to ridicule him; therefore, his reply should be viewed as hiding rather than manifesting his actual uptake.14 In my view, however, Harry’s recognition of Beth’s intention to ridicule him should be distinguished from his official uptake embodied and signaled in turn (7d). It is the latter, not the former, that is publically available and, as a result, contributes to determining the negotiated force of turn (7c). In short, Harry’s recognition of Beth’s intention is not his effective uptake. Nevertheless, it plays a role in the psychological mechanism that motivates him to resist Beth’s attack by responding to turn (7c) as if it were a mere question. In sum, turn (7c) manifests Harry’s effective uptake and, at the same time, is a

13 14

 I am grateful to Paolo Labinaz for suggesting that I clarify this point.  I am grateful to Laura Caponetto for raising this worry.

110 

M. Witek

faked or pretended signal that he has failed to recognize Beth’s intention to ridicule him. Finally, let us consider a situation in which Jim meets Juliette, his younger female colleague, and utters the following words: (8) You look great today, sweetheart. According to Aaron Bentley (2020, 16), from whom I borrow this example, it is possible that Jim intends his utterance to be a compliment,15 whereas Juliette may take his commentary on her appearance to constitute an act of subordinating her. In short, Jim and Juliette may ascribe different action types to his utterance of (8) and this divergence may invite interactional negotiation. By way of illustration, let us consider the following exchange of turns which might follow Jim’s utterance of (8): (9)  Juliette:  a.  There is more to me than how I look.        b.  Besides, I am not a “sweetheart.”    Jim:    c.  Come on, I just wanted to be nice.    Juliette:  d. I don’t think it is nice to say things about someone’s appearance.         e.  In fact, your remark was patronizing and subordinating. As Bentley (2020, 18) notes, one of the felicity conditions for the act of complimenting is that “what is offered as complimentary must be valued by the person to whom the compliment is offered.” In uttering (9a), then, Juliette suggests that this condition is not met and thereby defuses Jim’s purported act of complimenting her on how she looks.16 In uttering (9b), in turn, she thwarts his implicit attempt to make their relations less formal or, in other words, to shorten the social distance between them; more specifically, she blocks the accommodation of what is presupposed by his act of  It is instructive to stress that Jim’s intention to pay Juliette a compliment does not have to be Gricean. Viewed from the Austinian perspective, it is a mere intention-in-action whose presence renders Jim’s utterance of (8) an intentional act of paying a complement (provided other felicity conditions for complimenting are met). 16  If, by contrast, she sincerely said “Oh, thank you!” she would license Jim’s intended reading of his words. 15

  Interactional Negotiation 

111

calling her “sweetheart.” Jim’s response (9c) is intended to repair what he takes to be a misunderstanding: he says that he wanted to be nice and thereby implies that his utterance of (8) was a sincere compliment. Still, his attempt to repair fails: in uttering (9d) and (9e), Juliette rejects his intended reading of turn (8) and explicitly ascribes to it the forces of patronizing and subordinating.17 Her uptake prevails over Jim’s intention and she wins the negotiation. What gives her advantage is the fact that Jim can reliably be taken to execute a procedure for subordinating which involves one speaker making a positive comment on the appearance of another, where the former, because of his or her informal social group identity (Bentley 2020), is stereotypically and unjustly regarded as intellectually superior to the latter. In sum, the actual force or function of a conversational contribution is subject to interactional negotiation and at least in some cases it can be determined by the hearer’s uptake rather than by the speaker’s intention-­ in-­action,18 where the former is to be described as the hearer’s action ascription embodied in or presupposed by his response to the speaker’s act. The basic structure of interactional negotiation is represented by the initiation-response-evaluation sequence, whose third element blocks or licenses the accommodation of the action ascription presupposed or manifested by the second one.

3 Language Conventions as Lineages of Negotiated Precedents In the previous section, I have discussed Sbisà’s idea of interactional negotiation and the role it plays in her account of the conventional nature of illocutionary acts. Viewed from the perspective of her Austin-inspired  In other words, Juliette’s response signals that she takes Jim’s utterance of (8) to be a cat-call rather than a benign compliment. Compliments and cat-calls construed as acts of subordination have similar surface structures. For instance, in one context sentence (8) can be used to felicitously perform a compliment, whereas in another context it can take effect as a cat-call and subordinate the addressee. I am grateful to Laura Caponetto for drawing my attention to this topic. For a more extensive discussion of the relation between compliments and cat-calls, see McDonald (2021). 18  This position can be called “illocutionary agency externalism”; see Witek (2015c). It is instructive to stress that the speaker’s contribution to the process of negotiating the force of her utterance is her intention-in-action which should be distinguished from her Gricean communicative intention (if there is any). 17

112 

M. Witek

theory of speech acts, acting as conforming to an illocutionary convention is not an automatic process that makes use of form-meaning associations; in other words, illocutionary acts are conventional not because of the conventional means with which they are performed. Rather, following an illocutionary convention is a joint action of the speaker, the hearer, and possibly other relevant social agents, which normally takes the form of the initiation-response-evaluation sequence and involves interactional negotiation. In this section I suggest that the same should be said of following other language conventions, such as Austin’s (1975, 31) demonstrative and descriptive conventions—which contribute to the determination of the rhetic meaning (Austin 1975, 93) of a speech act— as well as phatic, rhetorical, and procedural conventions. My hypothesis is that the notion of interactional negotiation plays a key role in explaining the process whereby language conventions proliferate and shape our linguistic practice. Following Ruth G. Millikan (1998, 2005, cf. Witek 2015a, b, 2018), I assume that conventions are patterns of activity—for example, driving on the left, shaking right hands, using forks as eating utensils, using grammatical structures and lexical units, and so on—that proliferate, first, by reproduction and, second, because of the weight of their cultural precedents rather than due to their capacity to produce certain effects. Like Millikan, I hold that every language convention construed as a linguistic pattern can be represented as a lineage of its past uses which have been reproduced one from another. In my view, however, the two criteria posited by Millikan—that is, the reproduction condition and the weight-­ of-­precedent requirement—do not suffice to define the scope of language conventions. My contention is that the mechanism whereby linguistic conventional patterns are reproduced—or, in other words, whereby language conventions construed as lineages or families or reproduced items proliferate—necessarily involves interactional negotiation (for a more extensive discussion of this issue, see Witek 2022a). A key idea behind the account proposed in this section is that our linguistic activity is shaped by conventional patterns which operate at different levels of speech act organization: not only illocutionary, but also locutionary—including phonetic, phatic, and rhetic levels—as well as rhetorical and procedural. Of course the patterns exist in the form of

  Interactional Negotiation 

113

families of precedents; at a certain level of abstraction, however, we can refer to a certain conventional lineage by describing those aspects of its constituents which have been preserved during reproduction. For instance, to represent a certain phatic family, we focus on and refer to the syntactic and lexical properties of its members. By analogy, to describe a certain procedural lineage—which collects different executions of the same illocutionary procedure—we specify a set of felicity conditions and claim that it represents the procedure in question. Likewise, to describe a certain rhetorical lineage—whose members are different tokens of the same rhetorical relation—we can use the framework of Segmented Discourse Representation Theory (Asher and Lascarides 2001, 2003). For instance, pairs 〈(1), (2a)〉, 〈(4a), (4b)〉, and 〈(7c), (7d)〉 belong to the same rhetorical lineage which can be represented as a family of two-­ part sequences whose turns are connected by the rhetorical relation of Question-Answer-Pair.19 It is instructive to stress, however, that the different levels of speech act organization distinguished in the previous paragraph correspond to different abstract aspects of what Austin (1975, 148) called “[t]he total speech act in the total speech situation.” Let us first focus on the contrast between locutionary, illocutionary, and perlocutionary acts (see Sbisà 2013a, b). Roughly speaking, we distinguish between them by referring to their characteristic effects: an utterance described as a locutionary act produces a locution that represents a certain state of affairs in a certain illocutionary mode; if our focus is on how it affects the network of normative intersubjective relations, we describe it as an illocutionary act equipped with a certain more or less definite force; construed as a perlocutionary act, in turn, the utterance produces a certain consequential effect “upon the feelings, thoughts, or actions of the audience, or of the speaker, or of other persons” (Austin 1975, 101). In a similar vein, Sbisà rejects both the additive and the embedding model of the phonetic-­ phatic-­rhetic distinction. According to the former, the three acts under discussion are formed independently of one another and add up to the  Strictly speaking, turns (7c) and (7d) are linked by the rhetorical relation of Indirect-Question-­ Answer-Pair (IQAP); see fn. 12. For more examples of rhetorical relations which correspond to different lineages of two-turn sequences, see fn. 9. 19

114 

M. Witek

entire locutionary act; according to the latter, in turn, the rhetic act should be identified with the locutionary one. Following Austin (1975, 92–93), however, Sbisà (2013b, 230fn) argues that what he called the phonetic, phatic, and rhetic acts are abstract aspects of the locutionary act which can be distinguished by reference to their characteristic products: phones, phemes, and rhemes, respectively. In my view, the Austinian distinction between phonetic, phatic, rhetic, locutionary, and illocutionary acts corresponds to a distinction between five types of language conventions construed as families of reproduced items.20 For instance, a certain phatic lineage collects utterances which can be regarded as equivalent phemes with respect to their phatic meaning, where the phatic meaning of an utterance is defined in terms of its rhetic and illocutionary act potential (Forguson 1973; cf. Witek 2015b); a certain rhetic lineage, in turn, collects utterances that are taken to be equivalent with respect to their sense and reference, which are determined by what Austin (1950, 31) called descriptive and demonstrative conventions, respectively. In general, speech acts that are agreed to constitute cases of saying or doing the same—for example, as having equivalent effects—constitute a lineage of reproduced precedents. It is instructive to stress, however, that “[t]he same” does not always mean the same. In fact it has no meaning in the way that an “ordinary” word like “red” or “horse” has a meaning: it is a (the typical) device for establishing and distinguishing the meanings of ordinary words. Like “real”, it is part of our apparatus in words for fixing and adjusting the semantics of words. (Austin 1950, 29)

By extension, we can say that it is part of our apparatus in words for grouping speech situations into lineages of different types. In my view, besides phatic, rhetic, and illocutionary families of reproduced linguistic items, there are rhetorical and procedural lineages. The former are collections of reproduced pairs of discursive contributions which can be described in terms of rhetorical relations: Question For a more extensive discussion of this topic, see Witek (2015a, b, 2022a).

20

  Interactional Negotiation 

115

Answer-­Pair, Indirect-Question-Answer-Pair, Request-Elaboration, NotEnough-­Information, and so on.21 A procedural lineage, in turn, is a family of jointly reproduced executions of “the same” procedure, for example, the procedure for making a promise, issuing an order, paying a compliment, or subordinating the hearer. In sum, my hypothesis is that language conventions which are said to shape our verbal activity can be best understood as lineages of reproduced items. I also suggest that the mechanism responsible for the proliferation of language conventions so construed involves interactional negotiation. Even though past speech situations constituting a certain conventional family put constraint on what can count as its new member, they systematically underdetermine its exact form and properties. Following a convention, then, is never an automatic and algorithmic process. What makes a collection of linguistic items—structures, phrases, illocutionary acts, pairs of rhetorically connected discursive contributions, executions of procedures, and so on—a conventional lineage is the fact that the mechanism whereby they have been reproduced one from another necessarily involves interactional negotiation. For instance, the negotiated and hence conventional force of Jim’s utterance of (8) is subordinating rather than complimenting; by the same token, his verbal behavior contributes to executing the procedure for subordinating rather than that for paying compliments. The negotiated function22 of Beth’s utterance of (7c) is asking a question rather than ridiculing Harry’s statement; as a result, the rhetorical relation that holds between turn (7c) and turn (7b)—or, more appropriately, between the conversational move made in the former and the opinion expressed in the latter—is not Making-a-Ridiculing-­ Comment-on, but Asking-for-the-Source-of.

 See Asher and Lascarides (2001, 2003) as well as footnotes 9 and 12.  I use the term “function” rather than “force” because I would like to remain neutral on the question whether ridiculing is an illocutionary or perlocutionary act. In my view, it is neither illocutionary nor perlocutionary. My hypothesis is that ridiculing is a complex speech action whose point is to express one’s mockery attitude toward a certain target and thereby publically stigmatize it as unreliable or untrustworthy. However, a detailed discussion of this issue goes beyond the scope of the present chapter. 21 22

116 

M. Witek

Acknowledgments  The research leading to these results received funding from the National Science Centre, Poland, under grant no. 2015/19/B/HS1/03306. I thank Marcin Matczak for thoughtful comments and discussion on an earlier version of this work. I also thank Laura Caponetto and Paolo Labinaz for their insightful remarks which helped to improve the final version of this chapter.

References Asher, Nicholas, and Alex Lascarides. 2001. Indirect Speech Acts. Synthese 128: 183–228. ———. 2003. Logics of Conversation. Cambridge: Cambridge University Press. Austin, John L., 1950. Truth. Proceedings of the Aristotelian Society, Supplementary Volume 24: 111–128. Reprinted Proceedings of the Aristotelian Society. The Virtual Issue 1: 27–41. ———. 1975. In How to Do Things with Words, ed. James O.  Urmson and Marina Sbisà, 2nd ed. Cambridge, MA: Harvard University Press. Bach, Kent, and Robert M.  Harnish. 1979. Linguistic Communication and Speech Acts. Cambridge, Ma: MIT Press. Bentley, Aaron. 2020. Social Identity, Group Speech, and Negotiated Meaning. Language & Communication 72: 13–24. Caponetto, Laura. 2020. Undoing Things with Words. Synthese 197: 2399–2414. Clark, Herbert H. 1996. Using Language. Cambridge: Cambridge University Press. Corredor, Cristina. 2018. The Dynamics of Conversation: Fixing the Force in Irony. A Case Study. In Normativity and Variety of Speech Actions, ed. Maciej Witek and Iwona Witczak-Plisiecka, 140–158. Leiden: Brill. Fetzer, Anita. 2018. Speech Acts in Discourse. In Normativity and Variety of Speech Actions, ed. Maciej Witek and Iwona Witczak-Plisiecka, 101–121. Leiden: Brill. Forguson, Lynd W. 1973. Locutionary and Illocutionary Acts. In Essays on J.L. Austin, ed. Isaiah Berlin, 160–185. Oxford: The Clarendon Press. Gazdar, Gerald. 1981. Speech act Assignment. In Elements of discourse understanding, ed. Aravind K. Joshi, Bonnie L. Webber, and Ivan A. Sag, 64–83. Cambridge: Cambridge University Press. Harnish, Robert M. 2005. Commitments and Speech Acts. Philosophica 75: 11–41.

  Interactional Negotiation 

117

Jaszczolt, Kasia M., and Maciej Witek. 2018. Expressing the Self: From Types of de se to Speech-Act Types. In Expressing the Self. Cultural Diversity and Cognitive Universals, ed. Minyao Huang and Kasia M. Jaszczolt, 187–221. Oxford: Oxford University Press. Jefferson, Gail, and Jim Schenkein. 1978. Some Sequential Negotiations in Conversation. Unexpanded and Expanded Versions of Projected Action Sequences. In Studies in the Organization of Conversational Interaction, ed. Jim Schenkein, 155–172. New York: Academic Press. Labinaz, Paolo, and Marina Sbisà. 2021. The Problem of Knowledge Dissemination in Social Network Discussions. Journal of Pragmatics 175: 67–80. Langton, Rae. 2015. How to Get a Norm from a Speech Act. The Amherst Lecture in Philosophy 10: 1–33. http://www.amherstlecture.org/langton2015/. ———. 2018. Blocking as Counter-Speech. In New Work on Speech Acts, ed. Daniel Fogal, Daniel W. Harris, and Matt Moss, 144–164. New York: Oxford University Press. Levinson, Stephen C. 2012. Action Formation and Ascription. In The Handbook of Conversation Analysis, ed. Jack Sidnell and Tanya Stivers, 101–130. Chichester: Wiley-Blackwell. Lewis, David. 1979. Scorekeeping in a Language Game. Journal of Philosophical Logic 8: 339–359. Matczak, Marcin. 2019. Speech Act Theory and the Rule of Recognition. Jurisprudence 10 (4): 552–581. McDonald, Lucy. 2021. Cat-Calls, Compliments and Coercion. Pacific Philosophical Quarterly 103 (1): 208–230. ———. 2022. Reimagining Illocutionary Force. The Philosophical Quarterly 72 (4): 918–939. McHoul, Alexander W. 1979. The Organization of Turns at Formal Talk in the Classroom. Language in Society 7 (2): 183–213. ———. 1990. The Organization of Repair in Classroom Talk. Language in Society 19 (3): 349–377. Millikan, Ruth G. 1998. Language Conventions Made Simple. Journal of Philosophy 95: 161–180. ———. 2005. Language: A Biological Model. Oxford: Oxford University Press. Sbisà, Marina. 1984. On Illocutionary Types. Journal of Pragmatics 8: 93–112. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press.

118 

M. Witek

———. 1992. Speech Acts, Effects, and Responses. In (On) Searle on Conversation, edited by Herman Parret and Jef Verschueren, 101–111. Amsterdam: John Benjamins. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2002a. Speech Acts in Context. Language and Communication 22: 421–436. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2002b. Cognition and Narrativity in Speech Act Sequences. In Rethinking Sequentiality, edited by Anita Fetzer and Christiane Meierkord, 71–97. Amsterdam: John Benjamins. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2006. Communicating Citizenship in Verbal Interaction. Principles of a Speech Act Oriented Discourse Analysis. In Analysing Citizenship Talk, ed. Heiko Hausendorf and Alfons Bora, 151–180. Amsterdam: John Benjamins. ———. 2007. How to Read Austin. Pragmatics 17 (3): 461–473. ———. 2009. Uptake and Conventionality in Illocution. Lodz papers in Pragmatics 5: 33–52. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2013a. Locution, illocution, perlocution. In Pragmatics of Speech Actions, ed. Marina Sbisà and Ken Turner, 25–75. Berlin: De Gruyter. ———. 2013b. Some Remarks about Speech Act Pluralism. In Perspectives on Pragmatics and Philosophy, ed. Alessandro Capone, Franco Lo Piparo, and Marco Carapezza, 227–244. Heidelberg: Springer. ———. 2018. Varieties of Speech Act Norms. In Normativity and Variety of Speech Actions, edited by Maciej Witek and Iwona Witczak-Plisiecka, 23–50. Leiden: Brill. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. Schegloff, Emanuel A. 2006. Sequence Organization in Interaction. Cambridge: Cambridge University Press. Searle, John R. 1983. Intentionality. An Essay in the Philosophy of Mind. Cambridge: Cambridge University Press. Simons, Mandy. 2013. On the Conversational Basis of Some Presuppositions. In Perspectives on Linguistic Pragmatics, ed. Alessandro Capone, Franco Lo Piparo, and Marco Carapezza, 329–348. Heidelberg: Springer. Stalnaker, Robert. 2002. Common Ground. Linguistics and Philosophy 25: 701–721. ———. 2014. Context. Oxford: Oxford University Press.

  Interactional Negotiation 

119

Strawson, Peter F. 1964. Intention and Convention in Speech Acts. Philosophical Review 73 (4): 439–460. Thomason, Richmond, Matthew Stone, and David DeVault. 2006. Enlightened Update: A Computational Architecture for Presupposition and Other Pragmatic Phenomena. Available at: http://citeseerx.ist.psu.edu/viewdoc/dow nload?doi=10.1.1.76.8649&rep=rep1&type=pdf. Accessed 28 May 2022. Tsui, Amy B.M. 1989. Beyond the Adjacency Pair. Language in Society 18 (4): 545–564. Witek, Maciej. 2013. How to Establish Authority with Words: Imperative Utterances and Presupposition Accommodation. In Theory of Imperatives from Different Points of View (2), ed. Anna Brożek, Jacek Jadacki, and Berislav Žarnic, 145–157. Warszawa: Semper. ———. 2015a. An Interactional Account of Illocutionary Practice. Language Sciences 47: 43–55. ———. 2015b. Linguistic Underdeterminacy: A View from Speech Act Theory. Journal of Pragmatics 76: 15–29. ———. 2015c. Mechanisms of Illocutionary Games. Language & Communication 42: 11–22. ———. 2018. Coordination and Norms in Illocutionary Interaction. In Normativity and Variety of Speech Actions, ed. Maciej Witek and Iwona Witczak-Plisiecka, 66–97. Leiden: Brill. ———. 2019. Accommodation in Linguistic Interaction. On the So-Called Triggering Problem. In Philosophical Insights into Pragmatics, ed. Piotr Stalmaszczyk, 163–191. Berlin: De Gruyter. ———. 2020. Self-Expression in Speech Acts. Organon F 28: 326–359. ———. 2021. Illocution and Accommodation in the Functioning of Presumptions. Synthese 198: 6207–6244. ———. 2022a. An Austinian alternative to the Gricean perspective on meaning and communication. Journal of Pragmatics 201: 60–75. ———. 2022b. Irony as a Speech Action. Journal of Pragmatics 190: 76–90. Witek, Maciej, Sara Kwiecień, Małgorzata Wrzosek, Mateusz Włodarczyk, and Jakub Bondek. 2022. Prosody in recognizing dialogue-specific functions of speech acts. Evidence from Polish. Language Sciences 93, 101499.

Some Varieties of Illocutionary Pluralism Casey Rebecca Johnson

1 Introduction When theorizing about the use of language in communication, the aim, according to Marina Sbisà, “is to account for language use in all its irreducible complexity” (Sbisà 2001, 1792). In pursuit of that goal, Sbisà offers analyses of conversation, building from J.  L. Austin’s speech act theory, that are admirably sensitive to the real complexity and context sensitivity of language use, while remaining systematic and theoretically useful. One place that Sbisà’s commitment to this goal is evident is in her discussion of what she calls illocutionary pluralism—the view that a single token utterance can have more than one illocutionary force. In other words, illocutionary pluralism is a commitment to the claim that a

C. R. Johnson (*) Department of Politics and Philosophy, University of Idaho, Moscow, ID, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Caponetto, P. Labinaz (eds.), Sbisà on Speech as Action, Philosophers in Depth, https://doi.org/10.1007/978-3-031-22528-4_6

121

122 

C. R. Johnson

speaker can, by making a single utterance in a single context, make more than one conversational move.1 Illocutionary pluralism first comes into view in a paper by Sbisà (2013b). There, Sbisà sketches several ways that a single utterance could have multiple illocutionary effects. Her remarks on illocutionary pluralism are rich and provocative but do not constitute anything like a full-­ blown theory that could back a commitment to illocutionary pluralism. Fortunately, there have been two recent attempts to develop such a theory. In my own work, I defend a version of the illocutionary pluralism that I call illocutionary relativism (Johnson 2015, 2019).2 This view has much in common with Sbisà’s, but is distinct in ways I will explain below. Marcin Lewiński has also developed a version of illocutionary pluralism, which he contrasts with relativism, calling it illocutionary pluralism proper (Lewiński 2021). I will begin the chapter with some background on illocutionary force. In the second section, I will explain Sbisà’s sketch of pluralism. I will then explain my own and Lewiński’s developed accounts, dedicating a section to each.

2 Illocutionary Force Utterances, as Austin observed, are actions. As such, they can be used to do many things. We can use utterances to solicit information, to undertake future obligations, to mislead, or to testify. According to this understanding of language use, speaker’s speech actions can be successful along a number of different dimensions. Austin’s most famous distinction, perhaps, is that between performatives and constatives. That is, between speech actions that change people’s relationships and obligations (“I hereby pronounce you married”) and those that describe the world (“The priest married the couple”). This distinction will not be particularly important for our purposes here. Indeed, Sbisà argues that this distinction has proved something of a distraction in  I am grateful to Trevor Woodward, Laura Caponetto, and Paolo Labinaz for their help with this chapter. 2  Though this label has been applied to both my pluralism and Sbisà’s (Lewiński 2021). 1

  Some Varieties of Illocutionary Pluralism 

123

the literature following Austin. It should, Sbisà argues, be understood as a useful provisional distinction to get Austin’s audience to recognize that all speech is action (Sbisà 2007). The distinction between performatives and constatives, she argues, gives way to the distinction between locutionary acts, illocutionary acts, and perlocutionary acts.3 This latter distinction is central to our purposes, here. These three kinds of acts pick out three ways we can truly describe what a speaker does with her words. Take, for example, Anya who says to Barry, “The fan is on again.” In making this utterance she can truly be said to have uttered meaningful content. That’s the locutionary act. She can also truly be said to have made a certain kind of conversational move. Perhaps she reminds Barry to watch his head. Perhaps she informs Barry that his sister left the fan on. Perhaps she orders Barry to turn the fan off. Perhaps she commiserates with Barry over the cold. The conversational move is the illocutionary act, and reminding, informing, ordering, and so on are illocutionary forces. Finally, Anya’s act can truly be said to have some more distal effects—if Barry is enraged at being so reminded, she can truly be said to have enraged Barry by uttering. This is the perlocutionary act. The conventionality of the conversational move is an important feature of the illocutionary act, at least according to Austin. For Austin, there are several necessary conditions for a fully successful illocutionary act—for an illocutionary act to be “happily, successfully performed” (Austin 1975, 116).4 (A) The speaker must have the requisite authority; (B) there has to be a particular convention in the linguistic community for making the kind of force in question; (C) the speaker has to follow an appropriate procedure for making such a conversational move; (D) the speaker has to have the psychological state associated with the move in question; (E) and the hearer must give the speaker uptake5—that is, the utterance must have had the right sort of effect on the hearer. To see this  Austin offers even more fine-grained analyses of categories. I’ll leave that analysis aside here.  We should keep in mind, as Austin reminds us, that we ought to “not stress the normal connotations of these names!” (Austin 1975). 5  Austin did not conceive of uptake as a felicity condition, however he considered uptake to be necessary for an illocutionary act. I’ve included it in my list of conditions for fully successful illocutions because if the illocutionary act does not happen it cannot happen fully successfully. 3 4

124 

C. R. Johnson

more clearly, consider a question. For Carol to ask Danny a question, she has to (A) have the requisite authority. Luckily for Carol, most conversational contexts don’t require much in the way of authority to ask questions.6 (B) Carol and Danny must be in a linguistic community that has at least one conventionalized way to ask questions. (C) Carol has to follow a question-asking procedure. This too will be a matter of convention (or convention-approximation) in her linguistic community. (D) Carol must sincerely wonder about the answer to the question she is asking. More on this in a moment. And (E) Danny has to recognize that her utterance is a question. Failure of any of these conditions renders the utterance less than fully successful. This failure is not, however, always fatal for the illocutionary force. Carol can clearly ask a question even if she already knows, and so does not sincerely wonder about the answer. Perhaps less clearly, she can ask a question in a very unconventional way. This might, in turn, make it more difficult for Danny to meet condition E, but if he happens to catch on, it is not clear that her attempted question was a total failure—it was just not fully successful. I find Sbisà’s work on uptake (my condition E) illuminating. According to Sbisà, the procedure that Carol follows must be conventional, for fully successful illocution, but this is not the only place that convention enters into the speech act. For Sbisà, each illocutionary act has associated conventional effects. A question prompts different conventional effects in Danny’s and Carol’s linguistic community than an assertion would. When Danny is affected by Carol’s speech in the way that is conventional for questions, he takes her speech up as a question. As Sbisà puts it, “all illocutionary acts are conventional primarily because they all have conventional effects” (Sbisà 2007). Importantly, uptake is a necessary condition for illocutionary acts to be fully successful. However, the relationship between performing an illocution and performing it fully successfully is complicated and controversial (Langton 1993; Bird 2002; Kukla 2014). I agree with Sbisà, however,  We can, of course, think of contexts that do require authority to ask questions. Court rooms, classrooms, and so on, all provide examples. See Kukla and Lance (2009), Tirrell (2018), and McGowan (2019) for interesting discussions of this and related issues having to do with authority and conversational moves. 6

  Some Varieties of Illocutionary Pluralism 

125

that, “a speaker may perform a certain illocutionary act even if she violates one or more of its conditions” (Sbisà 2013a, 50). That is, full success is not necessary for illocutionary action.7

3 Sbisà’s Pluralism There are a number of things we might mean by speech act pluralism. On a straightforward reading of Speech Act Theory, nearly all speech action is plural. As we saw above, one of Austin’s most basic observations is that in making a single utterance a speaker can often truly be said to be doing three things: locuting, illocuting, and perlocuting. A more robust kind of speech act pluralism might be used to identify the fact, observed by Cappelen and Lepore, that “[n]o one thing is said (or asserted, or claimed, or …) by any utterance: rather, indefinitely many propositions are said, asserted, claimed, stated, etc.” (Cappelen and Lepore 2008, 199). Cappelen and Lepore argue that this is so by pointing out the contingency of content—our utterances depend for their content on many features of the conversation. This, they argue, means that there is rarely only one uniquely correct content, and so rarely only one unique speech act for any utterance. These speech act pluralisms, though worthy of the name, are not our focus here. Another possible understanding of speech act pluralism has to do with an observation from Searle (Searle 1975). Searle points out that sometimes speakers use the grammatical form typically associated with one illocutionary force to make a speech act with some other force. The paradigm example of this is when a speaker says, “Can you pass the salt?” to get her interlocutor to pass her the salt.8 In this case, the speaker’s utterance has the grammatical form of a question9 but she is using an ­utterance  This is a standard view in speech act theory, dating back to Austin (1975) and Searle and Vanderveken (1985). 8  It is possible to merely inquire after salt-passing abilities, but it is conventional in most of my linguistic communities to use this utterance to make a request. 9  Searle does not invoke grammatical form but rather what the sentence “literally says” (Searle 1975, 62). I find this talk of literality confusing so use the term “grammatical form” instead. Literal meaning, in Searle’s sense, could be substituted in this discussion without loss of argumentative power. 7

126 

C. R. Johnson

with that form to make a request. According to Searle, this indirect speech primarily has the force of a request and secondarily has the force of a question. Lewiński describes this as pluralism on the vertical axis—a speaker makes one conversational move by-way-of another (or by way of the grammatical structure most usually associated with another). We will not be focused on vertical axis pluralism. Instead, the kind of speech act pluralism we’ll take as our focus is described by Sbisà as illocutionary pluralism. This kind of speech act pluralism holds that a single token speech act—that is, a particular utterance made in a particular context—can have more than one primary illocutionary force. Sbisà holds that there are both conventional procedures and conventional effects for each illocutionary force. Illocutionary acts, on this view, bring about certain specific kinds of changes to interpersonal relationships. The changes in question are not the sorts of changes that result from having merely been in conversation together, or from being spatially proximate. She also does not mean the sorts of changes in attitude or behavior associated with perlocutions. Instead, Sbisà (2013b, 236) describes these conventional effects as follows: In speaking of the illocutionary act as bringing about “conventional” effects, I mean that its effects could not come into being without some kind of coming together or agreement, overt or (more often) tacit and by-­ default, of the relevant members of the relevant social group. Since such an agreement cannot be completely ad hoc, starting from scratch again every time, repeatable patterns and routines are needed, including possible standardized or even conventional means for performing the illocutionary act… In this perspective, the idea that illocutionary acts bring about changes in interpersonal relationships is to be qualified by adding that these changes should affect the conventional aspects of interpersonal relationships.

By this Sbisà seems to suggest that illocutionary acts can only be made in communities with conventions. These conventions make it possible for a speaker to bring about some specific conventional effects by performing some conventionalized action. This allows the illocutionary act

  Some Varieties of Illocutionary Pluralism 

127

to have useful, predictable effects10 and for act types like assertions to emerge. Imagine that it is conventional in our community to be able to apologize by saying, “I’m terribly sorry,” and that it is conventional in our community for the hearer to understand utterances of that type in the ways conventional for apologies (thinking about the offending act, considering forgiveness, etc.). If that is the case, then Emile can say to Frances, “I’m terribly sorry” and have a pretty good shot of affecting Frances in the way conventional for apologies. Frances’ changes in expectations and the obligations she assigns to Emile and to herself fit the profile of the effects of apologizing in this community. The way that Frances sees the deontic landscape has changed to reflect that Emile has apologized. Frances may also have some other responses which might change their relationship— Frances might be hurt that Emile brought up this offending event, or Emile might expect Frances to burst into tears, but neither of those changes is conventional for apologies.11 These are effects, but they’re not part of the illocutionary act, according to Sbisà, because they’re not conventional effects. This brings us to Sbisà’s remarks on pluralism. While she is not developing or defending a full theory of illocutionary pluralism, Sbisà does consider three ways in which, as she puts it, “the issue of illocutionary pluralism can be raised” (Sbisà 2013b, 236). The first way is a case in which the token utterance is ambiguous between more than one illocutionary pattern. Many of our illocutionary acts have overlapping conventional procedures. Sbisà offers an example: “Consider an utterance in which the speaker suggests to the addressee some joint action, which is alleged to be in the addressee’s own interest. Is it a proposal or an offer?” (Sbisà 2013b, 236). Sometimes, in a case like this, the addressee’s response reveals how the utterance is taken, thus disambiguating the illocutionary force. If the addressee thanks the speaker, the utterance counts as an offer, as expressions of gratitude are not conventional responses to proposals. If  These effects will be predictable in many, but not all cases—conventions can always be amended, exploited (as with Gricean implicature; Grice 1975), or abandoned. 11  At least in our community. We can imagine communities where these are parts of the conventional reaction to apology (or something apology-like). Communities can adapt to or create lots of different kinds of conventions and conventionalized actions. 10

128 

C. R. Johnson

the response is disambiguating in this way, the utterance has only one force. As Sbisà puts it, “One and the same utterance in the same context may display an ambiguous profile, which will be disambiguated by the response it gets” (Sbisà 2013b, 237). The second kind of case in which the issue of illocutionary pluralism might arise, is one in which the utterance’s conventional effect is consistent with more than one illocutionary pattern. Consider Gerald who utters, “I’ll never do that again” to Hattie. That utterance might have conventional effects of both a promise and an apology. Hattie might take Gerald to be undertaking an obligation not to harm her in the future. And, in addition, Hattie might also take Gerald to be reacting to and attempting to remedy some harm he’s already caused. Gerald’s utterance thus meets the conventions for two kinds of illocutionary acts. If Hattie treats these forces as necessarily entangled, this is not a case of multiple forces but rather what Sbisà calls a composite force. However, as Sbisà says, “it is easy for the audience’s response to shift from recognizing the performance of one composite illocutionary act to recognizing the performance of two illocutionary acts” (Sbisà 2013b, 238). This can occur when the addressee treats the utterance as a promise in some future contexts and an apology in some others (Sbisà 2001).12 This sort of case is likely fairly widespread and causes little to no conversational confusion. In many familiar cases, it seems we’re just comfortable with shifting illocutionary effects. Sbisà is not claiming that all of these cases are instances of utterances with plural forces. Regarding the first type of case she says, “insofar as the addressee or other participants, in manifesting their uptake of the speech act in their responses, select one illocutionary act as actually performed, the illocutionary act that is performed is only one” (Sbisà 2013b, 238). So, utterances that follow the conventions for more than one speech act only have plural forces if the conversational participants don’t select only one illocutionary act as performed. We’ll return to cases like this in the discussions of pluralism below. Regarding the second type of case, Sbisà  About this dependency on future uptake Sbisà (2001, 1801) says, “Such retrospective definitions of the deontic modal attributes of the speaker may merely confirm them, but they can also refine or alter them.” 12

  Some Varieties of Illocutionary Pluralism 

129

says, “[A] composite illocutionary effect is not therefore plural. But cases of this kind may come very close to a plural illocutionary performance, because it is easy for the audience’s response to shift from recognizing the performance of one composite illocutionary act to recognizing the performance of two illocutionary acts” (Sbisà 2013b, 238). If an utterance follows a procedure that is conventional for two (or more) illocutionary acts and generates a response for one illocutionary force in one context and the other in another context, then that utterance has multiple illocutionary forces. The third way an utterance might have more than one illocutionary effect has to do with indirect speech acts, or pluralism along the vertical axis. It occurs when an utterance is an indirect speech act, and the hearer’s response recognizes both the primary and secondary forces. Consider a typical indirect speech act: a speaker says, “Can you pass the salt?” and her addressee responds, “No, sorry, we’re out.” The “no” part of the response is elliptical for “No, I cannot” and is a response to the secondary force of the utterance—the question. The “sorry, we’re out” is an apology for failing to meet a request—and so is in response to the primary force. Each of Sbisà’s cases arises because of the ways a speaker and an addressee can tolerate or resolve illocutionary ambiguity. These remarks on the “issue of illocutionary pluralism” seem largely correct to me. However, Sbisà’s remarks do not constitute an account of illocutionary pluralism. And she makes clear that she does not take her sketch to exhaust the analysis of pluralism. To move forward in our understanding of pluralism, it will help to explore two more detailed accounts of illocutionary pluralism. In the next section I will explain my own pluralistic view, before turning to Lewiński’s more recent account.

130 

C. R. Johnson

4 Johnson’s Pluralism In each of Sbisà’s cases she focuses on conversation involving two conversers.13 There is a speaker and there is an addressee. Call this the simple conversation. To explain the kind of pluralism I’m interested in, I’ll consider more complicated conversations before returning to the simple conversation. In more complicated conversations a speaker makes an utterance in front of more than one other participant. I contend that all participants in the conversation can grant (or fail to grant) uptake in ways that are relevant to the illocutionary force(s) that the utterance has. I’ll say more about this below, but for now, let’s consider a conversation with three conversers. As before, in making an utterance, the speaker makes a move in the conversation. To be fully successful, the utterance must fulfill the requirements Austin gave us. As a reminder, these are: (A) The speaker must have the requisite authority; (B) there has to be a particular convention in the linguistic community for making the kind of force in question; (C) the speaker has to follow an appropriate procedure for making such a conversational move; (D) the speaker has to have the psychological state associated with the move in question; and (E) a participant must give the speaker uptake. Because of what we’ve learned from Sbisà’s work, we can now understand requirement E further—uptake involves the utterance having a conventional effect. And fully successful illocutionary force requires having the reaction that is conventional for the force in question. These are the conditions for fully successful illocutionary acts. But recall that these are not all necessary for the act to come off at all. Like Sbisà, I think that illocutionary acts can be performed even if they are not fully successful. And, like Sbisà, I think that an illocutionary act happens when a conversational participant’s perception of the deontic landscape is affected by an utterance the way that is conventional for some illocutionary act. The illocutionary act won’t, of course, be fully successful unless all the other conditions are met. That is, the act would go better (qua act of  She acknowledges that there may be other participants but does not dwell on those kinds of conversation. 13

  Some Varieties of Illocutionary Pluralism 

131

that type) if the speaker had the psychological state associated with the illocutionary force for which the participant’s deontic update was conventional. If, in other words, the speaker intends to ϕ, follows the procedure for ϕing, a participant updates their perception of the deontic landscape to reflect that she ϕ-ed. A mismatch between psychological state and uptake alone, though, does not mean that the illocution didn’t happen. An example may help. Consider a speaker, Iris, who utters “I’ll never do that again” but this time to two participants—Jackie and Kiana. Jackie responds in the ways conventional for a promise. Jackie takes Iris to be undertaking an obligation not to perform the action in question again. She takes Iris to be granting her a right against Iris’ doing so. Jackie also takes Iris to be granting her the authority to release Iris from this obligation. This effect on Jackie’s perception of the deontic changes makes Iris’ utterance a promise. Of course, I think Sbisà is correct that this could be complicated further. Jackie might react to Iris’ utterance as if it were a promise and an apology, but we can leave that aside for now. We don’t need Jackie’s reaction to be ambiguous to get my version of pluralism going. That’s because to get my version of pluralism going, all we need is for Kiana’s reaction to be different from Jackie’s. Imagine that Kiana reacts to Iris’ utterance as if it were a prediction. That is, she takes Iris to be undertaking an obligation to offer justification about the probability of her performing the action again. Jackie does not, however, take Iris to be undertaking such an obligation. Kiana reacts in the ways that are conventional for an illocutionary act but an illocutionary act that is different from the one that Jackie reacts to. And, if I’m right about illocutionary forces, Kiana’s reaction is sufficient to make the utterance be a prediction. So, in a case in which one participant reacts to the utterance in the ways conventional for promising, and another reacts in the ways conventional for predicting, I claim that the utterance is both a promise and a prediction. Better, the utterance is a promise from Jackie’s perspective, and a prediction from Kiana’s. Notice that if Sbisà is right about the second way an utterance can be multiple illocutionary acts, then this result should not be surprising. In Sbisà’s case a single addressee with more than one reaction made an

132 

C. R. Johnson

utterance more than one illocutionary act by toggling between reactions. I’m suggesting that two different participants have two different reactions that are more or less stable. If Sbisà’s case is one of illocutionary pluralism, the conversation between Iris, Jackie, and Kiana is as well. So much for the complicated conversation—what of the simple conversation? What if we return to a conversation between a speaker and a single addressee? It might seem that if the conversation were just between Iris, the speaker, and Jackie, her addressee, then we could only have the kind of pluralism Sbisà has in mind. That is, the only way to have multiple illocutionary forces for a token utterance in the simple conversation might seem to be if a hearer vacillates between multiple conventional reactions. I think this is not quite right. This is because speakers also react to their own utterances. So, imagine that Iris is only speaking to Jackie when she says, “I’ll never do that again.” Jackie reacts as if this utterance were a promise. Iris, however, reacts as if she’s made a prediction. The only obligation she thinks she has undertaken is an obligation to justify her prediction if asked. Her reaction is conventional for predictions. Iris might have this reaction because she’s aware of her intention to predict. She might intend to predict, make her utterance, and then react as if she’s successfully predicted. It might, however, float free from such an intention. Sometimes, I argue, we might intend some illocutionary act, and react as if we’ve made a different one. I might intend to suggest to my partner that he take out the trash, and then react as if I’ve criticized him once I’ve made my utterance (Johnson 2015, 2019). Whatever the relationship between the speaker’s intention and her reaction, the speaker has a reaction. This reaction is, like the reactions of the other conversers, indicative of illocutionary force. Because of this, in both simple and complicated conversations, there is an additional kind of multiplicity of illocutionary force—and therefore an additional kind of illocutionary pluralism. There are a couple of things to note, if I’m right. First, a speaker’s own reaction is sufficient but not necessary for her utterance to have some illocutionary force. If I react to my own utterance as to an assertion, then it is one relative to my perspective. Another converser might react to my utterance as an assertion even if I do not. And again, this is sufficient for

  Some Varieties of Illocutionary Pluralism 

133

me to have asserted relative to her perspective. This helps account for why children, or other speakers who may not be in a position to keep track of the state of the conversation can, nonetheless, make some illocutions. Perhaps children do not often count as promising, and almost never count as marrying, but we do sometimes perceive their utterances as assertions. And we can do so even if they do not yet have the concept of assertion, expectation, or obligation. Certainly, we sometimes have a strong intuition that a particular speaker asserted (or promised, etc.), no matter what anyone thinks. Sometimes we want to correct one another’s reactions. If illocutionary force is relative how are we to understand these intuitions? The answer, I think, lies in important facts about how we use language. These facts have figured in the above, but they warrant emphasis. While illocutionary forces are perspective-dependent, they also have weighty social consequences. There are often social costs when reactions differ,14 and we have, therefore, some motivation to align our reactions and expectations with others’. There are also social costs associated with relinquishing one’s opinion. If my reaction differs from yours, and I change it to bring it in line with yours, I admit to having miscounted—if only relative to my new position. I admit having made a mistake, from our now-­ shared perspective, in my understanding of the distribution of the obligations and permissions in the conversation. Because of these pressures, the costs of divergent reactions often motivate us to convince one another of the authority of our perspective.15 Similarly, sometimes our apparent disagreements over illocutionary force are actually disagreements over which social, practical, or political norms are operative on our conversational context. So, if you and I appear to disagree over whether or not an utterance was a joke, we might instead be disagreeing over whether or not that utterance should be treated as a joke. It might be that when I insist that a locution is a joke I’m actually recommending that we proceed as if the speaker was joking because it is  When reactions differ conversational participants might be confused, one participant might be out-grouped for diverging from convention, a participant might fail to be able to do what they want with their words. See Kukla (2014); Johnson (2020); Langton (1993) for examples. 15  There is even evidence that speakers who voice divergent perspectives are often identified as out group members—so perhaps there’s social pressure to align on that front as well (Sunstein 2000). 14

134 

C. R. Johnson

more practical, moral, or legal to do so. So we might be disagreeing about the best course of action rather than about the illocutionary facts. Also, notice what we do to develop intuitions about illocutionary force: we use our own impressions of the conversation—even if we were not participants. Say you’re telling me about a fight you had with your partner. In the midst of this fight your partner uttered, “Well, that’s because you’re just like your mother, isn’t it?” When I want to know if your partner’s utterance is an assertion or a question (or has some other force), I might inquire about your conversational history with your partner. I might ask about past conversations relating to you and your mother. I might inquire about tone. I might also inquire into how you responded or are disposed to respond to your partners’ locutions. In other words, I inform myself as to the state of the conversation, and then determine how to perceive the locution. By positioning myself to assess an utterance’s illocutionary status, I generate a reaction to that utterance. As a practiced converser, I have some justification for believing that others would share my reaction, as my reactions have aligned with others’ in the past. However, and as Sbisà points out, we are also familiar with disagreement over perceptions of token utterances. These disagreements can be fleeting, and resolved in the course of the discussion, or they can last, as they do in some conversations. One advantage of an account like mine is that all conversers can plausibly be right. A single locution might be perceived as an assertion, a conjecture, or some other illocution altogether, relative to different perspectives. There are a few things to notice in light of this. First, no perspective is authoritative in virtue of illocutionary facts. Recall that speakers themselves are affected in ways that are conventional (or not) for illocutionary acts. It is not as if my hearer has the only perspective on the state of the conversation. If he takes me to be obligated, and I do not take myself to be, then there is no reason to think that I am, thereby, ultima facie obligated. If my only hearer reacts to my utterance, “The window is open” as if it is an assertion, and I react as if it is a command, then there is a good chance that we will reach a conversational impasse sooner or later—and his attempt to elicit my justification or evidence will likely be met, at least at first, with confusion.

  Some Varieties of Illocutionary Pluralism 

135

The harder case is one in which many, all, or particularly powerful interlocutors differ from me in their reactions. In cases like this, I may well be persuaded to change my perspective because of practical considerations. One reason to retract an utterance is if I’ve asserted some content but cannot justify it—part of asserting is undertaking an obligation to retract if one is challenged and cannot offer justification. However, I may be given good reason to supply justification or retract my utterance, even if I do not take myself to have undertaken an obligation to do so with my utterance. If a thug or a henchman responds to my utterance “Is the cat on the mat?” with threats unless I can offer justification, then retraction seems like a good and practical option. But the motivation to retract, if there is one, does not come from the reactions of my interlocutors alone: it is due to the position they hold relative to me. No extra danger is undertaken by the difference in reactions that would not be undertaken by the holding of any minority opinion. Further, the power of the perspective of the majority affects how a speaker might change her own perspective, or her own behavior. However, it does not dictate the illocutionary facts. I might become convinced to update my perspective to reflect an assertion, to bring my perspective in line with that of the powerful party (presumably, this approximates one way in which we teach children how to individuate illocutions, e.g., how to tell a promise from a guess). Or, I might change my behavior, but maintain my original reaction. I might just act on what, from my perspective, is a bad reaction on the part of the powerful group. The second thing to note about the potential risks undertaken by speakers is that while reactions are relative to a perspective, they are not always under the conscious control of interlocutors. Conversations often (for better or for worse) proceed without much reflection on the part of the conversers. We do not, most of the time, think consciously about the state of the conversation unless something goes wrong.16 If the behavior of one of our interlocutors indicates that her conversational expectations have been violated, for example, we might pause and reflect. We occasionally make explicit the commitments and responsibilities we count  Even then, it can be difficult to put our finger on what exactly has gone wrong. See Johnson (2020) for examples. 16

136 

C. R. Johnson

speakers as undertaking, especially when we’re teaching people to converse. Most often, though, we react to a token utterance without reflecting on our reactions. We tend not to explicitly track for which illocutionary force our reaction is conventional.

5 Lewiński’s Pluralism Marcin Lewiński has developed a different detailed account of illocutionary pluralism. Lewiński describes the pluralism he has in mind as follows: it is the “intentional performance of a plurality of conventionally recognizable illocutionary acts via the same utterance token in one unique speech situation” (Lewiński 2021, 6692). As Lewiński argues, these performances can happen in conversations between two agents and in more complicated conversations as well. The distinctive features of the pluralism that Lewiński describes are that (a) the plural forces are always intentional, and (b) the plural forces are not dependent on uptake. One thing to note about my pluralism is that utterances with multiple illocutionary forces of the kind I’ve highlighted, are rarely fully successful. This is because there is often a mismatch between the speaker’s intention and the uptake the utterance receives. For Lewiński, this is not the case—sometimes speakers intend multiple forces. In Lewiński’s target cases of multiple forces for a token utterance, the speaker intentionally takes advantage of our ability to navigate conversations involving pluralistic forces. In the cases that Lewiński uses to motivate his pluralism, the speaker is either strategically ambiguous or illuminating a conversationally relevant decision point. In cases of strategic ambiguity, the speaker uses illocutionary ambiguity to avoid committing to a particular force. An interviewer for a job might, for example, be strategically ambiguous in saying, “I’m sure I’ll see you soon,” to a candidate. The speaker knowingly leaves it ambiguous whether she is predicting, reassuring, promising, and so on. As Lewiński puts it, this strategy allows the speaker to “maintain this ambiguity, thus keeping the space of illocutionary potential – and the related hopes, entitlements, and obligations – wide open” (Lewiński 2021, 6701). In cases involving decision points—what Lewiński calls “dilemmatic deliberations”—the speaker

  Some Varieties of Illocutionary Pluralism 

137

might use a single token speech act to illuminate various courses of action available in the conversational context. Here is Lewiński’s example: Two biologists are collecting samples from under the surface of a frozen polar lake. Thin, transparent ice means more precise data can be collected. But thin ice is also perilous. So, the scientists are facing one of the many data collection dilemmas (real good data are hard to come by). Additionally, the head of the project, John, explicitly advised them (was it just advice?) in the last project meeting to always consult him if they face some dilemma: walking over thin ice or otherwise. While sitting in their cabin-cum-lab, the following exchange emerges: Biologist 1: Shall we go out to collect samples? Biologist 2: The ice is thin. (Lewiński 2021, 6699)

Lewiński claims that Biologist 2 is using her indirect speech act to convey several different, equally likely, primary forces. She might be encouraging data collection, warning her research associate about dangers, or recommending that they defer to John.17 Each of these illocutionary forces produces different conventional changes in the deontic statuses of the conversational participants. The utterance has, then, incompatible, but nonetheless plausibly intentional, multiple forces. This case has quite a bit in common with Sbisà’s cases above. The speaker here, however, is deliberately taking advantage of the fact that distinct illocutionary forces can share a conventional profile. The force might be disambiguated by Biologist 1’s uptake, or by their future deliberations, or it might remain ambiguous. This case is also distinct from Sbisà’s cases because the illocutionary forces that Biologist 2 intends are far apart in their normative effects. Sbisà’s examples are of more or less compatible speech actions—proposing and offering, praising and complimenting—but Lewiński’s are not. Their incompatibility ensures that Biologist 2 cannot be correctly understood as, say, encouraging “by way of ” warning. If that were the case, the plurality of forces observed might be able to be modeled vertically. Instead, this kind of case demands what Lewiński would call a genuinely pluralist (horizontal) treatment. It would  Lewiński has the third illocutionary force here as a deferral, but I’m not convinced that’s a distinct illocutionary force. I don’t think much hangs on this disagreement. 17

138 

C. R. Johnson

also be surprising if Biologist 2 were making an utterance that could remain disambiguated—there is likely to be practical pressure to resolve the ambiguity in order to come to a decision. Lewiński also discusses pluralism in what I’ve called complicated conversations—those with more than two participants.18 However, unlike in Iris’ case of unintentional plural forces discussed above, Lewiński specifies that the speaker is intentionally making an utterance that is conventional for more than one force. And, because of contextually relevant facts about the other conversational participants, the speaker intends the participants to react in ways that are conventional for those different forces. As Lewiński (2021, 6703) puts it, in this kind of case, different communicatively intended illocutionary forces are directed at different audience members with the goal of producing different “conventional effects” for each of them. That is, the speaker is ready to stand by the illocutions conveyed and the hearers’ subjective interpretation of them has no decisive role to play.

So, for Lewiński, a speaker can produce an utterance with multiple illocutionary forces by intending to produce non-identical illocutionary effects in her hearers.19

6 Conclusion I will close by drawing out three details of the relationships between Sbisà’s remarks and these two more developed pluralisms. First, contrary to Sbisà’s view and in contrast to my own account, Lewiński denies that generating multiple uptakes is required for an utterance to have more than one force. For him, it is important that the speaker expects or

 What Lewiński calls “polylogues.”  In some places Lewiński says that the speaker intends to produce a different effect in each hearer, but this one-to-one expectation is not necessary—it is enough that the intended effects are non-­ identical. In other places, Lewiński recognizes this. 18 19

  Some Varieties of Illocutionary Pluralism 

139

intends multiple uptakes.20 And it is important that the speaker intends his hearers to recognize these intentions21 (though presumably it is not necessary that the speaker intend the hearers to recognize the plurality of illocutionary intentions). While I don’t have space to adjudicate this disagreement here, I have discussed some of the costs and benefits of basing force on speaker intention in other work (Johnson 2019). Second, other than the disagreement over the importance of uptake, Lewiński and my views are largely compatible. Lewiński’s observation that speakers sometimes intentionally exploit conversational participants’ tolerance of ambiguity is a good one. My own account is largely silent on speaker intention for reasons gestured at above, but this does not mean that my account rules out intentional performance of plural illocutionary acts. In that regard, Lewiński and I are highlighting different examples in which illocutionary pluralism is explanatory. Third, these ways of developing illocutionary pluralism helpfully illuminate important ways that conversers use language. Sometimes these uses are cooperative and sometimes they are antagonistic or even oppressive. In this, both of these accounts contribute to what Mary Kate McGowan calls “the linguistic approach to group-based injustice” (McGowan 2019). Lewiński uses his pluralism to help explain Dog Whistles (Lewiński 2021). I’ve used mine to help explain some kinds of Mansplaining (Johnson 2020). This suggests that it is worthwhile to develop illocutionary pluralism in some form, especially as we try to account for language in all of its irreducible complexity.

 In his discussion, Lewiński contrasts his view with the illocutionary relativism he attributes to Sbisà and myself (Lewiński 2021, 6703). 21  This is deliberately and explicitly echoing Grice: “Two crucial conditions for such illocutionary moves are that: (1) via various hearers’ uptake, they are expected to take distinct conventional effect; (2) the speaker intends his intention to illocute x, y, z to different speakers to be recognized as intended: by the exclusive subset of participants, by the inclusive subset, or by all the participants involved. These plural illocutionary forces are thus not only conventionally recognized but also intentionally performed  – in the sense of reflexive communicative intentions” (Lewiński 2021, 6706). 20

140 

C. R. Johnson

References Austin, John L. 1975. In How to Do Things with Words, ed. James O. Urmson and Marina Sbisà, 2nd ed. Cambridge, MA: Harvard University Press. Bird, Alexander. 2002. Illocutionary Silencing. Pacific Philosophical Quarterly 83 (1): 1–15. Cappelen, Herman, and Ernest Lepore. 2008. Insensitive Semantics: A Defense of Semantic Minimalism and Speech Act Pluralism. New Jersey: John Wiley & Sons. Grice, Paul H. 1975. Logic and Conversation. In Syntax and Semantics, Vol. 3: Speech Acts, edited by Paul Cole and Jerry Morgan. New York: Academic Press. Reprinted in Studies in the Way of Words (1989), 22–40. Cambridge, MA: Harvard University Press. Johnson, Casey Rebecca. 2015. Speech Acts and Silencing: A Social Account of Speech Action and Restrictions on Speech. PhD Dissertation, University of Connecticut. ———. 2019. Investigating Illocutionary Monism. Synthese 196 (3): 1151–1165. ———. 2020. Mansplaining and Illocutionary Force. Feminist Philosophy Quarterly 6 (4). Kukla, Rebecca. 2014. Performative Force, Convention, and Discursive Injustice. Hypatia 29 (2): 440–457. Kukla, Rebecca, and Mark Norris Lance. 2009. “Yo!‘and’Lo!”: The Pragmatic Topography of the Space of Reasons. Cambridge, MA: Harvard University Press. Langton, Rae. 1993. Speech Acts and Unspeakable Acts. Philosophy & Public Affairs: 293–330. Lewiński, Marcin. 2021. Illocutionary Pluralism. Synthese 199: 6687–6714. McGowan, Mary Kate. 2019. Just Words: On Speech and Hidden Harm. Oxford: Oxford University Press. Sbisà, Marina. 2001. Illocutionary Force and Degrees of Strength in Language Use. Journal of Pragmatics 33 (12): 1791–1814. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2007. How to Read Austin. Pragmatics. Quarterly Publication of the International Pragmatics Association (IPrA) 17 (3): 461–473. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2013a. Locution, Illocution, Perlocution. In Pragmatics of Speech Actions, ed. Marina Sbisà and Ken Turner, 25–75. Berlin: De Gruyter.

  Some Varieties of Illocutionary Pluralism 

141

———. 2013b. Some Remarks about Speech Act Pluralism. In Perspectives on Pragmatics and Philosophy, ed. Alessandro Capone, Franco Lo Piparo, and Marco Carapezza, 227–244. Heidlberg: Springer. Searle, John R. 1975. Indirect Speech Acts. In Syntax & Semantics, 3: Speech Act, ed. Peter Cole and Jerry L. Morgan, 59–82. New York: Academic Press. Searle, John R., and Daniel Vanderveken. 1985. Speech Acts and Illocutionary Logic. In Logic, Thought and Action, ed. Daniel Vanderveken, 109–132. Dordrecht: Springer. Sunstein, Cass R. 2000. Deliberative trouble? Why groups go to extremes. The Yale Law Journal 110 (1): 71–119. Tirrell, Lynne. 2018. Authority and Gender: Flipping the F-Switch. Feminist Philosophy Quarterly 4 (3): 1.

Speech Acts and Ventriloquation: The Contribution of Marina Sbisà to a General Theory of Action and Performativity François Cooren

1 Introduction For the past 50 years, Marina Sbisà has patiently shown us why it was worthwhile calling into question the standard conception of speech act theory to go back to Austin’s (1975) original ideas, as published in his posthumous opus, How to Do Things with Words (see Sbisà 1972). This return to the Austinian basics was far from easy, as Sbisà had to fight against a well-established reappropriation of Austin’s work, a reappropriation especially represented by the work of one of his most famous students, John Searle (1969, 1979). I would suggest that this patience is ultimately paying off, as Searle’s (1983) intentionalist perspective tends today to fall into disuse while Austin’s focus on the social effects of illocutionary acts keeps inspiring scholars interested not only in the detailed

F. Cooren (*) Department of Communication, Université de Montréal, Montreal, QC, Canada e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Caponetto, P. Labinaz (eds.), Sbisà on Speech as Action, Philosophers in Depth, https://doi.org/10.1007/978-3-031-22528-4_7

143

144 

F. Cooren

study of language and social interaction, but also in questions of performativity, more generally (Allard-Poesi and Cabantous 2021). But beyond this work of rehabilitation, it is her intellectual audacity that has always fascinated me the most. While most philosophers of language hardly venture outside of their sub-discipline, Sbisà never hesitated to mobilize ideas coming from other fields of study, such as narrative semiotics (Greimas 1987), conversation analysis (Schegloff 1991), or linguistics (Gazdar 1981). What could, at first sight, be denounced as a form of adventurous eclecticism was always meant to contribute to a better understanding of the various topics she addressed throughout her prolific career, whether these topics concern verbal interaction (Sbisà and Fabbri 1980), the discourse of childbirth (Sbisà 1996), or, of course, speech act theory itself (Sbisà 1984, 2013). In keeping with her Oxonian mentor, Sbisà is indeed an explorer, an explorer who keeps questioning well-established dogmas when these appear ill equipped to account for the phenomena that catch her attention. It is in this spirit of exploration that I would like to initiate a dialogue with Sbisà’s ideas throughout this chapter. In many respects, this exercise will give me the opportunity to acknowledge my intellectual debt to her, even if she might not always agree with everything I say. After having explored how Sbisà invites us to reinterpret the key notion of action, I will show how this reinterpretation leads us to an original way of conceiving of speech acts, which, according to this view, can be ascribed not only to human beings, but also to texts as well as other-than-humans in general. As I will try to demonstrate, Sbisà’s positions lead us, in my opinion, to recognize what I call the ventriloquial dimension of communication (Cooren 2010, 2020), which invites us to put forward a general theory of action and performativity.

2 Action: What’s in This Word? For most analytical philosophers (and I would add, for most scholars that I know, whether they come from communication, linguistics, sociology, or anthropology), the word “action” appears to be a technical term that should be used to refer to intentional acts and intentional acts only

  Speech Acts and Ventriloquation: The Contribution… 

145

(e.g., Bach and Harnish 1979; Davidson 1980; Searle 2002; Weigand 2021). This dogma is all the more surprising given that we usually do not hesitate to use the expression “unintentional acts” or “unintentional actions”1 to refer to situations where one is stepping on someone’s toe or insulting someone without meaning to (in this case, Davidson 1980 curiously speaks in terms of event instead of action). The very fact that we can talk about these acts as being unintentional should however suffice to demonstrate that such a restriction in the extension and intension of this concept is theoretically dubious and rather useless, practically speaking. Furthermore, and as I have tried to show elsewhere (Cooren 2000, 2010), many philosophical puzzles concerning a theory of action can be solved when intentionality is shown to be admittedly a key aspect of some specific actions, but not at the command of everything that is happening in a given situation. It is therefore not by chance that one of the key moves that Sbisà makes in her careful reinterpretation of Austin (1975) is to show that “all speech should be considered as action” (Sbisà, 2007, 462), where action does not necessarily refer to a doing that is performed intentionally. While Searle (1969, 1979) and his followers (Bach and Harnish 1979) keep analyzing speech acts in terms of intentionality, Sbisà (1984, 94) indeed points out that Austin shows a tendency towards 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.

This idea of social recognition is crucial as it positions Austin’s efforts in an attempt to understand the social dimension of speech acts, an aspect that will be examined later. However, if speech acts should first and foremost be analyzed in terms of action, one then needs to define what this term specifically means. This  For the sake of this discussion, and in keeping with Sbisà’s (2007) position, I will not create artificial distinctions between the words “acts” and “actions” even if some subtle nuances can, of course, always be introduced. 1

146 

F. Cooren

is where Sbisà (1984, 1985, 2003) again innovates by borrowing from the semiotician Algirdas Julien Greimas (1987; Greimas and Courtés 1982) his way of conceiving action, that is, “the change from one state into another, the responsibility for which is assigned to an agent” (Sbisà 1984, 94). For Greimas, “action” is indeed not a term that should necessarily be used to exclusively refer to an intentional act. On the contrary, this term can be used to depict not only what humans intentionally or unintentionally do, but even what other-than-humans do. Greimas (1987) was indeed especially interested in how narratives are structured, which led him to account for the various events and courses of action that make a story what it is. In any story, as he noted, human beings are often depicted as doing things, but these humans are also confronted to other-than-humans that partake in the development of the narrative. What are these other-than-humans? They can be of various ontologies and this is what makes narrative analysis so interesting. It can be a parchment that informs the heroine about the location of a secret passage; it can be an emotion that takes over one of her allies; it can be a magnetic stripe card that opens the door of a laboratory; or it can be a fake ID that authorizes the main character to penetrate a well-guarded compound. Although most philosophers would retort that these verbs—informing, taking over, opening, and authorizing—should not be considered as depicting actions in these specific contexts, Greimas (1987) would retort that they actually refer to transformations of state attributed to things (for lack of a better term) that can make a key difference in the course of events. For instance, without the parchment that she happens to find, the heroine of the story might be deprived of key pieces of information allowing her to move forward in her quest. Similarly, the emotion that is taking over one of her allies might have important consequences if this ally proves to be paralyzed by fear when he is supposed to intervene at some specific point. Although actions can definitively be intentional (looking for a document, explaining something to someone, telling an employee what to do, opening a session, etc.), the English language (as well as most languages I am aware of ) allows us to use the terms “action,” “acts,” or their equivalents to speak about things that other-than-humans do. For instance, the

  Speech Acts and Ventriloquation: The Contribution… 

147

Webster’s New Encyclopedic Dictionary, which lists the various usages of this verb, does not hesitate to speak about “the action of acid on metals” or “the action of a firearm” (Cooren 2010, 16). When Sbisà (2013, 231) invites us to conceive of action “by reference to what it does or what it brings about, that is, by reference to its effects,” she is therefore acknowledging the multiple usages of this key notion without restricting it to the sole intentional acts of human beings.

3 Speech Acts as Specific Types of Action But what are the consequences of this definitional move on Sbisà’s (1984, 2013) part? If we go back to the examples that I gave above in this chapter, we note that some of the actions that were mentioned can be identified with what would typically be called speech acts: a parchment informs the heroine about the location of a secret passage or a fake ID authorizes her to penetrate a well-guarded compound. In keeping with Austin (1975), Sbisà (2013, 242–243) writes: On my account, an action is the bringing about of a change in the world, a speech act (broadly intended) is the production of any such change on the basis of verbal behavior, and an illocutionary act is the production of a change in the conventional aspects of the interpersonal relationship among the participants.

Speech acts, according to Austin (1975), indeed include locutionary acts (which themselves include phonetic, phatic, and rhetic acts), illocutionary acts and perlocutionary acts (while Searle 1969 reduces, as Sbisà 2019 points out, speech acts to illocutionary acts). So, to what extent can we say that the sentence “a parchment informs the heroine about the location of a secret passage” describes a speech act? First, it is noteworthy that the term “speech act” is, to some extent, a misnomer in this context, given that we are not dealing, strictly speaking, with speech here, but with texts and discourse, more generally. The term “discursive act” could be considered more encompassing. If we then focus on the action that is taking place, we then note that the heroine has to

148 

F. Cooren

read the parchment in order to be informed of the location of a secret passage. This is an important point as it aligns with what Sbisà and Fabbri (1980) call “two-place model of speech acts (M2),” that is, a model that takes into account not only the speaker (here, the author of the parchment, who should rather be identified as a writer), but also the hearer (here, the reader). What Austin would call a locutionary act (the act of saying something) consists here in the writing of something on the parchment. While in the case of sound production, Austin talks about phonetic acts, one should rather speak here in terms of ichnoic acts (“ichno,” in ancient Greek (ἴχνος), means “trace”) performed by the author of the parchment. Given that these traces consist here in words “belonging to a certain vocabulary, in a certain construction, i.e., conforming to and as conforming to a certain grammar” (Austin 1975, 92), this author also performed what Austin would call a phatic act (the act of producing a pheme if by pheme we mean what is said, an expression that can also be sometimes used, as we know, to speak about what is written). Finally, since we can describe what happened by saying that this author wrote where the secret passage was located, one can also say that this person performed a rhetic act. In other words, he or she produced a rheme, that is, an utterance “with a certain more or less definite ‘sense’ and a more or less definite ‘reference’ (which together are equivalent to ‘meaning’)” (Austin 1975, 93). Although the phonetic/ichnoic, phatic, and rhetic acts are usually neglected in speech act theory, Sbisà (2013, 231) is right to point out that each can be considered as an act, or even an “action”, to be identified by reference to what it does or what it brings about, that is, by reference to its effects, for which the speaker can be ascribed responsibility. So, the locutionary act (the lowest step in the locutionary-illocutionary-perlocutionary distinction) is already one and many.

For a communication scholar like me, this is an important point, since it implies that any illocutionary act has, by definition, first to be materially produced under the form of a locutionary act, which itself can be described as a phonetic/ichnoic, phatic, and rhetic act. Communication

  Speech Acts and Ventriloquation: The Contribution… 

149

is indeed the establishment of a relation between two entities, which means that this relation has, by definition, to somehow materialize in order to be established. This is what a locution does or performs. A locutionary act can therefore be performed in producing sounds or written traces that belong to a certain vocabulary, have a sense and refer to something, but also in producing gestures or face expressions that conventionally mean something too (for instance, a finger pointing to a wall that would (illocutionarily) indicate where a secret passage might be located). What matters in this analysis is that any illocutionary act has, in order to be performed, to first take the form of a sign or series of signs that can be conventionally recognized by others. As we will see, this materialization and recognition are essential for the functioning of speech acts. If we now focus on the illocutionary act itself, we realize that the act of informing someone consists in a transformation of state (Cooren 2000): the heroine is given information about the location of a secret passage. In this regard, Sbisà (2007, 464–465) points out that the conventional character of illocutionary acts “comes from the conventionality of their effects, rather than from the conventionality of the means by which they are performed.” I will come back to this specific point, which constitutes an area of disagreement between the two of us, but she also adds that illocutionary acts can be seen as “bringing about change in interpersonal relationships” (Sbisà 2013, 233), which is indeed a crucial point. As we can see in this example, through this action, the heroine indeed becomes informed of the secret passage, while the author becomes an informer, whether this person meant to be one or not. What could be called an illocutionary relation between the author and the heroine has therefore been retrospectively established through this specific action. This analysis abides by Austin’s (1975) conception of illocutionary acts, when he notes that these types of acts are associated with three different kinds of effect. Sbisà (2001, 1796) summarizes these effects by listing them as follows: The securing of uptake The production of a conventional effect The inviting of a response or sequel.

150 

F. Cooren

As pointed out by Sbisà (2001, 1796), effect (i), which Austin (1975) calls uptake, “amounts to bringing about the understanding of the meaning and the force of the utterance.” In our case, the heroine understands that the parchment informs her about the location of a secret passage. As for effect (ii), it “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” (Sbisà 2001, 1796). In our case, the heroine can now be considered to be (conventionally) informed of the potential location of a secrete passage. In being informed, she can indeed rely upon the content of the speech act of informing in her subsequent verbal and non-verbal behavior. For example, she can (i.e., would be entitled to) tell that content to others, or use it as a premise in reasoning or as ground for decision making. Finally, effect (iii) “amounts to inviting a certain kind of subsequent behavior” (Sbisà 2001, 1796). In our case, this information can also be interpreted as an invitation to look for the location of this passage. In keeping with Austin (1975), Sbisà (2001, 1796) points out that the difference between illocutionary and perlocutionary acts is that “perlocutionary acts are the bringing about of changes in the natural course of events” while illocutionary acts are said to be conventional in their effects. In this regard, Sbisà (2007, 465) writes: 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 state 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.

In our example, if the actual location of the secret passage proves not to be where the author of the parchment says it is located (or if this secret passage simply does not exist), the new state of affairs, that is, the fact that the heroine has been informed about the location of the secret passage, could be considered null and void to the extent that she has, in fact, not be informed of this location. She was therefore misled or misinformed.

  Speech Acts and Ventriloquation: The Contribution… 

151

Note, however, that to nullify an act still means that something took place, which could still be identified as belonging to “the natural course of event” (an expression that, I have to say, is far from being clear). If this information happens to be fake, one could still say that the heroine was given a (false) information or misinformed, which corresponds to effect (i) (see also Caponetto 2020). In other words, until the uptake included, which is also conventional, we remain in courses of action that are not defeasible (what Austin would therefore call a natural course of action), while as soon as we consider effect (ii), we deal with courses of action that are indeed defeasible: the heroine thought that she had been informed about the location of the secret passage, but she realizes retrospectively that this is not the case. Similarly, one could think of situations where someone believes that he has been given authorization to penetrate a well-guarded compound, but realizes retrospectively that this is not the case (his authorization proves to be invalid). If we consider effect (iii), we also note that the inviting of a response or sequel also takes place and cannot be defeasible per se. Whether this information proves to be reliable or unreliable, giving information or an authorization conventionally implies that something be done with it. This effect cannot per se be defeasible to the extent that it cannot be considered as not having taken place even if the information proves to be false or the authorization proves to be fake. A better way to distinguish illocutionary from perlocutionary acts would consist, in my opinion, in noting that perlocutionary acts involves the recipient’s freedom of (re-) action (Cooren 2000, 2008b, 2015a). When the heroine understands that some information has been given to her (effect (i)), that she has therefore been informed of the (potential) location of the secret passage (effect (ii)), and that she is implicitly invited to do something about it (effect (iii)), she is not free not to react the way she did, as these effects are conventional, that is, their existence doesn’t depend on the heroine’s freedom of (re-)action: This is what giving information conventionally means in terms of effects, whether she likes it or not. However, perlocutionary effects take place as soon as her freedom of (re-)action is at stake. Will she do something with this piece of information? Will she actually go to the location indicated in the parchment? In terms of perlocution, one could for instance retrospectively say that the

152 

F. Cooren

author of the parchment led her to the wrong location by giving the heroine some false information. Or we could, on the contrary, say that this author led her to the proper location of the secret passage if the information proves to be reliable.

4 Textual Agency But something else is happening in this example, which is that the parchment itself can also be seen as an agent. When we describe what is taking place in this situation, we can, of course, say that the author of the parchment is informing the heroine, but we can also singularize the action of the parchment itself and position it as informing the heroine too. As Derrida points out in his comments on Austin’s speech act theory, to write is to produce a mark that will constitute a sort of machine which is productive in turn, and which my future disappearance will not, in principle, hinder in its functioning, offering things and itself to be read and to be rewritten. (…) For a writing to be a writing it must continue to “act” and to be readable even when what is called the author of the writing no longer answers for what he has written, for what he seems to have signed, be it because of a temporary absence, because he is dead or, more generally, because he has not employed his absolutely actual and present intention or attention, the plenitude of his desire to say what he means, in order to sustain what seems to be written “in his name”. (Derrida 1988, 8; my italics)

In other words, the parchment can also be seen as an agent to the extent that it does make a difference in this specific situation. Maybe the author of the parchment did not mean to inform the heroine of the location of the secrete passage, but this ultimately does not matter in this context, as this is what she or he ends up doing through the text she or he wrote in the parchment. This is what I have proposed to call elsewhere “textual agency” (Cooren 2004, 2008a, b, 2009), an idea that was also recently taken up by Sbisà (2017, see especially, 32–33). Although speech act theory as we know it today does not take into account this possibility, we see that Austin’s

  Speech Acts and Ventriloquation: The Contribution… 

153

(1975) actional perspective, as highlighted by Sbisà (1984), allows us to acknowledge the action texts (and, as we will see, signs, more generally) can be said to perform if we decide to singularize their contribution. Skeptics could retort that the parchment does not really inform the heroine of anything, as they would point out that it is its author who is actually performing this action. To this objection, I would respond that my point is not to eliminate the author of the parchment from the equation, but to acknowledge what the parchment also does. As pointed out by Derrida, to write is to produce a mark or series of marks that, whether we like it or not, will act in our name or for us. Again, this is an important point for a communication scholar as it highlights the transmissive or disseminative nature of communication (Krämer 2015; Peters 1999, 2015). Whenever people communicate with each other, they have, by definition, to produce sounds, traces, face expressions, gestures, and so on that have to travel from one point to another (through the form of waves in a specific medium—air or water— in the case of sounds; through the form of electromagnetic radiations called light in the case of traces, face expressions or gestures). As Latour (1996) points out, whenever one acts, others proceed into action, which means that action is always something that is shared with others (Cooren 2000). While the agency of a parchment is relatively easy to acknowledge (although some would not hesitate to deny its performative contribution), Derrida (1988) invites us to note that a certain graphematic logic also applies to speech. In other words, whenever we pronounce words in order to do something (asking someone for information, telling someone what to do, committing oneself, opening a session, apologizing, etc.), these words can make us say things that we did not necessarily mean to say. This concretely means that the words we pronounce have a form of agency that can sometimes surprise us. Most of the time, they seem to do what we believe or hope they will do, but sometimes they can do things that we had not anticipated. For instance, I could ask someone, “Do you know who is coming to our meeting tonight?” and mean my question as a simple request for information, that is, as a polite way to ask this person who is coming to a meeting that we are both supposed to participate in. However, my

154 

F. Cooren

interlocutor could hear what I said as what conversation analysts call a preannouncement (Schegloff 1988), that is, as a sequence that is usually meant to not only check whether my interlocutor knows about some news I am about to announce, but also create in him or her some form of expectation regarding what I am about to say (Cooren 2005). For this interlocutor, “Do you know who is coming to our meeting tonight?” therefore functions as a preannouncement and not as a request for information, a phenomenon that highlights the agency of the locution that was produced. This locution can indeed be conventionally used in English (as well as in French, Spanish, Italian, Portuguese) to both check whether someone knows about some news one is supposedly about to announce (preannouncement) or ask for a piece of information. Hence, the misunderstanding that can result from the possibility of this dual interpretation (even if this misunderstanding can then be subsequently repaired). Although the role our locutions play in communication is often ignored, their agency or action can be more easily acknowledged when things don’t work as expected. Most of the time, we tend to understand each other, which means that the locutions, utterances, or (more generally) signs we produce happen to do what they were expected to do, but sometimes we realize, to our own detriment, that what we said or wrote made us illocutionarily do things we did not mean to do (for instance, we are all familiar with situations where a simple observation such as “The apartment is quite messy, don’t you think?” can be heard as a reproach even if we did not mean it to be). To the extent that any form of communication is mediated (even so-called unmediated face-to-face interactions), any sign or string of signs we produce to communicate can itself be singularized as illocutionarily doing something that can be (correctly or mistakenly) ascribed to their producers. Derrida thus helps us understand the conventional character of speech acts, a conventionality that can be recognized not only in the effects, but also in the means used to communicate (a position that constitutes an area of disagreement between Sbisà and myself ). Because of the conventional character of the signs we produce, these signs benefit from a certain autonomy, even if this autonomy is, of course, never absolute. This autonomy is the condition of possibility of communication, but it is also, to some extent, the condition of its impossibility, as Derrida (1988) rightly

  Speech Acts and Ventriloquation: The Contribution… 

155

notes. Possibility because without the conventional character of speech acts, we could not understand each other. Impossibility because this conventional character—or relative autonomy—always puts us in a position where the locutions we produce can make us say things we did not mean to say. Recognizing this conventionality and autonomy also means that the role a context plays in any interaction should be reconceptualized. A context, according to this approach, indeed consists of aspects of a situation that can add elements of information or indications about what is actually performed, which then allows us to determine what is happening. Irony is an interesting case in this regard, as it shows that the conventional character of what is said has to be acknowledged for the irony to function (Cooren 2010). Imagine a cocktail party where someone has just openly made an obnoxious remark about one of the guests in front of everyone (e.g., “You look awful today, Larry!”). The person who has been the object of this comment then responds, “Thanks for the compliment, Charles!” As we see, this response consists in at least one illocutionary act, which can be identified as thanking Charles for what is presented as a compliment. Why can this response function as being ironic or even sarcastic? Because the situation the guests are facing speaks for itself to some extent (hence, its conventionality). All the guests who indeed heard Charles’s comment have probably understood it as an obnoxious comment about Larry, especially if Larry does indeed look awful (if this were not the case, e.g., if Larry was, in fact, impeccably dressed up, then Charles’s comment could have itself be heard as ironic, that is, as being meant to actually compliment Larry). As we see, the irony thus comes from the disconnection or contra-diction between what the situation obviously tells us about what happened (Larry looks awful and Charles mentions this openly to everyone) and what someone—in this case, Larry—says about this situation, which appears to be the opposite of what this situation (obviously) tells us. In saying “Thanks for the compliment, Charles!” Larry not only identifies what Charles just said as being a compliment (which everyone obviously understood as not being the case), but also thanks him for it, two illocutionary acts that can be heard as ironic because they appear to

156 

F. Cooren

(obviously) contradict what is the case and what actually happened just before. In order to reconcile this contra-diction, the participants therefore have to hear what Larry says as being ironic, that is, as saying the opposite of what he is saying: He is, in fact, not thanking Charles for his comment. On the contrary, he is deploring it, at best. What Larry is doing therefore contradicts what his locution is supposed to do. A context then communicates information to us conventionally (which, of course, does not mean that we all have to agree about what it tells us), which explains why we are talking about a logic of addition. To the performativity of the speech acts people produce, we therefore have to add the performativity of the context, which tells us things about what is happening and what should be done about it (Cooren and Matte 2010). For instance, the context of a hierarchical relation between a boss X and her secretary Y tells us why “Would you be kind enough to give me this file?” should be heard not only as a very polite request issued by X to Y, but also, and maybe especially, as a request that cannot be turned down by Y. Why? Because this context tells us (as observers) and the participants (X and Y) that as long as what Y is requested to do belongs to his task description, he is normally in no position to turn it down (even if this is what Bartleby, the scrivener, does not hesitate to do in Melville’s famous short story).2

5 Speech Acts and Ventriloquation This reflection on (con-)textual agency thus leads us, as we see, to acknowledge the agency of texts and locutions, but more generally the agency of all the signs we produce and mobilize and to which we react in our daily life. As we saw, a parchment speaks to the heroine, for instance, by giving her information about what to do next. As already mentioned, this performativity does not prevent us from acknowledging that this  I always hesitate calling a request that cannot be turned down an order. An order indeed usually consists of giving an authoritative instruction to do something, which is not the case when someone politely says, “Would you be kind enough to give me this file?” Maybe the verb “instruct” would be better in this case, as instructing is not as strong as ordering even if it is supposed to mark that what is requested cannot or should not be turned down. 2

  Speech Acts and Ventriloquation: The Contribution… 

157

parchment was also and obviously written by its author, an author who can therefore also be envisaged as speaking to the heroine through this document. Similarly, a fake ID that authorizes the same heroine to penetrate a well-guarded compound is also performing something when she is showing it to a guard or swiping it through a security system. This (con-)textual autonomy is even more obvious in the case of institutional documents, such as laws, contracts, forms, certificates, policies, and so on (Cooren 2000, 2010), what Sbisà (2017) also calls “normative texts.” For instance, why are we committed by the contracts we have signed? Because these contracts, whether we like it or not, are supposed to commit their signatories. One could, of course, retort that the agency of contracts depends on the legal systems that assure their validity and ultimately on the official judgments that would define what they are supposedly saying. My point, however, is not to deny that the agency of contracts depends on its recognition by other agents, especially human beings, as we saw with Sbisà (2007) that action, in this case, needs to be recognized in order to be considered as having taken place. (Hence, the principle of defeasibility that she puts forward.) So, the question then becomes: How can we recognize the agency of things such as texts, locutions, and signs in general while still recognizing the agency of the human beings who participate in the performance of these actions? A solution I have proposed consists of acknowledging what I call the ventriloquial dimension of communication (Cooren 2010, 2012, 2015b, 2016, 2020; Sbisà 2011). What interests me in this metaphor is the phenomenon that consists in making someone or something speak, which is, I believe, constitutive of any act of communication. As mentioned previously, and as acknowledged by Austin (1975) and Sbisà (1984) regarding speech acts, any act of communication is, to some extent, an act of tele-communication (Cooren 2000), that is, an act from a distance. Even in so-called unmediated face-to-face conversations, we are indeed relying on the transmission of sounds and photons to communicate verbally and nonverbally, respectively, which means that these sounds that we produce, the postures we hold or the gestures and face expressions that we make are, whether we like it or not, our delegates (they travel through specific mediums, that is, the air that we breathe for the sounds we produce with our mouths and the electromagnetic field for the

158 

F. Cooren

photons that bounce from our body or from the texts we write, creating specific images on our retinas). In other words, these signs make us say things as much as we make them say things. In this specific case, I propose to speak in terms of downstream ventriloquation. The point of reference here is the agent who is producing or emitting the signs, which are then identified as being metaphorically located “downstream” if by stream, we mean the flow of communication. This agent can be, of course, a human being, but it can also be, as previously seen, a parchment, a contract, a policy, and so on. It can even take, and this is an important point, the form of natural phenomena. For instance, smoke, through its mere presence, indicates that there is probably a fire located somewhere at its source, an indication that, as we see, relies on a phenomenon of double ventriloquation: (1) this smoke indeed makes us say/think3 that there is probably a fire, as much as (2) we make this smoke tell us that there is probably a fire. As we see, there is no absolute point of departure in the phenomenon of ventriloquation. Similarly, a contract commits its signatories because the person who is reading it—a lawyer or a judge, for instance—makes this contract say that this is indeed the case. However, we also see that this ventriloquation is possible because the contract can also be seen as making the lawyer or judge say that this is what the signatories are supposed to do. In other words, I am not passing over in silence the possibility that there could be a conflict of interpretations, but any conflict of interpretation precisely relies on what the contract (or the smoke, or anything that is interpreted) is supposed to be saying or indicating. With this idea of downstream ventriloquation, we can therefore acknowledge the agency of signs (words, gestures, face expressions, traces, etc.), produced intentionally or unintentionally, and show that this agency is also made possible because it is, of course, recognized, which is one of the key contributions of Sbisà’s (1984) faithful reading of Austin  Although I don’t have enough space to develop this point, it is noteworthy that thinking, in this case, is always a form of saying. Although thinking can take the form of images, what Peirce (1991) would call icons, when a thought is articulated under the form of symbols in the Peircian sense, that is, what Austin would call words, this thought always materializes, by definition, as a form of speech in our head. We always hear ourselves privately thinking, a phenomenon that is well translated by the expression “thinking to oneself,” which is symptomatically translated by “se dire” in French (literally speaking to oneself ). 3

  Speech Acts and Ventriloquation: The Contribution… 

159

(1975). In other words, Sbisà helps us, in my opinion, develop a relational perspective on communication and interpretation. Such a perspective allows us to deconstruct any attempt to reduce (1) an act of communication to what its producer intends to do or to what its recipient recognizes it to be and (2) an act of interpretation to what a recipient recognizes or to what the object of interpretation communicates. With a relational perspective, we see, for instance, that any interpretation always consists in making someone or something say something. But this act is only possible because this person or thing is also making us say something about him/her/itself. This brings me to the second form of ventriloquation, which is what I call upstream ventriloquation, if we, of course, still take the agent who is producing or emitting the signs as a point of reference. Through this production of signs, we can indeed recognize agents other than the agent who is directly producing the signs. For instance, in some contexts, when a CEO makes an announcement at a press conference, it is also the organization she represents that can be interpreted as making this announcement. Since ventriloquation is always about making someone or something say things (with or without words), we see that this logic can also go upstream in the flow of communication. When we acknowledge that a fake ID appears to authorize our heroine to penetrate a well-guarded compound, we have to recognize that it is not only the guard who authorizes her to do so (downstream: he makes this ID say that the heroine is indeed duly authorized), but also the organization that supposedly issued this ID (upstream, even if this is not the case in this example, since the ID happens to be fake), what Goffman (1981) would call the “principal.” If we study the details of interaction, we keep noting that these effects of ventriloquation are constitutive of any act of communication. Stating that vaccines are effective against a disease is, for instance, a way to hold this fact to be true, which means that this fact can also be seen, by proxy, as supporting an argument in favor of vaccination. Asking George to pass me the salt is also a way to convey that the alleged absence of salt in my meal requires that George kindly pass me the salt shaker. Promising Gail that I will come tomorrow is a way to give her my word

160 

F. Cooren

that I will be present at tomorrow’s meeting, which means that this word that I am giving to her is supposed to commit me to come. Apologizing for stepping on Larry’s toe is a way to express regret, which means that this regret is not only expressing itself through my apology, but also supposed to tell Larry that I am sorry about what happened. Declaring a session open means, by definition, that this session opens through this intervention if we are authorized to do so.

6 Conclusion As we see, adopting a ventriloquial view of communication, which is, in my opinion, one of the logical consequences of Sbisà’s theses, is a way to put forward what Austin (1975, 107) would call “a general doctrine about action.” To the performativity of speech acts, which is wonderfully analyzed by Sbisà and Austin, a ventriloquial view adds the performativity of the world that surrounds us and that we are part of. As we saw, this world speaks to us because we make it speak and it makes us speak. I could, of course, be criticized for ascribing the action of speaking to other-than-­ humans that do not actually utter words. What the ventriloquial thesis precisely allows us to show is that they don’t need to utter any words, precisely because we make them say things as much as they make us say things. In other words, they do things with words through us. In keeping with Greimas’s (1987) semiotic perspective, which is taken up by Sbisà (1984), a general theory of action accounts for the fact that we can explain the ordering of the social world without never leaving the terra firma of interaction (Cooren 2006), a position that is also implicit in Sbisà and Fabbri’s (1980) two-place model of communication. In this world, people certainly do things with or without words, but other-than-­ humans (animals, ecosystems, texts, architectural elements, emotions, etc.) do this too (because we make them speak and because they make us speak). It is toward this general theory of performativity that Austin and Sbisà, I believe, lead us. This is at least what I make them say, hoping that they would agree that this is also what they make me say.

  Speech Acts and Ventriloquation: The Contribution… 

161

References Allard-Poesi, Florence, and Laure Cabantous. 2021. Strategizing. In Handbook of Management Communication, ed. François Cooren and Peter Stücheli-­ Herlach, 195–211. Berlin: De Gruyter. Austin, John L. 1975. In How to Do Things with Words, ed. James O. Urmson and Marina Sbisà, 2nd ed. Cambridge, MA: Harvard University Press. Bach, Kent, and Harnish, Robert M. 1979. Linguistic Communication and Speech Acts. Cambridge, MA: M.I.T. Press. Caponetto, Laura. 2020. Undoing Things with Words. Synthese 197 (6): 2399–2414. Cooren, François. 2000. The Organizing Property of Communication. Amsterdam: John Benjamins. ———. 2004. Textual Agency: How Texts Do Things in Organizational Settings. Organization 11 (3): 373–393. ———. 2005. The Contribution of Speech Act Theory to the Analysis of Conversation: How Pre-sequences Work. In Handbook of Language and Social Interaction, ed. Kristine L.  Fitch and Robert E.  Sanders, 21–40. Mahwah, NJ: Lawrence Erlbaum. ———. 2006. The Organizational World as a Plenum of Agencies. In Communication as Organizing: Empirical and Theoretical Explorations in the Dynamic of Text and Conversation, ed. François Cooren, James R. Taylor, and Elizabeth J. Van Every, 81–100. Mahwah, NJ: Lawrence Erlbaum. ———. 2008a. Between Semiotics and Pragmatics: Opening Language Studies to Textual Agency. Journal of Pragmatics 40: 1–16. ———. 2008b. Speech Act Theory. In International Encyclopedia of the Social Sciences, ed. William A. Darity Jr., 2nd ed., 56–57. Detroit: Macmillan. ———. 2009. The Haunting Question of Textual Agency: Derrida and Garfinkel on Iterability and Eventfulness. Research on Language and Social Interaction 42 (1): 42–67. ———. 2010. Action and Agency in Dialogue: Passion, Incarnation, and Ventriloquism. Amsterdam: John Benjamins. ———. 2012. Communication Theory at the Center: Ventriloquism and the Communicative Constitution of Reality. Journal of Communication 62: 1–20. ———. 2015a. Speech Act Theory. In The International Encyclopedia of Language and Social Interaction, ed. Karen Tracy, Cornelia Ilie, and Todd Sandel. Oxford: Wiley.

162 

F. Cooren

———. 2015b. Studying Agency from a Ventriloqual Perspective. Management Communication Quarterly 29 (3): 475–480. ———. 2016. Ethics for Dummies: Ventriloquism and Responsibility. Atlantic Journal of Communication 24 (1): 17–30. ———. 2020. Reconciling Dialogue and Propagation: A Ventriloquial Inquiry. Language and Dialogue 10 (1): 9–28. Cooren, François, and Frédérik Matte. 2010. For a Constitutive Pragmatics: Obama, Médecins Sans Frontières and the Measuring Stick. Pragmatics and Society 1 (1): 9–31. Davidson, Donald. 1980. Essays on Actions and Events. London: Oxford University Press. Derrida, Jacques.1988. Limited Inc. Evanston, IL: Northwestern University Press. Gazdar, Gerald. 1981. Speech Act Assignment. In Elements of Discourse Understanding. Cambridge, ed. Aravind K. Joshi, Bruce H. Weber, and Ivan A. Sag. UK: Cambridge University Press. Goffman, Erving. 1981. Forms of Talk. Philadelphia: University of Pennsylvania Press. Greimas, Algirdas Julien, and Joseph Courtes. 1982. Semiotics and Language. An Analytical Dictionary. Trans. Larry Christ, Daniel Patte, James Lee, Edward McMahon II, Gary Phillips, and Michael Rengstorf. Bloomington: Indiana University Press. Greimas, Algirdas Julien. 1987. On Meaning: Selected Writings in Semiotic Theory. Trans. Paul J. Perron and Frank Collins. London: Frances Pinter. Krämer, Sybille. 2015. Medium, Messenger, Transmission: An Approach to Media Philosophy. Amsterdam: Amsterdam University Press. Peirce, Charles S. 1991. Peirce on Signs: Writings on Semiotics. Chapel Hill: University of North Carolina Press. Peters, John D. 1999. Speaking into the Air: A History of the Idea of Communication. Chicago: University of Chicago Press. ———. 2015. The Marvelous Clouds: Toward a Philosophy of Elemental Media. Chicago: Chicago University Press. Sbisà, Marina. 1972. Il problema della classificazione degli atti illocutori. In Ricerche di filosofia linguistica, ed. Renzo Piovesan, 3–41. Firenze: Sansoni. ———. 1984. On Illocutionary Types. Journal of Pragmatics 8: 93–112. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 1985. Manipulation et sanction dans la dynamique des actes de langage. In Exigences et perspectives de la sémiotique. Recueil d’hommages pour

  Speech Acts and Ventriloquation: The Contribution… 

163

Algirdas Julien Greimas, ed. Herman Parret and Hans-George Ruprecht, 592–538. Amsterdam: John Benjamins. ———. 1996. Feminine subject and female body in discourse about childbirth. European Journal of Women’s Studies 3: 363–76. ———. 2001. Illocutionary Force and Degrees of Strength in Language Use. Journal of Pragmatics 33: 1791–1814. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2003. Cognition and narrativity in speech act sequences. In Rethinking Sequentiality, ed. Anita Fetzer and Christiane Meierkord, 71–97. Amsterdam/Philadelphia: John Benjamins. ———. 2007. How to Read Austin. Pragmatics 17 (3): 461–473. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2011. Review of the Book “Action and Agency in Dialogue: Passion, Incarnation and Ventriloquism” by F.  Cooren. Language and Dialogue 1 (2): 320–327. ———. 2013. Some Remarks about Speech Act Pluralism. In Perspectives on Pragmatics and Philosophy, ed. Alessandro Capone, Franco Lo Piparo, and Marco Carapezza, 227–244. Heidelberg: Springer. ———. 2017. Implicitness in Normative Texts. In Pragmatics and Law, ed. Francesca Poggi and Alessandro Capone, 23–42. Dordrecht: Springer. Sbisà, Marina, and Fabbri, Paolo. 1980. Models (?) for a Pragmatic Analysis. Journal of Pragmatics 4: 301–319. Schegloff, Emanuel A. 1988. Presequence and Indirection. Applying Speech Act Theory to Ordinary Conversation. Journal of Pragmatics 12: 55–62. ———. 1991. Conversation Analysis and Socially Shared Cognition. In Perspectives on Socially Shared Cognition, ed. Lauren B. Resnick, John L. Levine, and Stephanie D. Teasley, 150–171. Washington, DC: American Psychological Association. Searle, John R. 1969. Speech Acts: An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. ———. 1979. Intentionality: An Essay in the Philosophy of Mind. Cambridge: Cambridge University Press. ———. 1983. Intentionality: An Essay in the Philosophy of Mind. Cambridge. UK: Cambridge University Press. ———. (2002). Consciousness and Language. Cambridge, UK: Cambridge University Press. Weigand, Edda. 2021. Dialogue: The Complex Whole. Language and Dialogue 11 (3): 457–486.

Towards a Unified Theory of Illocutionary Normativity Neri Marsili

1 Varieties of Illocutionary Norms Speech act theory has traditionally been concerned with a particular subset of linguistic rules: illocutionary rules. Illocutionary rules are rules that govern the performance of illocutionary acts.1 For example: “You should promise to do something only if you intend to do it,” or “Advising someone to do something is appropriate only if the action advised is in the audience’s interest.” Speech act theorists have been concerned with studying the nature and scope of these rules, and with identifying which rules govern which illocutionary acts. Throughout the years and in different areas of the world, various schools and traditions have emerged, each attempting to formalise and model the various norms governing illocutions.  For elaboration, see Searle (1969, Chap. 2.5), Alston (2000, Chaps. 3 and 7), and Sbisà (2019). In this chapter I will use the terms “rule” and “norms” interchangeably, unlike authors who adopt a narrower conception of rules (for discussion, Sbisà 2019, 44). 1

N. Marsili (*) Department of Logic, History, and Philosophy of Science, National Distance Education University (UNED), Madrid, Spain e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Caponetto, P. Labinaz (eds.), Sbisà on Speech as Action, Philosophers in Depth, https://doi.org/10.1007/978-3-031-22528-4_8

165

166 

N. Marsili

Although these different traditions share a common background, there are often terminological and theoretical differences. Often, theoretical convergence between schools is fairly evident despite differences in vocabulary. For instance, Searle’s opposition between “defective” and “unsuccessful” illocutions parallels (and derives from) Austin’s opposition between “abuses” and “misfires.”2 In other cases, however, it isn’t clear whether terminological divergences reflect actual theoretical differences. For example, to describe the rational expectations that govern conversational exchanges, Grice (1975) chooses the term “maxims” instead of “rules.” Whether this lexical choice reflects a concern for a genuinely distinct kind of normativity is up for dispute. Finally, there are traditions that use the same term in different ways. For instance, as we shall see, different schools characterise “constitutive rules” in radically different, incompatible ways. These terminological and theoretical differences can make the speech act theoretic literature difficult to navigate. Even experts often disagree on which differences are purely terminological (and were merely developed in parallel) and which are more substantial (and reflect genuine theoretical divides). The development of a unified framework and a common vocabulary would therefore be a substantial advancement in speech act theory. While several authors have tried to pitch their own terminology as a solution, few have tried to review extant work in order to develop a more neutral, unified conceptual framework. A promising exception is Marina Sbisà’s excellent 2019 essay “Varieties of Speech Act Norms.”

2 Sbisà’s Framework: The “Tripartite View” The Tripartite View Sbisà (2019) introduces three categories within which we can fit the “variety of illocutionary rules” independently developed in the speech act theoretic tradition. Sbisà’s three categories are constitutive rules, maxims, and objective requirements. In this introductory section, aided by some examples, I shall present and discuss each category of the tripartition.  The reader unfamiliar with these terms will find them explained in the next section.

2

  Towards a Unified Theory of Illocutionary Normativity 

167

Constitutive rules, the first category discussed by Sbisà, set conditions that need to be satisfied for a given illocution to be performed at all.3 For example, you can fire someone only if you have the authority to do so. If I tell you that you are fired but I have no authority to fire you, my speech act “fails” or “misfires”: I have not fired you at all. Generalising, if a constitutive rule isn’t met, the speech act typically falls flat: it is void or null. The category owes its name to Searle (1969, 1979), who distinguished “constitutive” rules from “regulative” rules. Regulative rules belong to the second family of rules identified by Sbisà, maxims. Violating a maxim doesn’t lead to illocutionary failure: you can violate the maxims for ƒ-ing and still ƒ. Compliance with maxims is only required for optimal performance (Sbisà 2019, 24, 29, and 33). A speech act that violates a maxim is defective, and open to “blame and disrepute”4 (Sbisà 2019, 47). For example, an insincere promise is still a promise, but the speaker can appropriately be criticised and scolded for failing to be sincere. This second category of norms includes Grice’s (1975) “conversational maxims,” from which it takes its name. Objective requirements are the third member of Sbisà’s taxonomy. They concern our assessment of the “correctness of the accomplished speech act with regard to the situation in the world to which it relates” (Sbisà 2019, 23). This notion derives from Austin, who noticed how different normative standards for “accomplished utterances” (Austin 1975, 139) apply to different illocutions. For instance, an assertion is assessed on the basis of whether it corresponds to the facts; an argument on the basis of its soundness; a piece of advice based on whether it is good or bad (Austin 1975, 139–44). Notably, objective requirements are different from  In passing, Sbisà (2019, 25) adds that “constitutive rules are widely recognized as rules without which a certain act type would not exist” (italics mine). I have argued elsewhere that this characterisation is problematic (Marsili 2019). For this reason, I shall stick to Sbisà’s characterisation of constitutive rules given in the main text—namely, as rules setting conditions that must be met for the speech act to be performed at all. 4  One might perceive a tension between the two characterisations of maxims offered by Sbisà: meeting maxims is merely needed for optimal performance (2019, 24, 29, and 33), but failure to follow them leads to a defective speech act, making the speaker liable to blame and disrepute (2019, 30 and 47). The former characterisation suggests that following maxims is merely supererogatory (preferable, but not required), the latter that it is required. Sbisà’s choice of examples favours the second reading: sincerity, for instance, is not supererogatory. Accordingly, I will henceforth prefer the second reading, and take maxims to describe conditions required for cooperative performance, rather than supererogatory conditions for optimal performance. 3

168 

N. Marsili

maxims, because an utterance can fail to meet its objective requirements even if it follows the maxims that regulate it. For instance, an assertion can be false even if it is sincere and warranted, a piece of advice can be bad even if it is in good faith, and so forth.

Applying the Tripartition to Existing Notions In the speech act theoretic tradition, several notions were independently developed for characterising the different kinds of normative constraints to which illocutionary acts are subject. It will be helpful to review how each of these notions fit into Sbisà’s tripartition. In most cases, we have close matches. The distinction between “success conditions” and “felicity conditions” drawn by Bach and Harnish (1979, 55–56) parallels Sbisà’s distinction between “constitutive rules” (success conditions) and “maxims” (felicity conditions). Austin’s (1975) A-rules (misinvocations of a procedure) and B-rules (misexecutions) are both reclassified as “constitutive rules” in Sbisà’s framework. Austin’s Γ-rules fall instead under the rubric of “maxims.” Within Γ-rules Austin draws an important distinction, on which we will soon return (§3.1). Γ1-rules (or “upstream norms,” cf. Macfarlane 2011) are rules for performance, which need to be met as the speech act is performed: for instance, you should believe what you assert as you assert it. Γ2-rules (“downstream norms”), by contrast, are rules for compliance, which need to be met once the speech act is performed: for instance, it is after promising to ƒ that you acquire an obligation to make sure that you in fact do ƒ.5 Both are maxims for Sbisà. The alleged “epistemic norm” that governs assertion, as discussed by Williamson (2000) and other epistemologists, is trickier to classify within Sbisà’s framework. This norm, which is said to take the form “assert p only if p satisfies condition C,”6 is often referred to as “the constitutive  Some authors place special emphasis on downstream normativity, arguing that speech acts can be characterised in terms of their essential effects (e.g. Brandom 1994; Sbisà 2007a, 2009; Geurts 2019). 6  What is “condition C”? For Williamson (2000), it is the property of being known by the speaker, so that one should assert only what one knows. A popular alternative is that an appropriate assertion requires rational belief in the asserted proposition (Douven 2006; Lackey 2007; Kvanvig 2009; Gerken 2012). Incidentally, the latter hypothesis is favoured by recent empirical research (e.g. Kneer 2018, 2021; Marsili and Wiegmann 2021). 5

  Towards a Unified Theory of Illocutionary Normativity 

169

norm of assertion” (or C-rule).7 Given that Williamson and other scholars call the C-rule “constitutive,” it is tempting to conclude that it is a constitutive rule also in Sbisà’s sense. But this would be a mistake. In Sbisà’s terminology, a constitutive rule cannot be violated while still performing the speech act (in contrast to regulative rules, which can). By contrast, Williamson’s C-rule can be violated while still asserting (for instance, if you assert what you are certain to be false), so it cannot be constitutive in Sbisà’s sense. Terminological traps like this one illustrate why reflecting on vocabulary is so important in speech act theory. Williamson and other epistemologists use the term “constitutive rules” to refer to their traditional antonym, “regulative rules.”8 Misunderstandings and confusion are likely to arise when the same term is used in radically different ways by different schools and traditions.9 Introducing a shared vocabulary can help alleviate these problems. If adopted, Sbisà’s taxonomy can serve this purpose. In this case, it reclassifies the C-rule: since it can be violated while still asserting, it cannot be a constitutive rule.10 How should we classify it, then? For Sbisà, this depends on how we spell out the C-rule. If the C-rule is non-­factive, and only establishes a truth-independent requirement for assertability (e.g. “assert only what you reasonably believe to be true,” “assert only what you believe with confidence”), it’s best classified as a maxim. If the C-rule is factive, and establishes that only true assertions are permissible (e.g. “assert only the truth,” or “assert only what you know”), it’s best classified as setting an objective requirement (Sbisà 2019, 39).11  For an overview, see Goldberg (2015) and Pagin and Marsili (2021, Sect. 5.1). Notably, the assumption that there is a rule that satisfies this description has been forcefully questioned in the literature (Cappelen 2011; Pagin 2016; Marsili 2019; Greenberg 2020). 8  See Hindriks (2007), Maitra (2011), and Marsili (2019). For a defence of Williamson’s conception of “constitutive rules,” see García-Carpintero (2022). 9  I discuss the misunderstandings arising from Williamson’s use of the term in Marsili (2019). 10  Sbisà concedes that we might call Williamson’s C-rule “constitutive” of assertion in a weak sense of “constitutive,” which differs from the strong use of the term that she prefers (Sbisà 2019, 28). I come back to this and other terminological complications in the next section. 11  Sbisà confirmed this point (mostly left implicit in her article) in personal communication. The classification of non-factive norms of assertions as maxims is in line with Grice’s terminology, since his Submaxims of Quality impose similar constraints. For more on the distinction between factive and non-factive norms, see Marsili (2018a). 7

170 

N. Marsili

Table 1  The three categories of rules introduced by Sbisà (2019)

To help keep track of each notion introduced so far, it can be helpful to have a brief summary at this point. Table 1 illustrates Sbisà’s tripartition, accompanied by a brief characterisation of each category,12 and a list of the kinds of norms it encompasses.13  These should be regarded as prima facie characterisations, not necessary and sufficient conditions. Sbisà (2019) carefully leaves open the possibility of exceptions. It is easy to imagine some. Violations of constitutive rules need not result in misfires, as when a group of people doesn’t notice that the speaker had no authority to issue an order, and executes it. Similarly, maxim-violation can lead to misfires, as when a maxim is violated so flagrantly that the audience doesn’t take the speaker to be attempting to perform the relevant illocution (as in Gricean irony). 13  Sbisà also describes some politeness norms as maxims. I have some doubts about this characterisation, but discussing them would lead us astray. For our purposes, politeness norms can simply be left out of the discussion. 12

  Towards a Unified Theory of Illocutionary Normativity 

171

Terminology While the aim of this chapter is to build upon Sbisà’s model, in what follows I will adopt a slightly different terminology. This is to avoid misunderstandings that may arise from the adoption of some terms. Consider the term “constitutive rule.” We have seen that epistemologists use this term to refer to a norm (the C-rule of assertion) that is not constitutive in Sbisà’s sense. This makes the term “constitutive rule” sub-optimal for our purposes: if the goal is developing a common vocabulary for speech act theory (one that can help us dispel confusion and avoid misunderstandings), it is better to avoid labels that have been used in radically different ways within different traditions. To complicate matters, the definition of constitutive rule adopted by Sbisà also departs from Searle’s foundational use of the term. For Sbisà, constitutive rules can take an imperative form (e.g. “Do not order someone to ƒ unless you have the authority to order them to ƒ”) and can be violated (e.g. if I have no authority to order you to ƒ). For Searle, constitutive rules do not take the form of imperatives (Searle 1969, 34 and 36) and cannot be violated (Searle 1969, 41). My aim here is not to enter exegetic debates:14 I am merely noting that the term “constitutive rule” has been interpreted differently by different authors. To avoid misunderstandings, adopting a different term would therefore be preferable. Henceforth, I will adopt “validity conditions” (or “validity rules”) instead of “constitutive rules.” This stipulative term doesn’t overlap with existing ones, and is self-explanatory: a speech act performance is invalid (null, void) if validity conditions are violated. Terminological worries also arise about “maxims.” In Gricean pragmatics, this label identifies constraints for performance that are not illocution-­specific;15 in Sbisà’s taxonomy, it identifies illocution-specific ones. This departure from tradition may equally lead to misunderstandings. Since in Sbisà’s tripartition maxims define the conditions under  I attempted to tackle some (primarily concerning different interpretations of Williamson’s notion of “constitutive rule”) in Marsili (2019). 15  Although a reasonable case can be made that Grice’s Quality maxims only apply to assertives, such a restriction isn’t explicit in Grice. Crucially, the other maxims (Relevance, Manner, Quantity) indisputably apply across the illocutionary board. 14

172 

N. Marsili

which a specific illocution is deemed appropriate and cooperative, I will adopt instead the term “cooperative rules”—which avoids terminological confusion, and maintains the Gricean flavour present in Sbisà’s original terminological choice. Summarising, I will depart from Sbisà in adopting the following terminological conventions: Constitutive rules   → Maxims      →

Validity conditions Cooperative rules

Revising and Extending the Model Sbisà has developed a valuable framework for organising and critically examining scholarly work on illocutionary norms. Her model allows us to better understand the connections between various traditions. It helps us highlight similarities that would otherwise be obscure, and better track progress in the discipline. It is a useful map to navigate the complex body of research on illocutionary normativity, and has potential to facilitate scholarly advancement in this field. Indeed, some authors have already adopted this framework to tackle specific problems in speech act theory (see, e.g., Shields 2020; Corredor 2021). While Sbisà’s tripartition does help us understand and classify the “varieties of norms” identified by scholars working on illocutionary normativity, it inevitably represents a starting point (rather than the finishing line) for the project of developing a unified framework for studying illocutionary normativity. The aim of this chapter is to make a further step in the same direction. To this end, the next section identifies some ways in which the tripartite model proposed by Sbisà could be refined and improved, whereas Sect. 4 introduces a novel, more ecumenical theoretical framework that aims to address the widespread disagreement concerning which rules govern which illocution.

  Towards a Unified Theory of Illocutionary Normativity 

173

3 Two Challenges for the Tripartite Model Upstream Rules and Downstream Obligations Sbisà’s tripartition identifies three kinds of norms: constitutive norms, maxims, and objective requirements. Ideally, each “kind of rule” historically identified by speech act theorists (regulative rules, A-rules, Γ-rules etc.) should fit neatly into one category or another. But perhaps there are some normative constraints that don’t fit the tripartition all that neatly— like “downstream” obligations that arise from the performance of the speech act (which partially16 overlap with Austin’s Γ-2 rules). Sbisà’s model classifies Γ-2 rules as “cooperative rules” (maxims). However, there are some important ways in which downstream obligations differ from genuine cooperative rules.17 First, unlike cooperative rules, downstream obligations display some striking commonalities with objective requirements. Consider promises. A promise meets its objective requirement if it is fulfilled (the speaker performs the promised act);18 similarly, a speaker meets its downstream obligations if they fulfil the promise (they perform the promised act). There is a sense, then, in which the condition set by the objective requirement coincides with the condition set by the downstream obligation. This, in turn, indicates that it is unclear whether downstream obligations are really at home inside the “maxim” category, or whether they would better fit the category of “objective requirements.” Second, and more importantly, the constraints set by upstream rules and downstream obligations are radically different. Upstream rules are  I write “partially” because Γ-2 rules only concern downstream obligations that the speaker has to discharge. 17  While Sbisà acknowledges that the Gricean notion of maxims would not include downstream normativity, her rationale for expanding the category is that upstream norms and Γ-2 norms have a feature in common: it is up to the speaker to comply with them (Sbisà 2019, 31). 18  I briefly motivate this claim in Sect. 3.2. Sbisà (2019, 35) stipulates a different requirement: that the promise turns out to be a “righteous action.” No argument, however, is provided for this claim. I find the idea counterintuitive, and inconsistent with other objective requirements (truth for asserting, soundness for arguing etc.). 16

174 

N. Marsili

rules for performance: they establish whether the speaker (not someone else) is in a position to perform the speech act. Downstream obligations are rules for compliance: they specify what one has to do once the speech act has been successfully performed, to fulfil the obligations generated by the speech act. If these two complications arise, I want to argue, it isn’t because of a genuine flaw in Sbisà’s tripartition. Rather, it is because downstream obligations belong to an altogether distinct realm of normativity—norms for compliance, rather than norms for performance. The idea that downstream obligations fall outside the taxonomy is already implicit in some of Sbisà’s comments. Her essay opens by stating that it is concerned with “the rules for the performance of illocutionary acts” (Sbisà 2019, 24, italics mine). This characterisation doesn’t include downstream obligations, which are rules for compliance, not for performance. I find this characterisation (which excludes downstream obligations from the tripartition) more plausible, and consistent with the work of authors who draw a sharper distinction between upstream normativity and downstream normativity (e.g. Rescorla 2009; MacFarlane 2011). We can therefore build on the tripartite model while acknowledging the special status of downstream obligations. My suggestion is to take the tripartition to only apply to illocutionary rules for performance. Downstream obligations are not norms for performance: they are norms for compliance. As such, they fall outside the trichotomy. They are a topic of investigation on their own, potentially with their own varieties of norms and subdivisions. The resulting model doesn’t depart much from Sbisà’s original proposal, and is summarised in Fig. 1. Upstream Normativity Illocutionary Norms (validity rules, cooperative rules, objective requirements)

Downstream Normativity regulate

Illocutionary Performance

generates

Illocutionary Obligations

Fig. 1  An account of illocutionary norms that distinguishes between norms for performance  (upstream normativity) and rules for compliance  (downstream normativity)

  Towards a Unified Theory of Illocutionary Normativity 

175

A Neglected Category: Aims Sbisà’s (2019) paper focuses on the normative constraints imposed by rules. Arguably, communication is normatively guided by its goals, too. This point is easily illustrated by drawing an analogy with games. In games, both rules and goals contribute to determining which moves are appropriate and which are not.19 Consider football. Clearly, rules guide the action of football players: they motivate (for example) players not to touch the ball with their hands, and to refrain from tackling their opponents too violently. Crucially, aims play an equally central role in shaping player behaviour. Players compete to kick the ball into the opponent team’s goalpost because that’s the aim of the game. If the aim of the game was different (e.g. the team who scores an odd number of points wins), players would behave differently. Both goals and rules play a normative role in games, shaping how agents behave and our evaluation of their actions. Are illocutionary performances like games, in the sense that they are regulated by both illocutionary rules and illocutionary goals? We can find a substantial case for a positive answer in the literature. Directions of fit are often interpreted as setting goals or success conditions for illocutions (Humberstone 1992; cf. also Searle 1976, 2–3). Thetic20 (fact-stating) speech acts are said to have truth as a success condition: these illocutions put forward descriptions whose purpose is to “fit” the way the world is. Telic (behaviour-directing) speech acts, by contrast, achieve their goal

 Oddly, some authors use ‘rules’ and ‘aims’ as synonymous. I take them to identify two distinct categories: rules define conditions for permissible action, goals for successful action (Marsili 2018a, 643–646). Sometimes I will prefer the term “aim” to “goal,” especially when confusion can arise (e.g. in relation to football). The reader should not conclude that I am talking about two different notions: for our purposes, the terms can be considered synonymous. 20  Here I adopt Humberston’s thetic/telic distinction instead of the classic (but lexically confusing) Searlean distinction between word-to-world/world-to-word directions of fit. 19

176 

N. Marsili

when they bring about the desired state of affairs. 21 Unlike cooperative rules, which set conditions for appropriate or permissible performance, directions of fit set conditions for successful performance. They set, in other words, goals for illocutions. Illocutionary goals are also discussed in the literature on assertion. Some authors (Dummett 1973; Williams 2002; Marsili 2018a, 2021a) argue that assertoric speech is governed by an illocutionary goal: truth.22 A point often emphasised in this literature is that illocutionary goals are speaker-independent. The idea is that a speaker making an assertion necessarily purports to be aiming to tell the truth, not that all speakers actually pursue this goal (after all, people lie). Also Grice (1975, 28–29) acknowledges the guiding role played by communicative goals. His Cooperative Principle places teleological normativity at the centre of communicative exchanges, characterising cooperative contributions as those that aim to meet the accepted goal of a conversation: Make your contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged. (Grice 1975, 26, italics mine)

While Grice leaves the purpose of the conversation unspecified, other authors identify more specific goals. Szabó (2020), for instance, takes23 conversations to be governed by a putative knowledge-goal:  In this chapter, I assume that the illocutionary goal of many directives is to get the addressee to do something. Sbisà (2013a, 36 and 60) has argued that getting the audience to do something is rather a perlocutionary object (a perlocutionary effect tightly linked to the illocutionary type; cf. Austin 1975, 118). To accommodate Sbisà’s suggestion, we could weaken the assumption (e.g. assume that directives only aim at inviting the addressee to do something). But there is another possibility: we might decide to call a goal illocutionary iff a speaker cannot perform the illocution without thereby representing themselves as aiming to achieve that goal. This characterisation is plausible, since it draws the right distinctions about (e.g.) assertions. It captures assertion’s goal, since I cannot assert a proposition without thereby presenting myself as attempting to tell the truth (Dummett 1973, 300–303), and excludes assertion’s perlocutionary objects, since I can (e.g.) assert without presenting myself as attempting to convince my audience (Alston 2000, Chap. 2). By the same principle, getting the addressee to do something is an illocutionary goal (rather than a perlocutionary object) of request and orders, since one cannot perform these illocutions without thereby presenting oneself as attempting to get the addressee to do something. 22  Mehta (2016) suggests that knowledge could be regarded as the aim of assertion instead. 23  Wrongly, in my opinion—for non-assertoric exchanges surely need not share this goal. 21

  Towards a Unified Theory of Illocutionary Normativity 

177

The goal of conversation is to share private knowledge pertaining to a topic of common concern. (Szabó 2020, 62)

A like-minded idea is found in Stalnakerian pragmatics, which regards discourse as a communal inquiry whose end or purpose is finding out “the way things might be” (the set of possible worlds that is compatible with the propositions that are mutually accepted as true in the conversation; Stalnaker 1978, 151, 2002, 704). This conversational goal, in turn, determines which contributions are appropriate and which are not (Roberts 2012, 4). Substantial scholarship, then, supports the hypothesis that communicative exchanges have speaker-independent goals. Now, goals can play a normative role in the practices within which they occur: they guide action, and ground our assessments of an action as good or bad, correct or incorrect—both in ordinary action and in communication. Taken together, these observations highlight an important connection between goals and objective requirements. So, for instance, an action in a game (say, shooting a penalty) can be assessed positively when it meets its purported goal (scoring a goal), and negatively when it doesn’t. Similarly, a speech act (say, asserting) can be assessed positively when it meets its presumed purpose (when it “fits” the way the world is), and negatively when it doesn’t (Searle 2007, 34). Given their role in grounding our assessment of illocutions as successful or unsuccessful, illocutionary goals establish “normative standards for ‘accomplished utterances.’” In other words, they are “objective requirements” in Sbisà’s (2019, 32) sense. Crucially, like objective requirements, illocutionary goals establish objective standards of assessment. Objective truth is the goal of assertion: if it seems to you that you have asserted the truth (but you haven’t), your assertion has not met its illocutionary goal. Similarly, if it seems to you that your request has been satisfied (but it hasn’t), your request has not met its illocutionary goal. Objective requirements and success conditions24 identify speaker-independent, objective standards of assessment.  I am using “success conditions” in the sense established above: conditions required for an action (like shooting a penalty) to be deemed successful. This usage is not to be confused with Bach and Harnish’s, who adopt the expression “success condition” to refer to a condition that must be met for the speech act to be performed at all. For more on my understanding of “successful” illocutions, see Marsili (2018a, 2021a). 24

178 

N. Marsili

Sbisà herself draws connections between objective requirements and illocutionary goals. To characterise objective requirements, she adopts a teleological vocabulary. She notes that objective requirements have to do with “the point of a certain type of illocutionary act” and that failure to meet them renders the illocution “unfit to contribute to the achievement of the goals of the speaker or […] other participants” (Sbisà 2019, 47, italics mine).25 A plausible case can be made, then, for marrying the notion of “objective requirements” with the notion of “illocutionary goals.” A first advantage of this reformulation is that it broadens Sbisà’s third category, which was originally only meant to capture some sparse Austinian remarks about the different terms we use for assessing speech (true/false, fair/ unfair etc.; cf. Austin 1975, 140–141). The notion of illocutionary goals casts a much wider net on the existing literature. We can accommodate fitness conditions (set by directions of fit), overarching conversational goals (as hypothesised by Grice, Szabó, or Stalnaker) and assertoric aims (as in Dummett and others). The resulting model is summarised in Table 2. The revised model also has the advantage of drawing its distinctions in a homogenous way. Each family of norms is identified by the kind of normative constraint (condition) that it establishes. The distinctions are simple and parallel each other. Validity rules set conditions for (actual) performance. Cooperative rules set conditions for cooperative performance. Illocutionary goals set conditions for successful performance. Finally, illocutionary obligations set conditions for compliance.26

 A further point of convergence is the following: just like illocutionary goals are speaker-­ independent, objective requirements also set standards that the speech act should meet “irrespective of the perspective of participants” (Sbisà 2019, 24). 26  We may still wonder (cf. Sect. 3.1) if the notion of ‘compliance’ and ‘success’ sometimes overlap. The success condition for commands, for instance, seemingly coincides with its compliance condition: both are satisfied when the deed is done. Even if they can be satisfied at the same time, however, these requirements capture different levels of normativity and different dimensions of assessment. About an order, for instance, we can say that A succeeded in getting B to do something (evaluating the accomplishment of a goal), but also that B complied with the order, or that B discharged their obligations (evaluating whether B has fulfilled their duty). 25

  Towards a Unified Theory of Illocutionary Normativity 

179

Table 2  A table visualising which family of rules falls under which category under the revised model. The triple line separates rules for performance (upstream normativity) from rules for compliance (downstream normativity)

A final advantage of this model is that goals can be regarded as the normative source for many of the cooperative rules governing each illocution. I will explore this suggestion in the next section, as I challenge “checklist accounts” of illocutionary normativity.

4 Cooperation, Rules, and Illocutionary Concepts Disagreement About Rules The proposed model is able to systematically classify a variety of illocutionary norms, overcoming differences between schools and traditions. Strong divisions persist, however, when we consider how each author characterises the norms governing specific illocutionary types.

180 

N. Marsili

Take, for example, Searle’s (1969) and Sbisà’s (2019) characterisation of the speech act of advising. Like most scholars, they agree that advising to ƒ is appropriate only if you believe that the hearer would benefit from ƒ-ing. But they disagree about other conditions. For Searle (1969, 66–67), advising H to do A is appropriate only if “it is not obvious to both S and H that H will do A.” Sbisà rejects this condition, but endorses some requirements that Searle had not included: for instance, that “the speaker must have authority over the addressee with respect to the field of activities with which the piece of advice is concerned.”27 Similar disagreements are widespread in speech act theory: given a speech act type SA, theorists often disagree about which rules regulate SA.

The Checklist View Scholarly disagreement about which precise set of norms governs each illocutionary act is usually regarded as unproblematic and predictable. It is, however, at odds with a foundational assumption accepted by most speech act theorists—namely, that speech acts are constituted by the unique set of rules that regulates them. According to this hypothesis, speech acts are regulated by their illocutionary rules by conceptual necessity. Necessarily, a given speech act is defined by the unique set of rules that governs its performance (Searle 1969; Pollock 1982; Searle and Vanderveken 1985; Williamson 2000; Alston 2000, Chap. 8; García-­ Carpintero 2022). To have a command of the concept of a speech act SA, then, is to know which unique set of rules governs SA. Just like a chess player cannot know what “castling” means unless they know what the rules for castling are, knowing what an illocution is and knowing its rules are on this picture the same thing.  Incidentally, Sbisà classifies this as a validity rule for advising: if violated, the speaker is not advising at all. However, this seems wrong: this is at most a condition for appropriate advising. The same could be said about many of the putative ‘constitutive rules’ described by Sbisà (2019, Sect. 5). For instance, for promises (Sbisà 2019, 35), the requirement that the speaker must be able to perform the promised fact (arguably, a promise to do something I cannot do is still a promise—as long as the audience is unaware of my inability to deliver); or, for congratulations (Sbisà 2019, 37), the requirement that the speaker must have an obligation to acknowledge the speaker’s achievement (I can congratulate you for your achievement even if I don’t have an obligation to do so). 27

  Towards a Unified Theory of Illocutionary Normativity 

181

Call this view the “Checklist View” of illocutionary normativity, since it presupposes a “checklist theory” (Fillmore 1975) of illocutionary concepts and their rules (that is, it stipulates a strict relation of ontological dependency between illocutionary types and a “checklist” of rules). As noted by some commentators (Pagin 2016; Greenberg 2020; Pagin and Marsili 2021), widespread scholarly disagreement about which norms govern each illocution is at odds with some assumptions of the Checklist View. If illocutionary acts are defined by their rules, having command of any given illocutionary concept requires knowledge of the set of rules that regulates it (just like knowing what castling is requires knowledge of its rules). Widespread disagreement among experts, then, should be difficult, if not impossible. Scholars should converge on identifying a similar set of rules, or at least they should come close in their characterisations.28 This isn’t what generally happens. Recall the above-mentioned disagreement concerning which cooperative rules regulate advising, and the disagreement about which exact norm regulates assertion.29 If there really is a unique set of rules that necessarily regulates each illocution (a set whose knowledge is required for having command of the illocution), scholars should converge on identifying a single correct solution. Since experts cannot agree on a unique set of rules, we must conclude that either the Checklist View is misguided or experts lack knowledge of these illocutionary concepts, and none of them really knows what “advising” or “asserting” is—just like a person who doesn’t know the rules for castling doesn’t really know what castling is. A similar point applies to ordinary speakers, who are also typically unable to articulate the rules governing a given illocution, if asked. If knowing what it is to perform a given speech act (promising, advising, asserting etc.) really presupposes knowledge of the unique set of rules that

 For like-minded criticisms to checklist theories, see Levinson (1979, 1983, Chap. 5), Harnish (2005), Harnish and Plunze (2006), Pagin (2016, Sect. 4), Green (2020, Sect.1). 29  As anticipated in footnote 6, scholars cannot agree as to whether appropriate assertion requires knowledge, truth, justification, belief, or something else altogether (Goldberg 2015; Pagin and Marsili 2021, Sect. 5.1). 28

182 

N. Marsili

regulates it, we must conclude that laypeople lack command of basic illocutionary concepts, too.30 To be sure, laypeople do exhibit knowledge of some illocutionary norms. There are some speech acts (like marrying, baptising, or condemning) whose rules are recorded (and enforced) by dedicated institutions, and whose conditions for appropriate performance people familiar with those institutions can explicitly articulate. Similarly, laypeople are clearly able to articulate the sincerity rules for basic speech acts, like promising and asserting (“don’t break a promise,” “don’t lie”). There is also empirical evidence that children master these rules at a relatively early stage of their linguistic development (Maas 2008; Isella et al. 2019). Laypeople’s general familiarity with sincerity rules (and rules for institutional speech acts) is in line with the predictions of the Checklist View. But it is doubtful that their competence extends to other rules and other illocutions. If asked under which conditions an assertion or a promise is sincere, competent speakers will give fairly confident and consistent answers. But their answers will be much less confident and consistent (if not utterly confused) if you ask them which further conditions are required for a promise or an assertion to be appropriate. Articulating a complete “checklist” of rules should prove difficult also for other illocutions, considered that even researchers disagree on the content of such checklists. Perhaps the Checklist View can be refined to overcome these difficulties. However, further problems lurk around the corner. Since illocutionary rules are said to regulate speech acts by conceptual necessity, checklist theorists typically assume that there is a unique set of rules that defines each speech act—rather than many equally good alternatives. However,  Perhaps the Checklist View could be interpreted as only requiring implicit (procedural) knowledge of the relevant rules. Implicit knowledge of rules could be understood (broadly) as a general disposition to behave in accordance with the rules, and to correct behaviour that violates them (cf. Chomsky 1965, 164–168; Searle 1969, 41–42; Mikhail 2011). Still, if competent speakers possessed such implicit knowledge, we should expect them to be able to spell out (if asked, and given sufficient time) the rules that uniquely define an illocutionary act—just like competent football players are able to work out the rules for a corner kick or a penalty kick, if prompted to. The observation that ordinary speakers (and experts alike) cannot articulate illocutionary rules (nor agree on their content) therefore still poses a challenge to the Checklist View. For further discussion, see Pagin (2016, 190–191). 30

  Towards a Unified Theory of Illocutionary Normativity 

183

this assumption is also controversial. It is doubtful that each speech act is regulated by precise, well-defined requirements that can be synthesised into uniquely correct formulas. A more plausible alternative is that, like moral norms for action (Watson 1996, 237–239) (and arguably norms for semantic reference; cf. Marconi 1997), the norms regulating speech acts are loose, imprecise, and not easily formalisable. Consider a simple case, like the sincerity rule for promising. It could take any of the following forms, since violating any of (1–5) results in an infelicitous promise: 1. Don’t promise to ƒ if you lack an intention to ƒ. 2. Don’t promise to ƒ if you actively intend not to ƒ. 3. Don’t promise to ƒ if you believe that you will not ƒ. 4. Don’t promise to ƒ if you believe that it is likely that you will not ƒ. 5. Don’t promise to ƒ unless you are convinced that you will almost surely ƒ. Since none of these rules can be violated felicitously, there seems to be no ground to claim that only one of these rules is the sincerity rule for promising.31 The Checklist View, however, seems committed to maintain that only one of these rules is the true sincerity rule for promising. The problem is generalisable. For every rule regulating a speech act, suitably similar alternatives can be derived, generating analogous dilemmas. Take the sincerity rule for requesting something. First, there is the issue of which mental state Ψ is mandated for making an appropriate request: a desire that the addressee complies with the request, or a wish, a hope, or a mere preference? Second, does one need to positively Ψ that the addressee brings about the state of affairs, or lacking the opposite attitude is enough? Third, we may wonder how precise the attitude should be: should the speaker Ψ that the addressee performs the exact deed specified in the request, or is it sufficient to Ψ that something in its ballpark is  I take these rules not to be reducible to each other, because I take intentions not to be reducible to beliefs or credences about the future (nor vice versa, see Marsili 2016, esp. Appendix II). A related (but independent) question concerns whether our ordinary concept of lying tracks the infraction of only one of these rules. For discussion, see Marsili (2016; cf. also 2018b, 2021b, 2022a). For a parallel argument concerning semantic norms, see Marconi (1997). 31

184 

N. Marsili

done? This list of questions can be expanded further. But my point here is that it is highly doubtful that a correct, ultimate answer can be given to this sort of questions. Illocutionary acts are likely governed only by loose expectations, not inflexible (and metaphysically necessary) imperatives that admit only one, uniquely correct periphrasis.

Cooperative Rules as Rational Expectations Summarising, checklist theories have trouble accommodating (A) widespread disagreement among scholars about which cooperative rules govern each illocution,32 (B) laypeople’s inability to articulate illocutionary rules, and (C) the looseness of the normative constraints imposed by illocutionary rules. None of these problems is necessarily decisive: perhaps we only need to refine checklist theories in order to accommodate (A-C). As they are, however, checklist accounts are not especially well suited to make sense of these data points.33 Are there alternative models that fare better? A promising suggestion comes again from Sbisà, who notes that cooperative rules are neither arbitrary nor conventional: they are based on “rational motivations” (Sbisà 2019, 47).34 This suggestion is also found in Grice, who regards maxims as derivable rationally from his Cooperative Principle (Grice 1989, 29–30).35 To illustrate the idea, consider how the Maxim of Relation (“Be relevant”) can be derived from the Cooperative Principle. Intuitively, one’s contribution cannot be cooperative if it is completely irrelevant to  It is not clear if this problem also arises for illocutionary obligations and validity rules. Here I am limiting myself to suggest that cooperative rules generate difficulties for checklist theories. 33  While I have no space to discuss it in detail, there is a further difficulty faced by the Checklist View. Illocutionary relativists argue that in some contexts it is indeterminate whether the speaker is performing an illocutionary act or another (or both) (Sbisà 2013b; Johnson 2019; cf. Witek 2015b; Lewiński 2021, 6689, 6703). If the relativist is right, we have further ground to endorse (C), because on this view illocutions are not always subject to determinate normative constraints. 34  Intention-based analyses (e.g. Bach and Harnish 1979) also avoid many of the difficulties of checklist theories, but are known to be subject to other compelling objections (Alston 2000, Chap. 2). 35  The Cooperative Principle was stated explicitly in Sect. 3.2. On deriving conversational expectations from the principle, see also Roberts (2012, 4). For more on Grice’s conception of rationality, see Sbisà (2006, 2007a) and Labinaz (2016). 32

  Towards a Unified Theory of Illocutionary Normativity 

185

“the accepted purpose or direction of the talk exchange.” To expect speakers to be cooperative, then, is to expect them to make relevant contributions to the conversation—that is, to follow the Maxim of Relation. Can we derive illocutionary rules from the Cooperative Principle in the same way? Grice passes silence on whether illocution-specific norms can be derived along the same lines. But if such a derivation is possible, the resulting model would represent a promising alternative to checklist theories. We would have that cooperative rules are not a matter of conceptual necessity, but rather rationally derivable expectations of cooperation. I have limited space here to develop this suggestion in detail, but I would like to sketch a tentative model of how such a derivation might work. Illocutionary goals are a natural starting point for deriving illocution-­ specific cooperative rules from the Cooperative Principle.36 Take, for example, the speech act of advising. If the purported goal of giving advice is to get someone to do something that is in their interest, it is only rational to expect a cooperative speaker to only give advice that they believe to be in the audience’s interest. After all, giving advice that you don’t believe to be in the audience’s interest is incompatible with meeting the presumed goal of giving advice. Just as the Maxim of Relevance can be derived from the Principle of Cooperation, the sincerity rule for giving advice can be derived from its illocutionary goal. The speech act of assertion provides another example of how cooperative rules could be derived from illocutionary goals. If saying something true is the aim of assertion, presumably a speaker cannot make a cooperative assertion unless they are trying to tell the truth. After all, unless you are trying to assert a true proposition, you are not trying to meet the goal you are presumed to have, and therefore you are not cooperating.37 Interestingly, this requirement (“Try to only assert propositions that are true”) ramifies into further ones. Asserting what you believe to be false is incompatible with attempting to tell the truth: a sincerity condition (“Don’t assert what you believe to be false”), then, can also be derived  Although Grice doesn’t discuss illocutionary goals, he suggests that conformity to the maxims isn’t just reasonable, it is reasonable given the aims of the conversation (Grice 1975, 29–30; cf. also Roberts 2012, 4), establishing a link between goals and maxims. 37  For discussion of this claim, see Marsili (2017, Sect. V.4.4, 2018a, 645, 2022b). 36

186 

N. Marsili

from assertion’s presumed goal. Further sincerity norms can be derived, too. Arguably, asserting what you believe to be likely false, or what you do not believe to be true, is equally incompatible with trying to tell the truth. If this is right, the corresponding injunctions (“Don’t assert what you believe to be likely false,” “Don’t assert what you do not believe”) can similarly be derived from assertion’s goal. Crucially, on the proposed view, none of these rules is the sincerity rule for asserting. I regard this as a strength of the proposed account, rather than a weakness, because it means that this alternative model can naturally accommodate (C) (the looseness of illocutionary rules). The idea envisaged here is that multiple normative constraints (dictated by considerations of consistency and rationality) can be derived from the Cooperative Principle, so that multiple cooperative rules are bound to arise for each illocutionary type—rather than a unique set of precise, well-defined rules, as the checklist theorist would have it. I have argued that some cooperative rules can be derived inferentially from illocutionary goals. I do not mean to suggest that all cooperative rules admit this sort of derivation. Consider Grice’s suggestion that a contribution should be “as informative as required.” The Searlean rule that advising H to do A is appropriate only if “it is not obvious to both S and H that H will do A” (Searle 1969, 66–67) is easily derivable from this assumption. The same goes for the requirement that an assertion is appropriate only if “it is not obvious to both S and H that H knows (does not need to be reminded of, etc.) p” (Searle 1969, 55–66). This suggests that not all cooperative rules are derivable from illocutionary goals. Similarly, we do not need to assume that all these expectations have the same strength and form. A more plausible view is that illocutions are governed by a loose spectrum of norms of different strength and nature.38 At one pole we have well-defined, stricter requirements, like institutional illocutionary rules, which are explicit and well-defined, enforced by  A similar assumption is also present in Grice (1975, 26–27; 1989, 371), who mentions that expectations of conformity to Quality are stronger than expectations of conformity to Quantity (especially the second maxim of Quantity). Presumably, the strength of a normative constraint will also be affected by whether the speaker has mitigated or reinforced the illocution with modifiers. Sbisà has written extensively on how these modifiers work, and how they modulate force and normative obligations (Bazzanella et al. 1991; Sbisà 2001a, 2014; Labinaz and Sbisà 2014). 38

  Towards a Unified Theory of Illocutionary Normativity 

187

dedicated institutions, and associated with precise sanctions. Close to this pole we also find rules that competent speakers can easily articulate, like the rules against lying or against breaking promises, which are actively policed by competent speakers of the language. At the lower end of the scale, by contrast, we find loose expectations that are not associated with clear penalties, that often admit exceptions, but that is still rational to expect other communicators to follow (like the Searlean rules against redundant illocutions). This account of illocutionary normativity can easily accommodate also (A) and (B). Given that the expectations falling at the weaker end of the spectrum are loose, defeasible, weak, and not associated with defined sanctions, we need not assume that competent speakers can articulate these expectations just in virtue of knowing what the relevant speech act is. Likewise, rules falling at this end of the spectrum will inevitably trigger scholarly dispute, since evidence for their existence (e.g. speakers’ dispositions to sanction violations) are bound to be harder to detect, and easier to challenge. On this view, illocutions are governed by loose normative expectations that are not easily reducible to a finite checklist of well-defined, strict rules. This model naturally lends itself to accommodate a more ecumenic, pluralistic approach to speech act theory, which regards dissenting theories as compatible attempts to highlight the various, multi-faceted normative constraints regulating different illocutions in different contexts. As such, this approach offers an ideal groundwork for developing a unified theory of illocutionary normativity, one able to overcome territorial divisions concerning which illocutionary rules govern each illocution. This model also offers a plausible genealogical story as to how illocutionary rules acquire the normative force that they have. Presumably, illocutionary norms can move along the spectrum as time passes. Some expectations (like sincerity rules) slowly sediment into stricter requirements, moving upwards in the hierarchy. They become associated with

188 

N. Marsili

clearer sanctions, crystallising into genuine, stricter rules.39 Less important expectations (like rules against redundant advising) are less likely to undergo this process, remaining at the bottom of the hierarchy. None of these norms, however, simply happens to regulate the relevant illocution in virtue of some mysterious, Platonic relation of metaphysical necessity, as some proponents of the Checklist View seem to assume.

5 An Open Project This chapter has offered some programmatic suggestions for developing a unified framework to classify, study, and analyse illocutionary rules. It has shown that Sbisà’s seminal work provides a fertile ground for bringing together under a single flag different lines of research on illocutionary normativity. It has argued that the notions of illocutionary goals and illocutionary obligations can enrich Sbisà’s model. Finally, it has developed an approach for deriving cooperative rules, delineating a way to model illocutionary normativity that avoids dubious appeals to relations of metaphysical necessity between illocutions and their norms. The project begun by Sbisà is ambitious; inevitably, many questions still remain open. This chapter has just attempted to make a further step in the same direction, laying down some groundwork for future research into foundational questions in speech act theory.

 Research on the evolution of linguistic communication usually makes assumptions along these lines, although details vary widely from theory to theory. Broadly, the idea is that repeated attempts to solve a coordination problem can lead to equilibria, which progressively come to be regarded as correct linguistic behaviour, slowly sedimenting into conventions and rules (see, e.g., Millikan 2005; Skyrms 2010; Steels 2011; Witek 2015a). Not all rules undergo this process—some, for instance, are the result of an institutional fiat (like laws against perjury or for marriage). We may also wonder whether the weakest and loosest of these requirements really deserve to be called “rules,” or are best described as mere ‘expectations’ or ‘constraints’ (Sbisà 2001b, e.g., denies the status of rule to the Cooperative Principle). I doubt that a straight answer can be given. For our purposes, all that matters is that these expectations all exercise some degree of normative guidance on illocutionary performances, and as such they fall under our domain of investigation. 39

  Towards a Unified Theory of Illocutionary Normativity 

189

References Alston, William P. 2000. Illocutionary acts and sentence meaning. Ithaca: Cornell University Press. Austin, John L. 1975. In How to Do Things with Words, ed. James O. Urmson and Marina Sbisà, 2nd ed. Cambridge Ma: Harvard University Press. Bach, Kent, and Robert M. Harnish. 1979. Linguistic Communication and Speech Acts. Cambridge, MA: MIT Press. Bazzanella, Carla, Claudia Caffi, and Marina Sbisà. 1991. Scalar Dimensions of Illocutionary Force. In Speech Acts: Fiction or Reality, ed. Igor Ž. Žagar, 63–76. Antwerp & Ljubljana: IPrA Distribution Center for Yugoslavia and Institute for Social Sciences. Brandom, Robert. 1994. Making It Explicit: Reasoning, Representing, and Discursive Commitment. Cambridge, MA: Harvard University Press. Cappelen, Herman. 2011. Against Assertion. In Assertion: New Philosophical Essays, ed. Jessica Brown and Herman Cappelen, 21–48. Oxford: Oxford University Press. Centola, Damon, and Andrea Baronchelli. 2015. The Spontaneous Emergence of Conventions: An Experimental Study of Cultural Evolution. Proceedings of the National Academy of Sciences 112 (7): 1989–1994. Chomsky, Noam. 1965. Aspects of the Theory of Syntax. Cambridge, MA: MIT Press. Corredor, Cristina. 2021. Illocutionary Performance and Objective Assessment in the Speech Act of Arguing. Informal Logic 41 (3): 453–483. Douven, Igor. 2006. Assertion, Knowledge, and Rational Credibility. The Philosophical Review 115 (4): 449–485. Dummett, Michael. 1973. Frege: Philosophy of Language. London: Duckworth. Fillmore, Charles J. 1975. An Alternative to Checklist Theories of Meaning. Proceedings of the Meeting of the Berkeley Linguistics Society 1: 123–131. García-Carpintero, Manuel. 2022. How to Understand Rule-Constituted Kinds. Review of Philosophy and Psychology 13 (1): 7–27. Gerken, Mikkel. 2012. Discursive Justification and Skepticism. Synthese 189 (2): 373–394. Geurts, Bart. 2019. Communication as Commitment Making: Speech Acts, Implicatures, Common Ground. Theoretical Linguistics 45 (1–2): 1–30. Goldberg, Sanford C. 2015. Recent Work on Assertion. American Philosophical Quarterly 52 (4): 365–380.

190 

N. Marsili

Green, Mitchell. 2020. Assertion and Convention. In Oxford Handbook of Assertion, ed. Sandford Goldberg, 347–370. Oxford: Oxford University Press. Greenberg, Alexander. 2020. There Is No (Sui Generis) Norm of Assertion. Philosophy 95 (3): 337–362. Grice, Paul H. 1975. Logic and Conversation. In Syntax and Semantics, Vol. 3: Speech Acts, ed. Paul Cole and Jerry Morgan. New York: Academic Press. Reprinted in Studies in the Way of Words (1989), 22–40. Cambridge, MA: Harvard University Press. ———. 1989. Studies in the Way of Words. Cambridge, MA: Harvard University Press. Harnish, Robert M. 2005. Commitments and Speech Acts. Philosophica 75: 11–41. Harnish, Robert M., and Christian Plunze. 2006. Illocutionary Rules. Pragmatics & Cognition 14 (1): 37–52. Hindriks, Frank. 2007. The Status of the Knowledge Account of Assertion. Linguistics and Philosophy 30 (3): 393–406. Humberstone, Lloyd. 1992. Direction of Fit. Mind 101 (401): 59–83. Isella, Margherita, Patricia Kanngiesser, and Michael Tomasello. 2019. Children’s Selective Trust in Promises. Child Development 90 (6): e868–e887. Johnson, Casey Rebecca. 2019. Investigating Illocutionary Monism. Synthese 196: 1151–1165. Kneer, Markus. 2018. The Norm of Assertion: Empirical Data. Cognition 177: 165–171. ———. 2021. Norms of Assertion in the United States, Germany, and Japan. PNAS 118 (37): 3. Kvanvig, Jonathan L. 2009. Assertion, Knowledge, and Lotteries. In Williamson on Knowledge, ed. Duncan Pritchard and Patrick Greenough, 140–160. Oxford: Oxford University Press. Labinaz, Paolo. 2016. Rationality in Linguistic Interpretation: From Charity to Cooperativeness. In Pratical Rationality in Politica Contexts: Facing Diversity in Contemporary Multicultural Europe, ed. Riccardo Martinelli and Gabriele De Anna, 77–98. Trieste: EUT. Labinaz, Paolo, and Marina Sbisà. 2014. Certainty and Uncertainty in Assertive Speech Acts. In Communicating Certainty and Uncertainty in Medical, Supportive and Scientific Contexts, ed. Andrzej Zuczkowski, Carla Canestrari, Ramona Bongelli, and Ilaria Riccioni, 31–58. Amsterdam: John Benjamins. Lackey, Jennifer. 2007. Norms of Assertion. Noûs 41 (4): 594–626. Levinson, Stephen. 1979. Activity Types and Language. Linguistics 17 (5–6): 365–399.

  Towards a Unified Theory of Illocutionary Normativity 

191

———. 1983. Pragmatics. New York: Cambridge University Press. Lewiński, Marcin. 2021. Illocutionary Pluralism. Synthese 199: 6687–6714. Maas, Fay K. 2008. Children’s Understanding of Promising, Lying, and False Belief. The Journal of General Psychology 135 (3): 301–322. MacFarlane, John. 2011. What Is Assertion? In Assertion: New Philosophical Essays, ed. Jessica Brown and Herman Cappelen, 79–96. Oxford: Oxford University Press. Maitra, Ishani. 2011. Assertion, Norms, and Games. In Assertion: New Philosophical Essays, ed. Jessica Brown and Herman Cappelen, 277–296. Oxford: Oxford University Press. Marconi, Diego. 1997. Semantic Normativity Without Semantic Norm. In Thought and Ontology, ed. M. Sainsbury, 57–125. Milano: Franco Angeli. Marsili, Neri. 2016. Lying by Promising. International Review of Pragmatics 8 (2): 271–313. ———. 2017. You Don’t Say! Lying, Asserting and Insincerity. PhD Dissertation, University of Sheffield. https://etheses.whiterose.ac.uk/19068/ ———. 2018a. Truth and Assertion: Rules versus Aims. Analysis 78 (4): 638–648. ———. 2018b. Lying and Certainty. In The Oxford Handbook of Lying, ed. Jörg Meibauer, 169–182. Oxford: Oxford University Press. ———. 2019. The Norm of Assertion: A ‘Constitutiveʼ Rule? Inquiry. https:// doi.org/10.1080/0020174X.2019.1667868. ———. 2021b. Lying: Knowledge or Belief? Philosophical Studies. https://doi. org/10.1007/s11098-­021-­01713-­1. ———. 2021a. Truth: The Rule or the Aim of Assertion? Episteme September, 1–7. DOI: https://doi.org/10.1017/epi.2021.28. ———. 2022a. Immoral Lies and Partial Beliefs. Inquiry 65 (1): 117–127. ———. 2022b. Affermazioni e verità fra regole e scopi, in “Rivista di filosofia” 3/2022, pp. 365–395. https://doi.org/10.1413/105394. Marsili, Neri, and Alex Wiegmann. 2021. Should I Say That? An Experimental Investigation of the Norm of Assertion. Cognition 212: 104657. Mehta, Neil. 2016. Knowledge and Other Norms for Assertion, Action, and Belief: A Teleological Account. Philosophy and Phenomenological Research 93 (3): 681–705. Mikhail, John. 2011. Elements of Moral Cognition: Rawls’ Linguistic Analogy and the Cognitive Science of Moral and Legal Judgment. Cambridge: Cambridge University Press. Millikan, Ruth Garrett. 2005. Language: A Biological Model. Oxford: Clarendon Press.

192 

N. Marsili

Pagin, Peter. 2016. Problems with Norms of Assertion. Philosophy and Phenomenological Research 93 (1): 178–207. Pagin, Peter, and Neri Marsili. 2021. Assertion. In Stanford Enciclopedia of Philosophy, Winter. https://plato.stanford.edu/archives/win2021/entries/ assertion/. Pollock, John L. 1982. Language and Thought. New Jersey: Princeton University Press. Rescorla, Michael. 2009. Assertion and Its Constitutive Norms. Philosophy and Phenomenological Research 79 (1): 98–130. Roberts, Craige. 2012. Information Structure in Discourse: Towards an Integrated Formal Theory of Pragmatics. Semantics and Pragmatics 5 (6): 1–69. Sbisà, Marina. 2001a. Illocutionary Force and Degrees of Strength in Language Use. Journal of Pragmatics 33: 1791–1814. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2001b. Intentions from the Other Side. In Paul Grice’s Heritage, edited by Giovanna Cosenza, 185–206. Turnhout: Brepols. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2006. Two Conceptions of Rationality in Grice’s Theory of Implicature. In Rationality of Belief and Action, edited by Elvio Baccarini and Snježana Prijic-Samarzija, 233–47. Rijeka: Filozofski fakultet u Rijeci. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2007a. How to Read Austin. Pragmatics 17. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2007b. On Argumentative Rationality. Anthropology and Philosophy 8 (1–2): 89–100. ———. 2009. Uptake and Conventionality in Illocution. Lodz Papers in Pragmatics 5 (1): 33–52. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2013a. Locution, Illocution, Perlocution. In Pragmatics of Speech Actions, ed. Marina Sbisà and Ken Turner, 25–75. Berlin: De Gruyter. ———. 2013b. Some Remarks About Speech Act Pluralism. In Perspectives on Pragmatics and Philosophy, ed. Alessandro Capone, Franco Lo Piparo, and Marco Carapezza, 227–244. Heidlberg: Springer.

  Towards a Unified Theory of Illocutionary Normativity 

193

———. 2014. Evidentiality and Illocution. Intercultural Pragmatics 11 (3): 463–483. ———. 2019. Varieties of Speech Act Norms. In Normativity and Variety of Speech Actions, edited by Maciej Witek and Iwona Witczak-Plisiecka, 23–50. Leiden: Brill. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. Searle, John R. 1969. Speech Acts. An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. ———. 1976. A classification of illocutionary acts. Language in Society 5 (1): 1–23. ———. 1979. Expression and Meaning. Cambridge: Cambridge University Press. ———. 2007. Illocutionary Acts and the Concept of Truth. In Truth and Speech Acts, ed. Dirk Greimann and Geo Siegwart, 31–40. New York: Routledge. Searle, John R., and Daniel Vanderveken. 1985. Foundations of Illocutionary Logic. Cambridge: Cambridge University Press. Shields, Matthew. 2020. Philosophical Speech Acts. Philosophy 95 (4): 497–521. Skyrms, Brian. 2010. Signals: Evolution, Learning, & Information. Oxford; New York: Oxford University Press Stalnaker, Robert C. 1978. Assertion. In Pragmatics, Syntax and Semantics, ed. Peter Cole, vol. 9, 315–332. New York: Academic Press. ———. 2002. Common Ground. Linguistics and Philosophy 25 (5): 701–721. Steels, Luc. 2011. Modeling the Cultural Evolution of Language. Physics of Life Reviews 8 (4): 339–356. Szabó, Zoltán Gendler. 2020. The Goal of Conversation. Aristotelian Society Supplementary 94 (1): 57–86. Watson, Gary. 1996. Two Faces of Responsibility. Philosophical Topics 24 (2): 227–248. Williams, Bernard Arthur Owen. 2002. Truth and Truthfulness An Essay in Genealogy. Princeton: Princeton University Press. Williamson, Timothy. 2000. Knowledge and Its Limits. Oxford: Oxford University Press. Witek, Maciej. 2015a. An Interactional Account of Illocutionary Practice. Language Sciences 47: 43–55. ———. 2015b. Linguistic Underdeterminacy: A View from Speech Act Theory. Journal of Pragmatics 76: 15–29. Young, H. Peyton. 1993. The Evolution of Conventions. Econometrica 61 (1): 57–84.

llocutionary Force, Speech Act Norms, and the Coordination and Mutuality of Conversational Expectations Sanford C. Goldberg

1 Introduction Marina Sbisà’s work in pragmatics has been an inspiration for many of us who think about the nature of speech acts. For my part, I would highlight the importance of her reflections on the normativity of speech acts, as well as her arguments emphasizing the inadequacy of speaker-based (intentionalist) accounts of illocutionary force. In this chapter, my aim is to suggest that the sorts of arguments she has offered in these connections may have implications beyond those she herself has noted. But my overarching ambition is to advance an analogy to which Sbisà has returned throughout her career, in her attempt to capture the nature of the illocutionary dimension of speech. Thinking of conversation as “a speech activity which is characterized by the bringing about of certain illocutionary

S. C. Goldberg (*) Department of Philosophy, Northwestern University, Evanston, IL, USA Arché Research Center, University of St Andrews, St Andrews, Scotland e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Caponetto, P. Labinaz (eds.), Sbisà on Speech as Action, Philosophers in Depth, https://doi.org/10.1007/978-3-031-22528-4_9

195

196 

S. C. Goldberg

effects of the expositive kind,” Sbisà (1992, 109) suggested that we characterize conversation by concentrating on the obligations and the rights assigned to participants by the fact that one of them is speaking … and the obligations and rights assigned to the speaker and hearer by the intersubjectively validated performance of illocutionary acts.1

The analogy here—between speech exchanges, on the one hand, and domains structured by rights and obligations, on the other—is highly suggestive. It is also one for which I have a great deal of sympathy: one of my guiding ideas is to think of conversation as a site of the generation, modification, and satisfaction of normative expectations.2 Whether we think in terms of rights and obligations or instead in terms of normative expectations, this sort of approach can be developed so as to underwrite Sbisà’s critique of intention-based accounts of illocutionary force. In this chapter I will combine this sort of approach with an appeal to the constitutive rules governing speech acts themselves. The result, I contend, is a more satisfying account both of the nature of illocutionary force itself and of what it takes to perform an illocutionary act successfully.

2 Normative Expectations in Speech Exchanges By analogizing speech exchanges to domains structured by rights and obligations, Sbisà is capturing a salient feature of speech exchanges: these take place in circumstances that are normatively structured, in the minimal sense that what actions are permissible or impermissible at a given point in the exchange evolves partly as a function of the speech acts that

 Sbisà (2009), which seems to endorse this sort of approach to taxonomizing illocutionary acts, argues that the association of an effect profile with each type of illocutionary act is itself a matter of convention. 2  For an approach that falls within this broad family, see Kukla and Lance (2009). I myself attempt to develop the basis for such an approach to conversation in Goldberg (2020). 1

  llocutionary Force, Speech Act Norms, and the Coordination… 

197

are being performed.3 If in the course of an ongoing conversation you ask me a question, then—all else equal, and other conditions (to which I return below) being satisfied—it is permissible for me to answer you, impermissible for me to ignore you. Alternatively, if you request that I do such-and-such, then—again, all else equal, and other conditions being satisfied—it is permissible for me to do such-and-such, impermissible for me to change the subject altogether without first responding to your request. This humdrum observation points to a research question: can we model what is permissible or impermissible at any given point in a speech exchange, while simultaneously acknowledging the necessary background conditions and the social complications that make this affair so messy? Can we do so in a way that captures the distinctive contributions the various types of speech acts make to these evolving permissions? I see Sbisà’s (1992) “analogy” as suggesting a vocabulary in which to frame our attempt at such a model: she recommends the vocabulary of rights and obligations. My own preferred vocabulary, which involves talk of normative expectations instead, can be seen as a variant on Sbisà’s. To a very rough first approximation, both rights and obligations can be conceived as types of normative expectations to which one is entitled, where to be entitled to these normative expectations is (among other things) to be entitled to redress in cases where the expectations themselves are violated. Thus rights can be conceived as a species of normative expectation one is entitled to have of others (namely, that they not prevent one from doing or having such-and-such, or perhaps even that they provide positive support to one in this regard), whereas obligations can be conceived as a species of normative expectations others are entitled to have of one (namely, that one do or refrain from doing such-and-such). What recommends proposals of this sort, I submit, is their ability to make sense of salient features of speech exchanges. Start with the idea that the performance of speech acts themselves makes certain things permissible or impermissible (for the speaker, for the addressee, for the audience at large). We might then observe that different speech act tokens  The idea that we think of speech exchanges as generating permissions of various sorts, which evolve over the course of conversation (as a function of the acts that are performed in it), is owed to Lewis (1979). More recently, it has been central to the work of Mary Kate McGowan. See McGowan (2019). 3

198 

S. C. Goldberg

have different effects, and that these differences can be grouped into classes. This would appear to bolster the hypothesis that there are speech act types: any token speech act can be typed according to its systematic effects on what is (im)permissible in the speech context. Using Sbisà’s language, the proposal would be to think of participants of a speech exchange as bearers of rights and obligations, and to see the performance of speech acts themselves as having systematic effects on the rights and obligations of the participants, where any token speech contribution can be typed according to these effects. Indeed, this is one natural way to articulate (and to make a case for the theoretical utility of ) the notion of an illocutionary act.4 We begin by distinguishing the various classes of systematic effects associated with the performance of the wide range of speech acts; we then label each of these classes (tentatively taking each one to be the profile of, and so to correspond to, a distinctive type of “illocutionary act”); and finally we proceed to ask how a speaker might pull off an act of each sort—that is, how she manages to perform an act with the profile in question—in such a way that her action is intelligible to her audience. Here, the proposed vocabulary—whether that of rights and obligations, or that of normative expectations—earns its keep by enabling us to capture generalizations that illuminate the dynamics of speech exchanges. On this sort of picture, what would be distinctive of the illocutionary dimension of speech—as distinct from the two other dimensions in Austin’s triad—would be the association with this range of effects, wherein a speech act is of a type whose tokens modify the normative relationship between speaker and audience in systematic (and predictable) ways. Such a research program would appear to be in keeping with the spirit of some of Sbisà’s more recent work on illocution. For example, Sbisà (2012, 47) proposes that we think of the illocutionary dimension of a speech act as the dimension which “produc[es] a change in the modalities affecting the speaker-addressee relationship.” This, she thinks, is a matter of

 As we will see below, something like this appears to be endorsed by Sbisà herself; see Sbisà (2009).

4

  llocutionary Force, Speech Act Norms, and the Coordination… 

199

whether the utterance makes itself recognizable as producing (or as apt to produce) a certain illocutionary effect … [that is,] an effect on the relationship between speaker and addressee, to be described in terms of modal predicates […]. (Sbisà 2012, 47)

In making this proposal Sbisà is implicitly addressing two important questions that will need to be faced by any adequate account of illocution. The first is what we might call the characterization question: what is the “relationship-changing” profile, that is, the relevant set of effects on the “speaker-addressee relationship,” with which we identify a given (type of ) illocutionary act? The second is what we might call the implementation question: how do speakers pull off acts of this type? In sum: it is fruitful to theorize about illocutionary force by treating the illocutionary dimension of speech as pertaining to the systematic modification of the normative relation between speaker and audience. Sbisà’s “analogy” (as I am understanding it) recommends that we capture the relevant range of facts—the facts we expect our theory of illocutionary force to predict and explain—in “modalized” language. More specifically, the illocutionary force of a speech act itself can be understood in terms of (some subset of ) the systematic effects it has on the rights and obligations (alternatively: the normative expectations) in play in the speech context itself. (Here illocutionary force itself is hypothesized as the source of these sorts of effects: particular speech acts have the effects they have because they have the illocutionary force they have.) Finally, as we pursue the project of characterizing illocutionary force in terms of the relevant range of effects as captured in the privileged vocabulary, we ought to keep in mind (as Sbisà herself does) the need to explain how speakers pull off acts that have the normative profile in question, and how audiences recognize this.

3 Constitutive Rules for Speech Acts: The Case of Assertion So far I am in agreement with Sbisà regarding how we might pursue the part of speech act theory that focuses on illocutionary force. I now want to argue this approach is well-advised to embrace a constitutive rules

200 

S. C. Goldberg

account of speech acts. Since this claim may be surprising, I inherit the burden of proof. In this section I attempt to meet that burden. In order to focus my discussion here, I will focus on the particular speech act of assertion. I do so primarily because assertion is the speech act about which I have spent the most time thinking (and so it is the one about which I have the most to say). But my hope would be that what I have to say about assertion might be said, mutatis mutandis, about other speech acts as well. (If this is not so, I would regard this as a major weakness of my proposal.) I will rely on an intuitive grasp of the category of assertion itself; roughly speaking, I have in mind the class of speech acts whose members constitute what pretheoretically we might call making or advancing a claim. The Sbisà-inspired strategy I am exploring is one according to which we aim to characterize the illocutionary force of a speech act in terms of its relevant effects on the speech context, where these effects are regarded as including “change[s] in the modalities affecting the speaker-addressee relationship,” and where these changes are to be described in the privileged modal vocabulary. Since the modal vocabulary I will use is that of (our entitlement to) normative expectations, I will speak of the effects in question as the “normative effects” of the speech act, and I will designate the characterization of a speech act’s normative effects as its “normative profile.” (In order not to confuse predictive with normative expectations, I will use a subscript—“expect”—when the kind of expectations I have in mind are normative expectations.) The hypothesis I am pursuing is that, qua illocutionary act, assertion can be characterized by its normative profile. Before addressing what that profile is, however, I should underscore that, while I agree with Sbisà that a speech act’s normative effects bear on the speaker-addressee relationship, I do not assume that this is as far as these normative effects extend. On the contrary, I take these effects extend to include all participants in the conversation (whether or not they are the explicit addressee), and some may even bear on anyone who observes the speech act (whether they are part of the conversation or not). In conceiving of a speech act’s normative effects in this expansive way, I am bearing in mind the implementation question: insofar as speech acts are public acts which are intended to be (as Sbisà herself regularly reminds us)

  llocutionary Force, Speech Act Norms, and the Coordination… 

201

publicly intelligible, we would anticipate that some of the effects at issue are fully public, and so that in principle they might bear on others who observe the exchange even if they are not being addressed themselves. To be sure, we still might want to distinguish between those normative effects that bear exclusively on the addressee, those that bear on conversational participants more widely, and those that extend further still to include “mere bystanders.”5 Even so, the normative profile that I will associate with the speech act of assertion includes effects that go beyond that of the speaker and the addressee. With this caveat in place, we can address our central question: what does it take to make an assertion? I submit that to assert is simply to perform a speech act with the normative profile distinctive of assertion.6 Thus, given a speaker S whose utterance U is observed by audience A, U counts as an assertion that p just in case the following (Normative Profile of Assertion) holds of U:

5  Relevant here is the distinction, owed to Kukla and Lance (2009), between two types of normative effect a speech act might have: agent-relative and agent-neutral. Agent-relative (normative) effects are those normative effects that pertain only to particular members of the audience. Consider the case in which a speaker demands that her friend do such-and-such: the normative expectation that the demand will be satisfied is an expectationN that her friend will do such-and-such (and does not bear on others who happened to overhear the demand being made). Contrast the normative effects of a speaker’s making an assertion. These effects (e.g., of authorizing belief in the content asserted) are “agent-neutral” in that they are relevant to anyone who observed the act (Kukla and Lance 2009, 18). 6  Here I draw on work I first presented in Goldberg (2015). I should flag, though, that my proposed answer differs from Sbisà’s in two important respects. Sbisà holds that there must exist some sort of “procedure” which the speaker “has to execute correctly and completely enough” in order to count as having performed the act in question (Sbisà 2020, 163). And she also holds that for any type of illocutionary act the procedure in question is associated with the characteristic conventional effects, and these are the normative effects that characterize the illocutionary type (Sbisà 2018, 2020). For assertion these effects involve the speaker’s incurring obligations (i) to be consistent with the content of her assertion and (ii) to offer reasons or evidence when called upon to do so, and the audience’s acquiring an entitlement (iii) to use the content asserted as a reason for action and belief. My own reasons for resisting these features will emerge below. (With thanks to Paolo Labinaz for suggesting that I make this contrast clear.)

202 

S. C. Goldberg

NPA S’s utterance makes it mutually manifest that S is entitled to expectN to be recognized as having presented herself as epistemically authoritative with respect to (what S is presenting as) the truth that p.7

A complete theory of assertion would then need to do at least three things. First, it would need to say what “epistemic authority” amounts to: must S know that p, be justified in believing that p, be epistemically certain that p, or what? Second, it would need to say how it is that, by uttering some words in a given context, S manages to pull off an act that has this normative profile.8 And third, it would need to say how a speech act that has this normative profile is apt to transmit knowledge.9 For the purpose of concreteness, I will assume the knowledge-based answer to the first question: a speaker’s performance entitles her audience to expectN the speaker to know whereof she speaks. Though this is not my actual view (see Goldberg 2015), not much hangs on the precise answer to this question for my purposes here. I go with the knowledge-­based answer both because it is the most popular answer these days10 and because it plays well in contexts in which one is interested in the aptness of assertion to transmit knowledge (the third matter above). I am more interested in the implementation question, which was the second matter above: how does a speaker pull off an act of this sort? How does a speaker do something that makes it mutually manifest that she is entitled to  It may be that this characterization of assertion’s normative profile needs to be more complicated, so that it reads: 7

S’s utterance (i) entitles A to expectN S to be epistemically authoritative with respect to (what S is presenting as) the truth of p, (ii) entitles S to expectN A to recognize that S has presented herself as epistemically authoritative with respect to the truth of p, and (iii) entitles both S and A to expectN that (i) and (ii) are mutually manifest to S and A. Whether the (abbreviated) characterization in the main text is equivalent to this longer characterization depends on whether one’s being recognized as presenting oneself as epistemically authoritative with respect to the truth of p is sufficient to entitle an audience to expectN one to be epistemically authoritative. Since the expectationN in question is normative rather than predictive, this seems plausible to me. 8  This is what I called the implementation question. 9  I attempt to do all three things in Goldberg (2015). The last of the three is complicated by the fact that the sort of “expectation” in the profile are normative expectations (expectationsN), not predictive ones. My own view is that several approaches to the epistemology of testimony are confused on just this point. See Goldberg (2011). 10  See Simion and Kelp (2020) for an overview of the state of the literature.

  llocutionary Force, Speech Act Norms, and the Coordination… 

203

expectN her act to be recognized as one in which she presented herself as epistemically authoritative with respect to the truth of a given proposition? It is in answering this question that the postulation of a constitutive rule for assertion earns its keep. In particular, we can explain how a speaker manages to pull off an act meeting the description above if we assume that 1 . assertion answers to a constitutive norm, 2. the norm itself requires that the speaker know (or be epistemically authoritative regarding) the truth of the proposition in question, and 3. this rule is a pragmatic rule of language (and so it will be mutually manifest that this rule is known by any pragmatically competent speaker of the language). In bringing this out I will be helping myself to a prior notion of a locutionary act with a propositional content,11 and I will be assuming that the audience in question can recover which locutionary act a speaker performed on a given occasion of speech. With this as background, my claim is that we can account for how S can “pull off” an act that meets the profile above, and so has the illocutionary force of an assertion, in terms of (i)–(iii). To see this, suppose that (i)–(iii) are true. Suppose further that audience A can work out what locutionary act speaker S performed on occasion O, and that S can take this for granted. In that case, we have a scenario in which it is mutually manifest that S (in producing utterance U) performed a locutionary act with the content that p. The question then arises for A: what was S’s point in performing such an act? A can approach this question by asking after the relevance of the content that p to the present circumstance. Suppose that it is manifest that there would be a practical value to having information whether p in the present circumstances. Then the pragmatic rule mentioned in (i)–(ii) would be not only relevant to but also salient in the circumstances: the performance of an act meeting that profile would enable S to do something that has this manifestly valuable effect. Given (iii), we can assume that the availability of such an act is mutually manifest to S and to A. And so we see that whenever having information regarding whether p is relevant to the  Of course a full theory of speech acts would need to characterize these acts as well, but it is beyond the purview of this chapter to do so. 11

204 

S. C. Goldberg

circumstances, A has (strong but defeasible) grounds for construing S’s utterance as a speech act with the normative profile captured by NPA— that is, the normative profile of an assertion. By interpreting S’s speech in that way, A would thereby construe S’s act as relevant to the present circumstances. To be sure, in any particular case in which an audience A is inclined to construe S’s act as an assertion, such a construal is defeasible: perhaps construing S’s act as an assertion fails to cohere with other features of S’s act; or perhaps there are other construals of the illocutionary force of S’s act which are also consistent with the facts and which are relevant in the circumstance. But absent such conflicting evidence or coherent alternative construals, the construal of S’s act as an assertion would stand. This is one possible basis on which A, having identified the locutionary act S performed, might come to be in a position to recognize that S’s speech act (has assertoric illocutionary force, and so) is an assertion. In this one an audience relies on the utility of information (were it to be had) to conclude that the point of the speaker’s speech contribution is to provide such information. Of course, there are other ways to arrive at a similar conclusion. Suppose that the content of your locutionary act is not information whose possession by an audience might be anticipated to have high utility. Suppose instead that the content is something regarding which it is obvious that you (the speaker) are knowledgeable: what you ate for dinner last night, what you did while on holiday, what thought just went through your head, etc. Then the fact that I take it to be obvious that you are knowledgeable about these things raises the prospect that you are performing an act governed by a rule that would require you to be knowledgeable—the rule mentioned in NPA’s (ii), for example. Now suppose that in such circumstances I can render the point of your speech contribution intelligible if I construe you as purporting to conform to that rule (since in that case I can see you as purporting to “inform me” of what you had for dinner etc., and I can imagine all sorts of reasons why this might be precisely what you want me to be take you to be doing). So long as there is no equally salient plausible construal of the point of your contribution, this would constitute a basis for construing you as having performed an act with the normative profile of assertion. (There may be other bases for this sort of construal as well.)

  llocutionary Force, Speech Act Norms, and the Coordination… 

205

The foregoing account carries an interesting suggestion. It suggests that, as a type of illocutionary act, the normative profile of assertion will loom large in our speech exchanges: we will be inclined to interpret many utterances as having such force. We can bring this out as follows. According to the account just described, we have a motive to construe a locutionary act as an assertion in any context in which there is practical value to having the information presented in the speaker’s locutionary act, as well as in any context in which the content of the locution is one regarding which the speaker can be presumed knowledgeable. We might say that in such contexts the normative profile associated with assertion (NPA) has an enhanced salience. Call these NPA’s “enhanced salience conditions.” I submit that NPA’s enhanced salience conditions are very common. It stands to reason, then, that whenever these conditions hold, the audience has a motive to consider whether the speaker’s locutionary act satisfies NPA. And when that profile does fit the facts of the speaker’s utterance, the utterance will (fallibly) be construed as an assertion. So to the extent that our speech exchanges take place in contexts in which we have strong interests in (epistemically high-quality) information, or in contexts in which it is obvious that the content of one’s saying is a content regarding which one is (or might reasonably be regarded as) knowledgeable, assertion will loom large as a construal of one another’s speech contributions. I would speculate that such an interpretation will be nearly irresistible in cases in which the speaker uttered a sentence in the indicative, under conditions in which she gave no clear indication of having some interest other than information exchange.12 By what right do I say that, in such cases, A would thereby come to be in a position to recognize that S’s speech act is that of an assertion? Here is where we do well to appeal to both assertion’s normative profile (NPA) and the constitutive rule of assertion (as captured in (i)–(iii) above). According to assertion’s normative profile,

 What should be said of performatives, such as “I order you to leave”? They are in the indicative; should they be regarded as assertions as well? I am not certain whether, in addition to their status as performatives, they should also be classified as assertions. Happily, I don’t think the matter is decided by what I say here about assertion, though I hope to return to this at a future time. (With thanks to Laura Caponetto for raising this issue.) 12

206 

S. C. Goldberg

NPA S’s utterance makes it mutually manifest that S is entitled to expectN to be recognized as having presented herself as epistemically authoritative with respect to (what S is presenting as) the truth that p.

Suppose an audience A has a motive to determine whether the normative profile of a given speech contribution is that of NPA.  How might A approach this? Well, A might reason as follows: S is entitled to the expectationN in question if and only if S did something to entitle herself to that expectationN. This would involve S’s actually presenting herself as epistemically authoritative with respect to the truth that p. Given (i)–(iii), A can discern how S might do just that. In particular, S might employ the speech act which is governed by the constitutive rule mentioned in (ii). Suppose S did so. Then her act would be proper—it would conform to the rule mentioned in (ii)—if and only if S was relevantly epistemically authoritative, that is, if and only if S knows that p. And given (iii), the fact that these are the propriety conditions on S’s act is itself something that is mutually manifest. But now when one performs an act regarding which it is mutually manifest that one’s act is proper only if one knows that p, then one can be taken to have presented oneself as knowing that p. And since this can be discerned on the basis of an audience’s pragmatic competence (as per (iii)), S is entitled to expectN to be recognized as having presented herself in this way. In sum, we have an account both of how a speaker can “pull off” an act with the normative profile of assertion, and of how any pragmatically competent audience can recognize this. It is worth noting here that, having followed Sbisà in offering a characterization of illocutionary force in terms of the normative effects of performing an act of the relevant sort,13 I went on to use this characterization to offer support for a particular hypothesis about the speech act of assertion. In particular, I have argued that we can account for assertion’s having the illocutionary force it has if we assume that it is governed by a constitutive epistemic rule. My argument on this score does not establish the constitutive rule hypothesis, but it does make an initial case for it, and it puts the burden on those who reject this characterization of assertion. Simply put, the present hypothesis enjoys a distinct advantage over  Bear in mind the differences between our accounts (for which see footnote 6).

13

  llocutionary Force, Speech Act Norms, and the Coordination… 

207

any alternative that can’t account for assertion’s normative profile. I mention this as a cautionary note in connection with Sbisà’s original hope for the “analogy” she offered. She wrote, 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. (Sbisà 1992, 109)

Here, Sbisà appears to be contrasting an account of speech acts that highlights their “normative profile” with one that assigns constitutive rules to them. However, if the foregoing considerations are sound, this sort of contrast is illusory. On the contrary, at least when it comes to speech acts with the illocutionary force of an assertion, it appears that, having characterized the illocutionary act itself in terms of its normative profile, our best account of that profile—of how speakers manage to “pull off” illocutionary acts with this normative profile, and to do so in a way that is intelligible to their audiences—is by appeal to constitutive rules.14

4 Mutuality in Speech Exchanges The foregoing picture has an implication worth highlighting, as doing so will enable us to underscore an important feature of conversational exchanges. This has to do with the mutuality of the illocutionary dimension of speech exchanges.15 When it comes to the facts that determine the illocutionary force of an utterance, neither the speaker’s intentions nor  In terms of Sbisà’s own account of illocutionary acts, we might say that it is by appeal to constitutive norms that we can account for how it is that speech acts have “conventional effects” (Sbisà 2002, 2009). It would appear that Sbisà herself has been moving toward this sort of view; see, for example, Sbisà (2018, 23). (With thanks to Paolo Labinaz for bringing Sbisà’s more recent position to my attention.) 15  The points I will be stressing have been emphasized previously by Sbisà herself (Sbisà 1984), as well as by Hornsby (1993, 2011) and Hornsby and Langton (1998). 14

208 

S. C. Goldberg

the audience’s beliefs regarding those intentions call the shots. On the contrary, illocution is the type of act that must be intelligible—on pain of what Austin might have called a “misfire.” After indicating how this is an implication of the foregoing picture, I will use this result to raise doubts regarding strongly intentionalist accounts as well as strongly audience-­based accounts of illocutionary force. The theoretical strategy I have been proposing is to characterize illocutionary force in terms of the various normative profiles (associated with types of acts of speech), and then to account for how speakers pull off acts with the relevant normative profile by appeal to constitutive (pragmatic) rules. If sound, the result of such a strategy would be an enumeration of the distinctive normative profile of each type of illocutionary act, together with an account of how speakers manage to perform acts of each type. In the latter respect, to perform of an act of the relevant type, it suffices that a speaker acts in such a way as to enable an audience to work out the normative effects of the act. I suggested that constitutive rules earn their keep precisely here: assuming that the constitutive rules of each of the illocutionary acts is a matter of common knowledge—something that holds if we assume that knowledge of these rules characterizes pragmatic competence—the speaker’s utterance in the speech context should guide the audience to draw on the right set of rules, that is, the rules that (when applied to the speaker’s utterance in the context) enable the audience to capture the normative effect(s) intended by the speaker. I submit that, when it comes to the successful performance of an illocutionary act, the foregoing picture places constraints on both the speaker and the audience. Let us consider first the perspective of the speaker. According to the foregoing picture, the speaker is not in a Humpty-Dumpty-type situation16 where she is free to intend her words to be taken however she wishes. On the contrary, if she is to be successful in pulling off the act in question, a speaker must do her part to enable her audience to work out  In Through the Looking Glass, Lewis Carroll characterized the Humpty-Alice exchange as follows:

16

“When I use a word”, Humpty Dumpty said in rather a scornful tone, “it means just what I choose it to mean – neither more nor less”. “The question is”, said Alice, “whether you can make words mean so many different things”.

  llocutionary Force, Speech Act Norms, and the Coordination… 

209

the intended normative effects.17 Performances in which no audience can work out these effects are to be described, not as cases in which there is a speech act with an intended illocutionary force which fails to attain audience uptake, but rather as failed acts, acts that have no illocutionary force to speak of. This point can be reinforced by considering my proposed characterization of assertion’s normative profile: NPA S’s utterance makes it mutually manifest that S is entitled to expectN to be recognized as having presented herself as epistemically authoritative with respect to (what S is presenting as) the truth that p.

As I noted above, if S is to be entitled to expectN to be recognized in this way, S—unlike Humpty Dumpty—must have done something that entitles her to such recognition. And this constrains what S can do if S aims to be making an assertion: in effect, to assert is (among other things) to do something which would make it reasonable for an audience to take one to have asserted. In short, I submit that what entitles a speaker to expectN the relevant sort of recognition—the sort of recognition that forms part of assertion’s normative profile—is that her performance has features which, given the presence of these features in the speech context, make it reasonable for A to take S’s contribution in the intended way, and so to attain the recognition in question. Strong intentionalism about illocutionary force is not an option. I just spoke of the “reasonableness” demands on the speaker: her speech contribution must have features which (together with the salient aspects of the speech context) make it reasonable for her audience to take her contribution in the intended way, and so to recognize the intended normative effects of that contribution. But these very demands on the speaker succeed in placing demands on the audience as well. To see this, note that the reasonableness demand on the speaker amounts to a sufficient condition for pulling off the speech act in question: so long as the speaker does produce a performance whose features (together with the salient aspects  This of course is a familiar theme from Grice (1957) as well. At least it has been part of his characterization of that part of speaker meaning that falls under “conversational implicature.” Following Searle (1965), I submit that the point holds of speaker meaning generally: one can’t speaker-mean what cannot be worked out by one’s audience. 17

210 

S. C. Goldberg

of the speech context) make it reasonable for her audience to take her contribution in the intended way, she has pulled off the act. If under these circumstances it turns out that the audience favored an idiosyncratic interpretation of her utterance, and so doesn’t attain the recognition of the act’s intended normative effects, this is too bad for the audience. Here the proper description is that the act has the relevant illocutionary force, but audience uptake fails. Accounts of illocutionary force that hew closely to the actual audience’s actual uptake are not an option. All of this makes clear how important the “reasonableness” requirement is. The facts that determine illocutionary force must be such as to make it “reasonable” that an audience confronted with such facts would attain the relevant recognition.18 It is for this very reason that the illocutionary dimension of speech exchanges constitutes a dimension of the mutuality of such exchanges: the speaker is under pressure to produce an act whose features make its normative profile one that a reasonable audience could recover, and the audience is under pressure to favor reasonable construals of a speaker’s performance. It is with this sort of mutuality in mind that we can criticize the dual tendencies, to give either too much authority to the speaker or, alternatively, too much authority to the actual audience (and its actual uptake), when giving an account of illocutionary force.19 Indeed, I speculate that the existence of speech act rules (of the sort described in Sect. 3) exist precisely to cut down on the dimensions of freedom available to the speaker and the audience as they produce and apprehend speech acts. Insofar as we think about illocutionary force  We might say that in this sense a proper account of context, as employed in a theory of illocutionary force, is more “objective” than “cognitive” in the sense used by Sbisà (2002). As Sbisà (2009, 39) remarks, such a notion of context is one in which “illocutionary successfulness […] 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.” Focus on “how the world actually is” can be misleading, though, since (on my view) the relevant features of the world include features that determine what is “reasonable” to assume in context, and this, presumably, depends on assumptions that are standardly made and exemplified in familiar patterns of thought exhibited by members of a speech community. 19  For a recent discussion of this dual tendency, see McDonald (2022). (With thanks to Laura Caponetto for bringing this paper to my attention.) 18

  llocutionary Force, Speech Act Norms, and the Coordination… 

211

(with Sbisà) in terms of a (modal) normative profile associated with the performance of acts of the various types, we will see the act of participating in a speech exchange as placing speakers and audiences alike into a normatively demanding context.20 But then it seems that they must be equipped to recognize the sorts of normative demands they place on ­others, and that others place on them, by performing the speech acts they do. And it seems that this sort of recognitional capacity itself characterizes (a part of ) our pragmatic competence as speakers. The hypothesis of constitutive speech act rules, then, is a proposal that aims to characterize the cognitive dimension of this competence. At the same time, it must be underscored that knowledge of constitutive speech act rules alone will not enable one to discern the illocutionary force of the speech acts one observes. A good deal of background knowledge, including such things as a speaker’s social or professional role (Sbisà 2012, 42), or other features in the common ground, will typically be called upon by the audience as she determines the normative profile of an observed speech act. The point I am making at present is only that the existence of speech act rules serves to further constrain the set of candidate normative profiles that she might ascribe to a given speech performance.

5 Normative Demands on Audience Uptake One final issue that requires attention has to do with Austin’s own view as to the bearing of audience uptake on the successful performance of an illocutionary act. Since Sbisà appears to endorse Austin’s view of the matter, our comments bear on her position as well. Austin (1975) addressed the bearing of audience uptake on successful illocution in several places. Early in the book he makes this remark about performatives in particular: It is obviously necessary that to have promised I must normally (A) have been heard by someone, perhaps the promisee; 20

 See Sbisà (2012, 43).

212 

S. C. Goldberg

(B) have been understood by him as promising. If one or another of these conditions is not satisfied, doubts arise as to whether I have really promised, and it might be held that my act was only attempted or was void. (Austin 1975, 22–23)

Here is how Sbisà summarizes the point: [P]erformative utterances can achieve the conventional effect envisaged by the procedure to which they belong only if they are so received by the relevant audience. (Sbisà 2009, 47; italics added)

From this it is clear that, at least when it comes to performatives, Austin held that they are successful as performatives—they “achieve the conventional effect envisaged by the procedure to which they belong”—only if there is audience uptake. This, it seems, was a special case of a more general thesis Austin endorsed regarding the success conditions of illocutionary acts as such. The following passages are particularly explicit: Unless a certain effect is achieved, the illocutionary act will not have been happily, successfully performed […]. (Austin 1975, 116; italics added) I cannot be said to have warned an audience unless it hears what I say and takes what I say in a certain sense […]. [T]he performance of an illocutionary act involves the securing of uptake. (Austin 1975, 117; italics added)

Once again, Austin’s view is that audience uptake is required for the successful performance of an illocutionary act. As is well known, several important papers in speech act theory— including Langton (1993), Hornsby and Langton (1998), and Kukla (2014, 2018)—have followed Austin in this respect. Their guiding idea is that a speaker’s failure to secure uptake in her audience—failure by the audience to recognize the speaker’s relevant intentions—is tantamount to the speaker’s failure to have pulled off the illocutionary act itself. Thus we have the notions of illocutionary silencing and illocutionary disablement, as found in the papers mentioned above. The examples used to illustrate these phenomena are familiar. Langton and Hornsby offer circumstances characterized by the sexual norms deriving from pornography, where

  llocutionary Force, Speech Act Norms, and the Coordination… 

213

women’s would-be attempts to refuse a sexual advance fail to obtain uptake, and so do not succeed in being seen as—and so fail to amount to—refusals. Kukla offers circumstances in which, owing to pervasive misogyny and sexist attitudes toward women in positions of authority, a female boss’s would-be orders are taken as requests, and so fail to obtain uptake as—and so fail to amount to—the orders they were intended to be.21 In both sorts of case, failure of uptake is treated by these theorists as a failure by the speaker to pull off the illocutionary act she intended to pull off. Since the woman’s intention to refuse the advance is not recognized by her audience, she has not succeeded at refusing; since the boss’s intention to order her employees to do such-and-such was not recognized by them, she has not succeeded at making an order. These proposals are in keeping with Austin’s thoughts about the act of warning: “I cannot be said to have warned an audience unless it hears what I say and takes what I say in a certain sense.” For her part, Sbisà too endorses the requirement of audience uptake on successful illocution. But she goes further: she offers an argument for this claim which takes as a premise the idea that illocutionary acts are to be conceived in terms of their corresponding normative profiles. Her claim is that this conception is viable only if actual uptake is required for successful illocutionary performance (Sbisà 2009). Since I accept her conception of illocutionary acts but reject Austin’s thesis regarding audience uptake, her argument is of great interest to me. In the remainder of this section I respond to that argument. I begin by acknowledging that Sbisà formulates her argument, not in terms of normative profiles, but rather in terms of the “conventional effects” of an illocutionary act. But it is clear that by speaking of the “conventional effects” of an illocutionary act, she has something in mind like what I mean by the normative profile of the act. She writes,

 Kukla (2014) analyses this case as one in which (owing to audience reception) the boss counts as having made a request. Consequently, unlike Langton and Hornsby (1998), who treat audience uptake as necessary but not sufficient for fixing illocutionary force, Kukla (2014) appears to hold that audience uptake is sufficient. (With thanks to Laura Caponetto for highlighting this difference to me.) 21

214 

S. C. Goldberg

[…] [T]he kind of effect that, according to Austin, illocutionary acts “take” […] 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 behaviour 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. (Sbisà 2009, 45; italics added)

Here it is clear that what Sbisà calls the “conventional effects” associated with the various types of illocutionary acts, though not precisely the same as what I have been calling the “normative profile” of the various illocutionary act types, are in the vicinity: the effects Sbisà highlights are changes in the normative standing of speakers and audiences, effected by the performance of the speech acts themselves. What is of interest to me is her case for the claim that one who thinks of illocutionary acts in these terms has grounds to endorse the Austinian thesis that uptake is required for successful illocution. Here is her argument: [T]he 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 […]. (Sbisà 2009, 48; first italics added)

She summarizes this argument a bit later on when she writes that “what is revealing of the conventionality of the illocutionary act (understood as the conventionality of its effect) is the need to secure uptake” (Sbisà 2009, 49). In sum, Sbisà’s contention appears to be this: one can think of illocutionary acts in terms of their “conventional effects” (roughly, their normative profiles) only if one treats actual audience uptake as required for success in performing one of these acts.

  llocutionary Force, Speech Act Norms, and the Coordination… 

215

I have two worries about the line of argument that is supposed to support this contention. First, even if it is true that conventional effects are brought about only when there is “agreement about their coming into being among members of the relevant social group,” it does not follow that actual uptake is required for successful performance. For we might think that the form taken by this “agreement” lies in the standards of what is considered reasonable among the members of the group, where the standards in question bear on how to interpret the relevant pragmatic (including illocutionary) features of utterances. Curiously, Sbisà herself appears to recognize as much. In a comment that comes between the two quotes cited above she writes, [W]hat 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. (Sbisà 2009, 48; italics added)

It is far from obvious that these conditions require actual uptake. On the contrary, it seems that “the effect which the utterance is designed to produce” can be “indicated clearly enough to be identifiable and possibly agreed upon” even in cases that lack actual audience uptake.22 To illustrate, suppose that people around here agree on what would be a “reasonable” way to construe Sally’s utterance, where this includes a construal of the normative profile of that utterance. (Ask any of them and they’d all agree that her utterance has the normative profile associated with, e.g., an assertion.) Then this would be all the agreement we need if we are to ascribe to her utterance an illocutionary force. More importantly, the fact that her utterance has features which, together with our assumptions about what is reasonable, lead us to ascribe a particular illocutionary force to that utterance, would support the idea that her utterance amounts to the successful performance of an illocutionary act. This would be so even on  Interestingly, in more recent work Sbisà appears to endorse an alternative way to understand Austin’s requirement on the securing of uptake, which version is more in line with what I am suggesting above. See Sbisà (2013, 31–32). (I thank Laura Caponetto for bringing this to my attention.) 22

216 

S. C. Goldberg

those occasions on which she is speaking to an audience whose idiosyncratic response is tantamount to a failure of uptake. Actual uptake, then, is not needed. My second worry about Sbisà’s line of argument (in defense of Austin’s actual-uptake condition on successful illocution) emerges from this first one. Simply put, when it comes to an account of the conditions on successful performance of an illocutionary act, there are additional options beyond (i) the requirement of actual uptake and (ii) the requirement of a mere intention to secure actual uptake. Indeed, the point is patent in cases like the one I just described: the speaker, intending to secure uptake, makes an utterance which by the prevailing standards of reasonableness are associated with a given normative profile. This, I submit, would suffice for successful performance of an illocutionary act, whether or not it attained actual uptake in the circumstances. And if this is so, then we have reason to think that, while the speaker’s mere intention to secure uptake is not sufficient for successful illocution, actual uptake is not necessary. It might be wondered how there can be the sort of agreement on which I am relying here: an agreement on the standards of reasonableness in interpreting speech acts. Partly this reflects the facts that language is a rule-­ governed activity, and those rules delimit the space of candidate interpretations. Still, as Sbisà herself has reminded us throughout her career, the facts that are potentially relevant to illocutionary construal are many, and language rules alone will rarely if ever succeed in enabling a unique construal. Return to the case of assertion. Given a speaker who performed a locutionary act by uttering a sentence in the indicative mood, no mere appeal to linguistic (syntactic, semantic, and pragmatic) rules will enable an audience to determine that the speaker made an assertion. After all, the speaker might have been advancing a guess (rather than making a claim); she might have been joking, or speaking ironically; intonation might make it clear that, while the sentence she uttered was in the indicative, she was actually asking a question; or what-have-you. What enables audiences to single out one illocutionary force, as the force that is to be assigned to an observed locutionary act, is (not just linguistic rules but also) what is part of the common ground, including the range

  llocutionary Force, Speech Act Norms, and the Coordination… 

217

of assumptions and social norms that are prevalent in her community. This range of assumptions and social norms inform what I am calling the standards of reasonableness in interpreting speech acts. These assumptions and social norms are part of the common playing field on which arbitrary members of the same speech community are playing, and it is with reference to these that we determine what is reasonable in the interpretation of speech—including in the interpretation of the illocutionary dimension of speech. This, I submit, is yet another face of the sort of mutuality that characterizes speech exchanges.23

References Austin, John L. 1975. In How to Do Things with Words, ed. James O. Urmson and Marina Sbisà, 2nd ed. Cambridge, MA: Harvard University Press. Goldberg, Sanford C. 2011. Putting the Norm of Assertion to Work: The Case of Testimony. In Assertion, ed. Jessica Brown and Herman Cappelen, 175–195. Oxford: Oxford University Press. ———. 2015. Assertion: On the Philosophical Significance of Assertoric Speech. Oxford: Oxford University Press. ———. 2020. Conversational Pressure: Normativity in Speech Exchanges. Oxford: Oxford University Press. Grice, Paul H. 1957. Meaning. Philosophical Review 66 (3): 377–388. Hornsby, Jennifer. 1993. Speech Acts and Pornography. Women’s Philosophy Review 10: 38–45. ———. 2011. Subordination, Silencing, and Two Ideas of Illocution. Jurisprudence 2 (2): 379–440. Hornsby, Jennifer, and Rae Langton. 1998. Free Speech and Illocution. Legal Theory 4 (1): 21–37. Kukla, Quill. 2014. Performative Force, Convention, and Discursive Injustice. Hypatia 29 (2): 440–457. ———. 2018. That’s What She Said: The Language of Sexual Negotiation. Ethics 129 (1): 70–97.  I would like to express my gratitude to Laura Caponetto and Paolo Labinaz for their very helpful comments and criticisms of an earlier version of this chapter. Any remaining errors or misinterpretations are entirely my responsibility. 23

218 

S. C. Goldberg

Kukla, Quill, and Mark Lance. 2009. “Yo!” and “Lo!”: The Pragmatic Topography of the Space of Reasons. Cambridge, MA: Harvard University Press. Langton, Rae. 1993. Speech Acts and Unspeakable Acts. Philosophy and Public Affairs 22: 293–330. Lewis, David. 1979. Scorekeeping in a Language Game. In Semantics from Different Points of View, ed. Rainer Bäuerle, Urs Egli, and Arnim Stechow, 172–187. Berlin and Heidelberg: Springer. McDonald, Lucy. 2022. Reimagining Illocutionary Force. The Philosophical Quarterly 72 (4): 918–939. McGowan, Mary Kate. 2019. Just Words: On Speech and Hidden Harm. Oxford: Oxford University Press. Sbisà, Marina. 1984. On Illocutionary Types. Journal of Pragmatics 8: 93–112. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 1992. Speech Acts, Effects, and Responses. In (On) Searle on Conversation, edited by Herman Parret and Jef Verschueren, 101–111. Amsterdam: John Benjamins. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2002. Speech Acts in Context. Language and Communication 22: 421–436. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2009. Uptake and Conventionality in Illocution. Lodz papers in Pragmatics 5: 33–52. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2012. Classifying Illocutionary Acts, or, a Tale of Theory and Praxis. In Pragmaticizing Understanding: Studies for Jef Verscheuren, ed. Michael Meeuwis and Jan-Ola Östman, 39–52. Amsterdam: John Benjamins. ———. 2013. Locution, Illocution, Perlocution. In Pragmatics of Speech Actions, ed. Marina Sbisà and Ken Turner, 25–75. Berlin: De Gruyter. ———. 2018. Varieties of Speech Act Norms. In Normativity and Variety of Speech Actions, edited by Maciej Witek and Iwona Witczak-Plisiecka, 23–50. Leiden: Brill. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2020. Assertion among the Speech Acts. In Oxford Handbook of Assertion, edited by Sandford Goldberg, 159–178. Oxford: Oxford University Press. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press.

  llocutionary Force, Speech Act Norms, and the Coordination… 

219

Simion, Mona, and Chris Kelp. 2020. Assertion: The Constitutive Norms View. In Oxford Handbook of Assertion, ed. Sanford Goldberg, 59–74. Oxford: Oxford University Press. Searle, John R. 1965. What is a Speech Act? In Philosophy in America, ed. Max Black, 221–239. Ithaca: Cornell University Press.

Speech in Non-ideal Conditions: On Silence and Being Silenced Alessandra Tanesini

1 Introduction Silencing is a ubiquitous occurrence. People who belong to groups that are marginalised or stigmatised often struggle to make themselves heard. Their contributions to debates or discussions are not taken seriously or dismissed. Oftentimes, out of an overwhelming sense of the futility of it all, they give up even trying to speak, or at least to talk to some people about some topics (Eddo-Lodge 2017). Their silence in these cases is a product of self-silencing. These are examples of what Rae Langton (1993) has called ‘locutionary silencing’; it occurs whenever some individuals remain silent because they feel intimidated, are assaulted or coerced, or finally because they think that no one is prepared to listen to them. These silences, however, speak volumes. They convey hostility, resentment, and disappointment.

A. Tanesini (*) School of English, Communication and Philosophy, Cardiff University, Cardiff, UK e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Caponetto, P. Labinaz (eds.), Sbisà on Speech as Action, Philosophers in Depth, https://doi.org/10.1007/978-3-031-22528-4_10

221

222 

A. Tanesini

In this chapter I argue that norms defeasibly entitling competent linguistic agents participating in a conversation to presume that silence indicates acceptance have silencing, combined with plausible deniability of responsibility, as their function. These norms would license the interpretation of silences, including those that are intended to be hostile, as indications of agreement. I thus disagree with Sanford Goldberg (2020), who, in his Conversational Pressure, has defended one such norm whilst acknowledging that in less-than-ideal circumstances it can be widely misapplied with resulting harmful effects. I also show Goldberg’s views to be an example of ideal theory in the sense defined by Charles W. Mills (2004). Because of his reliance on idealising assumptions Goldberg neglects the possibility that some conversational norms might be instituted by means of exercitive assertions whose felicitous performance depends on pre-existing unjust power differentials. Yet alertness to these possibilities is essential to uncover the ways in which discursive exchanges can be a source of injustice and oppression. In this chapter I draw inspiration from Marina Sbisà (2020) to argue that when some linguistic agents hold more power than their interlocutors they are often able to impose the agenda and the terms of the conversation. I hasten to add that I select Goldberg’s defence of this norm as my target because I think it offers one of the most careful arguments in support of norms of this kind and because it is clearly sensitive to the oppressive potential of assuming that silence indicates acceptance. The chapter consists of four sections. In the first I introduce Goldberg’s norm of no silent rejections (hereafter, NSR) and describe the evidence he marshals in support of the claim that our linguistic communities de facto conform to this norm. In the second section I provide empirical evidence from multimodal conversational analysis that speaks against Goldberg’s claim of de facto conformity. In the third section, after a description of Goldberg’s analysis of the role of NSR in the facilitation of silencing, I argue for an alternative explanation, inspired by Sbisà’s (2018, 2020) account of exercitive assertions, that attributes to NSR the function of silencing. In the final section, I examine Goldberg’s argument to the effect that NSR is a norm that we should rationally endorse. I show that Goldberg’s arguments are

  Speech in Non-ideal Conditions: On Silence and Being Silenced 

223

unconvincing and suggest that he might have been led astray by his commitment to ideal theory. I conclude the chapter with a brief articulation of an alternative proposal.

2 The Norm of No Silent Rejections (NSR) In this section I describe a social conversational norm defended by Goldberg (2020) in his Conversational Pressure. This is the norm of no silent rejections (NSR). Subsequently, I detail the evidence adduced by Goldberg in support of the claim that linguistic communities, like our own, de facto conform to NSR: NSR: In all speech exchanges which are Gricean conversations, all competent language users enjoy a default (albeit defeasible) entitlement to expect that an audience regarding whom it was manifest that he has remained silent in the face of a publicly made assertion has not rejected that assertion. (Goldberg 2020, 159)

NSR only applies to speech exchanges that are conversations in a narrow technical sense. The latter are activities which are cooperative because they are governed by Grice’s Cooperative Principle and its maxims (Grice 1975).1 For Grice, and for Goldberg, the principle sets normative standards to which we conform, and which we have reasons to endorse provided that we care for the goals of communication which include persuasion and the sharing of information (Goldberg 2020, 160–161). These reasons to comply with the Cooperative Principle are practical reasons; they are also pro tanto reasons, since participants in conversations might need also to consider reasons of other kinds. For example, moral reasons might justify lying and thus violating a conversational norm. In the context of speech exchanges that are cooperative activities, NSR licenses every linguistically competent participant defeasibly to presume that no other participant who has manifestly observed (and understood)  This principle prescribes that one makes one’s contribution to a conversation such ‘as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which [one is] engaged’ (Grice 1975, 26). 1

224 

A. Tanesini

an assertion would be silent if they harboured doubts about, or rejected, said assertion. That is, every participant to a conversation is (defeasibly) entitled to make a presumption about everybody else. NSR is thus not restricted to the addressees of the speaker’s assertion. Instead, it equally applies to everyone who was within earshot, of whom it is clear that they understood what was said, and who can plausibly be considered part of the conversation. The entitlement instituted by NSR is not merely the entitlement to predict that other participants are not silent when doubtful about, or in disagreement with, a claim (empirical expectation). Rather, it is also an entitlement to believe that they ought not to be silent in these circumstances, and to sanction them if they are (normative expectation). Hence, the entitlement conferred to every participant in a conversation by NSR is a practical, rather than an epistemic, entitlement (Goldberg 2020, 158). The flipside of this practical entitlement is the responsibility that every member of the audience has towards every other participant not to be silent if they have doubts or reject the assertion. This responsibility is the conversational pressure generated by NSR. Whilst Goldberg’s initial formulation of NSR is as a normative expectation of no silent rejections, he convincingly argues that this expectation is broadly equivalent to the presumption that silence signals acceptance, in the minimal sense of letting the claim stand for the purposes of the conversation without necessarily committing to believing it (Goldberg 2020, 166). Hereafter, I follow Goldberg in thinking of the entitlement instituted by NSR to be an entitlement to presume that silence indicates acceptance.2 Given NSR, any participant in a conversation is (defeasibly) within their rights to presume that when a participant is silent, that person accepts the assertion made. That is, participants in a conversation are (defeasibly) entitled to believe, in the absence of any positive evidence, of some silent participants to the same conversation that they accept the claim made. Hence, unless defeating conditions obtain, silent but doubtful participants are blameworthy for their silence.  I set aside Goldberg’s (2020, 167) additional arguments that in serious conversations we conform to a norm NSR+ which entitles participants to presume that silence indicates belief. 2

  Speech in Non-ideal Conditions: On Silence and Being Silenced 

225

Goldberg (2020, 176) identifies two kinds of defeating conditions: NON-CONVERSATION: the speech exchange is not cooperative and thus not a conversation. OUTWEIGHING EXPLANATION: the best explanation for the silence of some member of the audience is that they have some practical reasons to be silent which outweigh their conversation-generated reasons to be cooperative and thus voice any doubts or disagreements.

Whenever NON-CONVERSATION is true, the speech exchange is not a conversation. NSR simply does not apply in these cases. In circumstances of this sort there is no entitlement to presume that silence indicates acceptance or responsibility to voice one’s doubts.3 Some contexts that are characterised by general sexism or racism are, in Goldberg’s view, circumstances in which NON-CONVERSATION obtains (Goldberg 2020, 177).4 He also thinks that speakers who trap others into discussions when their audience is within their rights not to engage also forfeit their entitlement to presume that silence indicates acceptance. Their overbearing behaviour is sufficient to determine that the speech exchange is not a conversation. Whenever OUTWEIGHING EXPLANATION is true, participants’ entitlement to presume that silence indicates acceptance is outweighed in the case of some other participants by practical reasons that they have that justify their silence even when they doubt or reject some assertion that they have manifestly observed and understood.5 For example, a participant who witnesses many near simultaneous assertions might be excused for being silent without accepting each of these claims, especially if they are doubtful about several of them.   However, there is a difficulty here in characterising the conditions in which NON-­ CONVERSATION obtains. If it were true in any circumstance in which someone is not cooperative, then it would be impossible to violate NSR, since any behaviour that seemingly violates it would create circumstances that lie outside the range of application of the norm. 4   Other such contexts, whilst being conversations, are cases where OUTWEIGHING EXPLANATION obtains. 5  Plausibly OUTWEIGHING EXPLANATION also holds when these outweighing practical reasons excuse, rather than justify, silence. Goldberg (2020, 176, fn. 55) is somewhat vague about what kinds of reasons can function as defeaters. 3

226 

A. Tanesini

3 Some Doubts about NSR Goldberg seeks to argue that NSR is a norm to which we de facto conform but also that it is a norm which it is reasonable to have and which we should keep. In this section I argue that there is reason to doubt the descriptive adequacy of NSR. I develop an argument for the claim that NSR is not a norm that we should wish to have in sections four and five below. Goldberg offers five considerations in support of the claim that many societies de facto conform to NSR. These are: A. Many current and past societies have proverbs that state that silence indicates acceptance. Similar claims can be found in the work of philosophers and novelists (Goldberg 2020, 168–170). B. Many organisations use a silence procedure requiring dissenters to state their disagreement (Goldberg 2020, 172–173). C. Many prominent individuals have indicated their disapproval of those who kept silent in the face of grave injustices (Goldberg 2020, 170–172). D. There might be evolutionary reasons to think that acceptance to another’s testimony is a default state. If so, there might be evolutionary reasons why rejections should be marked (Goldberg 2020, 173–174). E. We ordinarily disapprove of those whom we discovered have been silent when harbouring doubts. This is evidence that we exhibit the normative expectations instituted by NSR (Goldberg 2020, 197). Goldberg takes these considerations as strong evidence that competent linguistic agents de facto conform to NSR, and also (although he does not say so explicitly) that they endorse this norm. That is, competent linguistic agents would typically believe that everybody ought to conform to NSR. These claims, I argue, are false. Since NSR (defeasibly) entitles speakers and other participants in a conversation normatively to expect that silence indicates acceptance, if NSR is widely complied with, we would predict that speakers and other linguistic agents regularly interpret silence as indicating acceptance. This

  Speech in Non-ideal Conditions: On Silence and Being Silenced 

227

is as a matter of empirical fact not the case. Speakers generally interpret complete silence as evidence of disengagement or even hostility (hence, the expression: ‘giving someone the silent treatment’). Work on paralinguistic aspects of speech exchanges has robustly demonstrated the existence of back channelling whereby participants to a conversation signal to a speaker their continual engagement, as well as offering cues as to the likely eventual agreement or disagreement. They use continuants such as mmh, uh? and yeah as well as nods and headshakes to indicate their responses (Stivers 2008).6 In general, empirical research in conversational analysis indicates that people tend to display their affiliation to the speaker to promote social cooperation, and thus voice their agreement as encouragement, and try to format their disagreements in a way that indicates that they are still affiliating (Lindström and Sorjonen 2013). What this evidence shows is that genuine silence in the face of an assertion is rare. When it occurs, it is interpreted not as acceptance but as uncooperative because it indicates lack of engagement or disaffiliation.7 These considerations show that linguistic agents do not interpret silence as indicating acceptance, as we would expect them to do if conformity to NSR were de facto in force. They do not adopt this interpretation because as a matter of fact assent (genuine or fake) is usually expressed in the back channel by means of nods and continuants such as yeah. When a speaker is faced with true silence her first reaction is that either she has bored her audience or the audience is hostile. A supporter of the descriptive adequacy of NSR could try to explain how these linguistic regularities are compatible with de facto conformity to NSR by invoking the fact that NSR is often, or perhaps even generally, defeated. Awareness of the prevalence of defeating conditions would explain why linguistic agents do not readily explain others’ silences as indicating acceptance. If this explanation were correct, we would expect there to be at least some paradigmatic cases where manifestly NONCONVERSATION and OUTWEIGHING EXPLANATION do not  These responses are also used to indicate understanding. Further, it might be unclear which specific propositional content has been nodded through. I suspect this is because conversants evaluate the overall gist of a conversational contribution rather than parse it into individual sentences. 7  I should add that there are many cultural variations in turn-taking, the use of continuants, and so forth. I gloss these over here because I do not think that such variation invalidates my main point. 6

228 

A. Tanesini

hold and in which participants readily interpret others’ silences as acceptance. The opposite is the case since speech exchanges that are fully cooperative involve a lot of back channelling communication. If these considerations are correct, then Goldberg’s anecdotal evidence of the existence of proverbs to the effect that silence indicates acceptance would be surprising. However, the facts adduced by Goldberg are easily explainable without invoking conformity to NSR. A. The prevalence of proverbs stating that silence indicates acceptance merely shows that in many societies powerful people have often found it convenient to claim that a norm in the neighbourhood of NSR should be adopted. That is, proverbs are a valuable ideological tool since they portray as folk wisdom an injunction that works in the service of re-enforcing current iniquitous relations of power. Goldberg presumes that these proverbs merely reflect a pre-existent conversational norm, adopted because of its rationality. But the evidence is equally compatible with the view that proverbs aim to impose NSR because of its ability systematically to silence members of some social groups. If this is right, the existence of proverbs in many societies, each of which are characterised by oppressive structures, could be explained either as reflecting widely adopted conversational norms or as the means to re-enforce existing inequalities by attributing a responsibility to dissenters to speak out, and relieving others from the responsibility to ascertain whether their views receive widespread endorsement. B. The existence of institutional procedures according to which it is presumed that a proposal is agreed unless any committee member publicly voices their objection is not evidence of widespread conformity to NSR, nor of the conversational or epistemic value of adopting such a norm. It is true that some organisations employ a silence procedure as an alternative to secret ballots when deciding whether to adopt some proposals. This procedure is used precisely because of its silencing powers. If members voted on proposals by secret ballot, less powerful individuals would be able to express their opinions. The requirement that they voice their disagreement in public greatly increases the costs associated with expressing their preference. It thus

  Speech in Non-ideal Conditions: On Silence and Being Silenced 

229

makes it much less likely that less powerful individuals voice any disagreement. Further, this silencing effect of the silence procedure is not a regrettable consequence of a procedure that is overall in the interest of each participant. On the contrary, the interests of less powerful individuals are generally better served when decisions are made by ballots (and especially secret ones). Its silencing effect is a feature, not a bug of the silence procedure. It has silencing effects because in realpolitik the function of the procedure is to silence. C. Goldberg thinks that NSR is analogous to a moral norm requiring that one prevents harms that one has witnessed, and one can stop. The existence of this analogous moral norm would offer some evidence in favour of NSR. It is unclear why this would be so. There are sufficient differences between the norms of justice and the norms of conversation not to warrant an expectation that a norm in one realm will have an analogue in the other. Be that as it may, I harbour doubts about the analogy, but even if an analogy exists it does not support a norm with the same features as NSR. The moral norm in question would indict of complicity those who are silent when witnessing some injustice that they could (and ought to) have stopped. Individuals who do not oppose injustice are complicit with, and thus partially responsible for, it because they have an obligation to intervene. Sometimes such intervention requires that one speaks out. In these cases, silence is a failure to intervene. It is because by being silent one behaves as one ought not to, that one is complicit with the injustice. Three points about this norm are significant for my purposes here. First, the obligation to speak in these instances is a moral obligation, it is unrelated to linguistic norms. Second, the obligation to intervene to stop an injustice might on occasion be fulfilled by remaining silent. Silent protests against censure and silencing have been held in numerous countries and by different groups (Ranjbar 2017; Woolley 2012). Third, the obligation to speak out is part and parcel of a more general obligation to contribute to justice being done. It is not restricted merely to doing something to stop injustices one witnesses. NSR could be thought to be analogous to the moral norm requiring that one acts to stop injustices. One might hold that when in a

230 

A. Tanesini

cooperative endeavour one witnesses some speech that by one’s own light is false or dubious one has an obligation to intervene and stop that speech having the harmful consequence of being accepted by the participants in the conversation.8 Hence, one ought to signal one’s disagreement or doubts by speaking out. In my view, this analogy between the moral norm to speak out against injustice and NSR is at best superficial. First, the moral norm only applies to non-­cooperative contexts, whilst NSR only applies to cooperative ones. Second, the moral norm requires that one contributes to justice, including by intervening against injustice. NSR instead entails that, when witnessing an assertion being made with which one agrees, one fulfils one’s responsibilities merely by not interfering. That is, the moral norm to speak out requires that we act to promote something good (namely, justice) and that includes stopping bad things from happening. NSR is limited to an obligation to block harm, it does not include an obligation to contribute to the good. Be that as it may, even though one grants the existence of an analogy, so that one endorses the existence of a conversational norm that one must signal one’s doubts about or disagreement with an assertion one has witnessed, it does not follow that silence cannot be that signal. More strongly, given that the empirical evidence offered above points to the rarity of genuine silence in conversation, it would seem that silence is suited for this role. It is only if we wrongly presume that silence is a nothing, that we might infer that it cannot serve as a signal. If this is right, whatever norm would be analogous in the conversational realm to the moral norm to speak out against injustice, that norm is not NSR. Goldberg (2020, 189) ignores the possibility that silence might signal disagreement because he defines silence as giving ‘no explicit indication of whether [one] has accepted or rejected [a] claim’. Hence, an interlocutor who gives the silent treatment to a speaker is not by this definition silent since she explicitly indicates her disapproval of the speaker. I do not follow Goldberg in adopting this definition.  Without an explicit signal the potentially harmful assertion would be seamlessly incorporated into the common ground (Stalnaker 2002). 8

  Speech in Non-ideal Conditions: On Silence and Being Silenced 

231

Instead, I use ‘silence’ to mean non-acoustic behaviour which is also non-verbal (Tanesini 2018, 111). I adopt this characterisation because, unlike Goldberg’s, it makes space for the possibility that some silences communicate. Despite this difference in our definitions of silence, I do not think that Goldberg and I are talking past each other on this issue. The linguistic facts about the backchannel I adduced above strongly support the conclusion that conversants are almost never silent in either definition of the term. Further, I have argued that when interlocutors who manifestly understood what is being said are non-verbal their behaviour often indicates hostility or boredom. If this behaviour does not count as silence by Goldberg’s lights, then his kind of silence is extremely rare in conversations to the point of being practically non-existent. D. Irrespective of whether acceptance of testimony has evolved to have some kind of default status, it does not follow that silence would indicate acceptance. Suppose that, as Goldberg claims, disagreement is usually the dis-preferred option by speakers and as such should be marked. As I indicated above, there is no reason not to believe that silence could be that marker. As it happens there is some inconclusive evidence that longer silences in turn-taking are markers of disagreement (Kendrick and Torreira 2015). Be that as it may, if silence in conversation is unusual, since it is tantamount to the absence of visual and auditory backchannels, it is well suited to serve as a marker. We would therefore expect disagreement (as the dis-preferred option) to be marked by silence (the unusual and uncooperative absence of backchannelling). Hence, the psychological evidence, albeit highly speculative, would point against Goldberg’s conclusions. E. Goldberg argues that we disapprove of those who disagree in silence. These reactive attitudes would indicate that we exhibit the kind of normative responses that one would expect if we conformed to, and approved of, NSR. In my view our reactive attitudes do not indicate that we endorse NSR. We do censor people who mislead us into believing that they agreed with us when they did not. Our disapproval is based on moral norms against lying or misleading people. It is not grounded on the violation of a conversational norm. I have argued above that participants in conversations engage in extensive backchan-

232 

A. Tanesini

nelling responses. These are cues that indicate to speakers and other participants that one is engaged, but also that one is likely to agree (yeah) or have doubts (mmhm) about the claims being made. Participants who mislead others are those who nod and offer other agreeing responses when they do not agree. It is this behaviour that is being censored. These reactive attitudes though do not indicate that linguistic agents endorse NSR.9 I conclude that none of the considerations presented by Goldberg in support of NSR speaks in favour of de facto conformity or endorsement of this norm. In addition, some of the facts invoked by Goldberg are better explained by the view that there is no widespread reliance on a norm that silence indicates acceptance. There are, however, contexts in which powerful individuals can implement procedures akin to NSR which function to stifle the ability of less powerful individuals to influence decision-making.

4 The Silencing Power of NSR If the arguments developed so far are along the right lines, NSR is not a conversational norm routinely followed in everyday linguistic practice. It does not follow, however, that there are no circumstances in which people are made to feel that if they do not speak, they will be presumed to have agreed with the speaker. As a matter of fact, there are linguistic contexts in which some speakers and/or participants seek to determine the terms of engagement so that to help themselves to a presumption that silence indicates acceptance. There are circumstances in which some linguistic agents have the legitimate authority to dictate the procedures to be followed in a speech exchange. Hence, for instance, a judge in a courtroom may determine who is allowed to speak and when. Similarly, the chair of a Q&A session  We also disapprove of people who do not resist oppression when they have an obligation to do so. In some cases, these people have a moral obligation to speak. Hence, we censor them for their silence. I have discussed these cases when considering Goldberg’s third piece of evidence for NSR. 9

  Speech in Non-ideal Conditions: On Silence and Being Silenced 

233

has the authority to call on questioners or to silence them if they arrogate too much of the floor. Ordinarily, however, no participant in a speech exchange has the formal authority to determine the terms of engagement without the consent of all participants. In these circumstances any attempt to impose an expectation that silence indicates acceptance is an exercise of power. If this is right, support for NSR in ordinary circumstances is ideological, since it is the unilateral attempt (by means of an exercise of power) to impose a novel prescription whilst framing it as merely reflecting pre-existing norms already widely accepted because of their practical utility. Goldberg is aware that NSR might appear tailor-made for silencing. He also agrees that in current oppressive conditions belief in NSR is instrumental to silencing individuals, and in making silenced individuals potentially complicit in their own oppression. Nevertheless, he thinks that these harmful effects are not the results of NSR itself but of its misapplication. Further, when properly followed NSR would be clearly seen as part of the solution rather than of the problem. In what follows I detail Goldberg’s arguments to the effect that the misapplication of NSR has silencing effects. I argue instead, drawing on Sbisà’s (2020, 74–75) account of exercitive assertions, that invocations of NSR are a piece of ideology whose function is to silence and to supply plausible denials of responsibility for silencing. Goldberg thinks that in sexist and racists societies, NON-­ CONVERSATION and/or OUTWEIGHING EXPLANATION is true of most speech exchanges between people of different races and genders. That is, NSR is very frequently, perhaps almost always, defeated. Thus, often there is no obligation to indicate one’s doubts and no entitlement to presume the agreement of the silent. Nevertheless, Goldberg thinks that individuals who are ‘oblivious’ to the existence of defeating conditions are likely to interpret silence as acceptance and thereby to contribute to silencing those individuals who in the given circumstances have no, all things considered, obligation to voice their doubts (Goldberg 2020, 194–195). These silencing effects would not be the result of NSR itself but of its misapplication. In addition, the false belief that someone’s silence indicates acceptance, grounded on the false belief in the applicability of NSR to the current circumstances, would generate a second, or

234 

A. Tanesini

double, harm. Silenced individuals might infer from the silence of others’ participants, whom they presume not to have been silenced, that an assertion they have all observed must be acceptable since it has been accepted by the audience. In this manner, even a participant who might have harboured doubts about an assertion might accept it having thus acquired higher-order evidence of its acceptability (Goldberg 2020, 194–195). Goldberg claims that the existence of these harms can only be explained if our community endorses NSR and widely misapplies it. More specifically, he argues that mere belief that silence generally indicates acceptance cannot explain those normative expectations, allegedly generated by NSR, and whose existence explains how people can be silenced by having their silence interpreted as acceptance.10 Goldberg is right that a belief that silence typically indicates acceptance merely licenses one to predict that those who are silent accept. It cannot license the thought that one is entitled to count them as accepting since if they had doubts, or wanted to disagree, they should have said so explicitly. This thought, however, can also be supported by the arrogation of an entitlement that one is able to enforce because one has the power to set the terms of the debate. There are many cases of asymmetry of power when one or more parties to a speech exchange can determine its directions and its rules of engagement. I have argued above that powerful members of organisations can often impose decision procedures that raise the costs of disagreement in order to give some form of legitimacy to their exercise of superior power. Similarly, when some participants in speech exchanges are more powerful than others, they might aim to shape the rules of a speech exchange to make dissent costly. They can achieve this goal in at least two related ways. First, they can rely on their superior power successfully to announce that any dissenting views should be voiced presently. These announcements are tantamount to introducing a silence procedure by relying on a silence procedure. These announcements are exercitive assertions; that is, assertions made in a manner that clearly indicates that speakers will not  This would be something akin to a discursive injustice that occurs when a linguistic agent performs a speech act other than the one she intends, and attempts, to perform despite legitimately expecting to be able successfully to perform it (Kukla 2014). Thanks to Laura Caponetto for pressing this point. 10

  Speech in Non-ideal Conditions: On Silence and Being Silenced 

235

tolerate disagreement (Sbisà 2020, 74–75). If successful, they enact a temporary norm of no silent rejections. Success occurs when all participants to the conversation accept that in that speech exchange silence counts as acceptance. Sometimes, the successful performance of this exercitive leads to accommodation of the belief that the speaker had the authority to perform it (Sbisà 2018, 42–43). In this manner, exercitives further exacerbates power differentials by conferring a vein of legitimacy to the alleged superiority of powerful speakers. But these speech acts can succeed without accommodation when they bring about the intended conventional effect of changing the norms governing the conversation, which they can do as naked exercises of power. The reason for this success lies in the content of the norm they seek to enact. Power is sufficient to silence opponents. But silence in the face of an announcement that silence indicates acceptance made by someone with the power to enforce it, leads to a situation in which everyone accepts (in the sense of being resigned to the fact) that silence has in that conversation the significance of acceptance. Second, powerful speakers can simply act and speak as if a norm akin to NSR is widely accepted. That is, they can simply presume that those who are silent accept, whilst being prepared, and able, to censure silent rejections.11 Such censure might involve invocations of NSR as explicitly formulated in the proverbs discussed by Goldberg. These invocations need not be explicitly performed, mere knowledge that they can be made and enforced can be sufficient to induce all participants to be resigned to allowing the norm to govern their conversation perhaps out of a conviction that resistance would be futile. In these and other ways, powerful speakers succeed in setting the terms of conversation and enact conversational norms of no silent rejections. That said, the conversational success of these exercitives has no tendency to counter the claim that these arrogations of entitlement are morally illegitimate. They are abuses of power because they are attempts to exploit differences in existing power relations to deny to some linguistic agents the equal role in speech exchanges to which they are entitled.  See Sbisà (1999) on the manner in which ideological content is often presupposed rather than explicitly articulated in speech exchanges. 11

236 

A. Tanesini

This alternative account of the role of normative expectations (and the temporary norms that they institute) to treat silence as acceptance can fully explain the role of these expectations in promoting silencing. It differs from Goldberg’s in arguing that these expectations are an arrogation of entitlement, rather than the misapplication of a genuine entitlement. One might argue that the explanation of the entitlement as an arrogation must be rejected because it cannot explain the double harm caused by the misapplication of NSR. There are, however, independent grounds to doubt Goldberg’s views about the alleged double harm. His account is predicated on the idea that individuals who have often experienced silencing when observing others who are silent rationally infer from that silence that other participants agree with claims these individuals are inclined to treat as doubtful. It is in my view extremely unlikely that victims of sexism and racism would be disposed to make the required inference. It is much more probable that they would, on the basis of their own and shared experiences, conclude that these other people are also being silenced. They would not treat others’ silence as higher-order evidence of the acceptability of some assertion about which they have reservations.12 One might also ask why if normative expectations that silence indicates acceptance are a form of arrogation of power, one would even bother with the pretence that others are offered a chance to disagree. The answer in my view is that invoking a norm according to which silence indicates acceptance offers plausible deniability for the silencing effects of presuming agreement unless others indicate otherwise. Harbouring normative expectations that others are not silent in rejection offers a standing justification for silencing behaviours. It is always possible for linguistic agents to deflect responsibility for silencing simply by arguing that those who disagreed should have said something or by invoking one’s misunderstanding of the others’ motivations for being silent.13 Since it is widely  Goldberg could respond that the double harm occurs only when a silenced individual is not aware that she has been silenced or that others have been. It is hard to think such a person exists. Since Goldberg’s arguments are driven by the desire to portray oppressed individuals as epistemically rational, it does not seem a gain to explain the double harm in terms of their alleged ignorance of blatantly obvious facts about oppression. 13  This is not to deny that sometimes silence is ambiguous and hard to interpret. 12

  Speech in Non-ideal Conditions: On Silence and Being Silenced 

237

agreed that silencing other people without the authority to do so is wrongful, the invocation of NSR is a useful means to exonerate oneself from this kind of wrongdoing.

5 NSR and Ideal Theory I have stated above that Goldberg is fully aware that, given widespread social discrimination, NSR is likely to be used in ways that promote silencing rather than remedy it. Hence, one might wonder why Goldberg thinks that NSR is part of the solution to the problem of silencing. I believe that he is led to this conclusion by his endorsement of the methodology of ideal theory. In what follows I briefly explain what I mean by ideal theory. I describe Goldberg’s argument for the ideal endorsability of NSR and show why it exemplifies ideal theory. Subsequently, I explain why NSR would not be part of the solution (even if it were one of our standing conversational norms) and why even an ideal system of communication would not include NSR. I conclude the section, and the chapter, with some brief remarks about a strategy to counter the silencing effects of the normative expectations associated with NSR. The distinction between ideal and non-ideal theory in normative philosophy marks a difference in methodology. Ideal theorists develop formulations of normative ideals that abstract away from significant, or even necessary, features of reality. It is not their subject matter concerning normative ideals (such as justice or justification) that makes these views a form of idealised theory. Instead, what marks these theories as idealised is their neglect of inescapable constraints such as the bounded nature of human rationality. These theories also ignore the (hopefully changeable) social realities of oppression and discrimination. Consequently, these theories are a kind of make-believe describing an imaginary reality where all the limiting conditions that characterise our circumstances have been stripped away. Hence, idealised theory lacks the resources to detail and explain actual circumstances for what they are. Instead, it can only describe reality in negative terms by measuring its distance and difference from the abstract idealised model that is the crucial component of the theory (Mills 2004).

238 

A. Tanesini

Goldberg’s defence of NSR exemplifies this approach. He argues that NSR is a norm that should be endorsed because its inclusion in an ideal system of communication whose goals are primarily epistemic can be defended on rational grounds. This ideal system of communication would include rules that contribute to the reliable achievement of the epistemic aims of persuasion and the sharing of information. Goldberg’s argument that this system would include a norm like NSR includes several steps and premises. My focus here is on the final two premises. The first of these is the claim that it is in the interest of every participant in a conversation to track the doxastic responses of other participants to assertions made during that conversation.14 The final premise is that a system of communication in which participants’ dis-preferred option is marked is more efficient than one in which both preferred and dis-preferred answers are signalled. These two claims support a conclusion that in an ideal system of communication participants to a conversation convey their doxastic responses to the assertions they witness to other participants in the same conversation. Further, they signal these responses efficiently by marking through speech only one of the two available options. This is the option that is likely to be dis-preferred by speakers: rejection. In conclusion, an ideal system of communication would have NSR as one of its rules because the presence of NSR makes it possible for silence to have epistemic significance (as an indication of acceptance) (Goldberg 2020, 204). Systems of communication without NSR would be less efficient in achieving their epistemic goals and thus would be improved by adding NSR to their norms. Further, the inclusion of NSR in the ideal system of communication warrants the further claim that a silent rejection is not a cooperative response, and thus participants in conversations are entitled not to expect a silent rejection, because they are entitled to cooperative responses (Goldberg 2020, 160).

 This claim presupposes that participants have formulated doxastic responses to the assertions they have witnessed. Goldberg (2020, 96–97) argues that participants in a conversation owe it to speakers to evaluate properly, in accordance with the relevant epistemic standards, speakers’ competence with regard to the asserted content. Participants who do not do this, barring excuses or justifications, are disrespectful to speakers. Hence, participants have an obligation to either reject or accept the asserted contents. That is, they have an obligation to formulate a doxastic response. 14

  Speech in Non-ideal Conditions: On Silence and Being Silenced 

239

Goldberg’s argument for the endorsability of NSR adopts the methodology of idealised theory. He abstracts from the limitations of our cognitive capacities. He is aware, for instance, that it might prove impossible to formulate doxastic responses to each assertion if we are faced with a barrage of claims. In addition, formulating these responses is cognitively effortful, so there might be instances in which one is too tired or too pre-­ occupied with something else to be able to respond properly. These constraints, due to the limitations of human cognitive abilities, are classified as defeating conditions as we would expect of a theory that relies on idealising assumptions. The premise that in a conversation it is in the interest of all participants to track each other’s doxastic responses is also predicated on some idealising assumptions. First, the premise only concerns conversations. Since these are cooperative activities, the premise is silent about speech exchanges that occur in conditions of oppression where it is not in the interest of several participants to reveal their mental states to other hostile linguistic agents. Second, the premise ignores that even when engaging in cooperative activities, human participants always have mixed motives (Henderson 2020). For instance, communicating is effortful for humans. Effort is a cost and as such it is a reason not to communicate. In circumstances in which many witness the same assertion there is little benefit accruing to others from being able to track the response of a single agent, and thus the agent might reason that that benefit is outweighed by the cost of the effort involved in communicating.15 Hence, whilst it might be in each person’s interest to track others’ doxastic responses, it might not be always in one’s interest to communicate one’s doxastic responses to others. If these considerations are along the right lines, Goldberg’s account, because it ignores some inescapable aspects of human cognition, cannot identify the sort of norms that would regulate a communicative system which is maximally successful in achieving the goals of persuasion

 Note that the benefit to each of knowing that 1000 participants agree as opposed to 999 is so small that it might be outweighed by the effort required to communicate. Communication is always effortful including when such communication is achieved by choosing to be silent. Goldberg acknowledges that something like this might be the case and classifies these circumstances as a defeating condition. 15

240 

A. Tanesini

and the sharing of information when adopted by creatures whose cognitive abilities are bounded. This point can be further driven home by noting that the addition to an imperfect system of norms of a norm that would figure in a perfect system is not always an improvement. For instance, the rule of always following arguments wherever they lead might be a good norm for a thinker unconstrained by limitations of cognitive capacities with regard to speed and tolerable cognitive load. Were a limited epistemic agent to add that norm to their current stock of zetetic norms, her epistemic performance would worsen rather than improve (Friedman 2020; Morton 2012). What this point shows is that even if Goldberg is right that a norm like NSR is part of an idealised system of communication that abstracts from the fact of human bounded rationality, it does not follow that its addition to a communicative system for creatures whose cognitive abilities are bounded constitutes an epistemic improvement. For this reason, and contra Goldberg, it does not follow that everyone has an interest in adopting a norm like NSR on the grounds that a community of idealised agents would have such interests.16 In addition, in my opinion Goldberg also overestimates the importance of the epistemic good of silence indicating acceptance. This epistemic good is one of efficiency (Goldberg 2020, 204, fn28). A system where only one of the two available options is marked would be more efficient than one in which both are.17 I have argued above that conversations are characterised by a stream of listener’s responses in the so-called backchannel. One of the functions of these responses is to indicate affiliation either in agreement or in friendly disagreement with the speakers. In human beings that often have reasons not to be cooperative, the continual display of affiliation would seem to promote the strengthening of social bonds. This strengthening leads to increased trust which is essential  This is the argument implicitly deployed by Goldberg (2020) on p. 204.  This argument for efficiency relies on assumptions that could be questioned. The first is that silence is not a mark. I have hinted above to reasons why this might not be true. The second is that silence is not effortful. This is also unlikely to be true since it might require resisting impulses to respond. In order properly to assess Goldberg’s claim one would need to be clear about his interpretation of efficiency. 16 17

  Speech in Non-ideal Conditions: On Silence and Being Silenced 

241

if speech exchanges are to proceed as conversations rather than confrontations. If this is right, there are good social and epistemic reasons in favour of a system of social norms that prescribes the expression of one’s doxastic response of whatever kind in the backchannel. These reasons indicate that such system would be epistemically superior to one in which acceptance is indicated by silence since it would be instrumental in increasing cooperation. To summarise, I have argued that NSR is not one of our norms in everyday friendly exchanges. I have also argued that there are instances in which some speakers invoke NSR to silence other linguistic agents. The arrogation of the entitlement of taking silence to be acceptance is successful even though the entitlement is morally illegitimate because it is secured by the effectiveness of sanctions which some speakers have the power to impose on others. The advantage for these speakers of invoking NSR, over the naked exercise of power, is the related ability plausibly to deny responsibility for silencing. Finally, I have argued that even if NSR were a norm one might wish to have in a fully idealised communicative system, it is not a norm we should wish to have when regulating communication between agents whose thinking is bounded and whose motives are mixed. In short, the arrogation of the entitlement is part of the problem and NSR would not be part of the solution. I wish to conclude by proposing a partial solution to the problems generated by the arrogation of the presumption that silence indicates acceptance. It is of course important to highlight that in ordinary conversations participants are rarely silent. When the exchange is a conversation, listeners display their friendliness by giving signs of encouragement and by preparing the speaker to a likely agreement or disagreement so that neither comes as a surprise. But I would also like to recommend that speakers, to avoid the risk of inadvertently silencing any participant to a speech exchange, never presume, when important matters are at stake, to know what others think without asking them for their views.18  I would like to thank the editors of this volume for their helpful suggestions. Jessica Brown, Catarina Dutilh Novaes, Greg Restall, and other members of Archè offered useful comments that greatly improved this chapter. Special thanks to Sandy Goldberg for making every speech exchange with him a conversation. 18

242 

A. Tanesini

References Eddo-Lodge, Reni. 2017. Why Iʼm no Longer Talking to White People about Race. London: Bloomsbury. Friedman, Jane. 2020. The Epistemic and the Zetetic. The Philosophical Review 129 (4): 501–536. Goldberg, Sanford C. 2020. Conversational Pressure: Normativity in Speech Exchanges. Oxford: Oxford University Press. Grice, Paul H. 1975. Logic and Conversation. In Syntax and Semantics, Vol. 3: Speech Acts, edited by Paul Cole and Jerry Morgan. New York: Academic Press. Reprinted in Studies in the Way of Words (1989), 22–40. Cambridge, MA: Harvard University Press. Henderson, David. 2020. Are Epistemic Norms Fundamentally Social Norms? Episteme 17 (3): 281–300. Kendrick, Kobin H., and Francisco Torreira. 2015. The Timing and Construction of Preference: A Quantitative Study. Discourse Processes 52 (4): 255–289. Kukla, Rebecca. 2014. Performative Force, Convention, and Discursive Injustice. Hypatia 29 (2): 440–457. Langton, Rae. 1993. Speech Acts and Unspeakable Acts. Philosophy and Public Affairs 22 (4): 293–330. Lindström, Anna, and Marja-Leena Sorjonen. 2013. Affiliation in Conversation. In The Handbook of Conversation Analysis, ed. Jack Sidnell and Tanya Stivers, 350–369. Chichester: Wiley-Blackwell. Mills, Charles W. 2004. Ideal Theoryʼ as Ideology. In Moral Psychology: Feminist Ethics and Social Theory, ed. Eggy DesAutels and Margaret M.U. Walker, 163–181. Lanham: Rowman & Littlefield Publishers. Morton, Adam. 2012. Bounded Thinking: Intellectual Virtues for Limited Agents. Oxford: Oxford University Press. Ranjbar, Marie A. 2017. Silence, Silencing, and (In)Visibility: The Geopolitics of Tehranʼs Silent Protests. Hypatia 32 (3): 609–626. Sbisà, Marina. 1999. Ideology and the Persuasive Use of Presupposition. In Language and Ideology. Selected Papers from the 6th International Pragmatics, edited by Jef Verschueren, 492–509. Antwerp: International Pragmatics Association. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2018. Varieties of Speech Act Norms. In Normativity and Variety of Speech Actions, edited by Maciej Witek and Iwona Witczak-Plisiecka, 23–50. Leiden: Brill. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press.

  Speech in Non-ideal Conditions: On Silence and Being Silenced 

243

———. 2020. Illocuzione e dislivelli di potere. In Linguaggio d’odio e autorità, ed. by Claudia Bianchi and Laura Caponetto, 63–86. Milano-Udine: Mimesis. Transl. as Illocution and Power Imbalance in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. Stalnaker, Robert. 2002. Common Ground. Linguistics and Philosophy 25: 701–721. Stivers, Tanya. 2008. Stance, Alignment, and Affiliation During Storytelling: When Nodding Is a Token of Affiliation. Research on Language and Social Interaction 41 (1): 31–57. Tanesini, Alessandra. 2018. Eloquent Silences: Silence and Dissent. In Voicing Dissent: The Ethics and Epistemology of Making Disagreement Public, ed. Casey R. Johnson, 109–128. New York and London: Routledge. Woolley, Susan W. 2012. The Silence Itself Is Enough of a Statementʼ: The Day of Silence and LGBTQ Awareness Raising. Anthropology and Education Quarterly 43 (3): 271–288.

A Speech-Act Theoretic Analysis of White (Prosocial) Lies Marina Terkourafi

1 Introduction A friend invites you over to show you their apartment, which they spent months renovating. Room after room, you find the new style of decoration abominable but don’t want to hurt your friend’s feelings; after all, you won’t be living there yourself. So you utter small phrases like “That looks really nice!” and your friend responds with a grateful smile, beaming happiness. Have you lied? Most people might say yes, inasmuch as you stated something you did not believe to be true. Were you wrong to lie? Most people might say no, inasmuch as telling your friend your honest opinion would achieve little under the circumstances beyond hurting their feelings (they do not have the resources to engage in further re-­ decorating anyway). Now consider an alternative scenario: friends are visiting you from out of town and you shower their young children with gifts. At the end of their visit, they forget to take their gifts with them and

M. Terkourafi (*) Centre for Linguistics, Leiden University, Leiden, The Netherlands e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Caponetto, P. Labinaz (eds.), Sbisà on Speech as Action, Philosophers in Depth, https://doi.org/10.1007/978-3-031-22528-4_11

245

246 

M. Terkourafi

when you remind them, the mother responds: “Our kids don’t play with such nonsense anyway.” Did she lie? Most people might say no, inasmuch as she voiced her true opinion about the gifts. Was she wrong to do so? Most people might say yes, inasmuch as she gratuitously hurt your feelings and displayed her lack of gratitude to you as a host. We can all think of similar situations where telling the truth might be less socially adept than telling a lie. Called “white lies” in English, such lies have been defined as falsehoods “not meant to injure anyone and of little moral import” (Bok 1978, 58). In what follows, I define white lies generally as lies that benefit the addressee and are uttered out of concern for them. Researchers have long recognized that white lies can be so routine as to constitute a “particular sort of communicative competence” (Camden et al. 1984, 321) and people’s ratings of the truthfulness of their own statements support this. Naïve subjects asked to rate the truthfulness of their statements in everyday conversation rated only about a third of their statements as completely honest (Turner et al. 1975), suggesting that the majority of what we casually say consists of various degrees of falsehoods. But if white lies are so common, are they still lies? This hinges on one’s definition of lying, as well as on the different “shades” of lying this allows.

2 Lying: A Locutionary Definition Despite centuries of philosophical, and more recently experimental, engagement with lying, no widely agreed upon definition of lying is available.1 Equally long-lived is the awareness that not all lies are morally reprehensible: according to Zembaty (1993, 29), Aristotle himself considered lies that “harm no one, involve no undeserved disrespect, and stem from excellence and self-sufficiency rather than some deficiency of character” as morally acceptable, yet he saw lies that simply benefit others as not justified (Zembaty 1988, 528 and 540). To justify the “lie” part of  Mahon (2019) provides a critical overview of several modern definitions.

1

  A Speech-Act Theoretic Analysis of White (Prosocial) Lies 

247

their name, then—that is, to provide a definition general enough to include malicious and white lies under the same umbrella—what we need is a locutionary definition of lying, such as the one provided by Saul (2012, 29; cf. Fallis 2009): (1) If the speaker is not the victim of linguistic error/malapropism or using metaphor, or irony, then they lie if and only if 1. they say that P 2. they believe P to be false 3. they take themself to be in a warranting context where a warranting context is one where sincerity is expected (as opposed to, e.g., being in a play, or telling a joke). Saul’s definition excludes (unintentional) errors like “She’s a great flamingo dancer,” where the speaker means flamenco dancer, as well as (intentional) floutings of Quality (“Do not say that which you believe to be false”; Grice 1975, 28), such as metaphor and irony, where correctly interpreting the speaker’s utterance necessitates recognizing that the speaker is speaking untruthfully. The definition in (1) is locutionary in the sense that what matters is the fact that the speaker has uttered the untruthful words.2 It can thus be used to distinguish between lying and misleading: under a locutionary definition such as (1), lying amounts to sayingloc p while believing p is false, whereas misleading involves meaning p (without sayingloc p) while believing p is false. On this definition, both malicious and white lies count as lies—because in both cases the speaker saysloc something they believe to be false—but false implicatures do not. Rather, false implicatures, where the falsehood is meant but not saidloc, fall under misleading.3 While this accounts for the observation that falsehoods

 On a locutionary sense of saying (Bach 2001), “S has saidlocutionary that p” does not entail meaning p. This is different from Grice’s (1975) sense of saying, in which saying p entails meaning p; that is, for Grice, saying p commits the speaker to the truth of the propositional content expressed. 3  For experimental support of this claim, see Weissman and Terkourafi (2018).

2

248 

M. Terkourafi

benefitting the addressee are characterized as “lies” cross-linguistically,4 it does not help us account for the difference between malicious, morally reprehensible lies and “innocent,” “pious,” “harmless” white lies. To account for this last difference, we need to consider the different intentions with which these two kinds of lies are uttered.

3 Two Kinds of Intentions Several researchers have raised the question whether lies must be intentional, and if so, exactly what the lying intention consists in: is it an intention to have the listener believe something that the speaker herself does not believe, or is it an intention to have the listener believe that the speaker believes something she does not in fact believe? At issue here is the content of the false belief that the speaker wants the listener to entertain: is it a false belief about the world or a false belief about the speaker’s beliefs about the world?5 Alternatively, lying can be a kind of perlocutionary effect, assessed depending on its results: if the target of lying (who may not always be the addressee)6 is deceived, then the speaker has lied, whereas if they are not deceived, then she has not lied.7 However, given the emphasis on lying as an indicator of the speaker’s morality cross-­ linguistically, it seems awkward to claim that lying can be judged exclusively based on its outcome, over which the speaker has only partial control. Therefore, I take the view that lying must involve some kind of  An informal polling of native speakers of different languages yielded the following terms (the original is given first, followed by an IPA transcription or transliteration in italics and English translation in parenthesis): Egyptian Arabic: ‫ اضيب ةبدك‬/kidba beːdˤa/ (white lie); Chinese: 圆场谎 yuánchăng huăng (lie smoothing over a situation); Danish: hvid løgn’ (white lie); Farsi: Duruq-e-­ Maslahati (lie with good intentions); French: pieux mensonge (pious lie), German: weiße Lüge (white lie); Greek: κατά συνθήκην/αθώα ψέματα kata sinθicin/aθoa psemata (lies by convention/ innocent lies); Hungarian: füllentés (white lie); Italian: bugie innocenti (innocent lies); Norwegian: hvite løgner (white lies); Russian: невинная ложь nevinnaia lozh (innocent lie); Spanish (Peninsular/Latin American): mentira (or mentirita) piadosa/bianca (pious/white (little) lie); Thai: เกรงใจ /kre:ŋ.tɕai/ (to be (too) courteous); Turkish: zararsız yalan (harmless lie); Urdu: ‫دیفس‬ ‫ ٹوھج‬sufaid jhoot (white lies). 5  Kupfer (1982) defends the former view while Chisholm and Feehan (1977) defend the latter. 6  According to the Oxford English Dictionary, lying to overhearers is possible (Mahon 2008, 218). 7  Coleman and Kay (1981) offer a definition of this type. 4

  A Speech-Act Theoretic Analysis of White (Prosocial) Lies 

249

lying intention on behalf of the speaker. Still, this leaves open what the lying intention involves. I propose that lies (both white and malicious) come with two kinds of intentions. The first intention all lies come with is a classic Gricean r(eflexive)-intention (Grice 1969), such as is necessary to invest an act of verbal behavior with meaning (Grice 1957). According to Grice (1969, 92), for the speaker to mean something by means of her utterance, she must have the following tripartite intention: (2) “U[tterer] meant something by uttering [expression] x” is true if [and only i]f, for some audience A, U uttered x intending: 1. A to produce a particular response r. 2. A to think (recognize) that U intends 1. 3. A to fulfill 1. on the basis of his fulfillment of 2. It is the third clause of this requirement that is usually identified as the r-intention: the speaker wants the listener to think about what the speaker is trying to achieve through her utterance and to use that information to help him figure out what her utterance means.8 This first, tripartite intention must be recognized in order to be fulfilled and is fulfilled simply by being recognized (i.e., it is not necessary for the listener to act on it). R-intentions are necessary for linguistic communication inasmuch as communication involves a change in both the speaker’s and the addressee’s cognitive environments.9 Nevertheless, lies also come with a second, “lying” intention. The content of this second intention is that the speaker wants the listener to believe something that she does not in fact believe herself. Inasmuch as this lying intention does not have to be recognized in order to be fulfilled, it is not a Gricean r-intention. Rather, it is a type of non-communicative intention similar to those that guide many of our solitary acts, such as the intention to eat, sleep, stand, and many other acts expressed as one-place  This is recast in terms of a communicative intention embedding an informative intention in Relevance Theory (Sperber and Wilson 1995). 9  On the Gricean picture, propositional contents that the listener entertains as a result of the speaker’s utterance but were not part of what the speaker wanted the hearer to understand are not linguistically communicated. 8

250 

M. Terkourafi

predicates. These intentions do not have to be recognized by someone else in order to be fulfilled. In everyday life, the word “intention’ is typically used in this more mundane, non-technical sense. Malicious and white lies differ in their treatment of this second intention. In the case of malicious lies, the lying intention must remain hidden from the addressee, or the lie will be unsuccessful (the speaker’s goal to deceive the addressee will not be achieved). However, in the case of white lies, it is permissible for this second “lying” intention to be recognized without jeopardizing the goal of the speaker. That is because the goal of the speaker this time is different. Rather than deceiving the addressee, in white lies, the goal of the speaker is first and foremost to constitute their face, and lying to them is one means of achieving that—this goal may still be achieved (for instance, by recognizing it directly) even if the means fails.10 In fact, recognition of the speaker’s lying intention in the case of a white lie can result in strengthening the politeness of the speaker’s utterance. That is because, assuming the speaker takes themself to be in a warranting context (see (1) above), recognition of the speaker’s lying intention by the addressee generates a search for reasons why the speaker transgressed expected standards of truthfulness. If as a result of this search the listener concludes that the speaker lied in order to constitute the addressee’s (and potentially also the speaker’s) face, then they end up attributing  A reviewer points out that in some cases recognition of the speaker’s lying intention in white lies can jeopardize the speaker’s goal not to hurt the hearer’s feelings, and offers the following as an example: suppose I have a new haircut and ask my partner if he likes it. In order not to hurt my feelings, he mumbles a “Yes.” Suppose also that I perfectly get that he’s lying: I can read it in his face. Moreover, the fact that he doesn’t tell me what he really thinks ends up hurting my feelings more than a criticism of my new look on his part would have. So in this case, the one who uttered a white lie ends up hurting the other’s feelings precisely because the other recognized that the former lied. I would argue that in this example the listener’s being hurt hinges precisely on her having first recognized the speaker’s “Yes” as a white lie, that is, her recognizing that the speaker’s goal was face-constituting. This upset her because, by prioritizing face-constituting, the speaker appears to have treated her as he might treat a stranger to whom politeness rather than sincerity is due. As his partner, she experiences his choice as creating distance between them and an affront to their relationship, and that is why her feelings are hurt. Rather than challenging the permissibility of the recognition of the lying intention in white lies, then, this example confirms that, when recognized, such lies are attributed to the speaker’s face-constituting goal. Whether that goal is felt to be appropriate under the circumstances or not is another matter; and it is precisely that appropriateness that the “white liar” seems to have miscalculated in this case. 10

  A Speech-Act Theoretic Analysis of White (Prosocial) Lies 

251

a face-constituting—that is, polite11—intention to the speaker.12 If I say that your hair looks nice and you recognize that I don’t really believe that, you realize that my reason for saying so is not a belief I hold about the world independently of how I feel about you (I think that your hair looks nice in the same way I think that the sun is shining), but that how I feel about you has precisely something to do with my saying that your hair looks nice: how I feel about you is my reason for saying I like your hair. In other words, politeness is in this case inferred as the reason why the speaker transgressed expected standards of truthfulness: if truthfulness is prized in conversation, then sacrificing it at the altar of politeness actively shows the lengths to which the speaker is willing to go to please (i.e., constitute the face of ) the addressee. Prima facie support for the claim that the lying intention is handled differently in the case of malicious vs. white lies comes from the fact that different responses are appropriate in case the lie is recognized. If a malicious lie is called out as such, the best response is generally to admit the lie. If the speaker keeps insisting they have not lied, they risk adding layers of lying and perjury to their initial lie, casting further doubt on their morality and the integrity of their character. However, if a white lie is called out as such, the best (and most common) response is denial. That is because admitting the lie in this second case—for which, once the lying intention has been recognized, politeness is the sole plausible motivation—has the effect of canceling that self-same politeness. That is, admitting a white lie would defeat the purpose of engaging in (white) lying in the first place. Isaacs and Clark (1990) propose a similar, two-level account for “ostensible invitations,” which they treat as part of a larger class of ostensible speech acts. According to them, ostensible speech acts are a type of nonserious language use which involves mutually recognized pretense and collusion by interlocutors to respond appropriately to the pretense. It is precisely this sense of collusion that would be destroyed if the utterer of  Face-constituting constitutes the core of a second-order definition of “politeness” (Terkourafi 2008).  This search may alternatively lead to the attribution of a face-aggravating intention to the speaker, as with bald-faced lies (Meibauer 2014). Formal and situational factors (see Section 4) help determine which of the two—the speech act of white lying (resulting in politeness) or a bald-faced lie (resulting in aggression)—is taking place. 11 12

252 

M. Terkourafi

a white lie were to admit the lie. Crucially, ostensible speech acts’ primary purpose (or perlocutionary effect) is face-management;13 yet, this remains off-record and is achieved by making evident that the perlocutionary effect canonically associated with the speech act (e.g., in the case of invitations, that the listener attend a certain event) is situationally blocked. However, ostensible speech acts also have a further feature setting them apart from other types of nonserious language use, namely their ambivalence: if asked, the speaker is unable to say categorically whether they intended to bring about the perlocutionary effect canonically associated with the relevant type of speech act or not; simply put, they are unable to say whether they “really meant” it. This is unlike other types of nonserious language use such as irony, where the speaker is clear about what they mean (and it is not what they say). This ambivalence regarding the speaker’s commitment to the meaning of their words creates the possibility that the speaker’s utterance may perform the speech act of white lying only up to a degree—that is, that the speech-act-hood of the utterance as a white lie will be graded (see Sect. 5). Isaacs and Clark’s account of ostensible invitations draws on the following exchange between Ross and Cathy, two students who had a date that same evening: (3) Ross: Cathy, Scott just called and told me that Brad and Dave and Rich and a lot of other guys from UCLA are going to be there tonight, so I guess we’re going to go a night early. [He explains the plans for the night.] Do you want to come? Cathy: That’s all right. I’ll pass. Ross: Okay. According to their analysis of this exchange, ostensible invitations involve

 This follows from Isaacs & Clark’s “P[erlocution]2. B comes to feel that A likes or approves of B to an extent consistent with P[erlocution]1” (Isaacs and Clark 1990, 502). 13

  A Speech-Act Theoretic Analysis of White (Prosocial) Lies 

253

two layers. At the top layer, Ross makes an invitation and Cathy declines it. At the bottom layer, Ross and Cathy take collusive actions toward each other with the mutual recognition that the top layer is a pretense. It is the discrepancy between the two layers that gives Ross’s and Cathy’s actions their functions. What gets put on record, via the pretense, is that Ross would like Cathy to go but that she can’t manage it. Off record, though, he breaks their date while assuring her he still enjoys her company, and she assures him she is not offended. (Isaacs and Clark 1990, 497)

In terms of conversational structure, this example empirically supports the claim made above that in the case of white lies (similar to Ross’s ostensible invitation), the preferred response format is reversed. While the preferred response to a genuine invitation is acceptance, Cathy colludes with Ross’s pretense by refusing, thereby indicating that she recognized his invitation as ostensible. What is more, Cathy’s polite refusal is immediately accepted by Ross, unlike in genuine invitations where an initial refusal normally prompts a repeat until the invitation is accepted (or refused for good reason). The reversal of the canonical pattern for invitations indicates that both interlocutors recognize the situation for what it is, and the ease with which they do this (both the refusal and its acceptance are formulated succinctly and without hesitation) signals that they know they are doing this. In sum, all lies (white and malicious) come with two types of intentions. The first is a classic Gricean r-intention. In virtue of this intention, lies, like all utterances, communicate some meaning about the world and perform a speech act such as an assertion, an invitation, and so on. The second is a lying intention, in virtue of which the speaker wants the listener to believe something she herself does not believe. While this second intention is not permissible to be recognized in malicious lies, it is permissible for it to be recognized in white ones. Differently put, the lying intention must not be recognized in the case of malicious lies but may be recognized (although it need not be) in the case of white lies. Permissibility of the recognition of the lying intention is different from its actual recognition and it is the former (whether permissibility is available as an option or not) that sets white lies apart from malicious ones. This means that, unlike the first intention all lies come with, which qua Gricean r-­intention

254 

M. Terkourafi

Fig. 1  The two types of intentions (top: Intention 1 = Gricean r-intention, bottom: Intention 2 = the lying intention) involved in malicious and non-malicious (gray and white) lies

must be recognized in order to be fulfilled, whether the second, lying intention is itself a Gricean r-intention varies for different types of lies: for malicious lies, the lying intention cannot be a Gricean r-intention, but for gray and white lies, it can (to different extents). Figure 1 represents this diagrammatically.14 On the left of this diagram, we have malicious lies, for which the lying intention (Intention 2) must not be recognized (that is why for these lies this intention lies below the level of awareness of the addressee and outside the domain of speech acts in the upper part of the diagram). As recognition of the lying intention becomes an option without jeopardizing the speaker’s goal (to the right of the red dashed line), we move toward the domain of white lies. However, we do so gradually, through shades of “gray” lies (Sect. 5). Once we acknowledge that permissibility of recognition of the lying intention is available for non-malicious (gray and white) lies but unavailable for malicious ones, we can both maintain a clear distinction between malicious and white lies (of various shades), and define white lies as a type of speech act that different utterances perform to various degrees.  For ease of exposition, Fig. 1 shows only assertions as a vehicle for (malicious or white) lies (that’s why Intention 1 at the top right of Fig. 1 corresponds to Assertion). That does not mean that non-­ assertive speech acts, such as invitations, compliments and so on cannot also be used ostensibly. Indeed, the type of two-layered account developed by Isaacs and Clark (1990) and here is meant to account for ostensible speech acts more generally, not just assertions, though that remains to be worked out in detail. 14

  A Speech-Act Theoretic Analysis of White (Prosocial) Lies 

255

4 White Lies Are a Type of Speech Act But Malicious Lies Are Not On Why There Is No Speech Act of Malicious Lying The above two-level analysis of the intentions involved in lying has interesting consequences for a speech-act theoretic account of lying in general, and of white lies in particular. Specifically, the non-permissibility of the recognition of the second, lying intention makes malicious lies different from speech acts such as promising, requesting, threatening, complaining, and so on. In promising, requesting, threatening, complaining, and so on, the speaker’s intention to promise, request, threaten, complain, and so on must be recognized by the addressee for the act to count as a promise, a request, a threat, or a complaint respectively—or, in Sbisà’s (2009) terms, in order to secure “uptake,” understood as the achievement of the characteristic conventional effect associated with this illocutionary act. However, in malicious lying, the speaker’s (non-Gricean) intention to lie must precisely not be recognized in order to be fulfilled (for the speaker to have successfully lied). Recognizing a lie as a lie makes a malicious lie unsuccessful. This means that, in the case of malicious lies, the only uptake the liar can aim for is as an assertion (or whatever other speech act the utterance is demonstrably performing)—corresponding to the recognition of the first, Gricean r-intention that all lies come with. Malicious lies are indistinguishable from assertions (or invitations, etc.) in this respect. To listeners, they are assertions (or invitations, etc.) tout court. While several theorists have proposed that lies are insincere assertions (Meibauer 2011; cf. Searle 1969, 42–49),15 others have been quick to point out the theoretical fraughtness of this move (see esp. Vincent and Castelfranchi 1981, 754, fn. 14; Reboul 1994). For instance, Reboul (1994, 297) writes: “[i]f a speaker produces an utterance which is a lie, it  Metaphors, ironies etc. can also be considered insincere assertions but there the insincerity is transparent and meant to be recognized for the speaker’s r-intention to be fulfilled. This is indicated by Grice’s phrase “make as if to say” in cases of metaphor etc., which indicates that their insincerity is communicative, that is, it is meant to be recognized by the addressee. 15

256 

M. Terkourafi

is necessary for the success of the lie that the illocutionary act of assertion should be successful. But if the perlocutionary act of lying is successful, then the illocutionary act of assertion is not successful [read: felicitous; A/N].” That happens, according to Reboul, because the speaker who lies is by definition not in the psychological state of belief toward the propositional content expressed by her utterance. In other words, the sincerity felicity condition, which stipulates that “a speech act is sincere only if the speaker is in the psychological state that her speech act expresses” (Green 2017) can never be met in the case of lies. On this view, an insincere assertion is necessarily an infelicitous assertion. But if lies are infelicitous assertions, how can they fulfil their perlocutionary goal of deceiving the addressee? The puzzle highlighted by Reboul arises only if we try to account for lies on a single level, as underlain by only one intention which is a communicative one. By contrast, on a two-level account such as the one proposed here, in which lies come with two types of intentions only the first of which is a Gricean r-intention intended to be recognized (see Fig. 1, Intention 1), while the second one is, in the case of malicious lies, a non-­ communicative intention that must remain hidden (see Fig. 1, Intention 2), a lie can be both a felicitous assertion (because the speaker sincerely wants the listener to believe something about the world) and a lie (because what the speaker sincerely wants the listener to believe is something she does not believe herself; see also Sect. 3).16 But if this second intention is not communicative, then in terms of the speech acts they perform, malicious lies perform only one type of speech act: they are assertions tout

 Reboul’s way out of this conundrum is to lower the bar on sincerity, so to speak (Reboul 1994, 298). In her view, all that is required for the speaker to be sincere is that she intend her utterance to be taken by the addressee as a reason to believe that she herself believes what she says. This is a condition on how the speaker would like the addressee to take her utterance (as evidence of belief ), not on how she actually positions herself toward the propositional content of her utterance (whether she believes it or not). On this revised (compared with the classical view reiterated by Green 2017) understanding of the sincerity condition, a lie is as sincere as any other assertion. The account I sketch immediately below also considers lies to be sincere assertions but additionally, by singling out the lying intention as a separate intention that is handled differently by malicious and non-malicious lies, offers the possibility of a unified account of these two types of lies. 16

  A Speech-Act Theoretic Analysis of White (Prosocial) Lies 

257

court.17 Lying is indeed something that is done through language but it is not done linguistically: it is not communicated.

The Speech Act of White Lying The fact that the lying intention in malicious lies is not communicative means that malicious lying cannot be a separate speech act, which in turn suggests that malicious lies perform only one speech act, that of assertion. However, one can go further and claim that, precisely because of the permissibility of the recognition of the lying intention in white lies, white lies can be a separate speech act. In this case, an utterance which is a white lie performs two speech acts at the same time: an assertion and a white lie (but see Sect. 5 on why this second speech act may be performed only up to a degree). Three types of evidence support this claim: (i) the availability of conventional means for their performance, (ii) their online processing, and (iii) their acquisition. Before proceeding, it is important to be clear about what we mean by “speech act.” In what follows, I will be adopting Sbisà’s (2009) neo-­ Austinian approach to speech acts, according to which a speech act is a socially constituted act that becomes meaningful against the background of existing norms and conventions in a group; it is a recognizable structured unit of communicative behavior aiming to bring about certain effects conventionally associated with this behavior. In other words, a speech act is a socially constituted tool for doing the things that a culture deems worthy of getting done linguistically.18 This neo-Austinian understanding of speech acts is especially apt in the case of white lies, since the extent to which it is permissible for white lies to be recognized is not up

 Meibauer (2011, 280) also correctly states that lies are not a separate kind of speech act. However, he considers them to be assertions, specifically insincere assertions, which is problematic for the reasons highlighted by Reboul. Moreover, he does not consider the (second) type of intention that makes an assertion additionally a lie or the fact that this lying intention is not meant to be recognized. 18  By contrast, the fact that speakers can use these tools in unexpected ways and listeners can understand unexpected meanings out of them goes beyond conventionality and into the realm of speaker’s intentions and perlocutionary effects achieved. 17

258 

M. Terkourafi

to the speaker to decide but rather hinges on the societal consequences of such recognition. These consequences, in the case of white lies, can include a range of positive outcomes. I have already commented in Sect. 3 on the possibility that, as a result of uttering a white lie, the speaker may be perceived as polite, and how this perception can be strengthened if the lie is recognized. Further to this possibility, research on trust has shown that prosocial lies (a type of white lie with high stakes) can enhance benevolence-based trust (Levine and Schweitzer 2015). Having asked experimental participants to play a trust game with a confederate who had previously either told them a prosocial lie or been selfishly honest, researchers found that participants trusted more those willing to tell a prosocial lie, despite recognizing them as deceptive, compared with those who stuck with the truth when it was selfish to do so. They explained these findings by appealing to “benevolence, demonstrating concern for others, [which] can be far more important for fostering trust than either honesty or selflessness” (Levine and Schweitzer 2015, 102). Findings such as these depend on the extent to which benevolence-based trust trumps other kinds of trust (e.g., integrity- and competence-based trust) in the situation at hand and need not generalize across cultures. Another positive outcome related with recognizing white lies concerns knowing how to respond to them. As Isaacs and Clark (1990, 496) note, with respect to example (3) above, it can be anything from embarrassing to hurtful not to recognize a white lie for what it is: Mutual recognition is important for several reasons. Without it, Cathy might genuinely accept the invitation, not realizing it was intended to be seen as merely a pretense. Or she might take it as obviously insincere, without realizing she was intended to see that, and she might feel insulted. Or if Ross made his invitation without expecting her to recognize the pretense, then it would be merely insincere, bearing the same relation to sincere invitations as lies bear to assertions. It would simply deceive.

The reversed preference organization in example (3), with Cathy’s unhesitating and brief refusal and Ross’s equally brief acceptance of it, helps them correctly identify what is going on.

  A Speech-Act Theoretic Analysis of White (Prosocial) Lies 

259

In sum, having one’s white lies recognized can be advantageous for several reasons including the attribution of politeness, benevolence, or simply eliciting an appropriate response. This observation provides a reason why languages/cultures should have recognizable ways of getting this done—in other words, it motivates the need for a speech act of white lying. Claiming the existence of a speech act, however, necessitates more than establishing a motivation for it. As Sbisà (2009, 46) reminds us: 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, 26–28)

The next three subsections provide evidence in this direction.

Conventionality of Means The first type of evidence supporting the existence of a speech act of white lying concerns the availability of conventional means for its performance. We may distinguish between two types of conventionality in this respect: conventionality of form and conventionality of content. Starting with conventionality of form, Brown and Levinson (1987, 172) highlight the existence of a high pitch in Tzeltal, a Mayan language, which functions precisely to signal a white lie: In Tzeltal, there is a highly conventionalized use of high pitch or falsetto, which marks polite or formal interchanges, operating as a kind of giant hedge on everything that is said. […] Use of it seems to release the speaker from responsibility for believing the truth of what he utters so that the presence of this falsetto in an otherwise normal conversation may well mark the presence of a social lie.

More recently, in a lab setting, Fish et al. (2017) identified a set of acoustic cues predicting the perception of a compliment as sincere or insincere (a white lie) in Canadian English. Keeping the wording of the compliment stable, they found that listeners rated as insincere those compliments which were spoken more slowly, started lower, and tended

260 

M. Terkourafi

to get louder as the utterance unfolded. The same stimuli were rated as less sincere if they had been elicited by an uncertain question (Asker asking for reassurance) compared to a confident one (Asker just curious about Responder’s opinion). In other words, not only the acoustic makeup of the compliment, but also the social requirements of the context (Asker’s expectations) affected listeners’ evaluation of a compliment as insincere. This finding supports the claim that white lies are expected in certain contexts, which has been independently advanced based on data about their online processing (see Sect. 4.2.2). The successful identification of acoustic cues for low-stakes white lies such as those investigated by Fish and her colleagues contrasts with the general difficulty of identifying reliable cues to lying behavior. According to Fish et al., this is because “control behaviours [taxed by prosocial lying] sometimes produce detectable vocal indications of the speakers’ insincere opinions, which are qualitatively different than when speakers are unfettered by the need to conceal their negative attitudes to the listener” (Fish et al. 2017, 157). It may well be that reliable cues are only available for white lies, whereas real lies are not consistently signposted by their utterers—precisely because, unlike white lies, they are not supposed to be recognized by their recipients. Coming to conventionality of content, perhaps the most conventionalized means for white lying can be found in Persian ta’arof, a highly ritualized practice involving figurative language and extreme “self-lowering” referring expressions that to outsiders can sound like “a collection of lies” (Miller et al. 2014, 15). By contrast, native speakers consider ta’arof to be a token of goodwill and respect, a strategic move to achieve particular perlocutionary goals, and/or an indication of the speaker’s good manners and upbringing (Miller et al. 2014, 19). In fact, ta’arof is thought to occur less in lower socio-economic classes and among villagers (Beeman 1986, 197). Example (4) illustrates this practice (for the original, see Izadi 2015, 86): (4) Ali and Reza are close friends returning home from an evening out. Reza has given Ali, who lives with his family, a ride home. It is almost midnight, not an appropriate time for receiving guests, especially if it is not prearranged.

  A Speech-Act Theoretic Analysis of White (Prosocial) Lies 

261

1. Ali: Come in. 2. Reza: Thanks a lot. 3. Ali: Come in. 4. Reza: Thanks. I’ve got to go. I have work. 5. Ali: Well, just come in for a minute then you can go. 6.  Reza:[extending hand] May I sacrifice for you. Don’t you need favor? 7. Ali: [refuses to shake hands] Are you doing taãrof? (standing on ceremony?) 8. Reza: (1.0) No (I swear) by God, convey my hello. 9. Ali: [extending hand] I’m at your service. 10. Reza: I’m your slave. Several features highlighted earlier as characteristic of white lies can be identified in this exchange, including the use of lexical formulae (“may I sacrifice for you,” line 6; “I’m your slave,” line 10) and the ceremonial denial of the white lie (lines 7–8). Being part of a leave-taking sequence, this exchange functions as an ostensible invitation “solicited by context” (Eslami 2005, 464). Nevertheless, it is not pointless: as Koutlaki (1997, 119) explains, “the fact that [the] speaker takes the trouble to use a socially enjoined formula indicates [his] intention to accord respect to [his] interlocutor and takes on therefore a phatic function.” Qua widely recognized cultural practice, ta’arof involves expressions that can be conventionalized to various degrees. While example (4) includes some highly conventionalized ones, similar contents can be expressed in more creative ways. What is recognizable in such cases is the attitude (of generosity, self-lowering, etc.) expressed by the utterance rather than the specific words used. The expression of these semantic meanings at particular moments during the interaction (especially during offers/invitations, requests/orders, thanking, complimenting, greetings and leave-taking) and in ways that are appropriate to one’s gender and age jointly co-constitute an exchange as an instance of ta’arof. In all of these ways, ta’arof resembles ritualized types of speech acts found in a variety of cultures (e.g., ritual refusals in Chinese; Yang 2008). Such acts are at one end of a continuum of ostensible speech acts (Isaacs and Clark 1990), the other end of which is populated by acts that are not necessarily as strictly

262 

M. Terkourafi

ritualized, yet can be conventionalized as regards their conversational make-up, sequential placement in interaction, objective felicity, and manner of delivery (Isaacs and Clark 1990, 499–502). The existence of these conventional signals suggests that speakers are aware of when their utterance is a white lie and have subtle ways of (consciously or subconsciously) signaling that to their addressees.

The Online Processing of White Lies Experimental results from the online processing of white lies suggest that in contexts where they are expected, white lies are treated no different from true statements in non-biasing contexts. Moreno et al. (2016) presented 26 female native speakers of Spanish with a context in which the truth was unpleasant (e.g., a host asking her guests whether they liked the dinner she had cooked for them, when the dinner had been burned). They then measured their ERP responses while reading one of three possible answers to the host’s question: (i) White Lies (WL) (e.g., The meat sauce was tasty); (ii) Blunt Truths (e.g., The meat sauce was overcooked); or (iii) Semantic Violations (e.g., The meat sauce was romantic). These answers had been previously elicited as likely continuations to The meat sauce was … in a paper-and-pencil task by a separate group of participants. They were then normed to obtain equally expected target words for each condition, factors such as word frequency biasing the results. What the researchers found was that white lies (The meat sauce was tasty) were not treated as semantically anomalous (did not induce an N400 response) in social situations where they were expected, whereas blunt truths (The meat sauce was overcooked) in the same situations were. This effect was heightened in situations strongly biasing toward white-­ lying behavior, that is, situations for which participants in the paper-and-­ pencil task agreed that a white lie was expected (34/93 scenarios in their study). Based on their findings, Moreno et al. (2016, 624) argued that the processing of white lies lacks of any semantic (N400) or interpretative (P600) difficulty. White lies are processed neither as false nor as ironic ­messages. Their immersion in a social context overrules the neural conse-

  A Speech-Act Theoretic Analysis of White (Prosocial) Lies 

263

quences that have previously been linked to the processing of factually untrue statements. Indeed, closer to truth statements [blunt truths] became more difficult to process, as indexed by an enhanced N400 for those sentences strongly/moderately biased toward white lying, as well as by a frontal late positivity in all cases.

In other words, taking N400 to be “not as sensitive to truth/falsity computations as […] an online predictor for upcoming information based on world knowledge” (Moreno et al. 2016, 617), white lies are treated as expected, at least in some situations. These results are interesting for a couple of reasons. First, both parts of the experiment (the paper-and-pencil task, and the online ERP task) confirm that there are specific moments in interaction which ratify the telling of a white lie and when participants expect a white lie to be told. This is in line with Fish et al.’s (2017) finding that exactly the same acoustic stimuli were judged as less sincere if prompted by an uncertain question of the Asker, that is, if the situation called for a white lie (irrespective of whether the propositional content of the utterance was truthful or not). The finding that white lies are what is normatively expected in some situations is also in line with Wilson and Sperber’s claim that “language use is not governed by any convention or maxim of truthfulness in what is said” but rather “expectations of truthfulness—to the extent that they exist—are a by-product of expectations of relevance” (Wilson and Sperber 2002, 583–584). The experimental evidence from the processing of white lies supports this claim, adding to our reasons for considering sincerity to be a maxim governing optimal performance of a speech act (along the lines of Sbisà 2018) rather than constitutive of its performance.

The Acquisition of White Lies Several studies have shown that children as young as three can produce white lies, that is, lies that show consideration for another’s well-being (Talwar and Lee 2002; Talwar et al. 2007). What is more, from a young age, children are able to do so appropriately: that is, they discriminate between situations that call for the truth versus those that call for a white lie and favor white lying in the latter (Walper and Valtin 1992; Bussey

264 

M. Terkourafi

1999). These findings are in line with Fish et al. (2017) and Moreno et al. (2016), who showed that there are specific moments in interaction when a white lie is likely to be told. By around age seven, children learn to align with adult preferences and “read” these moments correctly (Heyman and Lee 2012, 169). Crucially, these situations can be different in different cultures suggesting that children’s ability to read these situations correctly is not (just) part of cognitive maturation but rather of early socialization into the group norms surrounding them—that is, it is part and parcel of acquiring the repertoire of socio-culturally constituted speech acts available in their surrounding community. Specifically, although all children tend to value white lying when required by the social context (Xu et al. 2010), in cultures that place emphasis on autonomy, children justify the telling of white lies primarily with respect to their effect on the individual recipient’s emotional well-being, whereas in cultures prizing societal inter-­ dependence, they highlight the social implications for the recipient (Heyman and Lee 2012, 169). Focusing on the latter, experimental results from 7–11-year-olds in China suggest that their sensitivity to a public/private distinction and the greater risk of face loss in the former lead them to evaluate white lies less negatively in public settings vs. private ones, while they do the opposite for blunt truths (Ma et al. 2011, 314). Moreover, “children in all of the age groups rated white lies less negatively than harmful lies and rated blunt truths less positively than helpful truths. This finding suggests that even 7-year-olds are aware that the moral value of a verbal statement depends not only on its truthfulness but also on whether it serves to help or harm its recipient” (Ma et al. 2011, 314)—a claim that resonates with Wilson and Sperber’s (2002) argument that relevance, rather than truthfulness, is what is expected in interaction. Similar results were obtained in a study comparing classifications of lie- and truth-telling in prosocial situations by Persian and Canadian children aged 5 through 11 (Mojdehi et al. 2020). Not only did Persian children generally rate politeness and ta’arof lies more positively compared with the Canadians, as they got older, Persian children also

  A Speech-Act Theoretic Analysis of White (Prosocial) Lies 

265

evaluated untruthful statements in ta’arof situations less negatively than Canadians. All in all, the evidence from acquisition shows that children are not only able to produce white lies from early on but they do so in accordance with adult norms regarding the situational appropriateness of these lies, which can be different in different cultures. They thus support the claim that white lies constitute a type of speech act which different groups have developed different conventional procedures to signpost (Sbisà 2009).

5 The Context-Dependence of (White) Lies The three types of evidence surveyed above jointly suggest that white lies can be viewed as a type of speech act, understood as a socially constituted cultural practice aiming to bring about certain conventional effects along the lines of Sbisà (2009). However, it does not follow from this that white lies are always and unambiguously “white.” The existence of a continuum from malicious (real) to white lies has been empirically demonstrated by Bryant (2008), who combined interviews and focus-group discussion in an attempt to capture the perspective of the participants themselves. As Table 1 shows, next to real and white lies, participant views led Bryant to identify two categories of “gray” lies, meaning lies in which the various dimensions (intention, consequences, beneficiary, etc.) are at odds with each other, making it impossible to classify them as either real or white. An example of a gray lie is that of a vegetarian asking, after a meal that her friend had cooked, if the meal had contained any meat products and her friend assuring her that it did not, despite knowing that it did (since she had cooked it). Since the question was asked after the fact, when nothing could be done to rectify the situation, a positive answer would have only saddened or angered the questioner. Hence the falsehood can be said to benefit the addressee as well as the speaker and the relationship between them. Yet the lie is more consequential than an innocuous, white lie, making that label inappropriate.

266 

M. Terkourafi

Table 1  Bryant’s open-ended lie classification (from Bryant 2008, 32)

That what is a white lie depends on the context and is co-constructed between speaker and addressee is also supported by Fish et al.’s experimental results, who showed that there is no clear break between sincere and insincere compliments (i.e., white lies), but rather the interpretation of an utterance as a white lie also depends on the expectations of the Asker (2017, 156). As several scholars have noted, it is possible that, in the end, whether a lie is harmful or not depends on the recipient of the lie and how they feel about it: “what is a vicious, harmful lie for one person may be an act of loving concern for another. […] Lies can only ‘be’ as they are perceived by specific involved people” (Knapp and Comadena 1979, 271; cf. Bok 1978, 60). These empirical findings are especially pertinent to the claim that the hallmark of a white lie is permissibility of the recognition of the lying intention (Sect. 3) and bring us back to ambivalence as a defining feature of ostensible speech acts (Isaacs and Clark 1990). In Terkourafi (2014), I identified several reasons for such ambivalence, which extend beyond

  A Speech-Act Theoretic Analysis of White (Prosocial) Lies 

267

face-management served by ostensible speech acts: speakers may be happy to let addressees be the ultimate arbiters of what they mean, because it saves them the effort of spelling it out or because they do not want to take responsibility for it; but they may also not have a fully formed intention that they can communicate, as with children whose early vocalizations are requests only to the “ears” of caretakers who rush to attend to them and only gradually become requests (intended to be recognized) as the child observes their perlocutionary effects on others (Huls and van Wijk 2012, 92–97). To account for all of these cases we need a way of talking about an utterance’s performing a speech act as a graded notion. This idea is entertained in an early article by Lakoff (1992). Citing several legal cases where the courts decided differently, variably prioritizing the speaker’s (intended) meaning or the listener’s (perceived) one, Lakoff observes that “language and linguistic behaviour (and human behaviour in general) by their nature may not be reducible to yes/no decisions, nor amenable to being assigned to only one pile among many […]; to get anywhere near a deep understanding of the forms, functions, and properties of language, a theory which includes the possibility of continuous classification is necessary” (Lakoff 1992, 317; emphasis added). She then proposes such a classification, assigning utterances along a linguistic continuum that ranges from constatives to performatives corresponding to an action continuum that ranges from expression to conduct (Lakoff 1992, 320). What is useful to retain from this discussion is Lakoff’s observation that “words can have the effect of actions, be tantamount to action, sometimes […] but not all utterances are equally “active” that is, reality-changing” (Lakoff 1992, 318). A similar intuition is developed by Sbisà (2001, 1797), who writes of “a conception of illocutionary force according to which illocutionary acts have conventional effects” where such effects “may be described in terms of ‘deontic modality’, namely, as assignments to or cancellations from each one of the participants of modal predicates related to the necessity or possibility of actions with respect to norms.” In other words, speech acts, when performed felicitously, alter the rights and obligations of interactants and thereby the future courses of action they may plausibly take.

268 

M. Terkourafi

Since utterances can do this to different degrees, such a conception of illocutionary force 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. (Sbisà 2001, 1797)

Simply put, this means that an utterance may be a request, an invitation, a compliment and so on more or less; that is, it may carry the corresponding force to a greater or lesser extent. In the case of white lies, it would seem that, once it is permissible for the speaker’s lying intention to be recognized, what would otherwise be a malicious lie starts to become socially justifiable. Yet this leaves an enormous range over which language users can agree or disagree about the beneficial or damaging effects of an otherwise non-malicious (white or gray) lie.

6 Summary Potential lies come with two types of intentions: a Gricean r-intention, in virtue of which all lies are assertions; and a “lying” intention that must remain hidden in the case of malicious lies (i.e., is not a Gricean r-­intention) and is variably permissible to be recognized (i.e., may be a Gricean r-intention) in white ones. Due to their different handling of this second intention, malicious lies are not a separate kind of speech act, while white lies can be. Three types of evidence (conventionality of means, their online processing, and their acquisition) support analyzing white lies as a separate type of speech act. However, white lies are not always as clearly signposted in terms of where they occur in the exchange and the form they take, and a continuum of cases exists from the most innocuous white lie to the most malicious real one. While speaker and addressee(s) may disagree about whether the speech act of white lying or, conversely, malicious lying has occurred, the permissibility of the

  A Speech-Act Theoretic Analysis of White (Prosocial) Lies 

269

recognition of the speaker’s lying intention (i.e., the extent to which her goals would be served if her lying intention were recognized) provides a principled basis for analytically distinguishing between these two phenomena. Acknowledgments This chapter is based on presentations given at “New Directions in Linguistic Pragmatics” (NDLP2017), the International Symposium on Advances in (Im)politeness Studies 2018 at Fujian Normal University, and the workshop on Perspectives on Speech as Action in Trieste, in 2018 and its writing was made possible in part through a research leave granted by the Leiden University Center for Linguistics. I am grateful to the audiences on these occasions for their questions, as well as to the editors of this volume for their insightful comments on a previous draft. All remaining errors are my own.

References Austin, John L. 1975. How to do Things with Words. Oxford: Oxford University Press, 1962 (2nd ed.). Bach, Kent. 2001. You don’t say? Synthese 128: 15–44. Beeman, William. 1986. Language, Status, and Power in Iran. Bloomington: Indiana University Press. Bok, Sissela. 1978. Lying. Moral Choice in Public and Private Life. New York: Pantheon. Brown, Penelope, and Stephen Levinson. 1987. Politeness: Some Universals in Language Usage. Cambridge: Cambridge University Press. Bryant, Erin. 2008. Real Lies, White Lies and Gray Lies: Towards a Typology of Deception. Kaleidoscope 7: 23–50. Bussey, Kay. 1999. Children’s Categorization and Evaluation of Different Types of Lies and Truths. Child Development 70: 1338–1347. Camden, Carl, Michael Motley, and Ann Wilson. 1984. White Lies in Interpersonal Communication: A Taxonomy and Preliminary Investigation of Social Motivations. The Western Journal of Speech Communication 48: 309–325. Chisholm, Roderick, and Thomas Feehan. 1977. The Intent to Deceive. Journal of Philosophy 74: 143–159. Coleman, Linda, and Paul Kay. 1981. Prototype Semantics: The English Word Lie. Language 57: 26–44.

270 

M. Terkourafi

Eslami, Zohreh. 2005. Invitations in Persian and English: Ostensible or Genuine? Intercultural Pragmatics 2: 453–480. Fallis, Don. 2009. What Is Lying? Journal of Philosophy 106: 29–56. Fish, Karyn, Kathrin Rothermich, and Marc Pell. 2017. The Sound of (In) Sincerity. Journal of Pragmatics 121: 147–161. Green, Mitchell. 2017. Speech Acts. In Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta. URL = https://plato.stanford.edu/archives/ win2017/entries/speech-­acts/. Grice, Paul H. 1957. Meaning. The Philosophical Review 66, 377–88. Reprinted in Studies in the Way of Words (1989), 213–223. Cambridge, MA: Harvard University Press. ———. 1969. Utterer’s Meaning and Intentions. The Philosophical Review 78. Reprinted in Studies in the Way of Words (1989), 86–116. Cambridge, MA: Harvard University Press. ———. 1975. Logic and Conversation. In Syntax and Semantics, Vol. 3: Speech Acts, edited by Paul Cole and Jerry Morgan. New York: Academic Press. Reprinted in Studies in the Way of Words (1989), 22–40. Cambridge, MA: Harvard University Press. Heyman, Gail, and Kang Lee. 2012. Moral Development: Revisiting Kohlberg’s Stages. In Developmental Psychology: Revisiting the Classic Studies, ed. Alan Slater and Paul Quinn, 164–175. London: Sage. Huls, Erica, and Carel van Wijk. 2012. The Development of a Directive Repertoire in Context: A Case Study of a Dutch Speaking Young Child. Journal of Pragmatics 44: 83–103. Isaacs, Ellen, and Herbert Clark. 1990. Ostensible Invitations. Language in Society 19: 493–509. Izadi, Ahmad. 2015. Persian Honorifics and Im/Politeness as Social Practice. Journal of Pragmatics 85: 81–91. Knapp, Mark, and Mark Comadena. 1979. Telling It like It Isn’t: A Review of Theory and Research on Deceptive Communication. Human Communication Research 5: 270–285. Koutlaki, Sophia. 1997. The Persian System of Politeness and the Persian Folk Concept of Face, with Some Reference to Efl Teaching to Iranian Native Speakers. Ph.D. Dissertation, University of Wales. Kupfer, Joseph. 1982. The Moral Presumption against Lying. Review of Metaphysics 36: 103–126. Lakoff, Robin. 1992. Pragmatics and the Law: Speech Act Theory Confronts the First Amendment. In Papers from the 27th Regional Meeting of the Chicago

  A Speech-Act Theoretic Analysis of White (Prosocial) Lies 

271

Linguistic Society, Vol. 1: The Main Session, ed. Lise Dobrin, Lynn Nichols, and Rosa Rodriguez, 306–323. Chicago: Chicago Linguistic Society. Levine, Emma, and Maurice Schweitzer. 2015. Prosocial Lies: When Deception Breeds Trust. Organizational Behavior and Human Decision Processes 126: 88–106. Ma, Fengling, Xu Fen, Gail Heyman, and Kang Lee. 2011. Chinese Children’s Evaluations of White Lies: Weighing the Consequences for Recipients. Journal of Experimental Child Psychology 108: 308–321. Mahon, James. 2008. Two Definitions of Lying. International Journal of Applied Philosophy 22: 211–230. ———. 2019. Contemporary Approaches to the Philosophy of Lying. In The Oxford Handbook of Lying, ed. Jörg Meibauer, 32–55. Oxford: Oxford University Press. Meibauer, Jörg. 2011. On Lying: Intentionality, Implicature, and Imprecision. Intercultural Pragmatics 8: 277–292. ———. 2014. Bald-Faced Lies as Acts of Verbal Aggression. Journal of Language Aggression and Conflict 2: 127–150. Miller, Corey, Rachel Strong, Mark Vinson, and Claudia Brugman. 2014. Ritualized Indirectness: Explaining the Language Practice Taarof as a Reflection of Persian Speakers’ Cultural Values. Technical Report, Center for Advanced Study of Language. University of Maryland. Mojdehi, Atiyeh Shohoudi, Asadeh Shohoudi, and Victoria Talwar. 2020. Children’s Moral Evaluations of Different Types of Lies and Parenting Practices and across Cultural Contexts. Current Psychology. https://doi. org/10.1007/s12144-­020-­01059-­7. Moreno, Eva, Pilar Casado, and Manuel Martin-Loeches. 2016. Tell Me Sweet Little Lies: An Event-Related Potentials Study on the Processing of Social Lies. Cognitive Affective and Behavioural Neuroscience 16: 616–625. Reboul, Anne. 1994. The Description of Lies in Speech Acts Theory. In Pretending to Communicate, ed. Herman Parret, 292–298. Berlin, New York: de Gruyter. Saul, Jennifer. 2012. Lying, Misleading, and What Is Said: An Exploration in Philosophy of. Language and in Ethics. Oxford: Oxford University Press. Sbisà, Marina. 2001. Illocutionary Force and Degrees of Strength in Language Use. Journal of Pragmatics 33: 1791–1814. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press.

272 

M. Terkourafi

———. 2009. Uptake and Conventionality in Illocution. Lodz Papers in Pragmatics 5 33–52. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2018. Varieties of Speech Act Norms. In Normativity and Variety of Speech Actions, edited by Maciej Witek and Iwona Witczak-Plisiecka, 23–50. Leiden: Brill. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. Searle, John. 1969. Speech Acts. An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. Sperber, Dan, and Deirdre Wilson. 1995. Relevance. Communication and Cognition. 2nd ed. Oxford: Blackwell. Talwar, Victoria, and Kang Lee. 2002. Emergence of White-Lie Telling in Children between 3 and 7 Years of Age. Merrill-Palmer Quarterly 48: 160–181. Talwar, Victoria, Sean Murphy, and Kang Lee. 2007. White Lie-Telling in Children for Politeness Purposes. International Journal of Behavioral Development 30: 1–11. Terkourafi, Marina. 2008. Toward a Unified Theory of Politeness, Impoliteness, and Rudeness. In Impoliteness in Language: Studies on Its Interplay with Power in Theory and Practice, ed. Derek Bousfield and Miriam Locher, 45–74. Berlin: Mouton de Gruyter. ———. 2014. The Importance of Being Indirect: A New Nomenclature for Indirect Speech. Belgian Journal of Linguistics 28: 45–70. Turner, Ronny, Charles Edgely, and Glenn Olmstead. 1975. Information Control in Conversation: Honesty Is Not Always the Best Policy. Kansas Journal of Sociology 11: 69–89. Vincent, Jocelyne, and Cristiano Castelfranchi. 1981. On the Art of Deception: How to Lie While Saying the Truth. In Possibilities and Limitations of Pragmatics, ed. Herman Parret, Marina Sbisà, and Jef Verschueren, 749–777. Amsterdam: Benjamins. Walper, Sabine, and Renate Valtin. 1992. Children’s Understanding of White Lies. In Politeness in Language: Studies in its History, Theory and Practice, ed. Richard Watts, Sachiko Ide, and Konrad Ehlich, 231–251. Berlin: Mouton de Gruyter. Weissman, Benjamin, and Marina Terkourafi. 2018. Are False Implicatures Lies? An Empirical Investigation. Mind and Language 34 (2): 221–246. Wilson, Deirdre, and Dan Sperber. 2002. Truthfulness and Relevance. Mind 111: 583–632.

  A Speech-Act Theoretic Analysis of White (Prosocial) Lies 

273

Xu, Fen, Xuehua Bao, Fu Genyuen, Victoria Talwar, and Kang Lee. 2010. Lying and Truth-Telling in Children: From Concept to Action. Child Development 81: 581–596. Yang, Jia 2008. How to Say “No” in Chinese: A Pragmatic Study of Refusal Strategies in Five TV Series. In Proceedings of the 20th North American Conference on Chinese Linguistics (NACCL-20) 2, edited by Marjorie Chan and Hana Kang, 1041–1058. Columbus: The Ohio State University. Zembaty, Jane. 1988. Plato’s Republic and Greek Morality on Lying. Journal of the History of Philosophy 26: 517–545. ———. 1993. Aristotle on Lying. Journal of the History of Philosophy 31: 7–29.

Presupposition and Propaganda: A Socially Extended Analysis Michael Randall Barnes

1 Introduction There is a tendency within some strains of social philosophy of language to focus on the more subtle forms of subordinating speech. We see this in the recent explosion of work on concepts like microaggressions, code-­ words, dogwhistles, and more.1 And while there is certainly important philosophical work to do on these harmful uses of language, we should not lose sight of the fact that much of the speech that oppresses is not of this subtle kind, but is explicit, direct, and unmistakable.2 The philosophical literature, however, has been concerned largely with the subtle  On microaggressions, see Rini (2021). On code-words, see Khoo (2017). On dogwhistles, see Saul (2018). 2  For work on these more explicit instances of oppressive speech, see Tirrell (2012), Waldron (2012), among others. 1

M. R. Barnes (*) Humanising Machine Intelligence, Australian National University, Canberra, ACT, Australia e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Caponetto, P. Labinaz (eds.), Sbisà on Speech as Action, Philosophers in Depth, https://doi.org/10.1007/978-3-031-22528-4_12

275

276 

M. R. Barnes

side of this story, and for understandable reasons. The phenomena at play here interest philosophers and linguists even outside of their morally charged uses. The mechanics of not-at-issue content, accommodation, generics, and more, are complex, and when put to use in social/political contexts, deserve scrutiny.3 A related tendency lies in the focus on bystander silence as a mechanism for empowering hate speakers and propagandists. According to several recent theorists, the silence of bystanders can, in some cases, confer greater social power to a public hate speaker than they might at first seem to have. This is because the uptake given by hearers can, under certain conditions, play a significant role in constituting or securing the authority of a speech act, and silence—the failure to interject—can provide this uptake. So, if someone shouts racists remarks at a Black woman on the bus, uses slurs and tells her to “go home,” because she “doesn’t belong here,” and every other passenger averts their eyes or otherwise fails to express condemnation, those bystanders may be complicit in a subordinating (speech) act, partly because they contribute to the authority that backs up that speech act, even if unwillingly.4 The concept of authority plays a large role here, and building off Marina Sbisà’s (1999) suggestions, Rae Langton argues that it is through presupposition accommodation that authority may be accrued by hate speakers and propagandists (Langton 2018a, b).5 “Authority,” she claims, “can be obtained by accommodation, a default adjustment that occurs, without fuss, when hearers take on board what speakers presuppose” (Langton 2018a, 126). In this manner, accommodation “has the power to alter the illocutionary force of an utterance” (Langton 2018a, 144),  See Stanley (2015) for an analysis of the role of not-an-issue content in propaganda. See Leslie (2017) for discussion of the harmful effects of generics. Accommodation will be discussed throughout this chapter. 4  For various forms of this example, see Maitra (2012), Langton (2018a), McGowan (2012), Ayala and Vasilyeva (2016), Bianchi (2019); and see Barnes (2016) and Brown (2019) for critical elaborations on this example. 5  Langton cites Sbisà (1999) as one source of inspiration for this approach to presupposition, as she argues that “presupposition is suitable for transmitting a kind of content which might be called ideological” (Sbisà 1999, 493), as well as the attention she gives to blocking and explication as a means of resistance. See also, Sbisà (2002, esp. 430–431), which offers an early explanation of the attribution of speakers’ authority via accommodation. 3

  Presupposition and Propaganda: A Socially Extended Analysis 

277

and “hearers routinely accommodate what speakers presuppose, and the hearer’s omission is a quiet engine of the speaker’s success” (Langton 2018a, 127).6 While insightful, I argue Langton’s model focuses too strongly on individual speech acts isolated from their wider context, and for this reason cannot explain the authority of the bulk of propagandistic hate speech. I consider the limits of her conception of presupposition accommodation to clarify the audience’s role in helping empower a hate speaker. While Langton and other theorists tend to emphasize the role that bystander silence plays in these cases, I claim there are necessary mechanisms at play in multiple audiences that are much more active, and much more social. In other words, in addition to silently—and perhaps reluctantly—helping hate speakers via an omission, hearers also have a range of actions at their disposal that, when combined with other background conditions, are important to the conferral of authority to hate speech. My interest here is in speech acts whose subordinating force is secured by being a part of a broader group practice. The point I aim to make here is distinct from—though certainly related to—the more general one about oppression itself being a structural practice (Young 2011). In previous work (Barnes 2016), I criticized the tendency to isolate (oppressive) speech acts from the broader practices within which they are a part. Here, I elaborate on how these broader practices take shape, focusing on speech acts whose distinctive capacity to oppress—whose subordinating authority—depends on the utterances performed by other subordinating speakers. I sketch a conception of collective subordinating authority that puts an audience’s active participation at the center of an overall social, and not simply linguistic, phenomenon. Overall, my aim is to show how the entitlement to subordinate is not always located in individual speakers, but is more diffuse. This begins to fill an explanatory gap in the literature that, I believe, has had an unduly narrow focus so far. I begin by explaining the role of authority in hate speech and motivate the search for a stable form of informal authority (Sect. 2). Next I clarify  This builds on the influential work by David Lewis (1979), who argued that conversations follow a “rule of accommodation.” The speech act theory of J.L. Austin (1975) is another crucial conceptual ingredient. 6

278 

M. R. Barnes

the sort of nonideal authority at issue here (Sect. 3), and then examine Langton’s proposal that presupposition accommodation can supply this authority to hate speakers (Sect. 4). I then propose a model that has a more active role for some conversational participants and explain how this is more in line with Sbisà’s original approach to the normative aspects of presupposition, as well as her attention to the complexities of real-­ world communication (Sect. 5).

2 Authority in Hate Speech “Discursive authority,” Lynne Tirrell (2018, 15) writes, “is a situational power to make felicitous speech acts and gain a range of appropriate uptakes. A speaker’s discursive authority renders her speech acts socially meaningful.” Speaker authority, therefore, is what explains how some speakers can perform certain acts—do certain things with their words— that others cannot. Plausibly, one speech act that requires authority is to subordinate with words, including hate speech (Langton 1993). To be useful however, this notion of authority must be clarified. An initial distinction between formal (or institutional) and informal (or situational) authority helps to set the stage for the exploration of this tricky concept. A conception of formal authority can explain why the current President has some significant powers, and why their (executive) orders have the force to compel action that they do. This type of authority is properly recognized as institutional authority, and it’s by inhabiting institutional roles that some speakers are authorized to perform certain kinds of speech acts. Once we identify what role within a given institution a speaker occupies, we learn a great deal about their ability to perform authoritative speech acts. And, of course, this kind of authority can be used to subordinate. Indeed, this is the type of authority demonstrated in classic examples such as the South African legislator prohibiting Black citizens from voting (Langton 1993, 302–302), along with other cases where the legal rights and permissions are at issue—for example, slavery laws (Langton 2018a, 123–124). But when we’re considering informal, or non-institutional, conceptions of authority, things are much less clear cut. The mechanisms by

  Presupposition and Propaganda: A Socially Extended Analysis 

279

which speakers gain authority, along with the specific contours of the scope of its domain or the strength of its edicts are less obvious. This lack of clarity, however, sits alongside oppressive effects that remain clear. The harmfulness of more “everyday” hate speech is evident despite this lack of institutional authority. One promising direction is to understand informal authority as something that an audience can bestow on a speaker. Speakers make utterances whereby they attempt to move themselves into a position of authority and at the same time exercise this authority via the same utterance.7 How a speaker’s audience responds plays a significant role in whether such a move—and moreover, the exercise of authority—succeeds. If the audience goes along with the speaker, through their acts and omissions, it can appear that the speaker has effectively given orders, ranked options, and so on. Ishani Maitra (2012) uses examples like these to show how one path for gaining informal authority may occur via what she calls “licensing.” To explain this idea, she asks us to consider a group of friends trying to plan a hike together. As no one expresses any strong preferences in the logistics of the trip, they fail to make much headway in planning the outing. After having enough of this, one friend, Andy, decides to take charge and make decisions. He assigns specific tasks to each of the different members of the group. No one objects, the tasks are all completed, and the hike later takes place as Andy planned. Maitra’s claim here is that in this case Andy comes to have the authority to assign tasks to the group, and that his “instructions, moreover, are authoritative speech” (Maitra 2012, 106). In this way Maitra argues that a speaker who lacks authority prior to speaking may gain it when their speech acts go unchallenged. And this can play out in both innocuous situations (e.g., planning a hike, choosing a restaurant) and insidious ones. Consider the following example of ordinary hate speech, drawn from Maitra (2012, 100–101): An Arab woman is on a subway car crowded with people. An older white man walks up to her, and says, “F***in’ terrorist, go home. We don’t need  See Lance and Kukla (2013, 473–475), who explain how a single speech act can “do double duty” and function both as a meta-call that restructures “normative relationships and possibilities for making first-order calls” and as a first-order call at the same time. 7

280 

M. R. Barnes

your kind here”. He continues speaking in this manner to the woman, who doesn’t respond. He speaks loudly enough that everyone in the car hears his words clearly. All other conversations cease. Many of the passengers turn to look at the speaker, but no one interferes.

Using examples like this, Maitra argues that “licensing” is one route ordinary hate speakers can gain the authority they need for their subordinating speech acts to succeed.8 But notice the type of authority in these cases is situational, and as such is not stable across contexts. Andy may be able to give instructions relevant to the hike, but not outside of this narrow domain (Barnes 2016, 253). And while the hate speaker may possess the authority to subordinate in one moment, he may lose it as soon as he steps off the bus—or someone speaks up. So, unlike the President who retains their institutional authority from one day to the next and despite some challenges, speakers with this type of situational authority might lose it soon after they acquire it. The possibility of stable informal authority, however, deserves our attention. While the case of a random man on a bus may be explained by the situational authority afforded by something like licensing, other cases seem different. Successful propagandists and conspiracy theorists, who regularly and reliably spread hate, have distinct speech act powers. Whether their topic is the dangers of vaccines, the globalist plot around us, the threat of refugees and immigrants, or the gay and/or trans agenda, fringe speakers can cultivate a following and influence public discourse in ways that reveal the power of their speech. Not hindered by a lack of formal authority, these speakers nevertheless achieve a type of informal authority that is more stable than the situational kind, at least on certain—often controversial—topics. This points toward the importance of clarifying the domain of authority in question, as this explains why there are some topics within, and many beyond, a speaker’s authority.9 And once we recognize the importance of  While there are relevant differences, Langton adapts this example from Maitra to explain her own model of the presupposition and accommodation of authority. 9  Nearly all authority is limited to a domain, and so specifying its domain is a crucial task. The one exception is (perhaps) God’s authority in some religion. As Tirrell (2018, 16) notes, “Theists ascribe perfect authority to God, who, as the original speaker of performative utterances, has creative as well as absolute coercive power.” 8

  Presupposition and Propaganda: A Socially Extended Analysis 

281

domain in anchoring authority, we can notice a type of informal authority that is nonetheless fairly stable across contexts. For a first pass at this idea, consider a case where there is a distinct lack of rivals in a particular domain.10 Given a complete lack of sexual education in some schools, pornography may take on authoritative status in the domain of sex for adolescents. That is, absent rival candidates for authority, a speaker’s words might have the performative force of authoritative judgments. And this may be true for pornography as well as hate speakers. “If hate speakers are the only local voices,” Langton (2018a, 140) suggests, whether because of state-imposed restrictions, or technology-imposed echo-chambers, then hate speakers will have the authority of a monopoly. […] So a speaker may have authority relative to one field, where there are no rivals, but […] the speaker’s authority may be invisible to those looking from outside the bubble.

However, appeals to domain only go so far. It may make some sense to claim that pornographers have (stable informal) authority because of their near-monopoly status for many adolescents (Langton 2017, 2018a, 139). But it strains meaning to say that bigoted propagandists—like the one-time ubiquitous white supremacist Richard Spencer, for example— have authority in a domain like immigration because of an absence of comparative rivals. Here, there are rivals, and it’s not the case that his racist views are the only ones his audience has encountered. Indeed, it’s often because these views push back against what his audience sees as “the liberal orthodoxy of the elites” that he—or speakers like him—are given the uptake they do from a subset of their audience. Yet, since a not-­ insignificant portion of their audiences treats them as authoritative in this way, there is clearly something to be said here. Getting clearer on the sort of stable informal authority that hate propagandists like this wield, and how they accrue it, is my task below. And to begin, it’s useful to reflect on how this authority is—in more ways than one—a sort of nonideal authority. I turn to this in the next section.

 Langton explores the notion of authority’s relativity to comparative rivals in her (2017) and (2018a). 10

282 

M. R. Barnes

3 Nonideal Authority Distinguishing between practical and epistemic authority is a helpful way to reveal the decidedly nonideal nature of the authority that concerns us. Practical authority is authority for action, while epistemic authority is authority for belief.11 While analytically separable, practical and epistemic authority are usually exercised together. As Langton, borrowing an example from Joseph Raz, puts it: “doctor’s orders have their status as directives, in part because of what your doctor knows, or is taken to know: his practical authority has its source in his epistemic, or ‘theoretical’, authority” (Langton 2018a, 141). But the distinction between practical and epistemic authority helps to reveal something interesting about this sort of authority. This is because the conception of epistemic authority we’re interested in is more about credibility than it is about actual expertise. And while one does not become an expert simply because some people take them to be one, being taken as credible is often enough to get others to believe and act on those beliefs. Large swaths of the wellness industry would not exist were it not for the credibility—not expertise— of those who promote a variety of cures with little to no actual evidence. And this reliance on a receptive audience carries over to—and is perhaps even more straightforward in the case of—practical authority, since, as Langton (2018a) notes, “being put ‘in charge’ may be enough for being in charge.” In this way, the distinction between credibility and expertise clarifies that de facto authority is often enough. Our topic, in other words, is not authority as it ought to exist, but authority as it works in the real world.12 Credibility, not expertise. De facto, rather than de jure authority. This means moving away from the restrictive conception of formal speaker authority as it is depicted in Austin where, as Langton (2017, 33) notes, “authority is often a formal  The following discussion has its origins in Raz (2009, 9), who introduced the distinction between practical and epistemic (or “theoretical”) authority. Tirrell, in a similar though distinct way, distinguishes between “positional authority” and “expertise authority” (Tirrell 2018, 17). 12  As Langton (2018a) says, and I concur: “Our topic is not idealized authority, but something closer to home: structures of social authority, relative to practices, which enable the enactment of norms and hierarchies that are socially real—whether or not they exist in Plato’s heaven, or the better neighborhoods of Earth.” 11

  Presupposition and Propaganda: A Socially Extended Analysis 

283

matter: it is the authorized official who can name the ship, conduct the wedding, issue the sentence.” In contrast to this, Langton (2017, 33) observes, “the workings of informal authority are more subtle.” As we’ve already noted, authority is most often relative to a certain domain. Authority being further relative to a certain jurisdiction amounts to noting how it may be limited to certain audiences too. So, some, but maybe not all who hear its call will feel its normative pull. What the Pope or a local pastor says may have no grip on the atheist, but it can amount to a clear directive with clear authority to the devoted. Appealing to the idea of a relativity to jurisdiction is helpful because it “explains how a low type relative to one hierarchy could be, so to speak, a high type relative to another” (Langton 2018a, 139). And this is what a more on-the-ground, nonideal conception of authority permits. It is, in a sense, a descriptive account of who takes some speakers to be authoritative, rather than a prescriptive account of who ought to have authority. What we are attempting to describe, then, is a stable type of informal authority that recognizes that an anti-vaxxer blogger can be an authority for millions of parents, even though the blogger lacks all traditional markers of expertise on the subject.13 It is the fact that their speech functions like more traditional, authoritative speech for a portion of its audience that is the chief reason to still label this capacity “authority.” In other words, if some people take the “quack” doctor’s advice as credible, and shape their actions in accordance with their directives, we ought to understand their actions as responses to authoritative utterances, just like the person who takes the “legitimate” doctor’s words as sufficient reason for belief and action. In both cases, someone “grants authority to experts’

 The mechanisms at play here are related to the social epistemic notion of an “echo chamber,” which is aptly described by Thi Nguyen (2020, 146) as “an epistemic community which creates a significant disparity in trust between members and non-members. This disparity is created by excluding non-members through epistemic discrediting, while simultaneously amplifying members’ epistemic credentials.” While Nguyen is primarily interested in beliefs, and how these are impacted by idiosyncrasies of trust at play in certain communities, I have a slightly more expansive target in mind, owning partly to the fact that I’m interested in practical authority as well as theoretical authority. 13

284 

M. R. Barnes

speech acts by taking up the licenses they issue, while deferring responsibility for justification back to them” (Tirrell 2018, 19).14 And what explains the authority of “quack” doctors may also explain the authority of some speakers of hate propaganda. The authority the bigot sees in the white nationalist’s words may be as ineffective to the non-bigot as the Pope’s words are to the atheist, and just as invisible. And yet, within a certain community, that speaker might have real authority over these subjects. This may be most evident in propagandistic hate speech as it functions as ingroup speech, which occurs when one group is speaking to its own members, in the attempt to gin up hate for outsiders. However, its impact on outgroups is also significant. At this point, it’s helpful to remember that authority is satisfied with acquiescence, not necessarily (explicit) agreement. Speech, and especially authoritative speech, is about shaping the normative landscape for one’s audience—and, as I’ll elaborate, other third parties as well—and this can occur without explicit agreement.15 Rather, what it requires is an adjustment that takes on board what speakers do with their speech acts along with what they presuppose, whether this be the authority of the speaker, the content of their utterance, or both—that is, their uptake. And instead of explicit agreement, as Sbisà (2009, 50) reminds us, “uptake consists in a tacit agreement,” and that this may be “made manifest in the audience’s response (insofar as people act and speak upon what they take has been done up to that point).”16 Moreover, as Langton notes, “the social world can sometimes be accommodating, especially when helped along by background expectations”

 That is, the perlocutionary effects of some speech provide evidential support that these speech acts are of a certain illocutionary kind. This sort of methodology goes back to Langton (1993, 309), at least. 15  See, for example, Sbisà (2007, 466), where she notes that “in most cases implicit or even tacit agreement is enough” to bring about the conventional effects of an illocutionary act. 16  Earlier in that same paper she clarifies this process by appealing to the realm of norms rather than natural events, using the example of naming. “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 behaviour that may stem from the norm” (Sbisà 2009, 45, emphasis added). 14

  Presupposition and Propaganda: A Socially Extended Analysis 

285

(Langton 2018a, 134, emphasis added).17 And given background conditions of racism, sexism, and much more, the social world is often far too accommodating to subordinating speech, as many in its audience acquiesce to its authority. As the large and passionate following of many hateful speakers—like those of Richard Spencer, Alex Jones, and so on—show, there is often a significant portion of people who play the part of a willing, and compliant audience, who happily alter their corner of the social world. And this also impacts those around them. By responding in this way—by giving this uptake—an audience can demonstrate how a hate speaker’s words are backed up by authority, as this authority is embodied in their uptake. As Tirrell (2018, 21) puts it: “Authority is largely constituted by audience uptake, which is a matter of next moves.”18 If this is the case, then our target of analysis are the broad features of a situation that together inform what next moves occur. That is, in other words, “the total speech situation” that Austin (1975) concerned himself with. And as Sbisà (2007, 471) notes, this is not generally about information, but about roles. Austin, she points out, 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.

But, as I argue in the following section, the approach Langton uses to address these issues—where presupposition accommodation is the main mechanism—is focused too narrowly on audience silence as the main constitutive element that determines authoritative uptake. And, as I will go on to claim, there is much more besides this that our attention must turn to as well.  Langton goes on: “Accommodation is only a tendency, as Lewis said. It does not always work, and whether it works is itself sensitive to background social hierarchies” (2018a). This sensitivity to background social hierarchies is what I am attempting to investigate. 18  Tirrell directs us toward Bourdieu (1991, 116), who explains: “The symbolic efficacy of words is exercised only in so far as the person subjected to it recognizes the person who exercises it as authorized to do so, or, what amounts to the same thing, only in so far as he fails to realize that, in submitting to it, he himself has contributed, through his recognition, to its establishment.” 17

286 

M. R. Barnes

4 Presupposition Accommodation of Authority Presupposition accommodation is the default adjustment that occurs when a speaker says something that requires a change in the shared assumptions guiding the conversation.19 A classic example of this sort of accommodation occurs when a speaker says, “Even George could win,” and “straightway” the presupposition that George is an unpromising candidate “springs into existence, making what [the speaker] said acceptable” (Lewis 1979, 339). When this occurs, the hearer takes on board this new information, and future moves within this conversation are explained by the fact that hearers passively accept this low ranking of George. Langton’s intriguing suggestion is that this model may be straightforwardly applied to authority as well, including the authority of hate speech. “Speech acts,” she says, “including directives generally, and hate speech specifically, can acquire authority by an everyday piece of social magic: authority gets presupposed, and hearers let it go through, following a rule of accommodation” (Langton 2018b, 152).20 But extending this idea to the accommodation of the authority of hate speech is more difficult than Langton suggests,21 especially if we want to capture a wider  With some qualifications, these shared assumptions form the conversational score which can be thought of as “an abstract entity that tracks whatever is relevant to the proper development of the conversation (e.g., relevant topic, presuppositions)” (Ayala and Vasilyeva 2016, 257). For the original expression of this notion, see Lewis (1979). A related but distinct notion is that of the common ground, which tracks the participants of a conversation’s shared beliefs and other psychological states. See Stalnaker (1973, 1974, 2002). 20  See also: “Accommodation supplies a straightforward way for hate speech to gain authority” (Langton 2018a, 127). A similar account is developed by Ayala and Vasilyeva (2016). 21  Langton does acknowledge important differences between these two types of accommodation, but for her it is mainly a difference in effect (acceptability vs. truth), and the mechanism is largely the same. See: “When authority is what is presupposed, default uptake allows it to go through, and the speaker actually obtains authority, empowering the speaker to perform speech acts whose felicity requires authority. […] Their felicity depends on presupposed authority that becomes real, when passive hearers let it through. Observe here a difference between presupposed authority and other presuppositions. Unblocked presuppositions tend to become acceptable rather than true: “George is an unpromising candidate” may become acceptable, if unblocked, but that would not make it true. Authority is different. A presupposition of authority can become, not just acceptable, but true, because its existence, not only its acceptability, depends in part on what hearers do, or fail to do” (Langton 2018b, 155). And: “Authority’s existence comes into being through accommodation, not only its acceptability” (Langton 2018a, 144). See also Ayala and Vasilyeva (2016, 261) and Witek (2013). 19

  Presupposition and Propaganda: A Socially Extended Analysis 

287

set of cases beyond those like the subway example, which as I noted lacks stability. To be fair, Langton may not be interested in the wider set of cases that I aim to uncover, and her work is indeed focused most explicitly on the “back door” cases of hate speech that demonstrate the more sneaky and passive mechanisms of establishing an authority. So, I could (justly) be accused of changing the topic. It is, as she says, less controversial how active endorsement contributes to authority (Langton 2018a, 133). But I believe my point remains, and that the phenomenon that I’m interested in is something that should concern social philosophers of language. How chants of “No more refugees!” in one context contribute to the power of silence in another are part of the messy mechanisms that shape our discourse. My claim is that these are mutually sustaining practices. That the active support for hate speech by some enables the passive accommodation of others that Langton depicts. What’s required to capture the wider phenomena I’m pointing toward is a more complex understanding of the multiple audiences a speech act can have, along with greater appreciation for how (part of ) an audience can itself actively aid in the construction of authority, and not simply passively accept it. Once we acknowledge these features and modify the conception of accommodation to properly capture what is occurring in these cases, we are led to a novel type of speaker authority that I call collective subordinating authority. Consider Richard Spencer again, who rose to prominence by spearheading an unabashed white supremacist ideology. At the height of his popularity, he had legions of followers, some of whom he led in a Sieg Heil salute while he and the crowd chanted, “Hail Trump!”22 In my assessment, this utterance can be interpreted as a directive—much like how graffiti reading “Muslims Out!” is partly a directive (Waldron 2012)—as well as a show of strength. While strictly speaking, “Hail Trump,” could be interpreted as a mere greeting, this clearly fails to capture all of what is occurring. The Nazi-esque utterance (and gestures) are meant to evoke  See Wood (2017).

22

288 

M. R. Barnes

that particular history. Given that “Heil Hitler” was partly about displaying obedience, the implicit meanings of that utterance carry over to the present. And more than simply asserting one’s own obedience to a leader, it can also be interpreted to demand obedience as well. It is in this way it has a directive side to it as well, like “Muslims out!” and like much else hate speech too. This all makes the Spencer-led chant a clear case of (authoritative) subordinating speech and a useful example against which to test the limits of accommodation. But what is at issue here is not accommodation in the conversational sense as described by Lewis and taken up by Langton, but rather accommodation—if we call it that—in a much more extended sense: social accommodation.23 What a willing audience gives to empower hate speakers is significantly different than the accommodation that occurs within a single, relatively shorter conversation. And it demands a different explanation. Like Langton, I agree it is largely because of the audience uptake that this type of subordinating speech may be called authoritative and has the specific subordinating force that it does. But rather than focus on the passive segment of the audience—who either take on board their noxious presuppositions or at least fail to block them—I want to emphasize the role of those who take up these utterances willingly, and in doing so expand the reach of these speakers’ words even wider with their own speech acts. I propose a model that acknowledges this more active role for some conversational participants in the next section.

 A similar extension is proposed in Adams (2020). Though, for Adams, “social accommodation” is a mechanism that allows an audience to resist a speaker’s attempt at claiming authority by, instead, “changing the context.” My interest, on the other hand, is in cases where audiences offer a speaker assistance, through their active contributions, in establishing that authority. 23

  Presupposition and Propaganda: A Socially Extended Analysis 

289

5 The Social Accommodation of Subordinating Authority Part of my motivation for taking a more expansive look at the contexts that enable hate propaganda to subordinate lies in the observation that the rise to prominence of “successful” propagandists often takes time.24 Richard Spencer—or whoever—did not attain the status he did through a single speech act, but rather through a prolonged process involving countless utterances with many different audiences. In other words, even when this process feels sudden, it is without a doubt still much longer than a single conversation, and so is different than how the hike-planner or the restaurant-chooser attains authority in standard examples of accommodation.25 In this way it is also different than the situational authority of the hate speaker on the streets, and the stability of their authoritative status is what we should endeavor to explain.26 To capture this, I believe we must appeal to the notion of the shifting boundaries of permissibility as described by Tirrell (2012, 2017) and Jennifer Saul (2017, 2018).27 Like Langton, Saul is interested in how accommodation enables hearers to shift their understanding of what counts as acceptable, but Saul also notes how this doesn’t always occur. Instead, in some cases, openly racist utterances may be perceived to violate a general Norm of Racial Equality (NRE), and so aren’t smoothly taken on by hearers.28 In some social settings “many explicitly racist utterances will not normally be smoothly assimilated” (Saul 2017, 109)  One illustrative example is found in Jamie Bartlett’s The Dark Net, where the sudden rise of white nationalist groups in the UK is described to take place over the course of months and years, moving from the online to the streets and back again. See Bartlett (2014, esp. Ch. 2). For other accounts of real-life events of the rise (and fall) of prominent speakers in hate movements, see Strum (1999), Hategan (2004), and Saslow (2018). 25  See Maitra (2012), Thomason (1990), Ayala and Vasilyeva (2016), Witek (2013), Langton (2017, 2018a, b). 26  Matthew Shields (2021) explores a distinct but related phenomena that he calls “conceptual domination,” and there too the process is prolonged. 27  For example, as Tirrell (2017, 144) notes, “our speech acts also undertake a meta-level expressive commitment about the very saying of what is said. Expressive commitments are commitments to the viability and value of particular ways of talking, modes of discourse.” 28  See Saul (2017, 99–101) for a discussion of the Norm of Racial Equality (NRE), which she takes from the work of Tali Mendelberg. 24

290 

M. R. Barnes

perhaps because a hearer will interject and block it, or perhaps because the existing anti-racist norms in effect are strong enough to disrupt the process. If it’s the latter, silence will not represent assimilation—it may instead signify the mental distancing a hearer undergoes when confronted with an explicitly racist comment they reject and perhaps did not expect—as many contextual factors beyond mere silence are relevant to the question of whether a hearer takes on board noxious content. And if this were to occur, the presupposed racist norm would not be accommodated. Saul explores how this resistance can be bypassed through the use of covertly racist utterances—including dogwhistles—that do not so obviously violate the NRE. She also draws on accommodation as it is described by Langton and others, but makes the critical addition that these shifts in shared assumptions can be manipulated and effectively forced. This can occur through what Saul calls a “figleaf.” As she explains, a “racial figleaf is an utterance made in addition to one that would otherwise be seen as racist [that] provides cover for what would otherwise have too much potential to be labeled as racist” (Saul 2017, 103). A synchronic figleaf occurs at the same time as the original, problematic utterance. A diachronic figleaf is “one applied substantially later than the problematic utterance,” which can reconfigure a speaker’s earlier explicitly racist remark as seemingly unproblematic (Saul 2017, 105–106). Together, these are, in her view, tools that enable a speaker and their allies to provide cover for otherwise obviously racist statements, rendering them socially acceptable. Figleaves function to undermine standard inferences about the problematic interpretations of an utterance, and, in doing so, force the accommodation of new norms on an otherwise resistant audience. An effective figleaf allows explicitly racist utterances to be accommodated by a potentially unwilling audience. In the case of diachronic figleaves, this occurs “substantially later,” and furthermore, can be performed by someone other than the original speaker.29 It is cases like these that serve as a template for the type of social accommodation of hate speakers I’m interested in. Many speakers, in addition to the initial  Saul (2017, 110) also notes how “what becomes permissible within one community will not be permissible in another.” 29

  Presupposition and Propaganda: A Socially Extended Analysis 

291

speaker, help to solidify the normative shift embodied and presupposed by the original speech act, and in so doing extend its reach onto a larger (perhaps less sympathetic) audience. The sort of audience-contribution at issue here, then, is an active one, and demonstrates how some hearers of hateful utterances can play an important role beyond accommodation. They do so actively, not passively, with action rather than omission, and with intent, rather than unconsciously. In this way, this discursively forced accommodation is no “quiet engine of the speaker’s success,” but, like the group “Hail Trump” chant, is instead an often-loud assist. Acts of repetition, endorsement, and defense from criticism play important roles, then, in establishing a hate speaker’s authority in a way that is, in some cases, much more secure and stable than the situational authority of the bigoted bus rider. Despite stretching—perhaps too far—the conception of accommodation here, there are important senses in which this extended model is more in line with Sbisà’s (1999) approach to ideology and the normative aspects of presupposition. For starters, she points to “the reality of social communication processes, in which more subtle distinctions are involved” (Sbisà 1999, 492). And as she points out, real-world speech acts have complex audiences, and it may be hopeless to carve out informative uses of presupposition from non-informative uses. This leads Sbisà to distinguishing (at least) five different situations concerning whether the speaker and the hearer share the presupposed assumption and whether the speaker knows this or (incorrectly) believes this to be so. And, as she notes, this is further complicated by the fact that audiences are likely to be split, “partly sharing and partly not sharing” these assumptions (Sbisà 1999, 499). This appeal to many audiences is useful to remember in the context of hate propaganda, as well, and it inspires my reflections on cases where a hate speaker’s presupposition of authority may be passively accommodated by some, and actively endorsed by others. And, as Sbisà adds, a likely scenario is that all these possibilities will hold, maybe at the same time. As Sbisà (1999, 503) puts it, presupposition may be thought of as “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.” It is the notion of “constructing the

292 

M. R. Barnes

participants’ takes” that interests me here, especially when combined with the recognition of a heterogeneous audience. The combination of these two ideas suggests that one function of the ideological use of presupposition is to cultivate and construct an audience, or ingroup. And one way it can do so is by appealing to the shared assumptions that (should) bind that ingroup. Regarding a newspaper columnist who engages in the persuasive use of presupposition, she writes: 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. (Sbisà 1999, 497–498)

As this makes clear, while readers who don’t already share the presupposed assumption are the “main target” in one sense, other readers who already do are having their perspective bolstered, and this is itself an important function. Lots of propagandistic hate speech is similar, in that it flatters the enthusiastic supporter while denigrating outsiders. That both acts occur at the same time is what I aim to get across here. And my point is we shouldn’t treat these as wholly separate phenomena—in some cases hearers enthusiastically enshrine authority through their acts, in others, hearers passively permit authority through their omissions—and what I want to suggest is that they are related much more strongly. That is, the power of omission to encode authority partly comes from the fact that, in other cases and perhaps at the same time, the same norms (if not the exact same speaker) are being endorsed more actively, and this matters. As I argued in earlier work (Barnes 2016), the hate speaker in Maitra’s subway example is not starting from scratch, but is crucially operating within the domain of anti-Arab xenophobia. The robustness of this pre-­ existing domain is what allows a random hate speaker to take up the role of an authoritative speaker so quickly. And here I’m adding that part of what sustains these domains of hate (anti-Muslim hate, among many other forms of bigotry) is the “grassroots” support provided by hearerscum-speakers. It is by noting the heterogeneous responses of multiple

  Presupposition and Propaganda: A Socially Extended Analysis 

293

audiences that we open our eyes to this side of a hate speaker’s authority, and what they are doing with their words. Langton’s cases focus on the accommodation of a particular speaker’s presuppositions, and in particular their assumption of authoritative status relative to a given domain. But, in order to understand the power of propagandistic hate speech, we must acknowledge that much of its force pre-exists this particular speaker, and this particular utterance. This, again, is what is included in the “total speech situation” of these utterances, where racist and other bigoted values lurk in nearly all social situations, lying in wait of a speaker to make them salient. Whenever hate propaganda occurs, there are some who share its aims and others who, to differing degrees, oppose it in belief or action. Greater attention to this fact raises several worries, but also opens necessary lines of inquiry. It is in recognition of the general complexity posed by differing audience responses that Sbisà asks: “Can uptake be multiple (different participants may take an utterance in different ways) and what happens to be “done” in such cases?” (Sbisà 2009, 49). Her response points partly toward the approach I’m exploring here. She notes that this “depends on the kinds of illocutionary acts and on the structure of the social context” and also whether the speaker is most interested in securing the addressee’s uptake, the uptake both of the addressee and of some bystanders, or, most suggestively, “the uptake of ratified bystanders [who] may supplant the failure to achieve the addressee’s own uptake” (Sbisà 2009, 50). As I interpret this latter remark, in some contexts, the uptake of a subset of a speaker’s audience can do the work of establishing what act occurred through their uptake, and this has effects on the wider audience as well. And while Sbisà’s focus is on things like newspaper headlines and politician’s speeches (and how these are reported in the press), her attention to the “real world” uses of such constructions, particularly how they are used for persuasive means, and her recognition of multiple audiences are the lessons I am taking here. Moreover, the approach to presuppositions as “assumptions that ought to be shared” is importantly normative, and suggests that their function is not (only) about informing audiences of new information, but about constructing ingroups who accept that presupposition—whether it is content or the authority of the speaker—as this sets the stage for this “more agreeable” audience to play a role in

294 

M. R. Barnes

securing the speaker’s act as, say, an authoritative directive. All this draws attention to the means through which speakers and hearers sometimes forcefully articulate the assumptions that ought to be shared. Saul’s depiction of figleaves is one such instance where this “favored” interpretation is not left up to chance, and, in the case of third-party diachronic figleaves, the speaker’s allies make the role of the ingroup vivid. In closing, I want to note that one benefit of examining this socially extended sort of accommodation is that it allows us to see how speech acts in addition to those from the original source are also backed up by a similar type of authority. It’s a somewhat odd feature of the standard model to maintain that only the original utterance is (authoritatively) subordinating, while further utterances such as those occurring when an audience chants along to a racist mantra, or when others later repost similar messages online, are somehow not subordinating in a similar way. The force of these speech acts is left unexplained by Langton, as her account seems to focus on establishing single speakers as authoritative, rather than the mass of followers who may join in with acts of repetition. Because of the reliance of this type of subordinating speech on this part of its audience, whom we may see as collaborators and promoters, I view these speech acts as backed up by a type of collective subordinating authority. This sort of subordinating speech needs further examination, and my sketch of its mechanisms leaves many questions unanswered.30 My aim here, however, has been to show how the authority of subordinating speech is not always located in individual speakers, but often in their participation in harmful group practices. Sometimes, these speech acts occur in the context where a group of speakers take their lead from an individual (e.g., the Spencer-led Sieg Heil discussed above) and in other cases the operation is much more diffuse (e.g., Gamergate, though originating in a single-­authored manifesto, became a quasi-grassroots harassment campaign as it went on).

 I explain and defend this notion of collective subordinating authority more fully in other work (Barnes 2023). 30

  Presupposition and Propaganda: A Socially Extended Analysis 

295

6 Conclusion Writing after the Second World War, Jacques Ellul (1973, 118) argued against “a common view of propaganda.” On this view, propaganda “is the work of a few evil men, seducers of the people, cheats and authoritarian rulers who want to dominate a population.” According to this view, Ellul went on, “the public is just an object, a passive crowd that one can manipulate, influence, and use. […] In other words, this view distinguishes between an active factor–the propagandist—and a passive factor—the crowd, the mass, man.” “This view,” he added, “seems to me completely wrong.” Instead, he suggested that “the propagandist and the propagandee make propaganda together” (Bernstein 2021). In cases of informal authority speakers rely on their audience to recognize their authority and respond with the appropriate uptake to achieve this status. As Tirrell (2018, 17) puts it, “authority results in uptake, and uptake further entrenches authority.” There is therefore a reciprocal relationship between uptake and authority. And sometimes, perhaps often, hearers respond with speech of their own, and this is one type of uptake that can extend the reach of subordinating speech in troubling ways, and in some cases may lend some stability to the informal authority of hate speakers. Noting the significance of an active audience avoids the pitfall of reifying single speakers as the monstrous center of harmful speech. In this chapter, I argued there’s a way in which hate speakers gain authority that has so far remained on the periphery of the existing literature on this topic. While prevailing models tend to depict authority as primarily a feature of individual speakers, reality is more complex. Spokespersons for noxious views do indeed rise up, but their rise is supported by an army of collaborators. Thus, instead of a model that locates subordinating authority in particularly noteworthy individuals, I opt for a bottom-up approach that recognizes the power of the crowd. In these cases, the subordinating authority speakers draw on and make use of is not reducible to an individual position. Instead, these speech acts draw power and mutual support from taking place within a group practice. I believe this conception helpfully opens up new avenues for discussing real cases of subordinating speech.

296 

M. R. Barnes

References Adams, Nate P. 2020. Authority, Illocutionary Accommodation, and Social Accommodation. Australasian Journal of Philosophy 98 (3): 560–573. Austin, John L. 1975. In How to Do Things with Words, ed. James O. Urmson and Marina Sbisà, 2nd ed. Cambridge, MA: Harvard University Press. Ayala, Saray, and Nadya Vasilyeva. 2016. Responsibility for Silence. Journal of Social Philosophy 47 (3): 256–272. Barnes, Michael R. 2016. Speaking with (Subordinating) Authority. Social Theory and Practice 42 (2): 340–257. ———. 2023. Who Do You Speak For? And How? Online Abuse as Collective Subordinating Speech Acts. Journal of Ethics and Social Philosophy 25 (2). Bartlett, Jamie. 2014. The Dark Web: Inside the Digital Underworld. New York: Melville House. Bernstein, Joseph. 2021. Bad News. Harper’s Magazine. September. https:// harpers.org/archive/2021/09/bad-­news-­selling-­the-­story-­of-­disinformation. Accessed May 19 2022. Bianchi, Claudia. 2019. Asymmetrical Conversations: Acts of Subordination and the Authority Problem. Grazer Philosophische Studien 96: 401–418. Bourdieu, Pierre. 1991. Language and Symbolic Power. Trans. Gino Raymond and Matthew Adamson. Edited and introduced by John B. Thompson. Cambridge, MA: Harvard University Press. Brown, Alexander. 2019. The Meaning of Silence in Cyberspace: The Authority Problem and Online Hate Speech. In Free Speech in the Digital Age, ed. Susan J. Brison and Katharine Gelber, 207–223. Oxford: Oxford University Press. Ellul, Jacques. 1973. Propaganda: The Formation of Men’s Attitudes. Trans. Konrad Kellen and Jean Lerner. New York: Vintage Books. Hategan, Elisa. 2004. Race Traitor: The True Story of Canadian Intelligence Service’s Greatest Cover-Up. Toronto: Incognito Press. Khoo, Justin. 2017. Code Words in Political Discourse. Philosophical Topics 45 (2): 33–64. Lance, Mark, and Quill Kukla (writing as Rebecca Kukla). 2013. Leave the Gun; Take the Cannoli! The Pragmatic Topography of Second-Person Calls. Ethics 123 (3): 456–478. Langton, Rae. 1993. Speech Acts and Unspeakable Acts. Philosophy and Public Affairs 22 (4): 293–330. ———. 2017. Is Pornography like the Law? In Beyond Speech: Pornography and Analytic Feminist Philosophy, ed. Mari Mikkola, 23–38. New York: Oxford University Press.

  Presupposition and Propaganda: A Socially Extended Analysis 

297

———. 2018a. The Authority of Hate Speech. In Oxford Studies in Philosophy of Law, ed. John Gardner, Leslie Green, and Brian Leiter, vol. 3, 123–152. New York: Oxford University Press. ———. 2018b. Blocking as Counter Speech. In New Works on Speech Acts, ed. Daniel W. Harris, Daniel Fogal, and Matt Moss, 144–164. Oxford: Oxford University Press. Leslie, Sarah-Jane. 2017. The Original Sin of Cognition: Fear Prejudice, and Generalization. Journal of Philosophy 114 (8): 393–421. Lewis, David. 1979. Scorekeeping in a Language Game. Journal of Philosophical Logic 8 (1): 339–359. Maitra, Ishani. 2012. Subordinating Speech. In Speech & Harm: Controversies Over Free Speech, ed. Ishani Maitra and Mary Kate McGowan, 94–120. New York: Oxford University Press. McGowan, Mary Kate. 2012. On “Whites Only” Signs and Racist Hate Speech: Verbal Acts of Racial Discrimination. In Speech & Harm: Controversies Over Free Speech, ed. Ishani Maitra and Mary Kate McGowan, 121–147. New York: Oxford University Press. Nguyen, C. Thi. 2020. Echo Chambers and Epistemic Bubbles. Episteme 17 (2): 141–161. Raz, Joseph. 2009. The Authority of Law. Essays on Law and Morality. 2nd ed. New York: Oxford University Press. Rini, Regina. 2021. The Ethics of Microaggression. New York: Routledge. Saslow, Eli. 2018. Rising Out of Hatred: The Awakening of a Former White Nationalist. New York: Penguin Random House. Saul, Jennifer M. 2017. Racial Figleaves, the Shifting Boundaries of the Permissible, and the Rise of Trump. Philosophical Topics 45 (2): 91–116. ———. 2018. Dogwhistles, Political Manipulation, and Philosophy of Language. In New Work on Speech Acts, ed. Daniel Harris, Daniel Fogal, and Matt Moss, 360–383. New York: Oxford University Press. Sbisà, Marina. 1999. Ideology and the Persuasive Use of Presupposition. In Language and Ideology Selected Papers from the 6th International Pragmatics Conference, volume 1, edited by Jef Verschueren, 492–509, Antwerp: International Pragmatics Association. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2002. Speech Acts in Context. Language and Communication 22: 421–436. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press.

298 

M. R. Barnes

———. 2007. How to Read Austin. Pragmatics 17 (3): 461–473. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ——— 2009. Uptake and Conventionality in Illocution. Lodz Papers in Pragmatics 5 (1): 33–52. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. Shields, Matthew. 2021. Conceptual Domination. Synthese 199: 15043–15067. Stalnaker, Robert. 1973. Presuppositions. The Journal of Philosophical Logic 2: 447–457. ———. 1974. Pragmatic Presuppositions. In Semantics and Philosophy, ed. Milton Munitz and Peter Unger, 197–214. New York: New York University Press. ———. 2002. Common Ground. Linguistics and Philosophy 25: 701–721. Stanley, Jason. 2015. How Propaganda Works. Princeton: Princeton University Press. Strum, Phillipa. 1999. When the Nazis Came to Skokie: Freedom for Speech We Hate. Lawrence: University Press of Kansas. Tirrell, Lynne. 2012. Genocidal Language Games. In Speech and Harm: Controversies over Free Speech, ed. Ishani Maitra and Mary Kate McGowan, 174–221. Oxford: Oxford University Press. ———. 2017. Toxic Speech: Toward an Epidemiology of Discursive Harm. Philosophical Topics 45 (2): 139–161. ———. 2018. Authority and Gender: Flipping the F-Switch. Feminist Philosophy Quarterly 4 (3): Article 1. Thomason, Richmond H. 1990. Accommodation, Meaning, and Implicature: Interdisciplinary Foundations for Pragmatics. In Intentions in Communication, ed. Philip R. Cohen, Jerry Morgan, and Marta E. Pollack, 325–364. Cambridge, MA: MIT Press. Waldron, Jeremy. 2012. The Harm in Hate Speech. Cambridge, Ma: Harvard University Press. Witek, Maciej. 2013. How to Establish Authority with Words: Imperative Utterances and Presupposition Accommodation. In Logic, Methodology and Philosophy of Science at Warsaw University, ed. A. Brożek, 145–157. Warsaw: Warsaw University Press. Wood, Graeme. 2017. His Kampf. The Atlantic. https://www.theatlantic.com/ magazine/archive/2017/06/his-­kampf/524505/. Accessed 19 May 2022. Young, Iris Marion. 2011. Justice and The Politics of Difference. Princeton: Princeton University Press.

Replies to Contributors Marina Sbisà

I am extremely grateful to all the contributors to this volume, its editors, and Constantine Sandis, who suggested to plan it, for their attention to my work and my preferred themes. It is simply wonderful to receive such feedback and be offered such a great opportunity for dialogue. To give this dialogue the room it is worth, I should reply to each contribution with a whole chapter. To keep replies within the size planned for this volume, I shall limit myself to a few topics per contribution. Particularly, I shall select a few points of agreement and disagreement between each contributor and me, irrespective of how many other things there would be to say.

M. Sbisà (*) Department of Humanities, University of Trieste, Trieste, Italy e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. Caponetto, P. Labinaz (eds.), Sbisà on Speech as Action, Philosophers in Depth, https://doi.org/10.1007/978-3-031-22528-4_13

299

300 

M. Sbisà

1 Reply to Green Mitch Green discusses a philosophically significant aspect of my speech-­ act  theoretical perspective: my conviction that hosting propositions in the conceptual toolkit of speech act theory leads to viewing speech acts as consisting in the communication of attitudes concerning propositions rather than in the performance of actions affecting normative (or, in my idiom, deontic) aspects of social relations. While Green agrees with me that some attitudinal approaches are not suitable to account for the normative effects of speech acts, he implies that others might be. I will not tackle this problem directly here, but will offer a clarification concerning the relation between an act, its effect, and the consequences of the latter, which is perhaps marginal with respect to the current issue, but I consider somewhat preliminary to discussing it. In my view, an act (meant as the core move of an action as opposed to a unit of behavior within an activity) is the bringing about of an effect, and is identified and defined by reference to that effect irrespective of its consequences (which might in turn be construed as effects of other acts). So, we have to choose which effect to focus upon. The defining effect of a speech act (qua illocutionary act) is either an effect concerning the participants’ attitudes, or a change in the “normative landscape” (that is, in my terms, in the deontic properties of the participants). A speech act scholar might be willing to admit of changes in deontic properties as consequences of the illocutionary act’s effect on the participants’ propositional attitudes, but in this case, the illocutionary act would still be defined by the latter. Likewise, if one takes it, as I do, that the effect that identifies and defines an illocutionary act is its effect on the deontic properties of the participants, changes in propositional attitudes would not play any definitional role. This does not mean that changes in propositional attitudes play no role whatsoever in the overall dynamics of speech act performance. But in studying such changes, I claim, we are not focusing on the defining effect of the illocutionary act. Green’s main objection to my polemical attitude toward propositions is that they have various jobs to do in speech act theory. There would be aspects of speech acts or kinds of speech act we cannot track adequately

  Replies to Contributors 

301

without resorting to propositions. It is usually said, for example, that when we assert, we commit ourselves to a proposition (or to its truth) and that when we issue a command, we intend to get the addressee to do something—that is, to act on the world so as to make a certain proposition true. (It should be pointed out that Green sticks to Frege in holding that imperatives do not express propositions: but I will not tackle here the further problem of whether speech acts performed using declarative sentences and imperative sentences, respectively, deserve the same treatment.) I think that we can avoid referring to propositions in describing illocutionary effects. Consider my commitment that p as a commitment to speak and behave in ways compatible with the assertion I just made and (perhaps) to defend my assertion. Consider my command to do a as assigning the addressee an obligation to do a. Illocutionary effects impact on deontic properties, not propositions. However, Green’s objection may reappear in another form. We may as well avoid propositions, but there would still be no way to specify the commitment or the obligation (or the entitlement, right, license, authority) that are assigned to the relevant participants, apart from spelling out the contents to which they apply. This has been remarked by Oswald Ducrot in an essay (1993) in which, trying to dispense with the notion of content, he criticizes the notion of modality in that it is to be necessarily completed with that of content. Ducrot’s essay made me think a lot, but I have not found any plausible way to describe illocutionary effects in deontic terms without specifying their contents as well. So after all, I accept content, even if I do not think that “propositions” are objects that exist or, at any rate, are actually relevant to speech act theory. I conceive of the content of an illocutionary effect (which is, by the way, determined by the locutionary aspects of the utterance and by the context) as a type of  situation in the sense in which this phrase is used by Austin (1950, 1953, see Sbisà in preparation). A connected theme in Green’s chapter is the nature of illocutionary force indicators (ifids). Here, I agree with Green that ifids may be contentful and that the force/content distinction should therefore be revised (although not rejected). But I disagree with him on the connection he seems to make between the defeasibility of ifids and their being contentful. In my view, all ifids are defeasible, in principle at least, whether they are contentful or not. Moreover, some classic ifids such as verbal mood

302 

M. Sbisà

leave it largely underdetermined what illocutionary act the speaker is performing: were they indefeasible, they would still give an incomplete indication. The most indefeasible among ifids—the performative formula “I verb that/to …”—is at the same time contentful. Green calls it a “verbal signal” (I take this to mean that those words merely present the speaker as performing an act of the kind mentioned and are not used to perform an illocutionary act of their own). Finally, sometimes it is precisely the content of a speech act that makes the audience understand what procedure the speaker is invoking. You distinguish an answer from an objection in virtue of its content (more precisely, of the relation between its content and the content of some previous question or assertion). Likewise, it can be contended (as I suggest in my 2021a, as an alternative to the received Indirect Speech Act analysis) that when you say “I was delayed by the traffic,” you are apologizing for being late just because of the type of situation your utterance represents (its “content”).

2 Reply to Ambroise After examining my way of dealing with illocutionary acts and highlighting its possible convergence with that of Richard Moran, Bruno Ambroise concludes his chapter by crediting me, among others, with the “reinscription” of speech acts in the social world. This is indeed something I have been striving for all along my philosophical career. However, I disagree with something that Ambroise appears to be presupposing, that is, that speech acts derive their effectiveness from the social world. I think that he is downplaying the potential for social action that speech has. Nothing new would ever happen in the human world if the effectiveness of speech had to be passively derived from what is already given in a given society. But people constantly act in speaking and in responding to the speech of others, and do so in unforeseen manners. The openness to change (and progress) may be minimal, but it is not for philosophy to decree there is none. In the light of such considerations, I would like to reply to an objection that Ambroise raises against the idea, stated in my (2002a), that the context of a speech act (conceived as the set of facts against which a

  Replies to Contributors 

303

speech act is to be assessed) is ultimately set up by the participants in the speech situation. In my exposition, this idea is connected with the fact that felicity conditions are taken to be fulfilled by default, but, Ambroise argues, it cannot follow from such a premise for two reasons. First, if felicity conditions are not in fact fulfilled, the speech act cannot be performed. Second, the recognition of the “symbolic power” to perform a certain speech act to a certain speaker is itself socially conditioned and, therefore, preexists the occasion in which the speaker attempts to perform it. In response to the first reason that Ambroise provides, I should explain that the connection I see between the by-default functioning of felicity conditions and the constructed nature of the context is that the former makes the latter possible by allowing for the seamless accommodation, not merely of the assumption that the speaker possesses certain deontic properties, but of those deontic properties themselves. Conditions that consist in the distribution of deontic properties can indeed be fulfilled by accommodation, since their fulfillment depends on the de facto agreement that they hold, while this is not the case with conditions that consist in trivial facts such as the door being already open or the addressee actually having a preference the speaker is not aware of. However, since the relevance of such trivial facts to the ongoing speech activity is dependent on what the participants are up to, what aspects of the environment come into the speech act’s context also depends upon the participants. In response to the second of Ambroise’s reasons, I should like to grant him that the recognition of deontic properties to a speaker is socially conditioned. But that recognition is itself the source of the speaker’s “symbolic power” as it manifests itself in the speech situation. Without that recognition, even formally established authority might lose its grip on those who are expected to act as its subordinates. So, each participant is always partly free to accept or reject the distribution of “symbolic power,” or in my terms, deontic properties, apparently pre-existing with respect to the speech act. Were it not so, rebellion would be impossible. But rebellions happen, and sometimes even revolutions. That is to say, substantial change in the received social hierarchies is possible. Reflection on the social implications of speech act theorizing could be pursued further. An interesting contrast emerges from Ambroise’s

304 

M. Sbisà

comments on Moran, where he distinguishes the “social dimension” of speech acts from “mere intersubjectivity.” He is perhaps referring to a view of intersubjectivity that derives it from subjectivity and is opposed to a view of the social dimension in which the social takes priority over the individual (and therefore over what is subjective). But there might also be a view of intersubjectivity in which the intersubjective relationship is primary with respect to subjectivity. We live in a world we share with other subjects and we already live in it before being self-conscious or being recognized by others as subjects. One might claim that the classic conception of the Subject is a construction from manifold recognitions of subjectivity taking place in the primary intersubjective space. Taking intersubjectivity as primary, though, is not the same as believing in the autonomous existence of social structures; rather, it is the key to understand their ultimate dependency on the interaction among individuals. Ambroise also notices that—even though I accept that there are objective facts about language, speech acts, and social interaction—I avoid focusing on the role of judges (as third parties with respect to a dyad of interactants). Indeed, taking intersubjectivity to be primary commits me to assigning dyadic relationships the most basic character with respect to third-party mediated ones.

3 Reply to Bianchi Claudia Bianchi makes an original and challenging use of some of my ideas. I should like to touch upon three themes in her contribution: the first is political, the second pragmatic, and the third more generally philosophical. Bianchi maintains that claims about the role of uptake in illocution have political implications. In principle, I share this conviction. However, she argues that powerless speakers should be protected from perverse hearers and that a theory of illocution that makes success in illocutionary performance dependent on the hearer’s uptake hinders the illocutionary potential of members of underprivileged groups, thus undermining their agency. This, though, may be the case only when members of

  Replies to Contributors 

305

underprivileged groups play the role of speakers. When they are audience, a theory focusing on the speaker’s role leaves them little room for dissent and rebellion. When in the 1970s I started thinking about the role of the hearer’s uptake, there were diffuse political worries about consumerism and patriarchy and the ways they express themselves in public and private discourse. There was hope that a rebel audience might undermine consent to such discourse and, therefore, to the powers underlying it. Theories of illocution making the illocutionary act performed depend on the sole intentions of the speaker appeared to me not only wrong (I had, and have, other reasons against intentionalism), but also solipsistic and related to phallologocentrism. Be it as it may, I am still convinced that not only hearers, but also speakers can be perverse. Defense from unfair and unjust impositions in conversation is a political problem that the philosophy of language should take into due consideration: but there must be some way to support the powerless (or less powerful) in both their speaker and hearer roles. My second remark concerns the question of whose uptake matters (or whose uptake should be secured by the speaker). This is a question, discussed by Bianchi (and, in this volume, by Johnson, too), to which more attention should be paid in pragmatic research. We often speak as if there were one speaker and one hearer who is also the addressee and the main target of the illocutionary effect, but verbal interaction is usually more complicated. Participation roles are several and include bystanders and eavesdroppers; moreover, the addressee is often not an individual but an audience without clear boundaries, and there is also a distinction to be made between being an addressee of the utterance and being the target of the illocutionary act. Who is it, then, that is expected, in each speech situation and for each illocutionary act, to understand the meaning and the force of the speaker’s utterance, or (as Bianchi would put it) that should be led by the speaker to understand its meaning and force? I think that the reply may vary according to the kind of illocutionary act and speech situation. Bianchi is inclined to privilege the role of some neutral third party. I agree that neutral third parties are sometimes useful and even indispensable, but as I said above in my reply to Ambroise, I am wary of taking them as a basic element in the dynamics of illocution. Or

306 

M. Sbisà

perhaps, whenever such third parties come in as actual participants, they become part of the web of dyadic relationships on which the dynamic of illocution is based. Indeed, when a judge in a rape trial has to pass sentence, he or she is not merely a third part, neutral between rapist and victim: his/her sentence reflects his/her relationship to each of the protagonists of the episode under scrutiny. Formal thirdness is no guarantee of fairness: the judge’s relationship to the victim, for example, might be biased by assumptions about the way in which women should be dressed. Finally, I find the normative perspective put forward by Bianchi ambiguous in at least two respects. First, it is not clear what exactly Bianchi holds to be “normative”: on the one hand, there are illocutionary effects (consisting in changes to the deontic properties of the participants) and on the other, there is the normativity of language (including the normativity of linguistically indicated presuppositions) and of the illocutionary procedures in which language is involved. These two aspects of the normativity of speech as action, albeit connected, should be kept distinct not only for the sake of clarity, but also in order to account for their connection. Second, I wonder whether Bianchi’s normative perspective is still a form of intentionalism, albeit weak. On the one hand, Bianchi seems to make intentionalist assumptions, such as, for example, that speakers must have their own “original intentions,” but on the other hand, she contends that illocution involves “social uptake,” that is, a ratification of the changes in the normative landscape that is delivered by the relevant participants, who may but also may not include the speaker. So, when she writes that in her normative perspective, uptake is the way in which the utterance should be taken “considering the procedure invoked by the speaker or the speaker’s intentions, when adequately made manifest and public” (p. 76), what is the value of “or”? Are the disjuncts two substantive alternatives, possibly exclusive, or two alternative formulations of one thing—since invoking a procedure is doing something manifest and public that entitles or even obliges the audience to maintain that the speaker has a certain intention? If the correct reading is the latter, I agree. But I would not call that “intentionalism” at all.

  Replies to Contributors 

307

4 Reply to Witek Maciej Witek introduces innovations inspired (among others) by my work into mainstream philosophical pragmatics. In agreement with me, he admits of normative (or deontic) effects of illocutionary acts, recognizes that the audience’s uptake and response may play roles in the dynamics of illocution, and, in his contribution to this volume, examines conversational sequences according to a tripartite model Initiation-­ Response-­Evaluation, preferring it to the bipartite model offered by adjacency pairs which used to be the basic unit in conversation analysis. I greatly appreciate his attempt to use Millikan’s conception of convention to throw light on the controversial “conventionality” of illocution. However, I disagree in some important respects with the results of his research. He seems to conflate some distinctions that I regard as essential. First, he seems to use both the notion of speaker presupposition (pragmatic presupposition, in the sense of Stalnaker 1999) and that of utterance presupposition (presupposition as a precondition of speech act felicity or appropriateness). Of course, both these notions have a role to play in pragmatic theory as well as in the pragmatically oriented analysis of texts and conversations. But their roles should be carefully distinguished. Speaker presupposition is something that theoretical models of conversation must take into consideration, and which explains, for example, certain omissions or certain misunderstandings, but does not account for its own accommodation. It is therefore hardly useful when what is at issue is the analysis of an utterance or text searching for the implicit meaning they are imposing on the audience. Utterance presupposition can do the latter job, while, of course, there is no guarantee that what is presupposed by an utterance or text is actually part of the common ground of the participants. So, when Witek speaks of uptake as a presupposition of the hearer’s response, it is not clear whether he means that it is an assumption in the hearer’s mind (that the hearer possibly believes to be shared by other participants), which would assimilate uptake to a speaker presupposition, or an assumption that should hold in order for the response to be actually appropriate and would therefore be ceteris

308 

M. Sbisà

paribus accommodated. My choice would be the latter, in accordance with my research on presupposition accommodation (1999, 2007, 2021b). Second, Witek seems to feel free to conflate illocution with perlocution or even aetiolation (so-called non-serious speech). While these three phenomena all belong to the speech act (since they are aspects of speech as action), I maintain that they should be kept distinct in theory and analyzed separately in practice (without forgetting about their connections). As to illocution and perlocution, the weak conception of convention as pattern reproduction of Millikan (2005) fosters confusion between an utterance inviting a response, which depends on its illocutionary force, and it actually achieving a response (be it the designed one or not), which amounts to the speaker’s performance of a perlocutionary act (as the achievement of a perlocutionary object or the production of a perlocutionary sequel). Still, nothing in Millikan’s conception prevents us from making a distinction between changes in the deontic properties of the participants—among which are legitimate expectations or licenses—and the achievement of goals concerning a receiver’s psychological states or actual behavior. Moreover, in analyzing one of his examples, Witek deals with an ironic rhetorical question, which he says is aimed at “ridiculing” the addressee (pp. 108–109), and seems to put all these speech act phenomena (rhetorical questions, irony, derision) on the same level, while failing to mention the illocutionary act successfully performed by the speaker, that is, criticism. Indeed, it is clear enough that the ironic rhetorical question “Where did you find that, in a fortune cookie?” criticizes the interlocutor’s previous comment as trivial and ungrounded, and seems to be well understood as such by the interlocutor who responds in the same vein, defending himself from the criticism by citing a serious and authoritative source for his criticized comment. In so doing, he (perhaps) also avoids being ridiculed, but criticizing and ridiculing are two acts of different levels and in this case they are performed by means of aetiolated speech. I should also like to recall that there is still some sense in Austin’s remark that only verbs designating illocutionary acts have performative uses. Certainly, when a verb used to report or describe speech acts has no possible performative uses, there might be various reasons: it designates a perlocutionary act, or some kind of aetiolated performance, or (like “subordinate”) describes a consequence of an illocutionary effect,

  Replies to Contributors 

309

which, although being as normative as the illocutionary effect itself, is not the illocutionary effect of the utterance under consideration. Another remark I should make concerns Witek’s use of the initiation-­ response-­evaluation sequence. It is essential to my uses of this tripartite format (corresponding to the three steps of narrative sequences: manipulation, action, and sanction; see my 1989, 2002b) that it be conceived as lending itself to multiple applications even within the same conversation. Indeed, each conversational turn responds to the previous, and sometimes the first turn in a conversation responds to a previous event or action. Likewise, each conversational turn in a conversation expects the following to implicitly or explicitly evaluate it and sometimes the last turn in a sequence is itself evaluated non-verbally. So, if analysts want to highlight the complexities of each conversational turn (for example, its forward-looking and backward-looking aspects), they have to apply the tripartite format in each position in which it is heuristically relevant to do so, making it shift along the conversational sequence. While I agree with Witek that the third element of any initiation-response-evaluation sequence has, alongside other possible evaluative functions, the function of either blocking or licensing the uptake of the first turn that is made manifest by the second, I think that this very fact shows that one application of the tripartite structure is not enough to account for interactional negotiation. Indeed, the speaker making the first and third move in a three-turn sequence is in a position of de facto greater power (or the other way around: the speaker with greater power is more likely to have the first and third move at her disposal). To account for conflict between different perspectives and for its evolution in the direction of some compromise or balance, we need to make multiple uses of the tripartite structure.

5 Reply to Johnson Johnson’s view of illocutionary pluralism is based upon the idea that for each uptake that an utterance receives, there is an illocutionary effect and therefore an illocutionary act. In various works of mine (as Johnson recalls), I too have stressed the decisive role of the interlocutors’ uptake in informal speech situations (I still endorse this claim, while have become

310 

M. Sbisà

more cautious in generalizing from it). However, it is far from clear whether Johnson and I have the same conception of uptake and its role, and therefore, of the dynamics of illocution. Should the hearer’s uptake be considered as a necessary condition of the performance of an illocutionary act, a sufficient condition, or both? Johnson seems to think that it is a necessary condition. What she takes to vary is how many participants and which ones actually provide it. Some participant may not: interestingly, she argues that the speaker too, as a participant, usually “reacts” to her own speech act (which counts as uptake), but need not do so (therefore, we can ascribe illocutions to children who do not yet react to their own illocutionary acts as such). In my framework, uptake may act as a sufficient condition in certain cases, especially when, in informal speech situations, felicity conditions for the kind of illocutionary act at issue are either manifestly satisfied or easily accommodated, while it is a necessary condition that uptake be “secured.” In the case (mentioned by Johnson) of children who have just started to speak, we are especially flexible about what they have to do to “secure” uptake, since by our reactions which credit them with illocutionary performances, we are making the children participate in verbal interaction and learn how to participate in it more and more fully. Moreover, while Johnson includes uptake in her list of conditions for the performance of illocutionary acts, I do not think that uptake (or the “securing” of it) should be considered as a condition of illocutions in the same way as felicity conditions. The latter are act-­ specific, while, whatever “uptake” might consist in, it is a general feature of the dynamics of illocution. The confusion originates, perhaps, from the fact that Searle (1969) in his attempt to reach completeness, includes in the conditions for the performance of the illocutionary act of promising general requirements too (such as the obtaining of “normal” communication conditions). Inspired by Austin, who does not in general aim at exhausting the possible causes of infelicity and voidability, I prefer to proceed with by-default attributions of illocutionary acts, unless there are specific doubts about the satisfaction of some of their conditions. Finally, another possible divergence between my view and Johnson’s is about the role of uptake in intersubjectivity. More than one agent is affected by an illocutionary act: we may well imagine an illocutionary effect as giving shape to a “deontic landscape” (as Johnson calls it). But a deontic

  Replies to Contributors 

311

landscape depends (in part at least) on how participants make it (by agreeing, or converging, on how it is and letting their interaction proceed on that basis). It is therefore not a matter of “perceiving” something independent of us and then being affected by it or reacting to it. Johnson’s analyses of situations in which an utterance may be said to have “multiple” illocutionary forces illustrate scarcely explored complexities of what we do with words. What if—after I uttered words that might be taken as amounting to a commissive illocutionary act—you take me to be obligated, but I do not take myself to be? What if a participant promises to do something which is of interest to one of the interlocutors but not to others? What if the speaker has a reaction to their own speech act which differs from their previous intention? These and other similar questions are worth detailed and contextualized examination, which would also lead to more precise characterizations of the cases in which an utterance may be said to have “multiple illocutionary forces” and perhaps of the nature of illocutionary pluralism (provided it is a single phenomenon). At any rate, faced with the two distinct developments of illocutionary pluralism presented by Johnson—Lewinski’s illocutionary pluralism and her own illocutionary relativism—I have the following comments. First, illocutionary pluralism need not be intentionalist; rather, making force dependent on speaker intention greatly restricts our capability to account for those complexities of illocution that both Johnson and I find significant. Second, illocutionary relativism has some unwelcome, if not paradoxical, implications. One of those implications is that, if conversers who disagree about the force, and therefore, the illocutionary effect of an utterance, can all be right, their interaction appears to be doomed to impasse. In order for it to proceed, there must be a deontic landscape that is in fact accepted. If the interaction goes on without that, participants will be just talking (or acting) past one another until their relationship breaks down. Another implication of illocutionary relativism is the tacit assumption, often associated with relativist stances, that there be co-existence and universal visibility of the different points of view to which things (here, illocutionary forces) are “relative.” I would prefer contextualism, which, at least in one of the ways in which we can conceive of it, does not allow for any non-context-bound perspective, but allows for objectivity within each context: in our case, for one deontic landscape (however complex) from which subsequent interactional behavior proceeds.

312 

M. Sbisà

6 Reply to Cooren As François Cooren recalls, I argued in various occasions for a conception of action focused on its effects, which acknowledges the multiple uses of that notion without restricting it “to the sole intentional acts of human beings” (p. 147). In fact, I left it open what role intention plays in human action and in speech action in particular, with some sympathy (which calls for further development) with Austin’s idea that intention plays no causal role in action but accompanies it as the agent’s awareness of, or at least capability to realize, what they are doing. Be this as it may, I agree with Cooren that there are loose or extended uses of the notion of action, going beyond human intentional action, which retain its core feature, that is, the bringing about of a state of affairs under the responsibility of an agent. One of these uses concerns textual agency, on which Cooren rightly insists. Indeed, we attribute agency to utterances and even to sentences or words by transferring the speaker’s agency to them (likewise, Grice 1969 discusses how utterances and sentences can have meaning by tracing their “meaning” back to speaker meaning). But there is more. Once produced, a text has its autonomy, which plays a role in the dynamics of the speech act: we may say of it that it manifests the speaker’s intentions, secures uptake, mandates some presupposition, suggests some implication, or invites an inference or a response. It is also in virtue of this autonomy that the speaker is sometimes made responsible for some effect they did not foresee. But, while largely agreeing with Cooren on all that, I have an additional, typically philosophical worry, of which I cannot be completely forgetful. Are all agents “subjects” in the sense of centers of perspective? It does not seem so. When they are not, what resources does their agency rely upon? I should like to distinguish between the legitimacy of loose uses of the language of agency (with the consequential recognition of complex, diffuse, and deferred responsibilities) and the persistent need to reflect about what is the role of living human beings— flesh and blood—in all that. If not only non-human living beings, but also texts, signs, artifacts, and even natural phenomena can be recognized as agents, is recognizing human beings as agents exactly the same thing, or there is some kind of recognition which is owed to them in particular?

  Replies to Contributors 

313

Maybe this is not obvious, but I see these questions as connected to the issue of human rights. A more technical issue that Cooren touches upon is the conventionality of illocutionary acts. He disagrees on my idea that illocutionary acts are conventional primarily because of the conventionality of their effects (perhaps, he does not see how this idea fits with the conception of action he appreciates) and sets greater store on the conventionality of the means that enable us to perform illocutionary acts as well as on the conventional nature of language. I should perhaps clarify that I do not deny that there are “conventional means” for performing illocutionary acts and recall that Austin himself claimed that conventional effects cannot be achieved unless there are conventional means to do so. The problem with illocutionary procedures is, rather, in what sense and to what extent they are “conventional.” Indeed, there are sometimes motivations for adopting a certain procedure to achieve a certain (conventional) effect. Promising is about future action of which the speaker is in control, and there is clearly a reasonable connection between being entitled to commit oneself to something and being in control of what one commits oneself to. There is also an obvious connection between the use of some kind of future tense and the fact that the commitment that is the effect of a promise is about future action. Likewise, there are lots of non-arbitrary features in other illocutionary procedures. Are those non-conventional features compatible with the procedure being “conventional” and what does its conventional character consist in? Should the procedure have alternative versions in the same language/culture, or would it be enough for it to have alternative versions (if present at all) in other languages/cultures? To what extent would alternative versions of a procedure still be procedures for performing the “same” illocutionary act? The means for performing a certain illocutionary act should, at any rate, be recognizable as such and this requires some kind of social agreement or acceptance of a procedure as having certain effects. But if we track the conventionality of illocutionary procedures to this basic fact, then, we are connecting it again to the production of conventional effects. As to the conventional nature of language, I have nothing against calling it conventional if by that it is meant that it is rule-governed and that the rules of a given natural language are not the only theoretically

314 

M. Sbisà

available option for the corresponding function. But I would like to recall that Ferdinand de Saussure, the initiator of European structural linguistics, distinguished convention (meant as the coming to an agreement) from arbitrariness, on which he relies in presenting linguistic systems as imposing themselves on speakers (so that, for example, it takes generations of speakers to make them change). I think that it is the social agreement about the rules governing language—which cannot therefore be changed by individual speakers with a fiat even for very good reasons— that constitutes language as arbitrary in the Saussurean sense and therefore as autonomous from human desires and intentions in the way underscored by Cooren. This normativity of language is to be distinguished from the conventional and defeasible character of illocutionary effects as affecting the deontic properties of the participants (as I suggest also in my reply to Bianchi). I should like to highlight two contexts in which I make reference to that normativity: the conception of presupposition as presupposition of an utterance or speech act (or “text”), and the notion of an enunciator or enunciating subject (as distinguished from the speaker or utterer). In defining the presuppositions of a speech act or a text as assumptions that should be accepted by its audience (Sbisà 1999), I rely upon their being rule-governed in an autonomously linguistic way. The recognition of the enunciating subject of a text as an “author” (who is the source of what is expressed) and as a “principal” (who is responsible for the accomplished speech act; see Goffman 1981), and the assignment of attitudes to that subject, are also ultimately made possible (if not mandated) by the linguistic norms governing the uttered words. Still another interesting point raised by Cooren concerns the distinction between illocution and perlocution. He takes the effect of inviting a response, which Austin connected with the illocutionary act, as an effect of the illocutionary act and remarks that it is not conventional. Were he right, the illocutionary act would not be characterized, as I contend, solely by its conventional effects. But an effect of that kind does not seem to be associated to all illocutionary acts and so should not be counted as characterizing the illocutionary act as such. In my view, the effect that makes us identify and define an illocutionary act is the effect it takes conventionally—that is, thanks to the (often tacit) agreement that the speaker, in issuing her utterance, brought it about. As to the inviting of a

  Replies to Contributors 

315

response, however, a further distinction is to be made. The “inviting” itself may be considered as a conventional matter (perhaps a conventional consequence of the main conventional illocutionary effect), insofar as inviting someone to do something is assigning her a deontic predicate such as “should”. The achievement of the invited response is, instead, the achievement of the perlocutionary effect associated with the illocutionary act (its “perlocutionary object” in Austin’s terms). So, after all, the fact that certain illocutionary acts “invite a response” does not show that illocutionary and perlocutionary acts cannot be distinguished by the conventionality and, respectively, non-conventionality of their essential effects. It should be pointed out, though, that dissatisfaction with the conventional/non-­conventional divide leads Cooren to put forward a different way to make the distinction between illocution and perlocution (which, in my opinion, is not incompatible with the received one). He notices that a receiver is in principle free to react to a speaker’s utterance in accordance or in contrast with the speaker’s perlocutionary goals. But we have not the same freedom with respect to an utterance’s illocutionary effects. To put it as I prefer, if a procedure was executed in a recognizable way and without apparent infelicities, we should take it that the associated illocutionary effect was achieved, and if we do not, but others in our social group do, we can be proven to be wrong. On this point, Cooren’s contribution seems to me to converge with Bianchi’s normative perspective.

7 Reply to Marsili Marsili’s contribution takes the hint from a paper in which I distinguish constitutive rules, maxims and objective requirements (Sbisà 2019) and intends to promote further clarification of the field of speech act norms. I agree with him that further clarification is needed and that terminology too should be revised, in order to reflect conceptual distinctions in a more stable way. My conceptual and terminological worries, though, coincide only in part with those expressed by Marsili. Marsili remarks that rules of the kind that Austin dubs Γ2, which, in my (2019), are considered as maxims, are not genuine cooperative rules

316 

M. Sbisà

as Austin’s Γ1 rules seem to be. I think he is right and admit that I kept all Austinian Γ rules together on the basis of the fact that neither Γ1 nor Γ2 rules undermine, when violated, the illocutionary effect of the utterance. But that is a negative characterization and I can see it is insufficient. As to the norms I call “maxims,” Marsili claims that the maxims that are associated with illocutionary act kinds can be given reasonable foundations and, particularly, be derived from the illocutionary acts’ goals. This is a Gricean feature that might prove compatible with my Austinian framework. I should like to recall, in this connection, Grice’s way of dealing with rationality as “argumentative,” that is, characterized by the willingness to give justifications for what one says or does (Sbisà 2006a). Marsili also points out that “constitutive” rules, in the received sense in which I use this phrase in my (2019), and “constitutive” rules or norms in the context of the discussion, initiated by Williamson, on the norm of assertion, are “constitutive” in different senses, and I agree on this. Coming to the points of disagreement between Marsili and me, let me start by reconsidering Γ2 rules. When they are violated, what exactly is “infelicitous”? It has become usual to speak of illocutionary acts as “felicitous” or “infelicitous,” but this use is somewhat a transfer from the original one in which infelicity (and felicity) were meant as properties of (performative) utterances. An illocutionary act is performed in issuing an utterance, but it is not the same thing as that utterance. If a speaker who, in issuing a certain utterance, performed a certain illocutionary act violates the Γ2 rule for that act, that makes it possible to criticize the utterance as infelicitous. But the act itself was successful: it is the speaker’s subsequent behavior that is inconsistent with it (and with the utterance in issuing which the act was performed). Austinian Γ2 rules require the speaker to be consistent with the illocutionary effect of their own utterance and are especially salient when the deontic properties that the illocutionary act assigned to them include commitments or obligations. So, while I admit that Γ2 rules are not “maxims” in the same sense as Γ1 ones, I think that they should be understood as part of the illocutionary effect of the utterance, that is, the distribution of deontic properties among the participants that is the outcome of the procedure invoked. Therefore, they are “norms” only insofar as illocutionary effects are “normative.” Now, illocutionary effects in my sense seem to play no role in Marsili’s

  Replies to Contributors 

317

framework. This, in turn, might explain why he resorts to “illocutionary goals” with associated directions of fit to account for my “objective requirements.” Marsili thinks that a promise meets its objective requirement if it is fulfilled, that is (since illocutionary goals are associated with directions of fit), if its propositional content is (when the time comes) satisfied. But complying (or failing to comply) with the commitment one has undertaken is not the same as having said and therein done the right (or the wrong) thing. Achieving or failing to achieve the addressee’s compliance with a directive is even farther away from having said and therein done the right (or the wrong) thing when uttering an imperative sentence. I admit that this idea of a speech act being subject to an objective assessment as saying and therein doing the right (or the wrong) thing is rather vague and should be further specified. But let me recall where it comes from. Austin (1975, 140–145) claims that “accomplished utterances” of various kinds of illocutionary force, provided that they are illocutionarily felicitous, are subject to an objective assessment “similar” to the assessment of assertions according to so-called correspondence with facts, which has to do with facts “in complicated ways.” He leaves it open whether this kind of assessment extends to utterances of all illocutionary forces, implying that it might. It is clear from what Austin writes that he does not have in mind matters of compliance subsequent to the performance of the illocutionary act, but that performance itself, considered in its relationship to circumstances, audience, and purposes. I do not think that this idea, albeit formulated in a non-rigorous terminology (in the colloquial style characteristic of How to Do Things with Words as consisting of lecture notes), is haphazard and philosophically irrelevant. In my opinion, as I argued in my (2006b), it is an essential part of Austin’s view of speech as action, wholly consistent with his polemical attitude against the dichotomic opposition of value and fact. Assessment of words fades into assessment of actions and the other way around. Marsili’s “illocutionary goals” closely resemble Searle’s “illocutionary points” (1979), which I criticized on various occasions for their heterogeneity ranging from the deontic to the psychological level. I prefer to stick to Austin’s terminology and count Marsili’s “illocutionary goals” as “perlocutionary objects.” The adjective “perlocutionary,” in this case, does not exclude the association between these goals and kinds of illocutionary

318 

M. Sbisà

act, as I remarked above in my reply to Cooren. It means only that the achievement of the goal in question does not depend on the speaker’s performance and some kind of interpersonal or social agreement about it, but on the receiver’s alignment with the goals and attitudes of the speaker and the psychological or behavioral response it yields. In my framework, the illocutionary act is “successful” if it brings about its characteristic conventional effect (the effect I describe in terms of deontic properties). When a perlocutionary object is achieved (which means, in certain cases, that the commitment or obligation introduced by the illocutionary act is “complied with”), we also have the success of a perlocutionary act. All that is far away from my attempt to include norms that fix objective requirements for illocutionary acts—that is, criteria for the objective assessment of what was said and therein done in the light of circumstances, audience, and purposes—among the variety of speech act norms. Marsili might wonder how, in this framework, Williamson’s Knowledge Rule can be an “objective requirement” for assertions. But the utterance of  a declarative sentence felicitously and appropriately asserting something true on the basis of the speaker’s actual knowledge of its subject matter is certainly the right thing to say and therein do in its context.

8 Reply to Goldberg In his thoughtful essay, Sandy Goldberg credits me with a distinction (or even a contrast) between the normative profile of a speech act (in my terms, the distribution of deontic properties that characterizes its illocutionary effect) and the constitutive rules of the illocutionary act that the utterance is designed to perform. First, I should say that in the passage he quotes (from Sbisà 1992), I did not really intend to make this distinction. I was discussing the hypothesis that conversation might have constitutive rules analogous to those of illocutionary acts. Either turn-taking rules or Grice’s maxims as constitutive rules of conversation could play that role. I intended to claim that both analogies are inappropriate and fail to clarify how speech act theory is to deal with conversation. Second, while the distinction can be made and be illuminating up to a point, I do not think that the two ranges of phenomena it sets apart are without connection.

  Replies to Contributors 

319

Instead, I largely agree with Goldberg when he claims (on the basis of his analysis of assertion) that having characterized the illocutionary act in terms of its normative profile, there remains to account for how speakers manage to perform illocutionary acts with that profile, which in his opinion can be done by appeal to constitutive rules. Moreover, it seems to me that, since an illocutionary effect (the activation of a certain normative profile) is made possible by the fact that there are ways in which the illocutionary act bringing it about can be performed, there is a loop between these two aspects of illocution. A normative profile already presupposes the existence of the ways in which it can be activated. But beyond this partial agreement, there are potential sources of disagreement. What does Goldberg mean by “constitutive rules”? While I identify constitutive rules with the outline of the procedure to be executed in order to bring about the illocutionary effect, he identifies them with norms whose content and function (for the speech act that he examines, assertion) come very close to those of Williamson’s Knowledge Rule. But as already stated in Marsili’s contribution and in my reply to it, “constitutive rules” in Searle’s sense (and in Sbisà 2019) and in the sense of Williamson’s Knowledge Rule, are norms of two different kinds. It would be better to make this difference explicit, for example by calling the former “constitutive rules,” and the latter “constitutive norms.” If constitutive rules in the stricter sense are not followed, the illocutionary act is not “pulled off” (to use Goldberg’s words). If constitutive norms are not abided by, the act is performed and ceteris paribus successful, but subject to criticism. Those norms are “constitutive” not because a certain act is performed only if they are followed to a reasonable extent, but because the act is defined as an act of a certain kind by its being subject to them. We perform an act of that kind if and only if what we do is subject to that norm. However, if what Goldberg calls “constitutive rules” are in fact constitutive norms in the sense just defined, it is hard to see how they could establish the way in which the act is to be executed and made recognizable. If I understand Goldberg correctly, in order to make an assertion, a speaker has to utter a sentence displaying certain features in certain circumstances, so as to make it mutually manifest that they have a right to be recognized as making an assertion—that is, as presenting p as true

320 

M. Sbisà

with epistemic authority. But what features should that sentence have, and what should the appropriate circumstances be? This does not seem to be specified by constitutive norms, while it would be a matter of constitutive rules, or, with a phrase I prefer, of the accepted procedure for the performance of the illocutionary act at issue. Perhaps, Goldberg is assuming that constitutive rules (or even procedures) can be somehow derived from constitutive norms, but it is not clear how. Another theme in Goldberg’s contribution is that of the alleged indispensability of the audience’s uptake, which he rejects. He takes Austin as deeming actual uptake to be a necessary condition of the illocutionary act and assumes that I, following Austin, believe that one can think of illocutionary acts in terms of conventional effects (his “normative profiles”) only if one treats actual audience uptake as required for their success. As is clear from his contribution, but also from other contributions such as Claudia Bianchi’s, uptake is a controversial notion and, admittedly, both Austin and I are ambiguous about it. As to Austin, it should be recognized that he considers actual uptake as necessary to achieve the conventional effect when he discusses promises and warnings. On the one hand, he must have in mind formal promises, which, like bets, are not binding unless they are accepted; on the other, he has in mind informal warnings such as “Pay attention!” shouted out to an addressee who is already far enough, which, if not heard, remain mere attempts to warn (Austin 1975, 22 and 116). But informal commissives may be considered as binding by default, and formal warnings have to be posted according to certain legally established rules and if those who pass by do not read them, they nevertheless count as being warned. So, Austin’s stance in favor of the indispensability of actual uptake seems to be influenced by the particular examples he discusses. When he speaks more generally, he takes it that what is indispensable is “the securing of uptake” (which need not amount to achieving actual uptake). As to myself, it is true that I connect the conception of illocutionary acts in terms of conventional effects to the indispensability of “securing uptake,” and that I often insisted on actual uptake—especially, with a polemical function against speaker intention-­ centered conceptions of illocution, or in the analysis of informal conversations in which the speaker may leave up to the audience to decide what is the procedure invoked and what illocutionary effect is actually brought

  Replies to Contributors 

321

about. But I recognize that informal contexts of this kind do not exhaust informal speech situations and are not organized in the same way as more formal contexts. There are certainly many cases, formal and informal, in which the speaker’s behavior and words suffice to constitute the execution of a fairly precise illocutionary procedure (not to consider the cases in which reference to the procedure is specified by a performative verb), so that the speaker has a right to be recognized as invoking it. I still think that some sort of agreement is needed in order to bring about a conventional effect concerning deontic properties (the activation of a normative profile), but this agreement need not consist in “actual uptake”: for example, it might be achieved by default whenever the speaker’s performance is reasonably and univocally recognizable as the execution of a certain procedure. In that case, interaction proceeds on the basis of the achieved illocutionary effect unless problems that can suspend or annul it are raised. It should be pointed out that the same by-default mechanism is able to provide one of the possible solutions to cases of conflict between the speaker’s conscious intentions and the audience’s actual uptake (another solution being interactional negotiation).

9 Reply to Tanesini Tanesini argues that considering silence as a default reaction, which confirms the appropriateness of an utterance and its prima facie illocutionary force, is a way of silencing one’s interlocutor. In many cases, she seems to be right. Consider the following scenario: A makes a racist comment on C. B, who is present, disapproves of A’s behavior, but does not say anything (perhaps because she does not want to have a fight with A right now, or because she is not in a position to reprimand A: suppose A is B’s boss and can fire her). If silence is the default reaction to a comment, the interlocutor who remains silent manifests that she takes the comment as felicitous, thus accepting its preparatory conditions and other presuppositions as satisfied. So, if silence is the default reaction, B endorses A’s racism. This appears to be as an easy way for A to exercise authority over B and spread hate. But the same premises do not yield the same results in all contexts. Imagine that A makes sexual advances on B and B replies

322 

M. Sbisà

“No! Stop it.” A keeps silent. If silence is the default reaction to an apparent refusal, and therefore validates it, A, by remaining silent, has accepted B’s utterance as a refusal of his sexual advances and is bound to stop making sexual advances to B. Suppose that he goes on nevertheless and, later on, is charged with stalking. He cannot (truthfully) plead that B did not refuse his advances. In this case, the authority attributed to B, thanks to silence counting as the default reaction, protects (rather than endangering) the victim of the stalking. So, it seems to me that the perverse consequences of the adoption of the norm that an interlocutor must manifest her disagreement in some recognizable way if she intends to count as disagreeing on the speaker’s presuppositions or the felicity of her speech act, are limited to certain contexts and depend on the specific characteristics of speaker, interlocutor, and kind and content of the speech act. There might also be a difference, as my two examples suggest, between the effects of by-default endorsing a presupposition and those of bydefault accepting the indicated illocutionary force of an utterance. However, I agree with Tanesini that silence is rarely the default reaction to a speech act. If, in the first scenario outlined above, B is afraid of her boss and wants to conceal her disagreement on his racist presuppositions from him, she is likely to issue some back-channel signal expressing mild alignment (rather than just keep silent, which might reveal perplexity and therefore lack of agreement). In the second scenario, alas, the lack of verbal response on A’s part is unlikely to amount to an acceptance of B’s utterance as a felicitous refusal and more likely reveals that A does not care about her reaction altogether (as it becomes manifest if A goes on making sexual advances on B). In the former case, the interlocutor’s silence does not count as by-default acceptance of the utterance’s presuppositions, but the apparent acceptance is forced by a state of power imbalance. In the latter, not only does the interlocutor’s silence not count as by-default acceptance of the apparent force of the utterance, but there is no such acceptance altogether. Against Goldberg’s defense of the norm according to which the interlocutor’s silence cannot amount to any kind of rejection of a speech act, Tanesini assimilates the speech act introducing such a norm to an

  Replies to Contributors 

323

“exercitive assertion” as characterized in my (2020). Norms may be recalled, or even introduced, by means of simple declarative sentences (“When people introduce themselves to one another, younger people introduce themselves first,” “Vehicles circulate on the right”). Whenever the norm discussed by Tanesini is so recalled or introduced, the speech act that so does can be considered as an exercitive assertion. Hearers are assigned the obligation to believe that this is the way in which a conversation proceeds and that, therefore, the speaker is entitled to count them as accepting the felicity and appropriateness of what the speaker says if only they keep silent. But I am afraid that explicit introductions of the no-silent-­rejection norm are rare: such a norm is more likely to be introduced implicitly, in ways that are closer to McGowan’s “Conversational Exercitives” (2004) than to my exercitive assertions. Indeed, if you are explicitly informed that, if you do not say anything you will be taken as accepting what I say or presuppose or do, you may keep alert and react or attempt to eschew that danger. Proverbs such as our Italian “Chi tace acconsente” (“Who keeps silent, accepts”) are standardly cited after waiting for a reaction, not before a potentially questionable utterance. Another and more general remark made by Tanesini, with which I wish to express agreement, is that commitment to “ideal theory” may lead philosophers astray. If we attempt to describe speech act dynamics as we think it should be in ideal conditions (ideal speaker, ideal audience, ideal conversation, ideal common ground), it is doubtful that we will ever come to account for mistakes, misfires, misunderstandings, unjust impositions, or even the mere clash of interests and perspectives. These phenomena call for accounts taking into consideration cognitive limitations, bias, power imbalance and bounded rationality. I take the hint from Tanesini for claiming that idealizations may be useful in understanding what is going on in actual linguistic exchanges, only insofar as we do not take them to be descriptions of something actually existing or norms that must actually be followed whenever we speak, but use them as points of departure from which to reason about people’s expectations and the outcomes of interacting in conditions that are variously non-ideal.

324 

M. Sbisà

10 Reply to Terkourafi Terkourafi argues that, while lies cannot be considered as a type of speech act, so-called white lies are a type of speech act, and supports her claim with evidence from various research contexts. I agree with her. First, malicious lies are not a type of speech act, but speech acts that, while following a certain illocutionary procedure under other respects, covertly abuse of it by violating its sincerity-related maxim. The speaker may well be said to sincerely invoke an illocutionary procedure (such as asserting, assessing, or expressing an opinion), but fails to entertain the psychological state it implies, with the additional non-communicative intention to deceive. Second, it is clear from the evidence appealed to by Terkourafi that there are accepted conventional procedures (possibly, with differences across languages and cultures) which govern the utterance of white lies, making them recognizable, and that there are circumstances in which invoking such procedures is appropriate (so that participants expect a white lie to be told). I should like to add to Terkourafi’s analysis some considerations about the illocutionary effect of white lies and their place in the illocutionary act classification I endorse. It seems to me quite clear that white lies are not Verdictives (or verdictive Expositives), but Behabitives. They are often called “statements,” but what they share with statements is at most the declarative sentence form (by the way, they often contain evaluative lexical elements). When a speaker utters a white lie, they put themself in an initial position in which they owe something to the addressee, that is, they owe the addressee the taking of a stance that should protect and enhance the addressee's face. They satisfy that debt by offering the white lie. In uttering it, they also commit themself to a line of conduct that is coherent with the stance taken, which is another sort of commitment than that to justifying and defending the truth of an asserted content. Also sincerity requirements do not consist in actually believing that things are as said. I should like to suggest that white lies are sincere as the kind of behabitive illocutionary act they constitute, when the speaker has real concern for the addressee’s face. They are insincere when the speaker has no concern for the addressee’s face but engages in the conventional practice of white lying quite the

  Replies to Contributors 

325

same, perhaps because they know that this is expected. In this case the speaker might manage to protect or enhance the addressee’s face anyway, without entertaining the appropriate psychological states and particularly feelings. Further exploring the ways in which white lies relate to sincerity rules might shed light on the dark side of Behabitives that Austin already noticed (1975, 161) but did not tackle. It might also be interesting to reflect on what happens when a speaker performing a white lie is charged with insincerity (or confesses insincerity). I think that dealing with a white lie as an abuse would amount to taking it to be an assertion or other Verdictive (or verdictive Expositive) as opposed to a Behabitive. Another interesting point of Terkourafi’s is that she considers the “whiteness” of a so-called lie as something gradable. I read this particular case of gradability as a matter of the belonging of a speech act to one or other illocutionary type. Indeed, I conceive of illocutionary types as groups of procedures having conventional effects that are close enough to being instances of a certain pattern. A given procedure (with or without a specific name in the language, but at any rate, suitable for recognition and for bringing about conventional effects) may be a central case of a certain illocutionary type or a more or less marginal one; it may also be a marginal case of more than one illocutionary type. This allows for gradability in force assignment and in the assessment of the appropriateness and felicity of utterances, without suggesting that some utterances are more action-like than others. Indeed, I would not apply the notion of gradability to the capacity of words to be reality-changing as is done by Robin Lakoff in the paper cited by Terkourafi. I think that such an employment of gradability depends on an optical illusion, which we suffer from when we consider only certain kinds of speech acts as real actions. In the four illocutionary types I distinguish after Austin (to be completed with Expositives, to be distinguished within all the other types) there is a wide variety of patterns as to both procedure and resulting distribution of deontic properties, which should be enough to detect something that is done at the illocutionary level on any occasion in which language is spoken (apart perhaps from the occasions in which we can only detect attempts). It is true that much talk we do is repetitive in its illocutionary effects, which then confirm already attributed deontic properties rather than changing them, or reiterate some previously achieved effect. It must

326 

M. Sbisà

also be granted that some speech acts introduce tiny changes, while others introduce substantial ones. But this does not mean that nothing gets done. What I find most interesting in the gradability noticed by Terkourafi between malicious lies and white lies is that it offers a case in which the procedures that are designed to bring about illocutionary effects have fuzzy aspects and may be followed more or less completely and with a certain degree of freedom, lending themselves to different understandings and assessments, and, of course, to interactional negotiation.

11 Reply to Barnes I think that Barnes is right in maintaining that the entitlement to subordinate is not always located in individual speakers but may be collective. Indeed, collectives, too, are agents and speak, therein performing speech acts. The relationship between individual and collective agents/speakers has been analyzed by John Searle (1995) on the basis of the assumption that we are naturally endowed with “we-intentions.” In my approach, which is not intention-based, it is the ascription of agency to a collectivity that constitutes it as an agent, and that ascription is motivated by the identification of a state of affairs for which the collectivity is responsible. Collective agents may be defined formally (a class of students) or informally (a group of friends meeting in a park) and may be endowed with authority, like human individual agents. What if the man targeting the Arab woman on a bus with discriminatory speech had informal authority, in so doing, as the spokesperson of some alleged “public opinion”? Indeed, he speaks as if he were just that. Perhaps, introducing a collective agent—that is, (racist) public opinion—into our analysis of such episodes would help explaining why bystanders tend to keep silent. It is difficult to guess how many representatives of (racist) public opinion are on the bus and an attempt to defend the Arab woman might degenerate in a fight with uncertain outcome. Collective informal authority is certainly a very interesting topic to explore with the tools of a philosophical approach to speech that considers it as action. I should like to remark, though, that in Barnes’s

  Replies to Contributors 

327

contribution as well as in other literature on these issues, there is some tendency to use “authority” as an umbrella term covering all kinds of conventionally established power. In turn, power is an ambiguous notion, since there are de facto powers that enable certain agents to hurt or damage other agents, and this fact is often exploited to build up authority, which is a de iure or conventional matter. Moreover, it should not be forgotten that there are many kinds of deontic powers that do not coincide with authority, ranging from rights to licenses, legal (or otherwise recognized) capacities, legitimate expectations, and epistemic reputation. In discussing speech acts or discourses with interesting (or worrying) effects on the deontic properties of the audience or of certain target individuals or groups, this variety of deontic properties should be paid more attention. Indeed, it is connected to the variety of patterns that the distribution of deontic properties among the participants displays according to the four main illocutionary types of my Austin-inspired analysis. Barnes is also right in remarking that the ideological use of presupposition cultivates and constructs an audience or ingroup (who may then act as a collective agent). Indeed, I am convinced that utterance presuppositions, thanks to their being linguistically activated, indicate to the audience what common ground they should share. Those who actively endorse those presuppositions are, or become, the bulk of an ingroup, while the passive, the silent, and the skeptic remain aside, tempted to become part of that ingroup, if not for other reasons, in order to avoid openly challenging the speaker and their group. But it should be pointed out that the “normativity” of presupposition should not be confused with the “normativity” of illocution. In the latter case, obligations and other deontic properties (that might be said to compose a normative landscape) are brought about by the illocutionary act as its characteristic effect, while in the former, the speaker’s use of words or constructions (sometimes, but not necessarily, as illocutionary indicators) determines what assumptions the speaker and the audience should take for granted in order to justify their acceptance of the speaker’s utterance as appropriate and felicitous. Thus, presupposition (or presupposition accommodation) is part of the broader process of illocutionary dynamics: it is functional to it, but at the same time, has effects going beyond it that I, since my (1979), have maintained to be perlocutionary (amounting, more precisely, to

328 

M. Sbisà

perlocutionary sequels). I have recently included membership in a group (or group formation) as one of the functions that presuppositions may serve (Sbisà 2021b). I have some remarks about two examples discussed by Barnes. The first concerns his distinction of de facto and de iure authority. The Pope, he says, has de iure authority on believers but may have de facto authority on some atheists as well. I think, instead, that the authority a Pope may have on non-Catholics, non-Christians, or atheists is a matter of deontic properties and therefore a “conventional” matter as much as his formal authority on the Catholic Church. Recently, I have called that kind of informal authority “prestige” (Sbisà 2020, English translation). That sort of authority is granted to a speaker by their audience and enables them to issue certain kinds of exercitive speech acts, among which warnings, announcements, advice, recommendation, exhortation, incitement, exercitive assertion, and perhaps even commands. But it is true that formal authority and informal prestige do not always work together. Not only there are non-Catholics and non-Christians on whom the Pope has prestige, but also, there are Catholics who in fact disagree with the Pope, who then has no prestige (or informal authority) on them, while he remains their Pope and has therefore formal authority on them as members of the Catholic Church. I think that we should consider all kinds of authority (including prestige) as de iure authority (authority that someone is granted, because more than one agent agrees on his or her enjoying it), and contrast it with de facto powers which are mere abilities or opportunities to constrain or compel others, whether by physical or psychological means. The intriguing point is how these de facto powers interact with de iure deontic properties such as authority, prestige, but also rights, licenses, legitimate expectations. On the one hand, they often accompany those deontic properties, fostering their acquisition, maintenance, and efficacy; on the other, there must be a difference (albeit it is difficult to define) between brute power and power that is socially accepted and legitimized. Indeed, the latter might even not be grounded in de facto power. The second example I should like to comment upon is the utterance of “Hail Trump!” which Barnes analyzes as an authoritative directive speech act. In my framework, directives are not a class or type of illocutionary acts. The illocutionary act typology I endorse highlights illocutionary

  Replies to Contributors 

329

effects and not perlocutionary objects such as the goal of getting someone to do something. The utterance of “Hail Trump!” might well have a directive goal (the perlocutionary object of getting something done), but it should be further examined whether it is so because it assigns some obligation to its audience (as exercitive illocutionary acts do), or on the basis of some other illocutionary effect. “Hail Trump!” might well be a Behabitive or a Commissive. I am inclined to see it as a Behabitive: a reaction to Trump’s personality making manifest the speaker’s stance as a supporter of Trump and committing her to following his leadership. But then how can “Hail Trump!” be an authoritative speech act? What kind of an authority is involved? Utterances of the “Hail x!” form primarily confer authority to x, not to the speaker. It is x who is honored as the leader. So, first of all, it is made manifest to the audience that x (Trump?) is a leader, has followers, is recognized as powerful and might exercise authority. This is kind of threatening for those who are unwilling to recognize that authority. When a crowd shouts out (or writes on social networks) “Hail x!” you, the skeptic, know that an authority has been established and the person endowed with it or their representatives are likely to use it to impose themselves on others. This might explain, in part, how “Hail x!” acquires a directive goal, but does not explain why it sounds itself authoritative. A sort of exercise of authority comes in, though, when saying “Hail x!” is presented as a choice. The speaker, in uttering that sentence, does not merely take a stance in reaction to a situation, but makes a decision in favor of something as well as against something else. Making a decision is a feature that “Hail x!” shares with exercitive illocutionary acts. Moreover, it might be added that the pride of being a follower of x expressed in saying “Hail x!” is compatible with presenting one’s saying “Hail x!” as “exemplar” behavior: behavior that is displayed in order to be imitated. In this way, “Hail x!” albeit remaining basically a Behabitive, acquires both an authoritative feature and a directive goal. Collective (speech) agency and individual (speech) agency are intertwined in ways that Barnes’s contribution invites us to further explore. I should like to insist, however, perhaps more in accordance with Langton than Barnes, that it remains morally and politically relevant to trace social facts (including collective informal authority) back to individual

330 

M. Sbisà

attitudes and behavior. It is not only a matter of understanding how certain social facts may arise, but also of urging people to monitor what they themselves contribute to those facts.

References Austin, John L. 1950. Truth. Proceedings of the Aristotelian Society, Supplementary Volumes 24, 111–128. Reprinted in John L.  Austin (1979), Philosophical Papers. 3rd edition, edited by James O. Urmson and Geoffrey J. Warnock, 117–133. Oxford: Oxford University Press. ———. 1953. How to Talk. Some Simple Ways. Proceedings of the Aristotelian Society 53: 227–246. Reprinted in John L.  Austin (1979), Philosophical Papers. 3rd edition, edited by James O. Urmson and Geoffrey J. Warnock, 134–153.Oxford: Oxford University Press. ———. 1975. In How to Do Things with Words, ed. James O.  Urmson and Marina Sbisà, 2nd ed. Cambridge, Ma: Harvard University Press. Ducrot, Oswald. 1993. A quoi sert le concept de modalité? In Modality in language acquisition, ed. Norbert Dittmar and Astrid Reich, 111–130. Berlin: de Gruyter. Goffman, Erving. 1981. Footing. In Forms of Talk, ed. Erving Goffman, 124–157. Philadelphia: University of Pennsylvania Press. Grice, Paul H. 1969. Utterer’s Meaning and Intentions. The Philosophical Review 78. Reprinted in Studies in the Way of Words (1989), 86–116. Cambridge, MA: Harvard University Press. McGowan, Mary Kate. 2004. Conversational Exercitives: Something Else We Do with our Words. Linguistics and Philosophy 27: 93–111. Millikan, Ruth G. 2005. Language: A Biological Model. Oxford: Oxford University Press. 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. Federico Albano Leoni and M.  Rosaria Pigliasco, 37–60. Roma: Bulzoni. ———. 1989. Linguaggio, ragione, interazione. Per una teoria pragmatica degli atti linguistici [Language, Reason, Interaction. Towards a Pragmatic Speech Act Theory]. Bologna: Il Mulino. 2nd edition reprinted as an e-book (2009). Trieste: EUT.

  Replies to Contributors 

331

———. 1992. Speech Acts, Effects, and Responses. In (On) Searle on Conversation, edited by Herman Parret and Jef Verschueren, 101–111. Amsterdam: John Benjamins. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 1999. Ideology and the Persuasive Use of Presupposition. In Language and Ideology. Selected Papers from the 6th International Pragmatics Conference, edited by Jef Verschueren, 492–509. Antwerp: International Pragmatics Association. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2002a. Speech Acts in Context. Language and Communication 22, 2002: 421–436. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2002b. Cognition and Narrativity in Speech Act Sequences. In Rethinking Sequentiality, edited by Anita Fetzer and Christiane Meierkord, 71–97. Amsterdam: John Benjamins. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2006a. 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, edited by Elvio Baccarini and Snježana Prijić-Samaržija, 233-247. Rijeka: University of Rjieka, Faculty of Arts and Sciences and Croatian Society for Analytic Philosophy. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2006b. Speech Acts Without Propositions? Grazer Philosophische Studien 72: 155–178. ———. 2007. Detto non detto. Le forme della comunicazione implicita. Roma– Bari: Laterza (2nd ed. 2010). ———. 2019. Varieties of Speech Act Norms. In Normativity and Variety of Speech Actions, edited by Maciej Witek and Iwona Witczak-Plisiecka, 23–50. Leiden: Brill. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press. ———. 2020. Illocuzione e dislivelli di potere. In Linguaggio d’odio e autorità. Lezioni milanesi per la Cattedra Rotelli, edited by Claudia Bianchi and Laura Caponetto, 63–86. Milano-Udine: Mimesis. Transl. as Illocution and Power Imbalance in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press.

332 

M. Sbisà

———. 2021a. (Im)politeness and the Human Subject. In The Philosophy of (Im)politeness, ed. Chaoqun Xie, 157–177. Cham: Springer. ———. 2021b. Presupposition and Implicature: Varieties of Implicit Meaning in Explicitation Practices. Journal of Pragmatics 182: 176–188. ———. In preparation. Austinian Themes. Illocution, Action, Knowledge, Truth, and Philosophy. Oxford: Oxford University Press. Searle, John R. 1969. Speech Acts. An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. ———. 1979. Expression and Meaning. Cambridge: Cambridge University Press. ———. 1995. The Construction of Social Reality. New York: Free Press. Stalnaker, Robert. 1999. Context and Content. Essays on Intentionality in Speech and Thought. Oxford: Oxford University Press.