The Cambridge Handbook of Language in Context 9781108839136, 9781108989275, 9781108984461

Citation preview

The Cambridge Handbook of Language in Context For more than a decade, linguistics has moved increasingly away from evaluating language as an autonomous phenomenon and toward analyzing it “in use,” showing how its function within its social and interactional context plays an important role in shaping its form. Bringing together state-of-the-art research from some of the most influential scholars in linguistics today, this Handbook presents an extensive picture of the study of language as it used “in context” across a number of key linguistic subfields and frameworks. Organized into five thematic parts, the volume covers a range of theoretical perspectives, with each chapter surveying the latest work from areas as diverse as syntax, pragmatics, psycholinguistics, applied linguistics, conversational analysis, multimodality, and computer-mediated communication. Comprehensive, yet wide-ranging, the Handbook provides a full description of how the theory of context has revolutionized linguistics, and how its renewed study is crucial in an ever-changing world. J E S Ú S R O M E R O - T R I L L O is Professor of Pragmatics at Universidad Autónoma de Madrid. He is the founder and Editor-in-Chief of Corpus Pragmatics, and since 2008 he has edited seven volumes on pragmatics.

Published online by Cambridge University Press

CAMBRIDGE HANDBOOKS IN LANGUAGE AND LINGUISTICS

Genuinely broad in scope, each handbook in this series provides a complete state-ofthe-field overview of a major sub-discipline within language study and research. Grouped into broad thematic areas, the chapters in each volume encompass the most important issues and topics within each subject, offering a coherent picture of the latest theories and findings. Together, the volumes will build into an integrated overview of the discipline in its entirety.

Published titles The Cambridge Handbook of Phonology, edited by Paul de Lacy The Cambridge Handbook of Linguistic Code-switching, edited by Barbara E. Bullock and Almeida Jacqueline Toribio The Cambridge Handbook of Child Language, Second Edition, edited by Edith L. Bavin and Letitia Naigles The Cambridge Handbook of Endangered Languages, edited by Peter K. Austin and Julia Sallabank The Cambridge Handbook of Sociolinguistics, edited by Rajend Mesthrie The Cambridge Handbook of Pragmatics, edited by Keith Allan and Kasia M. Jaszczolt The Cambridge Handbook of Language Policy, edited by Bernard Spolsky The Cambridge Handbook of Second Language Acquisition, edited by Julia Herschensohn and Martha Young-Scholten The Cambridge Handbook of Biolinguistics, edited by Cedric Boeckx and Kleanthes K. Grohmann The Cambridge Handbook of Generative Syntax, edited by Marcel den Dikken The Cambridge Handbook of Communication Disorders, edited by Louise Cummings The Cambridge Handbook of Stylistics, edited by Peter Stockwell and Sara Whiteley The Cambridge Handbook of Linguistic Anthropology, edited by N. J. Enfield, Paul Kockelman and Jack Sidnell The Cambridge Handbook of English Corpus Linguistics, edited by Douglas Biber and Randi Reppen The Cambridge Handbook of Bilingual Processing, edited by John W. Schwieter The Cambridge Handbook of Learner Corpus Research, edited by Sylviane Granger, Gaëtanelle Gilquin and Fanny Meunier The Cambridge Handbook of Linguistic Multicompetence, edited by Li Wei and Vivian Cook The Cambridge Handbook of English Historical Linguistics, edited by Merja Kytö and Päivi Pahta The Cambridge Handbook of Formal Semantics, edited by Maria Aloni and Paul Dekker The Cambridge Handbook of Morphology, edited by Andrew Hippisley and Greg Stump The Cambridge Handbook of Historical Syntax, edited by Adam Ledgeway and Ian Roberts The Cambridge Handbook of Linguistic Typology, edited by Alexandra Y. Aikhenvald and R. M. W. Dixon The Cambridge Handbook of Areal Linguistics, edited by Raymond Hickey The Cambridge Handbook of Cognitive Linguistics, edited by Barbara Dancygier The Cambridge Handbook of Japanese Linguistics, edited by Yoko Hasegawa The Cambridge Handbook of Spanish Linguistics, edited by Kimberly L. Geeslin

Published online by Cambridge University Press

The Cambridge Handbook of Bilingualism, edited by Annick De Houwer and Lourdes Ortega The Cambridge Handbook of Systemic Functional Linguistics, edited by Geoff Thompson, Wendy L. Bowcher, Lise Fontaine and David Schönthal The Cambridge Handbook of African Linguistics, edited by H. Ekkehard Wolff The Cambridge Handbook of Language Learning, edited by John W. Schwieter and Alessandro Benati The Cambridge Handbook of World Englishes, edited by Daniel Schreier, Marianne Hundt and Edgar W. Schneider The Cambridge Handbook of Intercultural Communication, edited by Guido Rings and Sebastian Rasinger The Cambridge Handbook of Germanic Linguistics, edited by Michael T. Putnam and B. Richard Page The Cambridge Handbook of Discourse Studies, edited by Anna De Fina and Alexandra Georgakopoulou The Cambridge Handbook of Language Standardization, edited by Wendy AyresBennett and John Bellamy The Cambridge Handbook of Korean Linguistics, edited by Sungdai Cho and John Whitman The Cambridge Handbook of Phonetics, edited by Rachael-Anne Knight and Jane Setter The Cambridge Handbook of Corrective Feedback in Second Language Learning and Teaching, edited by Hossein Nassaji and Eva Kartchava The Cambridge Handbook of Experimental Syntax, edited by Grant Goodall The Cambridge Handbook of Heritage Languages and Linguistics, edited by Silvina Montrul and Maria Polinsky The Cambridge Handbook of Arabic Linguistics, edited by Karin Ryding and David Wilmsen The Cambridge Handbook of the Philosophy of Language, edited by Piotr Stalmaszczyk The Cambridge Handbook of Sociopragmatics, edited by Michael Haugh, Dániel Z. Kádár and Marina Terkourafi The Cambridge Handbook of Task-Based Language Teaching, edited by Mohammed Ahmadian and Michael Long The Cambridge Handbook of Language Contact: Population Movement and Language Change, Volume 1, edited by Salikoko Mufwene and Anna Maria Escobar The Cambridge Handbook of Language Contact: Multilingualism in Population Structure, Volume 2, edited by Salikoko Mufwene and Anna Maria Escobar The Cambridge Handbook of Romance Linguistics, edited by Adam Ledgeway and Martin Maiden The Cambridge Handbook of Translation, edited by Kirsten Malmkjær The Cambridge Handbook of Chinese Linguistics, edited by Chu-Ren Huang, YenHwei Lin, I-Hsuan Chen and Yu-Yin Hsu The Cambridge Handbook of Intercultural Pragmatics, edited by Istvan Kecskes The Cambridge Handbook of Role and Reference Grammar, edited by Delia Bentley, Ricardo Mairal-Usón, Wataru Nakamura and Robert D. Van Valin Jr. The Cambridge Handbook of Historical Orthography, edited by Marco Condorelli and Hanna Rutkowska The Cambridge Handbook of Third Language Acquisition, edited by Jennifer Cabrelli, Adel Chaouch-Orozco, Jorge González Alonso, Eloi Puig-Mayenco, Sergio Miguel Pereira Soares and Jason Rothman

Published online by Cambridge University Press

Published online by Cambridge University Press

The Cambridge Handbook of Language in Context Edited by Jesús Romero-Trillo Universidad Autónoma de Madrid

Published online by Cambridge University Press

Shaftesbury Road, Cambridge CB2 8EA, United Kingdom One Liberty Plaza, 20th Floor, New York, NY 10006, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 103 Penang Road, #05–06/07, Visioncrest Commercial, Singapore 238467 Cambridge University Press is part of Cambridge University Press & Assessment, a department of the University of Cambridge. We share the University’s mission to contribute to society through the pursuit of education, learning and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781108839136 DOI: 10.1017/9781108989275 © Cambridge University Press & Assessment 2024 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press & Assessment. First published 2024 A catalogue record for this publication is available from the British Library. Library of Congress Cataloging-in-Publication Data Names: Romero-Trillo, Jesús, editor. Title: The Cambridge handbook of language in context / edited by Jesús Romero-Trillo. Description: Cambridge, United Kingdom ; New York, NY : Cambridge University Press, 2024. | Series: Cambridge handbooks in language and linguistics | Includes bibliographical references and index. Identifiers: LCCN 2023029574 | ISBN 9781108839136 (hardback) | ISBN 9781108984461 (paperback) | ISBN 9781108989275 (e-book) Subjects: LCSH: Context (Linguistics) | LCGFT: Essays. Classification: LCC P325.5.C65 C36 2024 | DDC 401/.43–dc23/eng/20230721 LC record available at https://lccn.loc.gov/2023029574 ISBN 978-1-108-83913-6 Hardback Cambridge University Press & Assessment has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Published online by Cambridge University Press

Contents

List of Figures List of Tables List of Contributors Acknowledgments Language in Context Studies Jesús Romero-Trillo Part I Language in Context: A Sociohistorical Perspective 1 Conversation Analysis John Heritage 2 Context in Historical Linguistics Elizabeth Traugott 3 Context in Discourse Analysis Cedric Deschrijver and Alexandra Georgakopoulou

page ix x xi xiii 1

9 49 71

Part II Philosophical, Semantic, and Grammatical Approaches to Context 4 Philosophy of Language and Action Theory María de Ponte, Kepa Korta, and John Perry 95 5 A Functional Approach to Context Rebekah Wegener and Lise Fontaine 116 6 The Grammar of Incremental Language Production in Context J. Lachlan Mackenzie 140 7 Cognitive Linguistics and Context Dirk Geeraerts 160 Part III Pragmatic Approaches to Context 8 The Role of Context in Gricean and Neo-Gricean Pragmatics Jacques Moeschler 9 Sociopragmatics and Context Sophia Marmaridou 10 Natural Semantic Metalanguage and Context Anna Gladkova 11 Relevance Theory and Context Deirdre Wilson

Published online by Cambridge University Press

183 205 225 247

viii

Contents

12 The Interplay of Linguistic, Conceptual, and Encyclopedic Knowledge in Meaning Construction and Comprehension Istvan Kecskes 268 13 Corpus Pragmatics Karin Aijmer 289 14 Prosodic Pragmatics in Context Jesús Romero-Trillo and Yalda Sadeghi 314 Part IV Applications of Context Studies 15 Language Learning and Assessment in Context Dale Koike and Elisa Gironzetti 16 Linguistic Creativity and Humor in Context Delia Chiaro 17 Context in Translation and Interpreting Studies Luis Pérez-González 18 The Role of Context in Clinical Linguistics Louise Cummings Part V Advances in Multimodal and Technological Context-Based Research 19 Nonverbal Communication and Context: Multimodality in Interaction Pauline Madella and Tim Wharton 20 AI, Human–Robot Interaction, and Natural Language Processing Ian McLoughlin and Nitin Indurkhya 21 Social Media and Computer-Mediated Communication Francisco Yus Index

Published online by Cambridge University Press

331 353 371 393

419 436 455 477

Figures

2.1 Heine’s (2002) and Diewald’s (2006) models compared (based on Traugott 2012: 230) page 54 2.2 Grammaticization as context expansion (Himmelmann 2004: 33) 55 5.1 A Functional approach to context within a theory of linguistics (adapted from Halliday 2013) 122 5.2 Language and context, system and instance (adapted from Halliday 1991/2007: 275) 124 12.1 Recent linguistic theory of language 269 12.2 Kecskes’ definition of language 269 12.3 Conceptual knowledge and encyclopedic knowledge 277 14.1 The function of Adaptive Context 320 18.1 Communication disorders 397 19.1 Prosodic inputs 423 20.1 High-level diagram of a dialogue system showing major components and context level 438 21.1 Discourse and contexts across the online/offline divide 459 21.2 Personal, interactive, and social contexts on a Facebook profile 462

Published online by Cambridge University Press

Tables

7.1 Fundamental recontextualizing characteristics of Cognitive Linguistics 10.1 Semantic primes, English exponents (after Goddard and Wierzbicka 2014) 11.1 Interpretation of my utterance I have to finish a paper 15.1 Historical overview of the relationship between context and L2 learning

Published online by Cambridge University Press

page 175 227 259 341

Contributors

Aijmer, Karin. University of Gothenburg, Sweden Chiaro, Delia. University of Bologna, Italy Cummings, Louise. The Hong Kong Polytechnic University, Hong Kong De Ponte, María. University of the Basque Country, Spain Deschrijver, Cedric. Ming Chuan University, Taiwan Fontaine, Lise. Université du Québec à Trois-Rivières, Canada Geeraerts, Dirk. University of Leuven, Belgium Georgakopoulou, Alexandra. King’s College London, UK Gironzetti, Elisa. University of Maryland, USA Gladkova, Anna. Australian National University, Australia Heritage, John. UCLA, USA Indurkhya, Nitin. Singapore Institute of Technology, Singapore Kecskes, Istvan. State University of New York, Albany, USA Koike, Dale. University of Texas at Austin, USA Korta, Kepa. University of the Basque Country, Spain Mackenzie, J. Lachlan. CELGA-ILTEC, University of Coimbra, Portugal Madella, Pauline. University of Bedfordshire, UK Marmaridou, Sophia. National and Kapodistrian University, Greece McLoughlin, Ian. Singapore Institute of Technology, Singapore Moeschler, Jacques. University of Geneva, Switzerland Pérez-González, Luis. University of Agder, Norway Perry, John. Stanford University and UC Riverside, USA Romero-Trillo, Jesús. Autónoma University of Madrid, Spain Sadeghi, Yalda. CU Escuni, Spain

Published online by Cambridge University Press

xii

List of Contributors

Traugott, Elizabeth. Stanford University, USA Wegener, Rebekah. Paris Lodron Universität Salzburg, Austria Wharton, Tim. University of Brighton, UK Wilson, Deirdre. University College London, UK Yus, Francisco. University of Alicante, Spain

Published online by Cambridge University Press

Acknowledgments

First and foremost, I want to thank my colleague Dr. Yalda Sadeghi for her excellent job as an editorial assistant. Her meticulous work has been fundamental during the compilation and production stages of the volume. I am also grateful to the reviewers of the chapters. Their comments have helped the authors to write excellent contributions to The Cambridge Handbook of Language in Context.

Published online by Cambridge University Press

Published online by Cambridge University Press

Language in Context Studies Jesús Romero-Trillo

The Purpose of the Volume Context is everywhere. Context is everything. Context is whatever contributes, consciously or unconsciously, to the understanding of reality to facilitate language processing in human interaction. We continuously construct and constrain context in our minds to understand and be understood. The Cambridge Handbook of Language in Context describes how context interacts with language across different traditions and theories, and the chapters in the volume will answer some critical questions in context studies that have been puzzling linguists and scholars from other related fields, such as how much context goes into a specific linguistic model or what facets of contextual information are indispensable in a specific theory. To answer some of these burning questions, the volume brings together some of the most influential scholars in linguistics and provides a comprehensive guide to language in context from a multifaceted perspective. For this reason, the present Handbook does not consider context as an ancillary feature for the interpretation of language, as was sometimes invoked in the past, but makes context the focal point to highlight its essential dynamic role for the interpretation of reality. Linguistics describes language and its users from various viewpoints: interactional, textual, historical, cognitive, etc., so this volume departs from the various theories that describe these perspectives to find the emergent aspects that can illuminate a renovated description of the whole, that is, of context. In this sense, the contributors to the volume have made a creative tour de force in their reflection of how context can sharpen the theories they are describing vis-à-vis the recent developments in linguistics and other fields such as technology, philosophy or psychology, amongst others. Context can be described in the digital era as the sum of the linguistic, physical, psychological, and technological factors that influence humans in their interpretation of reality, being aware that the addition of all these factors

https://doi.org/10.1017/9781108989275.001 Published online by Cambridge University Press

2

JESÚS ROMERO-TRILLO

amounts to a more complex result than the mere sum of its parts. The understanding of communication has broadened thanks to the ubiquity of the Internet and the pervasiveness of social media. Technology has changed face-to-face communication irreversibly. Nowadays, access to global information is available with our smartphones within seconds, and conversations no longer rely uniquely on the knowledge of the interlocutors but on the verdict of the web. Context has a polyhedric perspective, and the chapters in the volume weave a crisscrossing safety net that describes context as the background scenario of any communicative exchange without searching for a consensus between the different approaches. Like a mosaic, only the individual tiles can render the whole picture intelligible. For this reason, the Handbook pays attention to the evolution of context throughout history and identifies the various conceptualizations behind each theory to specify their differences, similarities, and complementary views. The chapters in this volume essentialy align with two main traditions: The first has a cognitive foundation and maintains that context is related to the information processing that allows humans to understand reality. The second has a sociological anchor and focuses on the analysis of the elements that allow speakers to engage in the co-construction of conversation as the essence of society. Ultimately, the objective of The Cambridge Handbook of Language in Context is to (re)propose context studies in linguistic communication following the same principle as the negative in analogic photography: most of the image space is occupied by the background and this context is essential to put the focus on the silhouettes. The volume is organized into five topical sections. Part I is entitled “Language in Context: A Sociohistorical Perspective.” In Chapter 1, John Heritage describes the historical development of Conversational Analysis and how the discipline revolutionized linguistics and sociology with its novel principles and methods. The author gives a detailed account of the discipline and its anchor to context. Chapter 2, “Context in Historical Linguistics” by Elizabeth Traugott, investigates the contexts that are most relevant to language change, with particular attention to the contextual changes that enable change and those that follow from change. The author highlights some of the key phenomena in the field: grammaticalization, semantic/pragmatic change, and constructionalization. The last chapter of Part I, “Context in Discourse Analysis” by Cedric Deschrijver and Alexandra Georgakopoulou, provides an overview of the main theoretical concepts and empirical tools used to investigate context in Discourse Analysis and focuses mainly on the most significant text-based conceptualizations of context: context as dynamically evolving, context as interactionally achieved, and context as productive and constitutive of speakers’ positions. Part II describes the philosophical, semantic, and grammatical foundations of context studies. It starts with the chapter by María de Ponte, Kepa Korta, and John Perry entitled “Philosophy of Language and Action Theory.” The

https://doi.org/10.1017/9781108989275.001 Published online by Cambridge University Press

Language in Context Studies

authors reflect on the transition in philosophy of language when scholars started to think about natural languages as opposed to what were called “perfect languages” – giving a detailed account of this change of paradigm, present action theory, and its tenets in context studies. In Chapter 5, Rebekah Wegener and Lise Fontaine describe the functional approach to context and its enormous impact on context studies. They mainly focus on the Systemic Functional Linguistics framework and explore its main parameters as a comprehensive theory of language, giving an account of the historical development compared with other related functional grammars. Chapter 6, “The Grammar of Incremental Language Production in Context” by J. Lachlan Mackenzie, describes the incrementalist approach, which considers how an utterance reflects the language user’s communicative needs and the specific context. The contribution describes the way in which incremental grammars interact with psycholinguistics to demonstrate how multisensory situational contexts are integrated into the speaker’s processing strategies. In Chapter 7 Dirk Geeraerts presents the role of context in Cognitive Linguistics by giving a complete overview of this theoretical approach to context. His main argument is that even if (re)contextualization provides a unifying perspective on the contribution of Cognitive Linguistics, it lacks a unifying perspective based on two main approaches: a universalist tendency under the assumption that the human body is the same for everyone and a sociocultural conception emphasizing the diversity and historicity of the human experience. Part III of the Handbook revises the main pragmatic theories and their understanding of context. In Chapter 8, “The Role of Context in Gricean and Neo-Gricean Pragmatics,” Jacques Moeschler gives a full account of the role of context in the processing of meaning by making the distinction between particularized and conventional pragmatic implicatures in Gricean and postGricean pragmatics. He advocates that context is the last resort for computing conversational implicatures. Chapter 9, by Sophia Marmaridou, approaches the relationship between sociopragmatics and context. Her contribution describes pragmatic meaning as the tool to assess participants’ social distance, the language community’s social rules and appropriateness norms, the discourse practices, and the accepted behaviors. The author incorporates in her description some of the main domains of analysis within this field in their understanding of context and compares the research angle with other pragmatic approaches, like pragmalinguistics. Anna Gladkova outlines, in Chapter 10, the understanding of context within the Natural Semantic Metalanguage, described by the authors as a humanspecific symbolic system for expressing meaning. In their approach, context includes the syntagmatic and paradigmatic relations of a word and the social and cultural contexts it functions in, therefore providing a holistic approach to meaning.

https://doi.org/10.1017/9781108989275.001 Published online by Cambridge University Press

3

4

JESÚS ROMERO-TRILLO

In Chapter 11 Deirdre Wilson examines the relation between context and Relevance Theory and reflects upon the crucial, yet controversial, concept of context due to its importance in the different approaches to linguistics. However, she attributes the lack of a shared theory not only to the linguists’ diverging understandings but also to the fact that there seem to be no limitations to the contextual information that can be triggered in understanding utterances. In her opinion, the question at stake for relevance theory is how hearers can find the necessary contextual information and succeed in recognizing the speaker’s informative intention. In Chapter 12, “The Interplay of Linguistic, Conceptual, and Encyclopedic Knowledge in Meaning Construction and Comprehension,” Istvan Kecskes describes how these three components of meaning collaborate with each other. To delve into this proposal, the author suggests analyzing communication from an intercultural perspective as, even if it is the case that linguistic knowledge is relatively easy to encode across languages, they present essential differences in terms of their cultural approach to very similar concepts. Chapter 13, “Corpus Pragmatics” by Karin Aijmer, describes this recent discipline as a combination of the methodology of corpus linguistics and pragmatics as a window to human thought. For the author, corpus linguistics provides the content matter and pragmatics of its explanation. Although corpus pragmatics is still a novel discipline, it is already achieving great insights into new ways of exploring language use and linguistic theory, with promising implications in some new directions, including variation and change and cultural studies. The last chapter of Part III, “Prosodic Pragmatics in Context” by Jesús Romero-Trillo and Yalda Sadeghi, explains that the relationship between context and prosody is very intuitive but, at the same time, extremely challenging to delineate because it is based on acoustic cues that last milliseconds. However, the liaison between prosody and context is so strong that speakers of a language can understand their interlocutors’ emotional and cognitive status through their prosodic realization. The chapter describes the theory of prosody in the light of pragmatics and context, with special reference to how speakers and hearers use suprasegmental information to encode and decode a message respectively and take full notice of speakers’ intentions. Part IV of the Handbook deals with the applications of context studies. In Chapter 15, “Language Learning and Assessment in Context,” Dale Koike and Elisa Gironzetti discuss some of the most significant historical, methodological, and social issues in context and language learning and assessment in first, second, and heritage languages, making an accurate dissection of how context influences learning and assessment in such different social and linguistic domains. Linguistic creativity and humor in context is the topic of Delia Chiaro’s contribution in Chapter 16, in which she underlines the awareness of context for creating and understanding humor. For the author, although humor is primarily dependent on language to be recognized in its intentionality, it is also

https://doi.org/10.1017/9781108989275.001 Published online by Cambridge University Press

Language in Context Studies

heavily context-dependent, with interesting side implications regarding the recipient’s sense of humor and the moral attachment to the object of our humor. Chapter 17, Luis Pérez-González’s contribution, explores how scholarly thinking on context has informed research in translation and interpreting studies since the early 1990s to proceed to a critique of different approaches on context in translation scholarship. In his view, although there is a widely accepted premise that translators and interpreters are fully immersed in crosslinguistic and cross-cultural contexts, the theoretical construct of the concept has not been systematically approached by translation and interpreting theories until relatively recently. The last chapter of Part IV, “The Role of Context in Clinical Linguistics” by Louise Cummings, examines how clinical linguistics intersects with context. The author’s main focus in the chapter is to show that for people with language disorders, context can be both a barrier to communication but also a powerful resource for the compensation of impaired language skills. The author concentrates on five context-based themes: the nonnormative use of context in children and adults with language disorders; context as a barrier to, and facilitator of, linguistic communication; the role of context in the language disorders clinic; context and the ecological validity of language assessments; and context in the setting of therapy goals and the generalization of language skills. Part V of the volume presents relevant advances in multimodal and technological context-based research. In Chapter 19, “Nonverbal Communication and Context: Multimodality in Interaction,” Pauline Madella and Tim Wharton advocate for the joint vocal-cum-visual approach to linguistic analysis to understand human linguistic behavior, which sometimes expresses more about the emotional state of our interlocutor than the words pronounced. For this reason, nonverbal modalities cannot be ignored, as our word melodies and gestures are essential to guide the hearer to our intended meanings. In Chapter 20, “AI, Human–Robot Interaction, and Natural Language Processing,” Ian McLoughlin and Nitin Indurkhya describe the structure and processes of an AI-driven speech or dialogue system. Following the three stages of the process: speech processing (phonetic level), natural language processing (lexis, grammar, and syntax), and dialogue processing, the authors analyze the implications within the three related layers of context that they describe to reach the highest level of conversation and topic complexity. The authors hypothesize on the conditions for the future arrival of a pragmatic context level in which beliefs, desires, and intentions could also be processed. Chapter 21, the last chapter of the volume, is entitled “Social Media and Computer-Mediated Communication.” For the author, Francisco Yus, context in cyberpragmatic communication is the information that is brought when the coded input reaches some interpretations. As in other contextualization models, the mind must select the appropriate information for meaningful

https://doi.org/10.1017/9781108989275.001 Published online by Cambridge University Press

5

6

JESÚS ROMERO-TRILLO

interpretations. Bearing this in mind, the author describes the challenges for social media and internet communication in online contexts, like the relationship between an interface and utterance contextualization, the role of the physical–virtual interface, also the differentiation between the user’s personal, interactive, and social non-virtual contexts. As readers can observe, The Cambridge Handbook of Language in Context presents a comprehensive description of how the theory of context has evolved in the history of linguistics and how its renewed study is crucial in an everchanging world. I am grateful to the authors for their insightful contributions and to Cambridge University Press for accepting the challenge of publishing the volume. I am sure that The Cambridge Handbook of Language in Context can suggest guidelines and be of assistance to future linguistic research in the (re)creation of context studies.

https://doi.org/10.1017/9781108989275.001 Published online by Cambridge University Press

Published online by Cambridge University Press

1 Conversation Analysis John Heritage

1.1 Introduction In 1964 Harvey Sacks (1935–1975) began giving lectures on conversation at UCLA and circulated them privately on request, in the process creating his “most successful and prolific form of scientific communication” (Schegloff 1992a: xix). The time was a propitious one for innovation in the human sciences. Human cognition and reasoning had been recently rehabilitated as respectable topics of academic study, having been banished for the previous forty years as subjective and unobservable by the now-waning edict of logical positivism and behaviorism. In psychology, pioneering work by Jerome Bruner (Bruner et al. 1956) and George Miller (Miller et al. 1960) led the way and, in linguistics, the behaviorism that had held sway since the days of Leonard Bloomfield was destroyed in Chomsky’s (1959) devastating review of Skinner’s Verbal Behavior. In anthropology, ethnological research on nonWestern classification systems began to expand (Conklin 1959; Frake 1961), while in sociology there arose a widespread research program premised on the social construction of reality (Schütz 1962; Berger and Luckmann 1967; Holzner 1968). Finally, strongly influenced by the approach to communication that focused on context analysis (McQuown 1971; Leeds-Hurwitz 1987; Kendon 1990), Goffman’s research into the interaction order (1955, 1956) was a major influence on Sacks and his colleagues (Schegloff 1992a; Clayman et al. 2022). The philosophical research associated with these movements, particularly the work of Alfred Schütz (1962) and Ludwig Wittgenstein (1958), made it clear that the human sciences had a range of difficult issues to confront. These may be summarized as follows: First, social actions are meaningful and involve meaning-making. Second, this is achieved through a combination of the content of actions and their contexts. Third, action involves shared or intersubjective meanings that may be far from perfect but are normally enough to allow the participants to carry on. Finally, as Schegloff (1988: 442) observed,

https://doi.org/10.1017/9781108989275.002 Published online by Cambridge University Press

10

JOHN HERITAGE

“social life is lived in single occurrences.” Because they embody unique and singular conjunctions of content and context, actions are themselves unique and singular. While possessed of general and generalizable characteristics, these contents and contexts nonetheless intersect in mutually elaborative synergies to enable meanings that are also particular (Garfinkel 1967: 76– 103; Garfinkel and Sacks 1970). The consensus among those who were determined to study social interaction – in particular, Harold Garfinkel (1952, 1967) and Erving Goffman (1955, 1964, 1967, 1983) – was that none of these problems could be analyzed by appealing to psychological processes or other essentially private states of mind. Rather, the task was to focus on the surface of action – the visible order of society (Livingston 2008) – since it is in public action and public communication that meaning-making and comprehension are managed. Within the emerging Conversation Analysis (henceforth CA) paradigm, the focus was not on the “propositions” of positivistic philosophy, nor on the “messages” of communication science, nor yet again on the “meanings” that might emerge from cultural analysis. Rather CA was premised on the notion that, so far as human interaction is concerned, the job of speech is to deliver actions. For example, if A tells B that “Someone just vandalized my car” (Wilson 1991), B has the task of determining whether this statement is intended as a complaint, a request for help, or as an excuse (for a late or nonarrival). This task is of pressing significance because B must frame a response to this information then and there, and this response (for example, an offer of assistance, or an expression of sympathy, or of forgiveness) will unavoidably portray B’s analysis of the action that A’s utterance implemented. It follows that conversational participants have an urgent and compelling interest in the performance and analysis of social actions, and this in turn mandates the corresponding CA focus on action and sequence organization (Schegloff and Sacks 1973; Sacks 1987; 1992). The leading idea that dominated the early development of CA, as well as its subsequent evolution, was Garfinkel’s (1967) notion that social action is methodical, that is, organized by reference to shared methods of acting and reasoning about action that are the fundamental resources for meaningmaking. Moreover, the same methods are deployed both in producing recognizable actions and in recognizing them. As Sacks put it: “A culture is an apparatus for generating recognizable actions; if the same procedures are used for generating as for detecting, that is perhaps as simple a solution to the problem of recognizability as is formulatable” (Sacks Lectures, Fall 1965, Appendix A, p. 226, emphasis in original). The program of research should, accordingly, center on the methods for producing and recognizing action, with these methods construed as a structured set that, like language itself, functions in broad independence from personal sentiments and proclivities. With these insights in hand, CA thinking slowly developed toward the kind of research that we witness today (Clayman et al. 1922). At first, the going was very slow. By the time of Sacks’s death in 1975 in an automobile accident,

https://doi.org/10.1017/9781108989275.002 Published online by Cambridge University Press

1 Conversation Analysis

fewer than twenty CA papers had been published. Starting in the 1980s, the field began to grow and by 2020 the number of publications per annum had increased approximately fortyfold. Starting from an exclusively anglophone base, CA methods have come increasingly to be applied across a broad range of languages.1 This growth has also been associated with an institutionalization of the field together with the broadening of CA’s disciplinary context (Stivers and Sidnell 2013). Initially motivated by fundamental sociological concerns with the analysis of social action and intersubjective understanding bequeathed by Schütz (1962), CA rapidly found an audience within the field of discursive psychology (Potter and Wetherell 1987) and now forms the basis of the more recently founded field of interactional linguistics (Couper-Kuhlen and Selting 2018). It also intersects with a variety of other fields, including computer science and neuroscience, and has a very wide range of applications in the fields of medicine, education, law enforcement, media analysis, second language acquisition, and research into disabilities involving autism, stroke, neuro-degenerative illness, and other forms of brain damage.

1.2 A Revolution in Methods A field that started with the preoccupations of CA – a focus on social action, the achievement of intersubjectivity, and the lodging of both in an ever-shifting envelope of context – could hardly proceed using the older methods of discourse analysis. Speech act theory, with its invention of single sentence examples, devoid of any real context save that stipulated by the theorist, coupled with a top-down analytic process conceived in terms of rules and conditions, was already foundering as an approach to action and proved to be largely without merit as an approach to discourse (Levinson 1981, 1983, 2013, 2017). From the outset, Sacks argued that research based on imagined sentences and hypothesized contexts was essentially constrained by what audiences would accept as “reasonable” (1984: 25). Moreover, these imagined scenes would lack the detail and complexity of the real world of action. Indeed, as he put it, from “close looking at the world we can find things that we could not, by imagination, assert were there. We would not know that they were ‘typical’ . . . Indeed, we might not have noticed that they happen” (p. 25). As Sacks’s reference to “close looking” suggests, he adopted the stance of a naturalist going into the field, and finding and describing elements of social actions and observing their “habitats,” the contexts in which they appeared. Recordings enabled him to study them repeatedly and to allow interactional 1

As summarized in Clift and Raymond (2018), these embrace a wide variety of typologically distinct languages: ǂAkhoe Haiǀǀom, Arabic, LSA (Argentinian Sign Language), Bequian Creole, Bikol, Cebuano, Cha’palaa, Chintang, Danish, Duna, Dutch, Finnish, French, Garrwa, German, Guyanese Creole, Hebrew, Hungarian, Icelandic, Ilokano, Indonesian, Italian, Japanese, Korean, Kri, Lahu, Lao, Mandarin, Murrinh-Patha, Norwegian, Polish, Portuguese, Quiche, Russian, Siwu, Spanish, Thai, Tsou, Tzeltal, Tzotzil, Yélî Dnye, Yucatecaya, and Yurakaré.

https://doi.org/10.1017/9781108989275.002 Published online by Cambridge University Press

11

12

JOHN HERITAGE

materials to be examined in concert with others. A vital result was “that anyone else can go and see whether what was said is so. And that is a tremendous control on seeing whether one is learning anything” (1984: 25). The tenor of Sacks’s comments on methods here, made at the inception of CA, is very much that of an explorer who does not know what is out there to be found. Accordingly, Sacks was generally opposed to the use of methods including coding, experimental procedures, role playing, etc. (cf. Heritage and Atkinson 1984) that would involve top-down generalization or would otherwise presuppose what is likely to be found. Such methods have the potential to obscure the particularity, precision, and underlying organization of observable human conduct. This stance remains a fundamental one within CA, though the increasing expansion of CA findings has more recently eventuated in the use of statistical, experimental, and neuroscience methods as adjuncts to CA investigations (see Robinson et al. 2022).

1.3 The Basic Structure of CA Research At the present time, CA research may be characterized in terms of three levels of analysis.

1.3.1 Sequence The most fundamental of these levels is that of sequence. All actions are built by reference to some “place” in talk and exploit this positioning as a sense-making resource (Schegloff 1984). Moreover, a current action will normally “project” the relevance of some next action, albeit with different degrees of constraint on what should be done next (see below). Combining these two points we may say that each action is context shaped and context renewing (Heritage 1984a: 242). In addition, each next action will normally display some analysis of what has just been said. (1) below is taken from the opening of a conversation between a community nurse (HV) and a new mother, at the mother’s home. The baby is chewing on something: (1) [HV:4A1:1] ((HV: health visitor; M: Mother and F: Father)) 1 HV: He’s enjoying that [isn’t he. 2 F: [°Yes, he certainly is=° 3 M: =He’s not hungry ’cuz (h)he’s ju(h)st (h)had 4 ’iz bo:ttle .hhh 5 (0.5) 6 HV: You’re feeding him on (.) Cow and Gate Premium.=

Here the HV’s apparently casual observation about the baby at line 1 is treated in distinctive ways by the child’s father and mother. The father treats it as something to be agreed with, which he does promptly at line 2. The mother’s

https://doi.org/10.1017/9781108989275.002 Published online by Cambridge University Press

1 Conversation Analysis

response, however, is notably defensive. She is at pains to claim that her baby’s behavior is not due to hunger (lines 3–4), and in this way treats the HV’s observation as potentially critical of her parenting. These two different responses likely reflect the division of labor in the household, at least as far as childcare is concerned. It is in these responses, that preceding speakers can see how they were understood and can respond by validating or correcting the understandings displayed in second turns. It follows that, as we shall see, sequences are also a fundamental vehicle for the management of intersubjective understanding between interactants.

1.3.2 Practices The actions that make up sequences are the objects of numerous practices of turn construction (Drew 2013). To be successfully identified, these practices will normally be (a) recurrent, (b) specifically situated within a turn or sequence, and (c) differentiate the turn from one that is comparable but does not embody the practice in question. Here are some examples: (i) prefacing a turn with an address term, which selects a next speaker to speak next: (2) A: John, do you want another drink?

In this case, the address term selects John as the recipient of the offer. (ii) prefacing a turn with the word well, projecting a turn that will be expanded to more than one unit of talk (Heritage 2015), and that will likely be negatively valenced (Davidson 1984; Heritage 2015; Kendrick and Torreira 2015). In (3) Edgerton is offering to help a family stricken with an injury: (3) [Heritage:0II:4:11–19] 1 Edg: =Oh:hh Lord.< And we were wondering if there’s anything 2 we can do to help< 3 Mic: [Well that’s] 4 Edg: [I mean ] can we do any shopping for her or 5 something like tha:t? 6 (0.7) 7 Mic: Well that’s most ki:nd Edgerton .hhh At the moment 8 no:. Because we’ve still got two bo:ys at home. 9 Edg: Of course.

After receiving the offer to help with the shopping (line 4), Michael begins his response with well at line 7. His response comprises two turn constructional units: an appreciation preceded by well, and a rejection of the offer accompanied by an account. The conjunction here between negative valence and turn expansion is scarcely accidental. In fact, the well preface is critical because, without it, that’s most kind might have been understood as an acceptance of the offer, rather

https://doi.org/10.1017/9781108989275.002 Published online by Cambridge University Press

13

14

JOHN HERITAGE

than as an appreciation headed toward rejection. Here the well preface instructs the recipient that there will likely be more to come (Heritage 2015) (iii) Incorporating the word any into a polar (yes/no) question. (4) [Heritage 2010] 1 Doc: Do you have any drug aller:gies?

This communicates the questioner’s expectation that the response to the question is likely to be negative (Heritage et al. 2007) and is therefore an aspect of the epistemic stance of the questioner (Heritage and Raymond 2021). As previously noted, practices must be identified as recurrent and as specifically situated. This kind of identification is essential for analysis. For example, the import of practices such as well-prefacing shifts with respect to the word’s placement both within a turn (Lerner and Kitzinger 2019), and within a sequence (Heritage 2015, 2018).

1.3.3 Organizations of Practices Conversation Analysis is founded on Goffman’s (1955, 1964, 1967, 1983) notion that social interaction comprises an institutional order in its own right, and one that is embedded in a wide variety of other social institutions, while structuring participation within them. As an institution, social interaction has its own internal organization and, as Schegloff (2006: 71) notes, “where stable talk in interaction is sustained, solutions to key organizational problems are in operation, and . . . organizations of practice are the basis for these solutions.” Collections of practices are readily found in relation to a range of systemic organizational problems in interaction, all of which are putatively universal (Schegloff 2006). These include the following: (i) Turn-taking: General resources are available to solve questions concerning who should talk next, and when they should do so. These embrace practices for the construction of units of talk (turn constructional units) and projecting their conclusion, for their one-at-a-time allocation, and for allocating next turns to next speakers. These were described in Sacks et al.’s (1974) foundational paper on the topic and have been greatly elaborated in subsequent decades. Recent comparative research strongly suggests that the issues raised by the Sacks et al. study have universal relevance across languages (Stivers et al. 2009). (ii) Sequence organization: This will be elaborated in more detail later in this chapter, but the sequence organizational problem concerns the resources and constraints through which turns at talk are lined up into coherent trajectories of interaction. Although many of the practices involved are specific to particular kinds of actions and sequences, the generic notion of paired actions has been a central resource for understanding how

https://doi.org/10.1017/9781108989275.002 Published online by Cambridge University Press

1 Conversation Analysis

(iii)

(iv)

(v)

(vi)

sequences work and is also likely to be a conversational universal (Kendrick et al. 2020). Repair: Interaction can run awry when failures of speaking, hearing, and understanding undermine the mutual intelligibility on which it is based. The organization of practices for dealing with such problems is described under the heading of “repair.” Once again, the basic elements were sketched out by Schegloff et al. (1977) with an important addendum by Schegloff (1992b), and this analysis has undergone considerable refinement in the intervening years. Again, many of the basic principles of repair organization are highly general, if not universal (Dingemanse et al. 2015) Social solidarity: If the first three levels of organization are addressed to the coherent management of interaction itself, the following groups of interactional practices are concerned with the management of the relations between speakers. In the case of social solidarity, these practices are focused on the limitation of conflict and the avoidance of its escalation. Here the organization of “preference” (Pomerantz 1984; Sacks 1987; Schegloff 2007: 58–97; Pomerantz and Heritage 2013; Pillet-Shore 2017) mandating the (delayed) timing and (mitigated) expression of disaffiliative actions is central (see below). Knowledge management: Different speakers bring distinctive kinds of knowledge to bear in the course of their interactional engagements. A wide variety of interactional practices inform the expression of this knowledge: these practices concern its novelty, how the speaker knows it, and whether the speaker has primacy in access to it (Heritage and Raymond 2005). Deontic rights: speakers may vary in their deontic rights: the right to issue directives to others or to require actions from them (Stevanovic and Peräkylä 2012). A range of practices are associated with the management of these rights and with the concrete expression of directives under different circumstances (Stevanovic and Svennevig 2015).

If these six domains of interaction have attracted a good deal of research, there are others that are also managed through clusters of practices. These include references to persons and places, gaze and body behavior, the expression of emotion, storytelling and many more.

1.4 Major Concepts in Conversation Analysis 1.4.1 Sequence Unquestionably, the fundamental and animating concept in CA is that of sequence. Every action (including the first) will necessarily occur at some “place” in an interaction and its content will unavoidably be produced and understood by reference to that place. This means that the most fundamental context in which an action, housed in a turn at talk and/or bodily behavior, will

https://doi.org/10.1017/9781108989275.002 Published online by Cambridge University Press

15

16

JOHN HERITAGE

be understood is the immediately preceding turn. It is sequences of action which serve as the medium through which every other aspect of context – both local and distal, social and psychological – emerge at the surface of social interaction (Schegloff 1987, 1991, 1992c). It cannot be sufficiently stressed that, for CA, sequence is the engine room of interaction, and it is the vehicle through which other elements of context such as social identities and social institutions are realized (Heritage and Clayman 2010). Accordingly, very many resources are deployed in the management of sequential coherence, including the “fitting” of elementary actions to one another (Schegloff and Sacks 1973), the management of coherence and cohesion across turns (Halliday and Hasan 1976), and the use of a variety of particles such as oh, well, and by the way to fine tune the relationship between a current turn and its immediately preceding one (Schiffrin 1987; Fischer 2006; Heritage and Sorjonen 2018). The initial work on sequences was focused on adjacency pairs: actions tightly organized as first pair parts and second pair parts in which the production of a first part strongly obligates a second speaker to produce a second “next” (Schegloff 1972; 2007: 13–27; Schegloff and Sacks 1973). Greetings, “goodbyes,” and question–answer pairs are paradigmatic examples. This apparently simple conceptualization has far-reaching implications. The normative requirement for a fitted second pair part to occur “next” creates a context in which failures to respond appropriately become “noticeable absences,” which in turn can be sanctioned by a first pair part producer who may repeat the first pair part in the hope of response, or complain about the absence. Negative inferences may also be drawn from absent second pair parts: the absence may be attributed to inattention, rudeness, hostility, or an unwillingness to respond due, for example, to a desire to avoid self-incrimination.2 The close-ordering constraint characteristic of adjacency pairs is associated with both the aim of “getting things to happen” in interaction, and processes of understanding that are associated with this. As Schegloff and Sacks (1973: 297–298) summarized the point: by an adjacently positioned second, a speaker can show that he understood what a prior aimed at, and that he is willing to go along with that. Also, by virtue of the occurrence of an adjacently produced second, the doer of a first can see that what he intended was indeed understood, and that it was or was not accepted. Also, of course, a second can assert his failure to understand, or disagreement, and inspection of a second by a first can allow the first speaker to see that while the second thought he understood, indeed he misunderstood. It is then through the use of adjacent positioning that appreciations, failures, corrections, etcetera can be themselves understandably attempted.

2

A variety of countries, including the United States, have laws enabling witnesses to refuse responses to questions without a lawyer present, and without this refusal being admissible as (potentially incriminating) evidence in court.

https://doi.org/10.1017/9781108989275.002 Published online by Cambridge University Press

1 Conversation Analysis

It is these functions that also necessitate the close ordering of adjacency pair parts. Again Schegloff and Sacks (1973: 297) pinpoint the fundamental issue: Given the utterance-by-utterance organization of turn-taking, unless close ordering is attempted there can be no methodic assurance that a more or less eventually aimed-for successive utterance or utterance type will ever be produced. If a next speaker does not do it, that speaker may provide for a further next that should not do it (or should do something that is not it); and, if what follows that next is “free” and does not do the originally aimedfor utterance, it (i.e., the utterance placed there) may provide for a yet further next that does not do it, etc. Close ordering is, then, the basic generalized means for assuring that some desired event will ever happen. If it cannot be made to happen next, its happening is not merely delayed, but may never come about. The adjacency pair technique, in providing a determinate “when” for it to happen, i.e., “next,” has then means for handling the close order problem, where that problem has its import, through its control of the assurance that some relevant event will be made to occur.

These observations further suggest a second vital dimension of close ordering and sequence in general. This concerns the use of sequential organization in the management of intersubjectivity, to which we now turn. 1.4.1.1 Sequence Organization and Intersubjectivity More or less explicit in the preceding discussion is the idea that sequence organization is a vehicle for the management of mutual understanding and social accountability. Central to this process is what may be termed the “three-step model of intersubjectivity” (Heritage 1984a: 254–264; 2022). This is based on the following two notions: (i) a second turn response to a prior turn communicates, if only indirectly and inferentially, a treatment of the prior action that allows a first speaker to see whether (or how) they were understood; and (ii) a third turn from the first speaker is necessary for the second speaker to see whether the understanding conveyed in Turn 2 was appropriate. “Turn 3” is thus the point at which an intersubjectively shared understanding of “Turn 1” is first available and, just as important, is known to be shared and accountably so by the participants (Heritage 1984a; Schegloff 1992b). It follows that each and every next turn at talk is subject to this three-step consolidation of understanding. Thus, rather like a tracked vehicle, an ongoing process of intersubjective consolidation – an incarnate “running index” (Heritage 1988) of the state of the talk – is continuously and unavoidably extruded through successive turns at talk. An important consequence of this line of reasoning is that intersubjective understanding is neither an exclusively subjective, nor an instantaneous event. Rather, it is a social process unfolding across multiple turns at talk.

https://doi.org/10.1017/9781108989275.002 Published online by Cambridge University Press

17

18

JOHN HERITAGE

However, “intersubjective understanding” is not the explicit project of sequences. Rather, the actions that compose sequences are accountable in their own right as actions and as competent understandings of what has gone before. In this context then, every turn simultaneously embodies both an understanding of the prior turn and constitutes an action in its own right. This double role for turns at talk is nicely illustrated in (5) below. Here Marty approaches Loes, who is at a desk, with a request (line 1): (5) [Schegloff 1992b: 1321) 1 Marty: Loes, do you have a calendar, 2 Loes: Yeah ((reaches for her desk calendar)) 3 Marty: Do you have one that hangs on the wall? 4 Loes: Oh, you want one. 5 Marty: Yeah

Loes’s responsive action (line 2) is interdicted by Marty’s second request (line 3), at which point Loes, at line 4, articulates the understanding that Marty wants to have a calendar to take away, rather than merely to consult one. This “display of understanding” simultaneously conveys a revised understanding of line 1, while also covertly accounting for the previous misunderstanding. Accountability in the two fundamental senses of understanding and action are seamlessly interwoven here. Adjacency pair sequences can be expanded. These expansions can precede the adjacency pair proper (pre-expansions) and in this position are frequently used to establish the appropriateness of the “base” adjacency pair sequence (Schegloff 2007: 28–57). Insert expansions, occurring between a first pair part and a second, are mainly occupied with initiating repair on the first pair part or establishing conditions that will be necessary for the provision of the second (Schegloff 2007: 97–114). While both these types of expansions are normally themselves carried out through adjacency pairs, the third type of expansion – the post-expansion – is often more loosely structured and can take a wide variety of forms (Schegloff 2007: 115–180). 1.4.1.2 Relaxing the Notion of Sequential Constraint Conversations, of course, are not exclusively implemented through sequences of adjacency pairs. Rather the latter are a particularly tightly organized form of sequential constraint, in which recipients have few options or “room for maneuver.” However, the very idea of sequential organization suggests that talk is always organized by a relationship of “nextness” to a prior. And, even though this relationship may be looser in terms of the constraint placed on a next turn by a prior, it is still “central to the ways in which talk-ininteraction is organized and understood. Next turns are understood by coparticipants to display their speaker’s understanding of the just-prior turn and to embody an action responsive to the just-prior turn so understood” (Schegloff 2007: 15).

https://doi.org/10.1017/9781108989275.002 Published online by Cambridge University Press

1 Conversation Analysis

Consider the following case. Here a daughter Lesley (Les) has recommended garlic tablets to her mother, and enquires whether she has followed up on the suggestion: (6) [Field 1:1:89–94] 1 Les: Uh didyuh get yer garlic tablets. 2 Mum: Yes I’ve got them, 3 Les: Have yuh t- started tak[ing th’m 4 Mum: [I started taking th’m t’da:y 5 Les: Oh well do:n[e 6 Mum: → [Garlic’n parsley. 7 Les: ↑THAT’S RI: ght.[BY hhoh-u-Whole Food? 8 Mum: [( ) 9 (0.3) 10 Mum: Whole Foo:ds ye[s, 11 Les: [YES well done, 12 (0.3) 13 Mum: ( ) 14 (0.6) 15 Les: ‘s I’ve got Katharine on: th’m too: now,

This sequence begins with two adjacency pair sequences that are “chained” both logically and in terms of anaphoric reference. The sequence is apparently concluded with an oh-prefaced assessment (line 5), which is a frequent method of sequence closure (Heritage 1984b; Schegloff 2007: 118–136). However, at line 6 Mum initiates a further expansion of the sequence with “Garlic’n parsley” that is tied to Lesley’s initial turn that describes the tablets as “garlic” and is understandable, and understood, as a minor correction of that earlier characterization. When Lesley confirms this at line 7, she slightly mischaracterizes the name of the supermarket chain that sells the tablets, attracting a further correction from Mum at line 10. Lesley then confirms the now corrected name and moves to close the sequence with another assessment (line 11). At line 15, she initiates a further expansion on the subject. The point about this chain of turns is that, after line 5, none of these contributions is mandated by the turn preceding it. Rather, each is “volunteered” and achieves its sense as an action through its inferred connections to preceding talk. In this looser form of sequential organization, action is both constructed and understood through the retrospective application of relevance rules that convey the general sequential implicativeness of turns at talk (Heritage 2012b).

1.4.2 Preference Organization It is a readily observed fact about conversational interaction that participants generally avoid actions that disaffiliate with others. That is, confronted with an offer, request, invitation, or an evaluation of something, participants will

https://doi.org/10.1017/9781108989275.002 Published online by Cambridge University Press

19

20

JOHN HERITAGE

generally manage their conduct to avoid or mitigate responsive actions that disagree with or refuse what came before. Conversation analysts use the term “preference organization” to describe the range of practices that are associated with this outcome (Davidson 1984; Pomerantz 1984; Sacks 1987; Pomerantz and Heritage 2013; Pillet-Shore 2017). The term was coined by Sacks in a public lecture in 1973 (later published in 1987, some years after his death). It does not refer to the private, personal, or psychological desires of individuals, but rather to institutionalized practices of talk that are generally expected under such circumstances (Heritage 1984a: 265–280; Schegloff 2007: 58–96). In his original paper, Sacks argued that both first speakers and second speakers engage in practices that contribute to the minimization of potential or actual disaffiliation. A primary resource open to first speakers is the use of pre-sequences to check out the availability of recipients for some form of joint action, as in (7): (7)

[SB:1 (Schegloff 2007:31)] 1 John: Ha you doin- ‘knight’; pejoration as in the shift from cnafa ‘boy, servant’ > ‘knave’; broadening as in cyning ‘king, absolute monarch’ > ‘(figure)head of government’; narrowing as in deer ‘animal’ > ‘specific

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

51

52

ELIZABETH TRAUGOTT

kind of ruminant animal’; metaphor as in extension of king to any person or entity considered to be the best at something (king of portfolios); and metonymy as in wheels used to refer to car (see, e.g., Ullmann 1962; Bréal [1900] 2018). For the most part, these changes were studied out of context, although social factors like taboo and referential changes such as changes in cars (from carriages) were mentioned. The “cognitive turn” in the later part of the twentieth century led to many new conceptualizations of semantic change (e.g., Györi 2002; Geeraerts and Cuyckens 2007). Two aspects of the cognitive turn are particularly relevant for context. The fundamental shift in thinking was that meaning does not change. Rather, language users attribute new meanings to forms over time. A second important shift in thinking brought attention to pragmatic factors such as conversational implicatures (Traugott and Dasher 2002). The study of semantic change is now in large part understood as the study of semantic/ pragmatic change, not semantic change alone (Hansen and Visconti 2009). Interest shifted from changes in lexical meaning to changes associated with grammaticalization.

2.2.3 Diachronic Construction Grammar The third research domain that will figure large in what follows, especially in Section 2.4 on actualization of change, is diachronic construction grammar. While work on grammaticalization focuses on the question of how grammatical material comes into being, work on diachronic construction grammar focuses on how form–meaning pairings (constructions) come into being (e.g., Traugott and Trousdale 2013; Barðdal et al. 2015). There is inevitably some overlap in the domains studied, since grammatical units such as tense, aspect, and modality are constructions. However, the domain of diachronic construction grammar is larger, encompassing argument structure constructions such as ditransitives (e.g., She gave Melissa a book) and benefactives (e.g., She bought Melissa a book) (Zehentner and Traugott 2020); “nominative and infinitive” constructions like She is said to have bought Melissa a book (Noël 2008); and the way construction (e.g., Melissa elbowed her way through the crowd) (Israel 1996). Because studies of grammaticalization and diachronic construction grammar ask different questions, they are complementary and have much to offer each other where areas overlap (Coussé et al. 2018). There are several models of construction grammar (Hoffmann and Trousdale 2013). The model that has received most attention to date in historical work is that of Goldberg (1995, 2006). The foundational thesis of construction grammar is that knowledge of language consists of knowledge of form–meaning pairings. Therefore, in doing diachronic construction grammar, the researcher has to focus on change in both form and meaning, although in practice meaning often predominates (see Israel 1996; Enghels 2018).

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

2 Context in Historical Linguistics

2.3 Contextual Factors That Enable Change 2.3.1 Ambiguity as an Enabling Factor Two highly influential papers theorizing enabling contexts in grammaticalization (also known as “grammaticization”) were published in 2002: Heine (2002) and Diewald (2002). The focus in both is on types of context that enable grammaticalization and on proposing scenarios in which these types are ordered as “stages” in change. These stages are not discrete but on a continuum (Heine 2002: 86). They can be thought of as sequences of contextual changes. Both Heine and Diewald posit a stage of ambiguity prior to grammaticalization in which the new “target” meaning can be inferred as well as the original “source” meaning. However, the stages have different names and are conceptualized somewhat differently. Heine (2002: 85–86) posits an unconstrained Stage I, a specific ambiguous “bridging context” which gives rise “to an inference in favor of the new meaning” (Stage II), a new “switch context” in which the source meaning is incompatible with the source meaning (Stage III), and a “conventionalization” Stage IV in which the new meaning “no longer needs to be supported by the context that gave rise to it” and uses in new contexts can occur. The switch context is the context in which grammaticalization becomes manifest. Heine suggests that switch contexts “can be viewed as a filtering device that rules out the source meaning” (p. 86). Diewald (2002: 103; 2006) posits a Stage I in which the unit in question begins to be used in “untypical contexts” where it has not been used before and new meanings may arise as conversational implicatures. At Stage II there may be a “critical context, which is characterized by multiple structural and semantic ambiguities and thus invites several alternative interpretations, among them the new grammatical meaning” (Diewald 2002: 103, italics original). This is the stage “in which, because of its multiple structural and semantic ambiguity, the grammaticalization process is triggered” (p. 116). At Stage III “isolating contexts” develop, in which the new reading ceases to be context-dependent. This is the stage in which grammaticalization becomes manifest: there are now two separate readings: the original lexical one and the new grammatical one (p. 117). Diewald points out that this conceptualization of “isolating contexts” makes them different from Heine’s “switch” contexts, in which the source lexical item is not mentioned. The similarities and differences can be summarized in Figure 2.1. Note that in both models grammaticalization (Gzn) is manifest at Stage III. The main differences are that Heine’s model includes conventionalization, a postgrammaticalization stage, and that Diewald posits structural as well a semantic and pragmatic ambiguity at Stage II. As she points out (Diewald 2002: 117), Heine focuses on meaning as a major enabling factor, while she finds “changes in related categories and subsystems” important too. Unlike Heine’s bridging contexts, these critical contexts are not necessarily structurally adjacent. In the

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

53

54

ELIZABETH TRAUGOTT

Heine

Diewald Stage 0: “normal” use

Stage I: “normal” use Stage I: “untypical” context

Stage II: “bridging” context

Stage II: “critical” context

(pragmatic, semantic)

(multiple opacity; pragmatic, semantic, structural)

Stage III: “switch” context

Stage III: “isolating” context (reorganization and differentiation)

Stage IV: conventionalization

Figure 2.1 Heine’s (2002) and Diewald’s (2006) models compared (based on Traugott 2012: 230)

particular case she studies, the grammaticalization of German modals, the enabling structural context “does not persist in later stages.” However, in other cases, structural enabling contexts may persist.

2.3.2 Contextual Expansion as an Enabling Factor Heine’s and Diewald’s work on context, as discussed above, is on stages of development, including the stage at which grammaticalization becomes manifest to other language users (and the linguist!). In another important paper on contextual factors in grammaticalization, Himmelmann (2004) focuses on the fact that many examples of grammaticalization (which he calls “grammaticization”) were in the early days presented in terms of the development of a bare lexical into a grammatical item without attention to context, e.g., of Latin ille ‘that’ into the French definite article le. However, “It is never just the grammaticizing element that undergoes grammaticization. Instead, it is the grammaticizing element in its syntagmatic context which is grammaticized” (Himmelmann 2004: 31, italics original). From the perspective of the lexical item, it diminishes in the process of grammaticalization: Latin ille is reduced in substance (two syllables to one, and, in the context of a following vowel, to the single segment l) and in meaning (from a distant demonstrative to an article). But from the perspective of the immediate context, that context expands.

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

2 Context in Historical Linguistics

(Xn) An B | Kn → (Xn+x) An+x b | Kn+x where A and B represent full lexical items, b a grammaticized element and the following three types of contextual changes occur: a. host-class formation: An → An+x (e.g., common nouns → common and proper nouns) b. change of syntactic context: Xn → Xn+x (e.g., core argument position → core and peripheral argument positions) c. change of semantic/pragmatic context: Kn → Kn+x (e.g., anaphoric use → anaphoric and associative anaphoric use) Figure 2.2 Grammaticization as context expansion (Himmelmann 2004: 33)

Himmelmann proposes that the proper domain of grammaticalization is what he calls “constructions” (syntactic constituents). In many cases “the class of elements the gram is in construction with, i.e. the host class, may be expanded” (p. 32). For example, when demonstratives become articles the nouns with which they may occur (the “host-class”) may be expanded from common nouns to uniquely referring nouns like the sun, or proper names like The Mississippi. Further, pragmatic expansion may occur to “associative anaphoric uses (a wedding – the bride, a house – the front door), contexts in which use of demonstratives is impossible” (p. 33). The syntactic contexts in which the construction is used are often also relevant post-grammaticalization, for example, whether the construction is used as a core argument (subject or object) or a peripheral one (adverbial).2 Figure 2.2 presents Himmelmann’s formula for “grammaticization as context expansion.” It accounts for three kinds of context expansion: host-class expansion, syntactic expansion, and semantic/pragmatic expansion, and exemplifies with prototypical definite article development. Himmelmann regards semantic/pragmatic expansion as “the core defining feature of grammaticizing processes” (2004: 33). In Figure 2.2, A and B are lexical items, An is a class of lexical items, x is the expanded set, and b represents a particular grammaticalized lexical item B. X represents syntactic contexts, and K semantic/pragmatic contexts. The important point is that host-class expansion is itself contextualized within larger syntactic and semantic/pragmatic domains. In Himmelmann’s paper, the term “host-class formation” evokes the process of grammaticalization and appears to be conceptualized as occurring at roughly what Diewald calls the “critical context” stage. No temporal relationship between use of the three types of context is proposed. However, in later work on host-class expansion from the perspectives of grammaticalization theory and construction grammar, e.g., Hilpert (2008), host-class expansion, or rather “collexeme” expansion, is conceptualized chiefly as part of the postgrammaticalization actualization of a new grammatical unit (see Section 2.4.2). Although the models described above say little explicitly about exactly what changes may occur after grammaticalization has taken place, research on this question has flourished recently in quantitative research (e.g., Hilpert 2008). 2

This sketch ignores the fact that the rise of article may be part of the development of a determiner phrase (DP).

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

55

56

ELIZABETH TRAUGOTT

Much of it has been conducted in terms of constructional changes discussed below in Section 2.4.

2.3.3 Inferences as Contexts Concomitant with research in grammaticalization in the later part of the twentieth century was increasing interest in pragmatic contexts for change (note Heine 2002 refers to the importance of inferences in “bridging contexts” at Stage II of grammaticalization). Traugott and Dasher (2002: 34–40) developed the Invited Inferencing Theory of Semantic Change (IITSC), according to which conversational implicatures ambient in contexts may be exploited and used innovatively. Eventually they may become conventionalized, in that they are normally associated with an expression. In this process, pragmatic implicatures play a significant enabling role. The IITSC draws on but is not identical with ideas developed in Grice ([1957] 1989) and Levinson (2000). As Hansen (1998: 28) points out, many implicatures are highly contextdependent. According to the IITSC, the implicatures that arise in the flow of SP/W’s speech are metonymic to speech because they are contiguous to it. If individual expressions and the implicatures associated with them are replicated and Addressees/ Readers (AD/Rs) repeatedly draw similar inferences from them, they may become conventionalized and enable new meanings to arise. One of Grice’s Maxims has played a significant role in historical semantic/ pragmatic work. It concerns the quantity of information to be provided and has two submaxims (Grice [1957] 1989: 26): (1) Quantity 1: Make your contribution as informative as is required (for the current purposes of the exchange). Quantity 2: Do not make your contribution more informative than is required.

Reformulating these as SAY AS MUCH AS YOU CAN and SAY NO MORE THAN YOU MUST, Horn (1984) articulated an important hypothesis about diachronic lexical narrowing and broadening. In this work Horn paid little attention to discursive contexts. Traugott and Dasher’s (2002) contextual approach to meaning change builds on Horn’s conclusion that SAY NO MORE THAN YOU MUST leads to enriched interpretations. In this approach, the context can be local, for example, the constituent in which an expression occurs, or clausal, or even larger complex discourses. As an example, consider the development of the quasi-modal ought to (Nordlinger and Traugott 1997). In Present-Day English ought to expresses both deontic (obligation) meaning, as in You ought to write a note of thanks, and epistemic conclusion ([after a knock at the door] That ought to be Jill). Ought to has its origins in Old English agan ‘to have, own, possess,’ as in eall þæt heo ahte on golde ‘all that she had in gold’ (1042). In the larger syntactic context of non-finite complements with non-referential,

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

2 Context in Historical Linguistics

implied future meaning, and in which the purposive was locally expressed by geldan ‘to pay,’ agan was interpreted as ‘to owe,’ as in: (2) he spæc þa wið ænne he spoke then with a to geldene ten þusend pundæ to pay ten thousand pounds (c. 1000 ÆHomM [DOE agan II.A.I])

mon man

þe that

him him

ahte had

In similar passages in DOE ahte to geldene is used to translate Latin debebat ‘owed.’ This reinterpretation of the lexical verb agan by hypothesis arose out of replication of similar scenarios where it is implied that in order to pay a debt someone will be obliged to possess (temporarily) a certain amount of money. Uses of this kind in which future possession is expected for purposes of fulfilling an obligation are not unique to agan; they are also found associated with reinterpretation of have, as in have to do something. Obligation is only an implicature of agan in (2) and other examples prior to the eleventh century; it is contextually dependent on the future orientation of the non-finite clause. By the beginning of the eleventh century, some new uses appear that according to DOE “shade into auxiliary uses.” The obligation may concern a king’s right to taxes, but it may also concern moral and social obligations, as in (3), where possession is ruled out (Diewald would consider this evidence of an isolating context). (3) ic eow wylle eac eallswa cyðan, þæt man ah I you wish also similarly say that one ought seoce men to geneosianne and dead bebyrian sick men to visit and dead bury ‘I also want to say to you that one ought to visit sick people and bury the dead’ (early 11th century Hom U, 46 [DOE agan II A.2.a; Nordlinger and Traugott 1997: 307; Traugott and Dasher 2002: 140)

(3) exemplifies just one hypothesized source context for the deontic obligation: a non-finite clause with non-referential, future-oriented meaning, implicating that possession is an obligation. Further investigation would no doubt reveal additional contexts.

2.3.4 Constructional Assemblies as Contexts One of the contributions of work on diachronic construction grammar to the understanding of contexts that enable morphosyntactic change is that there may be “assemblies” of contexts, in other words, over time contexts that potentially enable a change may accumulate. For example, Petré (2019) investigates what he calls the first “episode” of the development of the auxiliary [BE going to INF] out of a motion construction, specifically the rise of the relative future in the early seventeenth century. In construction grammar of the Goldberg variety, utterances are conceptualized as combinations of constructions, e.g. What did Liza buy the child? is analyzed in Goldberg (2003: 221) as a combination of the constructions in (4):

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

57

58

ELIZABETH TRAUGOTT

(4) 1. 2. 3. 4. 5. 6.

Liza, buy, the, child, what, did the ditransitive construction the question construction the subject-verb auxiliary construction the VP construction the NP construction

Petré (2019: 166) identifies three constructions relevant to the first episode of the development of auxiliary [BE going to INF]:3 (5) 1. 2. 3.

the progressive construction [[BE V ing] ↔ [ongoing activity]] the verb [[GO] ↔ [‘go’]] construction purpose construction [[to INF] ↔ [intended activity]]

Each of these can be used by itself. They are attested combined in: (6) you thinke I am going to market to buy rost meat, de ye not? (1592, Wilson [Petré 2019: 166])

Here the utterance is fully compositional and is about motion with a purpose. Petré (2019) considers the progressive to have been a crucial factor in the development of [BE going to INF] future. Petré (2016) argues that [BE Ving] was originally a stative expression. By the sixteenth century it had been reinterpreted as an expression of ongoing activity especially in past tense adverbial clause environments in narratives and extended to use in present tense main clauses. It was used especially frequently with the verbs come and go (Petré 2016: 47). In the early seventeenth century. some topicalized examples are attested in which, although motion is relevant to the scene, it is backgrounded to intention to do something. The utterance is no longer fully compositional: intention predominates. (7) HORT: FLO:

. . . I must pick it out of him by wit. . . . what mary-bone of witte is your iudgement going to pick now? ‘. . . what essential part of wit are you, a person capable of good judgment, going to pick now?’ (1608, Day [Petré 2019: 173])

Additional contexts have been identified, such as passive complements of to, e.g., when he is going to be whipped (1628). What is important here is that by hypothesis somewhat frequent replication of assemblies of several contexts contributed to the first phase in the development of [BE going to INF] future. Furthermore, each of the constructions that came to be used in the assembly contributed to deprofiling of motion (Petré 2019: 185).

2.3.5 Ambient (Inter)subjectivity as the Context for (Inter)subjectification Language use is intersubjective. Speakers select what they want to say (subjectivity) to someone (intersubjectivity). Benveniste ([1958] 1971: 230) made a distinction that has proved useful in understanding subjectivity between the 3

[[Form] ↔ [Meaning]] formalizes “form–meaning pairing.”

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

2 Context in Historical Linguistics

syntactic subject (sujet d’énoncé) and speaking subject (sujet d’énonciation). The (inter)subjectivity of talk and text creation is the context for changes known as subjectification and intersubjectification. (Inter)subjectification has largely been associated with actualization. However, use as a procedural by definition involves a shift toward some subjectivity, since procedurals cue SP/ W’s perspective on the argument structure of the clause and on relationships between clauses. Therefore, I include subjectivity here in the section on contextual enabling factors and will point to examples of actualization in Section 2.4 below. There are currently two main views of (inter)subjectivity (see Cuyckens et al. 2010; López-Couso 2010). One is mainly synchronic and has been theorized by Langacker (e.g., 1990, 2006; see also Athanasiadou et al. 2006). On this view, (inter)subjectivity concerns how content of an utterance is construed. Cuyckens et al. (2010: 9) summarize construal as: (i) choice of degree of specificity, (ii) focus, or foregrounding and backgrounding, (iii) degree of prominence, as in A is above B vs. B is below A, and (iv) perspective, the viewing arrangement, especially the location of the speaker or hearer in the spatial domain at the time of speaking. All these are speaker-dependent, and therefore subjective. Langacker (1990: 23) calls the viewing arrangement in An earthquake is going to destroy that town “subjectification.” Here SP/W is offstage, not profiled as an argument of the clause. The other, mainly diachronic, perspective on (inter)subjectivity and especially (inter)subjectification has been theorized by Traugott (e.g., 2003, 2010; see also Davidse et al. 2010). Subjectivity is conceptualized as ambient in language use, but especially in certain domains of linguistic structure, such as grammatical markers, performative verbs (promise), and evaluative adjectives (probably). Intersubjectivity is conceptualized as SP/W’s expression of awareness of AD/R’s attitudes, beliefs, and especially “face” or “self-image.” Subjectification and intersubjectification have been characterized in Traugott (2003: 124–125) as the mechanisms of language change by which meanings tend: (8) a. “to become increasingly based in the SP/W’s subjective belief state or attitude toward what is being said” (subjectification) b. “to encode SP/Ws’ attention to the cognitive stances and social identities of addressees” (intersubjectification)

(Inter)subjectification may precede or follow grammaticalization/grammatical constructionalization. Focus in Traugott’s work on subjectification is on ways in which particular largely objective expressions come to be associated with increasingly subjective meanings, for example, when a circumstance adverbial of space (e.g., by the way ‘along the route’) or time (e.g., while ‘length of time’) is used as a connective, it is modestly subjectified. It no longer refers to space or time but is used to signal SP/W’s perspective on the relationship between discourse segment 1 and discourse segment 2. This kind of change is enabled by use of

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

59

60

ELIZABETH TRAUGOTT

the circumstance adverbial in syntactic positions such as topicalized initial position (9): (9)

but by the way your late stock being spent, here are ten peeces towards a supply. (1653, Brome, The mad couple well matched [Traugott 2020: 125])

Since procedurals by definition cue SP/W’s perspective on the relationship between arguments, temporality, connectivity, etc., early stages of grammaticalization (or grammatical constructionalization) are also by definition examples of (low) subjectification.4 Once procedural, slightly subjectified uses have come to be conventionalized, further subjectification may occur, as illustrated immediately below.

2.4 Contexts for Actualization Changes are dynamic and typically continue. In many cases they involve extension to new contexts. As a result, the contexts themselves may change. In earlier work it was argued, though somewhat controversially, that distinct polysemies may arise (Section 2.4.1). In more recent quantitative work, it is assumed that the use is modulated by extension to more contexts without a distinct polysemy arising (Section 2.4.2).

2.4.1 Contexts in the Rise of Polysemies Briefly, polysemy posits two or more related but distinct meanings, for example, out of context couple means (among other things) ‘two’ or ‘two people in a partnership relationship.’ Likewise, out of context ought to means ‘is obligated to’ or ‘I conclude that.’ However, in context, the distinctness may appear to be rather minimal, and distinguishing polysemy from contextual modulation can prove problematic even in broad-scaled work. Arguments against polysemy were a cornerstone of early work in Relevance Theory, an inferential approach to pragmatics (e.g., Blakemore 1987; Sperber and Wilson [1986] 1995), but see more recently, Carston (2019). In the early model, words are monosemous, which means that they are thought to have core meanings, and differences in meaning are considered to be minimal and non-discrete. They are modulated by context. There are only a few accounts of semantic/pragmatic change in Relevance Theoretic terms (e.g., Nicolle 2011). However, several historical linguists who work on pragmatics have adopted aspects of Relevance Theory. For example, Hansen (2008) argues for the Relevance Theoretic view that the semantics/pragmatics distinction is not between truth-conditional and non-truth-conditional meaning but between 4

Traugott and Dasher (2002) argued that some cases of subjectification also involve contentful, lexical meaning, e.g., use of promise ‘give assurance of’ as a performative verb. As De Smet and Verstraete (2006) point out, many of these are interpersonal, especially performative use of a speech act verb. The context in cases like this is social and not co-textual. It is not discussed further here.

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

2 Context in Historical Linguistics

elements of meaning that are coded and those that arise inferentially from them in context. She nevertheless recognizes the importance of polysemy and proposes that polysemy “is to a large extent the result of diachronic sense extensions” that come to be “at least partially structured by family resemblances” (Hansen 2008: 35). Polysemy has been found to be an essential analytic tool in most work on semantic change, whether the polysemy is semantic or pragmatic (see Horn 1984; Sweetser 1990). It can be conceptualized as a relatively stable extension of earlier meaning. At least for a while, older and newer meanings coexist polysemously, a phenomenon known as “layering” (Hopper 1991: 22–24), and provide the context for further developments. In English several auxiliaries have come to be polysemous over time. BE going to V was first used in the early seventeenth century to express relative future (the later of two events). It was polysemous with BE going to V (motion with a purpose). After the beginning of the eighteenth century, it was extended to deictic uses based in the speaker’s perspective (subjectification). This is clearest in the context of an inanimate subject or there-construction (10), where SP/W is typically “offstage” (see Section 2.3.5): (10)

told him there was going to be an Inquisition made in some Accounts (1701, Anonymous [Petré 2019: 168])

Another example of the development of polysemy is the rise of epistemic uses of ought to meaning ‘it can be concluded that.’ Although a few examples appear in Middle English, this use is not well established until Early Modern English and is infrequent overall. As in the case of other auxiliaries with both obligation and epistemic readings (should and must), the context for epistemic readings is often an inanimate subject and the verb is BE as in (11). In (11) the first speaker expresses contempt at the unfashionableness of that little dowdy and uses ought to in the obligation sense, while the second speaker, Gertrude, defends her using ought to meaning ‘it can be concluded that’ (her trousseau will be correct because of who the tailor is): (11)

“that little dowdy who doesn’t even know how to dress decently! Common respect ought to(deontic) teach her about mourning!” “Her trousseau ought to(epistemic) be right: it was made by Madame Vauban,” interposes Gertrude. (1883, Douglas, Floyd Grandon’s Honor [COHA])

The developments of deictic future BE going to V and epistemic ought to V are both examples of changes in the domain of auxiliary verbs. Both are highly dependent on two contexts: (i) subject of the clause, (ii) type of V. Both are examples of subjectification, first weakly in the sense that use as a procedural always involves at least a small degree of subjectivity, and then more strongly as deictic and epistemic uses emerged respectively. Van linden (2012) explores the lexical development of modal meanings of adjectives like critical and essential both qualitatively and quantitatively in an analysis that refers to polysemy. She details how essential was borrowed from

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

61

62

ELIZABETH TRAUGOTT

French in the fifteenth-century meaning ‘by true nature’ (a defining property) as in þe Escencyalle Ioy es in þe lufe of Godd ‘the definitive joy is in the love of God’ (1440) (Van linden 2012: 90). This meaning was extended to relational use as in essentiall to every living creature (1656) (p. 91). A new modal meaning, ‘indispensible for,’ is attested in the early seventeenth century, as in Government is essential to formed and regular Societies (1681–1686) (p. 93), and a deontic ‘morally necessary’ meaning in the nineteenth century, as in it is essential to see whether (COBUILD Corpus) (p. 216). These changes exemplify increase in subjectivity and involve shifts in syntactic contexts that are so finegrained that they might be more properly thought of as revealing “modulations” of meaning.

2.4.2 Contexts for Modulations of Uses Hilpert (2008) explores a fine-grained approach to expanding contexts using diachronic distinctive collexeme analysis to track before changes in the preferred collocates of individual constructions” (p. 41). Because polysemies like those cited in Section 2.4.1 are context-dependent, he queries the validity of the hypothesis that there are two or more senses (p. 137) but does not completely reject polysemy. Combining the perspectives of grammaticalization and diachronic construction grammar, he shows that the top ten verbs in the complement of BE going to from 1710 on, the period when deictic function had already been established, shifted in the 1850s from primarily telic, dynamic, and intentional verbs such as say, fight, and give to more stative verbs such as be, have. By comparison, future auxiliary gaan ‘go’ in Dutch is first found in the context of imperfective motion verbs and is later extended to cognitive and experiential verbs such as denken ‘think’ and voelen ‘feel.’ Hilpert’s conclusion is that generalized, abstract schematic “paths” of grammaticalization are misleading. At a fine-grained contextual level, there may be significant differences in how particular grammatical units are conceptualized and used. Hilpert (2013) has refined the diachronic distinctive collexemic method using “variability-based neighbor clustering.” Important here is the distinction between changes in particular types of collocation and changes in the token frequency with which a particular collocation type is used. These are both contexts that allow for further modulations.

2.4.3 Contexts Enabling Analogical Change A much-studied motivation for change in recent years is analogy. In thinking about analogy it is useful to distinguish analogical thinking, a motivation that may result in activation of the mechanism of analogization, the creation of new uses based on pattern similarity (Traugott and Trousdale 2013: 37–38). De Smet (2012) shows how actualization “proceeds from one environment to another on the basis of similarity relations between environments” (p. 601). Actualization is understood as the process by which “an item’s new syntactic

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

2 Context in Historical Linguistics

status manifests itself in new syntactic behavior” (p. 601). This new syntactic behavior may be based on local surface forms rather than on abstract generalizations. Actualization may be “sneaky” (p. 608) because “language change often advances most easily where it is least obtrusive, apparently thriving on structural ambiguities and (possibly superficial) resemblances to existing patterns” (p. 607). Among De Smet’s examples is the rise of all but as a downtoner meaning ‘not quite,’ using CLMETEV (pp. 611–617).5 Initially, all but meant ‘everything except’ and was used in the context of NPs (1766, all but his misfortunes). Contexts were extended to gerund nominals (1773, all but the whining end of a modern novel), where all but could be interpreted as ‘almost.’ De Smet hypothesizes that the semantic reinterpretation was due “to a pragmatic implicature that if something is ‘everything but not X’, it is ‘nearly X.’” (p.611). This reinterpretation enabled a new use in the nineteenth century as a downtoner in the context of adjectives (1864 now all but complete) and verbs (1947 he all but knocked over a young woman). De Smet (2012: 615) shows that in the CLMETEV data all but appears with predicative noun phrases like They’re all but a parcel of Pigeons (1773) before it appears with attributive adjectives as in I really have been all but disconcerted (1835) and suggests this is because of pattern similarity. In the shift from predicative to attributive environments, the type of syntactic subject remains constant, as do the predicative function and the pattern BE all but. This is a “sneaky,” almost imperceptible shift. Such analogical extensions are the context for further extensions. As De Smet (2013: 255) points out, one extension opens up new possibilities for further extensions: “one change only becomes possible as a result of a previous change.” In this case, an “analogical chain” has arisen. Collexemic and inferential contexts are contiguous contexts in the linear flow of speech or writing on what is called the “syntagmatic dimension.” Analogical contexts are patterns and potential variants in what is known as the “paradigmatic dimension.” A prototypical paradigm is the set of inflections that may be used in an inflectional language for case or tense. This kind of morphological set is typically restricted in number of types and comes to be reduced over time. For example, in the system of Old English verbs, past singular and past plural indicative and subjunctive tenses were distinguished; a relic of the singular–plural distinction is to be found in was–were, but elsewhere it has been lost (for an account of the gradual reorganization of morphological paradigms in English, some of it due to prescriptivism, see Lass 2006). Typically, adjustments occur over time that generalize one variant over another. For example, of the various forms of the past tense in Old English, -ed was selected. Diewald and Smirnova (2012) suggest that the formation of paradigms is a fourth stage of grammaticalization in addition to the three posited in Figure 2.1. This is a rather restrictive view of grammaticalization (Hilpert 2013: 11), but it has been fruitful in thinking about how patterns 5

A “downtoner” is an expression that reduces the force of its complement.

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

63

64

ELIZABETH TRAUGOTT

develop, whether of a homogeneous kind such as inflections alone or of a mixed kind like inflections and prepositional phrases (e.g., case marking). In particular, because Diewald and Smirnova conceptualize grammatical markers as constructions, it has been fruitful in thinking about integration of constructions into networks. These networks are the contexts for change (Sommerer and Smirnova 2020).

2.4.4 Testing Hypotheses about Analogization De Smet’s examples support the analogization hypothesis because the contexts are so similar. Sometimes, however, there may be competing hypotheses about a development. Hilpert (2013) suggests that variability-based neighborhood clustering can be used to test for evidence of analogization (he calls it “analogy”) in morphosyntax. His example (pp. 180–198) is use of concessive parentheticals with while and if meaning ‘although,’ as in (12): (12) While young, Reed is rated as a top player.

The question is whether this type of concessive parenthetical could have been derived by reduction from full concessive clauses (while he is young) or analogized on temporal uses of while as in While young, I was taught that anger is bad (p. 181). Hilpert argues that if the reduction hypothesis is supported, the semantic and structural behavior of the concessive parentheticals should be reasonably similar to that of the full clauses, in other words, there should be considerable overlap in contexts. Similarly, if the analogization hypothesis is supported, the semantic and structural behavior of the concessive parentheticals should be reasonably similar to that of temporal parentheticals. Again, the contexts should overlap. Synchronic data from the TIME corpus does not support the hypothesis of analogization with temporal uses on several grounds identified by the variability-based neighborhood clustering methodology. For example, concessive parentheticals with while are favored mainly in clause-medial position but temporal parentheticals with while are favored in final position and with -ing forms (e.g., while texting). The results suggest that the hypothesis of reduction from a full concessive clause is better supported than the analogization hypothesis for concessive parentheticals like (12).

2.5 Suggestions for Further Research As has been shown in this overview, in usage-based historical linguistics context is a factor in all change, even if research is limited to co-textual contexts. A recurrent question in the research mentioned above is whether particular kinds of change can be predicted from context, and indeed whether one can predict the kinds of context that will be relevant in a particular type of change.

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

2 Context in Historical Linguistics

These questions require further study. However, the more detailed the analysis, the clearer it becomes that little can be predicted beyond the fact that change is highly context-dependent (Hilpert 2013). Furthermore, contexts continually change. Some correlations have been observed between discursive contexts and change, but they tend to be rather specific. For example, Lenker (2010: 38) found that in English many adverbials used in the context of verbs of locution (say, tell) gave rise to discourse markers, e.g., in addition, by the way. Lenker (2010) and Haselow (2013) have also noted that post-clausal position is often associated with expressions of concession or correction (see, e.g., postcausal uses of however, then), but this cannot be generalized to all discourse markers, e.g., by the way. Some other areas of research that are called for include further work on patterns and networks as contexts for change, as suggested in Section 2.4.3. This will require conceptualizing patterns and networks as contexts rather than as independent components of knowledge of language. Most importantly, in-depth integration of work on co-textual contexts with various types of “external context” is essential for a full understanding of how context contributes to change. Hilpert (2013) includes sex, age, register, and other well-known sociolinguistic factors in his variability neighborhood clustering and binary linguistic regression models. But there is a wealth of other factors to consider since language is a complex adaptive system (Beckner et al. 2009). As mentioned in Section 2.1, among them are the role of turn-taking and other interactional factors in change (Hansen and Visconti 2009), and of multimodal, including gestural contexts (Hata 2016). Schmid (2016) suggests a model for integrating pragmatics, sociolinguistic, and cognitive linguistics. Such models are called for to inform the variety of approaches to language systems.

Corpora CLMETEV

COBUILD COHA

DOE

TIME

The Corpus of Late Modern English Texts Extended Version, compiled by Hendrik de Smet. University of Leuven. https:// perswww.kuleuven.be/~u0044428/clmet.htm Collins Bank of English. 1990–1999. Compiled by John Sinclair. University of Birmingham. Corpus of Historical American English. 1810–2009. Compiled by Mark Davies. Brigham Young University. www.englishcorpora.org/coha/ Dictionary of Old English. A–I online. 2018. Angus Cameron, Ashley Crandell Amos, and Antonette diPaolo Healey. Toronto: Dictionary of Old English Project. Time Magazine Corpus. 2007–. Compiled by Mark Davies. Brigham Young University. http://corpus.byu.edu/time

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

65

66

ELIZABETH TRAUGOTT

References Athanasiadou, A., Canakis,C., and Cornillie, B., eds. (2006). Subjectification: Various Paths to Subjectivity. Berlin: Mouton de Gruyter. Barðdal, J., Smirnova, E., Sommerer, L., and Gildea, S. eds. (2015). Diachronic Construction Grammar. Amsterdam: John Benjamins. Beckner, C., Ellis, N. C., Blythe, R., Holland, J., Bybee, J., Ke, J., Christiansen, M. H., Larsen-Freeman, D., Croft, W., Schoenemann, T., and Five Graces Group (2009). Language is a complex adaptive system: Position paper. Language Learning, 59(Suppl. 1), 1–26. https://doi.org/10.1111/j.1467-9922.2009.00533.x. Benveniste, É. [1958] (1971). Subjectivity in language. In Problems in General Linguistics (trans. M. E. Meek) (pp. 223–230). Coral Gables, FL: University of Miami Press. Blakemore, D. (1987). Semantic Constraints on Relevance. Oxford: Blackwell. Bréal, M. [1900] (2018). Semantics: Studies in the Science of Meaning (trans. Mrs. Henry Cust). New York: Dover. Bybee, J. L. (2010). Language, Usage and Cognition. Cambridge: Cambridge University Press. Bybee, J., Perkins, R., and Pagliuca, W. (1994). The Evolution of Grammar: Tense, Aspect, and Modality in the Languages of the World. Chicago: University of Chicago Press. Carston, R. (2019). Ad hoc concepts, polysemy and the lexicon. In K. Scott, B. Clark, and R. Carston (eds.), Relevance, Pragmatics and Interpretation (pp. 150–162). Cambridge: Cambridge University Press. Comrie, B. (2003). Reconstruction, typology, and reality. In R. Hickey (ed.), Motives for Language Change (pp. 234–257). Cambridge: Cambridge University Press. Coussé, E., Andersson, P., and Olofsson, J., eds. (2018). Grammaticalization Meets Construction Grammar. Amsterdam: John Benjamins. Croft, W. (2000). Explaining Language Change. Harlow: Longman, Pearson Education. Cuyckens, H., Davidse, K., and Vandelanotte, L. (2010). Introduction. In K. Davidse, L. Vandelanotte, and H. Cuyckens (eds.), Subjectification, Intersubjectification and Grammaticalization (pp. 1–26). Berlin: De Gruyter Mouton. Davidse, K., Vandelanotte, L., and Cuyckens, H., eds. (2010). Subjectification, Intersubjectification and Grammaticalization. Berlin: de Gruyter Mouton. Dehé, N., and Wichmann, A. (2010). Sentence-initial I think (that) and I believe (that): Prosodic evidence for use as main clause, comment clause and discourse marker. Studies in Language, 34, 36–74. De Smet, H. (2012). The course of actualization. Language, 88(4), 601–633. De Smet, H. (2013). Spreading Patterns: Diffusional Change in the English System of Complementation. Oxford: Oxford University Press. De Smet, H., and Verstraete, J.-C. (2006). Coming to terms with subjectivity. Cognitive Linguistics, 17(3), 365–392. Diewald, G. (2002). A model for relevant types of contexts in grammaticalization. In I. Wischer and G. Diewald (eds.), New Reflections on Grammaticalization (pp. 103–120). Amsterdam: John Benjamins. Diewald, G. (2006). Context types in grammaticalization as constructions. Constructions SV1–9. Diewald, G., and Smirnova, E. (2012). “Paradigmatic integration”: The fourth stage in an expanded grammaticalization scenario. In K. Davidse, T. Breban, L. Brems, and

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

2 Context in Historical Linguistics

T. Mortelmans (eds.), Grammaticalization and Language Change: New Reflections (pp. 111–134). Amsterdam: John Benjamins. Enghels, R. (2018). Towards a constructional approach to discourse-level phenomena: The case of the Spanish interpersonal epistemic stance construction. Folia Linguistica, 52(1), 107–138. Fetzer, A. (2012). Contexts in interaction: Relating pragmatic wastebaskets. In R. Finkbeiner, J. Meibauer, and P. B. Schumacher (eds.), What Is a Context? Linguistic Approaches and Challenges (pp. 105–128). Amsterdam: John Benjamins. Finkbeiner, R., J. Meibauer, and P. B. Schumacher, eds. (2012). What Is a Context? Linguistic Approaches and Challenges. Amsterdam: John Benjamins. Frank, R. M., Dirven, R. Ziemke, T., and Bernárdez, E., eds. (2008). Body, Language and Mind, Vol. II: Sociocultural Situatedness. Berlin: De Gruyter Mouton. Geeraerts, D., and Cuyckens, H., eds. (2007). The Oxford Handbook of Cognitive Linguistics. New York: Oxford University Press. Givón, T. [1979] (2018). On Understanding Grammar. New York: Academic Press. Goldberg, A. E. (1995). Constructions: A Construction Grammar Approach to Argument Structure. Chicago: University of Chicago Press. Goldberg, A. E. (2003). Constructions: A new theoretical approach to language. Trends in Cognitive Sciences, 7, 219–224. Goldberg, A. E. 2006. Constructions at Work: The Nature of Generalization in Language. Oxford: Oxford University Press. Grice, H. P. [1957] (1989). Logic and conversation. In Studies in the Way of Words (pp. 22–40). Cambridge, MA: Harvard University Press. Györi, G. (2002). Semantic change and cognition. Cognitive Linguistics, 13(2), 123–168. Hansen, M.-B. Mosegaard (1998). The Function of Discourse Particles: A Study with Special Reference to Spoken Standard French. Amsterdam: John Benjamins. Hansen, M.-B. Mosegaard (2008). Particles at the Semantics/Pragmatics Interface: Synchronic and Diachronic Issues: A Study with Special Reference to the French Phasal Adverbs. Amsterdam: Elsevier. Hansen, M.-B. Mosegaard, and Visconti, J. eds. (2009). Current Trends in Diachronic Semantics and Pragmatics. Bingley: Emerald Group. Haselow, A. (2013). Arguing for a wide conception of grammar: The case of final particles in spoken discourse. Folia Linguistica, 47, 375–424. Hata, K. (2016). On the importance of the multimodal approach to discourse markers: A pragmatic view. International Review of Pragmatics, 8(1), 36–54. Heine, B. (2002). On the role of context in grammaticalization. In I. Wischer and G. Diewald (eds.), New Reflections on Grammaticalization (pp. 83–101). Amsterdam: John Benjamins. Heine, B., Claudi, U., and Hünnemeyer, F. (1991). Grammaticalization: A Conceptual Framework. Chicago: University of Chicago Press. Hernández-Campoy, J. M., and Conde-Silvestre, J. C., eds. (2015). The Handbook of Historical Sociolinguistics. Chichester: Wiley & Sons. Hilpert, M. (2008). Germanic Future Constructions: A Usage-Based Approach to Language Change. Amsterdam: John Benjamins. Hilpert, M. (2013). Constructional Change in English: Developments in Allomorphy, Word-Formation and Syntax. Cambridge: Cambridge University Press. Himmelmann, N. P. (2004). Lexicalization and grammaticization: Opposite or orthogonal? In W. Bisang, N. P. Himmelmann, and B. Wiemer (eds.), What Makes

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

67

68

ELIZABETH TRAUGOTT

Grammaticalization: A Look from Its Fringes and Its Components (pp. 21–42). Berlin: Mouton de Gruyter. Hoffmann, T., and Trousdale, G. eds. (2013). The Oxford Handbook of Construction Grammar. New York: Oxford University Press. Hopper, P. J. (1991). On some principles of grammaticization. In E. C. Traugott, and B. Heine (eds.), Approaches to Grammaticalization (Vol. I, pp. 17–35). Amsterdam: John Benjamins. Hopper, P. J., and E. C. Traugott [1993] (2003). Grammaticalization, 2nd ed. Cambridge: Cambridge University Press. Horn, L. R. (1984). Toward a new taxonomy for pragmatic inference: Q-based and R-based implicature. In Deborah Schiffrin (ed.), Meaning, Form, and Use in Context: Linguistic Applications; Georgetown University Round Table’ 84 (pp. 11–42). Washington, DC: Georgetown University Press. Israel, M. (1996). The way constructions grow. In Adele Goldberg (ed.), Conceptual Structure, Discourse and Language (pp. 217–230). Stanford, CA: CSLI Publications. Jacobs, A., and Jucker, A. H. (1995). The historical perspective in pragmatics. In Andreas H. Jucker (ed.), Historical Pragmatics: Pragmatic Development in the History of English (pp. 3–33). Amsterdam: John Benjamins. Kemmer, S., and Barlow, M. 2000. Introduction: A usage-based conception of language. In Michael Barlow, and Suzanne Kemmer (eds.), Usage-Based Models of Language (pp. vii–xxviii). Stanford, CA: CSLI Publications. Kuryłowicz, J., [1965] (1975). The evolution of grammatical categories. In Jerzy Kuryłowicz, Esquisses linguistiques (Vol. II, pp. 38–54). Munich: Fink. Kuteva, T., Heine, B., Hong, B., Long, H., Narrog, H., and Rhee, S. (2019). World Lexicon of Grammaticalization, 2nd ed. Cambridge: Cambridge University Press. Langacker, R. W. (1990). Subjectification. Cognitive Linguistics, 1(1), 5–38. Langacker, R. W. (2006). Subjectification, grammaticization, and conceptual archetypes. In A. Athanasiadou, C. Canakis, and B. Cornillie (eds.), Subjectification: Various Paths to Subjectivity (pp. 17–40). Berlin: Mouton de Gruyter. Lass, R. (2006). Phonology and morphology. In R. Hogg, and D. Denison, (eds.), A History of the English Language (pp. 43–108). Cambridge: Cambridge University Press. Lehmann, C. [1995] (2015). Thoughts on Grammaticalization, 3rd ed. Berlin: Language Science Press. Lenker, U. (2010). Argument and Rhetoric: Adverbial Connectors in the History of English. Berlin: De Gruyter Mouton. Levinson, S. C. (2000). Presumptive Meanings: The Theory of Generalized Conversational Implicature. Cambridge, MA: MIT Press. Lightfoot, D. (1999). The Development of Language: Acquisition, Change, Evolution. Oxford: Blackwell. Lindblom, J. and Ziemke, T. (2002). Social situatedness: Vygotsky and beyond. Adaptive Behavior, 11(2), 79–96. López-Couso, M. J. (2010). Subjectification and intersubjectification. In A. H. Jucker and I. Taavitsainen (eds.), Historical Pragmatics (pp. 127–163). Berlin: De Gruyter Mouton. Meillet, A. [1912] (1958). L’évolution des formes grammaticales. In A. Meillet, Linguistique historique et linguistique générale (pp. 130–148). Paris: Champion.

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

2 Context in Historical Linguistics

Milroy, J., and Milroy, L. (1985). Linguistic change, social network and speaker innovation. Journal of Linguistics, 21(2), 339–383. Narrog, H., and Heine, B., eds. (2011). The Oxford Handbook of Grammaticalization. New York: Oxford University Press. Newport, E. L. (2019). Children and adults as language learners: Rules, variation, and maturational change. Topics in Cognitive Science: Topic 2017 Rumelhart Prize Issue honoring Lila Gleitman. Wiley Online Library. https://doi.org/10.1111/tops.12416 (accessed January 26, 2021). Nicolle, S. (2011). Pragmatic aspects of grammaticalization. In H. Narrog and B. Heine (eds.), The Oxford Handbook of Grammaticalization (pp. 401–412). New York: Oxford University Press. Noël, D. (2008). The nominative and infinitive in Late Modern English: A diachronic constructionist approach. Journal of English Linguistics, 36(4), 314–340. Norén, K., and Linell, P. (2007). Meaning potentials and the interaction between lexis and contexts: An empirical substantiation. Pragmatics, 17(3), 387–416. Nordlinger, R., and E. C. Traugott (1997). Scope and the development of epistemic modality. English Language and Linguistics, 1(2), 295–317. Petré, P. (2016). Grammaticalization by changing co-text frequencies, or why [BE Ving] became the “progressive.” English Language and Linguistics, 20(1), 31–54. Petré, P. (2019). How constructions are born: The role of patterns in the constructionalization of be going to INF. In B. Busse and R. Möhlig-Falke (eds.), Patterns in Language and Linguistics: New Perspectives on a Ubiquitous Concept (pp. 157–192). Berlin: De Gruyter Mouton. Schmid, H.-J. (2016). Why Cognitive Linguistics must embrace the social and pragmatic dimensions of language and how it could do so more seriously. Cognitive Linguistics, 27(4), 543–557. Sommerer, L., and Smirnova, E., eds. (2020). Nodes and Networks in Diachronic Construction Grammar. Amsterdam: John Benjamins. Sperber, D., and Wilson, D. [1986] (1995). Relevance: Communication and Cognition, 2nd ed. Oxford: Blackwell. Sweetser, E. E. (1990). From Etymology to Pragmatics: Metaphorical and Cultural Aspects of Semantic Structure. Cambridge: Cambridge University Press. Traugott, E. C. (2003). From subjectification to intersubjectification. In R. Hickey (ed.), Motives for Language Change (pp. 124–139). Cambridge: Cambridge University Press. Traugott, E. C. (2010). (Inter)subjectivity and (inter)subjectification: A reassessment. In K. Davidse, L. Vandelanotte, and H. Cuyckens (eds.), Subjectification, Intersubjectification and Grammaticalization (pp. 29–71). Berlin: De Gruyter Mouton. Traugott, E. C. (2012). The status of onset contexts in analysis of micro-changes. In M. Kytö (ed.), English Corpus Linguistics: Crossing Paths (pp. 221–255). Amsterdam: Rodopi. Traugott, E. C. (2020). The development of “digressive” discourse–topic shift markers in English. Journal of Pragmatics, 156, 121–135. Traugott, E. C., and Dasher, R. B. (2002). Regularity in Semantic Change. Cambridge: Cambridge University Press. Traugott, E. C., and Trousdale, T. (2013). Constructionalization and Constructional Changes. Oxford: Oxford University Press.

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

69

70

ELIZABETH TRAUGOTT

Ullmann, S. (1962). Semantics: An Introduction to the Science of Meaning. Oxford: Blackwell. Van linden, A. (2012). Modal Adjectives: English Deontic and Evaluative Constructions in Diachrony and Synchrony. Berlin: De Gruyter Mouton. Walkden, G. 2019. The many faces of uniformitarianism in linguistics. Glossa: A Journal of General Linguistics, 4(1), 52, 1–17. Zehentner, E., and Traugott, E. C. (2020). Constructional networks and the development of benefactive ditransitives in English. In L. Sommerer and E. Smirnova (eds.), Nodes and Networks in Diachronic Construction Grammar (pp. 168–211). Amsterdam: John Benjamins.

https://doi.org/10.1017/9781108989275.003 Published online by Cambridge University Press

3 Context in Discourse Analysis Cedric Deschrijver and Alexandra Georgakopoulou

3.1 Introduction Despite being inescapable in socially orientated discourse studies, the notion of context has remained highly ambiguous and elusive. Though it is nowadays acknowledged that discourse should be analyzed in view of its situated production and interpretation, a generally accepted definition or operationalization of context, satisfying the needs of various approaches within discourse studies, is still lacking. Within the field, different views on context often relate to the methodological and epistemological foundations of various research traditions. Work rooted in Conversation Analysis, for example, has adopted a relatively restricted view of context: Its features are only relevant if and when the interlocutors audibly or visibly orient to them (e.g., Schegloff 1997). Critical Discourse Studies, by contrast, may assume certain social structures as immanently present, and textually attestable, features of discourse context (e.g., Fairclough 1995). It would not do justice to the field to provide necessarily brief and limited examples of how “context” has been operationalized within different approaches. Instead, this chapter seeks to focus on facets of convergence that, despite a lack of overall theoretical consensus, have become increasingly accepted within an ongoing, productive discussion in the field. This convergence is summarized in Section 3.2.1 as a view of context as dynamically and interactionally constituted. In Section 3.2.2, we show how this dynamically and interactionally constituted context is retrievable in discourse through linguistic features that reflect the self-referential qualities of language, which, in turn, makes context recursive (Section 3.2.3). Following from that we show how, due to its complexity, “context” is often analyzed through different layers (Section 3.2.4), and we present the latest developments in the aforementioned dynamic view of context in Section 3.2.5. By adopting this approach to analyze the contentious notion of “context,” the chapter allows us to acknowledge and take stock of how far the field has

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

72

CEDRIC DESCHRIJVER AND ALEXANDRA GEORGAKOPOULOU

come in developing a critical mass. Only three decades ago, for instance, in Duranti and Goodwin’s (1992) seminal volume Rethinking Context: Language as an Interactive Phenomenon, many features of context that will be discussed in this chapter, especially its dynamic nature, seemed at best incipient. The features of context will be illustrated and made more tangible with reference to the framework of small stories research in Section 3.3. Finally, after discussing the issue of delineating context in Section 3.4, Section 3.5 highlights contemporary social trends that may inform the future study of context. Specifically, these trends suggest an analytical focus on fragmented, complex communicative situations as prime areas for further theoretical work.

3.2 Theoretical Elements and Illustrations 3.2.1 Early Dynamic and Interactional Conceptions of Context Early works in linguistics have attested that linguistic forms can only be meaningful and fully understood in reference to, and presence of, a certain context (see “context of situation” in Malinowski 1923). A case in point are deictic features of language (e.g., she, there, earlier), which are only interpretable within a certain context. Through deictics, context is marked and encoded in language use. For instance, the deictic here in an utterance may be understood as a trace of the utterance being produced in the same place to which it is referring. At the same time, while deictics are traces of context, language also “points to” language users and the situation in which communicative signs occur (Bar-Hillel 1954), thus making the text–context relation a mutually constitutive one. Such early articulations of the text-context relationship, however, may by now be viewed as narrowly conceived accounts, emphasizing a well-defined extralinguistic context to which language intrinsically referred. Schneider (1993), as cited in Auer (1996), describes this issue as the “semantization of pragmatics”: a truth-conditional assessment of the traces in discourse left by an external context that relies on representational theories of language. To be sure, it is difficult to deny an extralinguistic objective reality of aspects of context, such as day–night cycles or other natural phenomena, especially when these “pre-linguistic” features are necessary to interpret certain utterances (see Verschueren 1999: 109ff.). Nor is it necessarily desirable to remove the idea of an objectively verifiable reality from discourse-analytical endeavors (see, e.g., Hacking 1999). Yet an analysis that predominantly focuses on extralinguistic aspects of context as a separate, surrounding environment of discourse cannot adequately address social aspects of that same context that only come into being through discursive action; nor can it adequately explain nonreferential aspects of meaning, such as those that depend on interaction between participants. Rather than a static view of context (as preexisting to acts of communication and unchangeable by them), a dynamic view of context, emphasizing that

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

3 Context in Discourse Analysis

some of its aspects are brought into being through language, is nowadays commonly accepted in discourse studies. This dynamic nature of context takes two main forms: (i) discourse can construct features of context, and (ii) participants may more or less strategically highlight, negotiate, or equally resist certain aspects of context. Foundational work paving the way for this understanding is that of Goffman (e.g., 1964, [1974] 1986). He started with the observation that talk is guided and managed by the type of social situation that participants find themselves in: A card game requires different forms of speaking from a formal, institutional encounter. Yet, for a particular context to have a bearing on talk, it must first be mutually ratified by participants in interaction (1964), i.e., dynamically brought into being. Participants, furthermore, need not acquiesce to the requirements of whatever social situation they find themselves in. Rather, they may “frame” the context of talk. They may also change their “footing” toward a certain frame: They may alter their alignment toward themselves and others “as expressed in the way [they] manage the production or reception of an utterance” (Goffman 1981: 127), as in Goffman’s (1981) famous example in which the then US president Nixon marked a break from “business” by engaging in small talk with a reporter. While Goffman’s observations were primarily derived from face-to-face, oral interactions, they have shaped discourse-analytic approaches to both oral and written discourse, especially those influenced by linguistic ethnography. Other perspectives also acknowledge the dynamic influence of language on context. In Systemic Functional Grammar (SFG), for example, Halliday and Hasan (1976) adopt Malinowski’s (1923) terminology and argue that features of a “context of situation” are associated with specific linguistic attributes, which collectively constitute a register. As Martin explains, in SFG, a text does not operate as removed from context, but dynamically construes readers’ “understandings and expectations” (2003: 35) about social context, allowing for readers’ interpretation of context. SFG served as the basis for earlier works of Critical Discourse Analysis (CDA), which sought to analyze discourse’s ideological frameworks that may, for current purposes, be considered contextual (and see Section 3.4). Another influential conceptualization of context as dynamic, which emphasizes that context is also an interactive achievement of the participants, is found within Conversation Analysis, where context is fundamentally defined as cotext. The premise of this definition is that “the events of conversation have a sense and import to participants which are at least partially displayed in each successive contribution, and which are thereby put to some degree under interactional control” (Schegloff 1997: 165). It is in these moment-bymoment contributions and “endogenous orientations of the participants” (p. 167) that conversation analysts urge us to look for the constitution of a context relevant for a given interaction. Context is essentially constituted by speakers’ actions which are in turn “doubly contextual.” Specifically, “a speaker’s action is context-shaped: its contribution to an ongoing sequence of

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

73

74

CEDRIC DESCHRIJVER AND ALEXANDRA GEORGAKOPOULOU

events cannot be understood but by reference to the context, including specifically the immediate configuration of actions, in which it participates. But it is also context-shaping: it shapes a new context for the action that will follow” (Heritage 1984: 242). The idea that talk-in-interaction has a defensible sense of its own reality and that, in this sense, the analyst’s identification of relevant contexts for a given interaction should both be grounded in participants’ orientations and be ‘an outcome, a finding, a result of analysis, [not] a presupposition of analysis [or] a definition of what analysis should be’ (Schegloff 1997: 170, emphases in original), currently holds sway over diverse approaches to context in discourse analysis.

3.2.2 Contextualization Cues, Indexicality, and Discourse Markers The combination of Goffmanian insights with the toolkit of linguistic anthropology has engendered influential theoretical models that have established the role of a wide range of linguistic items in constructing context. Of relevance here are Gumperz’s (1982) notions of “contextualization cues” and Tannen’s (1993) subsequent elaboration on how these cues frame discourse. Gumperz’s “contextualization cues” refer to “constellations of surface features of message form . . . by which speakers signal and listeners interpret what the activity is, how semantic content is to be understood and how each sentence relates to what precedes or follows” (1982: 131, emphasis in original). The description “surface features” is aptly broad, as these cues encompass lexical and syntactic choices, phonological features, and a myriad of formulaic expressions and strategies. Gumperz, then, not only analyses linguistic features in terms of their contribution to the text-context interface, but also grounds these in participants’ understandings of context, revealing if and how participants achieve their situated conversational goals. This explains why, in his Discourse Strategies (1982), the analytical focus lies on miscommunications, where relevant contextualization cues are either not recognized or misinterpreted. Contextualization cues, and the associated processes of contextualization and decontextualization, are not merely in the purview of communicative participants; they also aid the analyst in “discern[ing] how the participants themselves determine which aspects of the ongoing social interaction are relevant” (Bauman and Briggs 1990: 68). As such, they point to the aspects of context which participants dynamically orient to. This allows analysts to retrace participants’ dynamic meaning-making, through pinpointing aspects of context that are treated as relevant by participants. This methodological approach remains influential, especially in discourse studies that draw on linguistic anthropology (though see Section 3.2.5 for most recent discussions). Its main benefit consists in delimiting context for analytical purposes, thus avoiding the “dizzying effect” of the “apparent unboundedness of context” (Verschueren 1999: 108). To restrict this analytical remit of “context,” the concept of indexicality has emerged as playing a key role. Originally a concept of linguistic anthropology,

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

3 Context in Discourse Analysis

it has been applied to the study of context building upon early attention to the deixis–context relation (see, e.g., Hanks 1992). Indexicality involves the creation of semiotic links between linguistic devices and social meanings that are (more or less) implicit, direct, mediated, and associative (cf. Ochs 1992). In this way it moves us away from referential, propositional meaning-making to nonreferential, implied relations between text and context. These implied relations are construed in discourse with processes as disparate as categorization of social actors, stance taking (i.e., expressing an attitude toward a proposition), change in prosody, code-switching (i.e., switching between different languages or language varieties in a single communicative event), and several others. Gumperz’s contextualization cues may be seen in the same light: they signal a participant’s recognition of, and orientation toward, nonreferential meanings. A comparable framework to that of Gumperz, in terms of providing distributional analyses of discursive elements in the constitution of context, is to be found in Schiffrin’s (1987) influential account of discourse markers (see also Maschler and Schiffrin 2015). Discourse markers encompass conjunctions (such as and, but, because), temporal indicators (now, then) as well as various relatively idiomatic expressions such as oh, you know or I mean. In Schiffrin’s (1987) analysis, discourse markers are multi-functional and versatile: They may operate on one or more of five “planes of talk”: “information state,” “participation framework,” “ideational structure,” “action structure,” and “exchange structure” (Schiffrin 1987: 315ff.). A discourse marker such as but may function on an ideational structure (linking ideas), an action structure (offering a counterargument), and an exchange structure (where but signals a new turn where the utterer has the floor). The discourse marker well is one that may function on all five planes. In pointing to more than one aspect of context simultaneously, discourse markers unite several structural features of discourse (e.g., ideational, actional) while also indexing the roles of communicative actors in a participation framework. A marker such as but may thus occasion an indexical relation with a previous text (to which the following text is in opposition) while also indexing that the utterer locates themselves in an oppositional role in a participation framework. Both contextualization cues and discourse markers are grounded in the theoretical notion of metapragmatics, a functional modality of language use that, in simplified terms, links aspects of language use with its indexical/ pragmatic context through participants’ own awareness of language use (Silverstein 1993; Verschueren 2000). Since discourse markers clarify framing and participation frameworks, they may also have a metacommunicative function (as in talking about the situated communication). Both Gumperz’s and Schiffrin’s frameworks convincingly show that even relatively minor linguistic choices may contribute to the dynamic construction of context. In doing so, these linguistic choices both reflect participants’ own interpretation of the contextual situation and provide a means for analysts to reconstruct how participants interactionally construct context.

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

75

76

CEDRIC DESCHRIJVER AND ALEXANDRA GEORGAKOPOULOU

3.2.3 Recursivity of Dynamic Contexts and Frames The studies cited above show that context is not (merely) an external factor affecting discourse production and interpretation; rather, it is continually negotiated, evoked, and altered through ongoing interaction. Central to this process are participants in communication, who, through metapragmatic awareness, recognize and retrieve in discourse relevant features that interconnect with context. Likewise, the analyst, by closely attending to textual features that point to or index aspects of context, may retrace how participants orient to, interpret, and dynamically create these same aspects. Yet metapragmatic awareness itself is, theoretically speaking, dynamic and recursive. For instance, a participant may talk about a term’s definition; another participant may then comment on that definition; another participant may then dispute the comment on the definition, and so on (e.g., Berry 2005; Bridges 2017). Given that interpretations of contextualization (either the participant’s real-time ones or the analyst’s retrospective ones) rely on metapragmatic processes, it follows that these interpretations, too, are subject to the same recursive possibilities. As Cook-Gumperz and Gumperz (2011) point out, for example, even the notion of “frame,” itself fundamental for explaining the multifaceted nature of context, is not immune to contextualizing processes. The authors suggest to newly imbue the concept with dynamism and move toward a conceptualization of “framing.” Becoming “part of the communicated message, [framing] can serve to tie together not only linguistic but sociocultural knowledge available in interactive exchange, so taking on part of the previous communicative territory inhabited by many notions of context and contextualization” (Cook-Gumperz and Gumperz 2011: 285). In this sense, the concept itself acquires a “meta” dimension, as participants can utilize their emic understanding of “frames” to achieve their communicative goals.

3.2.4 “Layers” and Dimensions of Context Section 3.2.2 above discussed the role of discourse markers in constructing context, emphasizing their multi-functionality in discourse; as mentioned, in Schiffrin’s terms, discourse markers operate on different “planes of talk.” Postulating levels and more broadly taxonomic divisions of text and context is, in fact, a common feature in discourse-analytic approaches to context as a means for facilitating more detailed analyses. An early example of this approach is Hymes (1974), who developed his influential mnemonic SPEAKING (referring to Setting/Scene, Participants, Ends, Act Sequence, Key, Instrumentalies, Norms, and Genres) so as to group relevant features of context that influence meaning. A similar subdivision of context into dimensions or “layers” may by now be found across theoretical perspectives, where this division often includes contextual parameters that are considered to be “extralinguistic” (see, e.g., Auer 1996; Fetzer 2004).

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

3 Context in Discourse Analysis

Within Critical Discourse Studies, Wodak proposes four “heuristic ‘levels of context’” (2011: 628f. and cf. Wodak 2007: 211): the immediate co-text; intertextual/interdiscursive relationships between and within the texts; the extralinguistic social, environmental, and institutional context; and the broader sociopolitical and historical context. Analyzing the linguistic patterns and strategies relevant to each of these four levels reveals how participants strategically make use of, as well as co-construct, relevant contextual features. Wodak’s Discourse-Historical Approach (DHA), furthermore, lends itself to a combination with ethnography (e.g., Clarke et al. 2012), aiding a triangulation of research findings. Establishing findings on multiple different layers of context will strengthen a unified interpretation (Reisigl and Wodak 2001). In practice, this means relying on, and comparing findings among, various research methods and data. Textual analysis may illuminate, for example, how antisemitism is (re)produced in a particular text, while a broader analysis of institutional and legal characteristics, which strictly speaking transcends textual evidence, may demonstrate how this textual reproduction should be understood in its institutional circumstances. Simultaneously, this institutional analysis may strengthen the claims of the textual analysis (Reisigl and Wodak 2001). This type of approach has often been formulated in terms of macro- and micro- levels of context (at times supplemented by a meso-level). Under this formulation, approaches rooted in, for example, Conversation Analysis would be mainly concerned with locally situated aspects of context (see our discussion in Section 3.2.1 of context as co-text), while Critical Discourse Analysis or language-ideological approaches would be concerned with macro-aspects of contexts. These micro- and macro-levels are subsumed, respectively, in Hanks’s (2006: 121) notions of “emergence” and “embedding.” Whereas the former comprises discourse stemming from “production and reception as ongoing processes,” and thus focuses on “context as phenomenal, social, and historical actuality,” the latter draws attention to “the relation between contextual aspects that pertain to the framing of discourse, its centering or groundedness in broader frameworks” (p, 117). This micro/macro distinction has led to several debates on the links between discourse and “macro” aspects, which will be discussed in Section 3.4. At the same time, many have argued that an analysis of context requires moving beyond the distinction, as will be discussed in the following section.

3.2.5 Toward Timescales and Trans-contexts Concurrent with a more detailed focus on participant roles in the dynamic constitution of context, scholars have increasingly focused on metapragmatics and metalanguage (see Section 3.2.3) as the interface that reconciles “decontextualized” language data and the contextual features that make the data meaningfully relevant. These contextual features, moreover, may be wideranging, transcending localized context and participation frameworks and linking up to more macroscopic layers of context.

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

77

78

CEDRIC DESCHRIJVER AND ALEXANDRA GEORGAKOPOULOU

Yet the macro–micro divide itself has been criticized as being too simplistic, by not considering the complex ways in which several aspects of context are simultaneously indexed (see, e.g., Goodwin and Duranti 1992). Silverstein (1993), for example, explains that Gumperz’s concept of “conceptualisation cues” may be a “narrow” view of contextualization, still emphasizing denotation and relying on “various assumptions about the effective reduction of social organization to localized agentive interaction based on individual rational (or logically reconstructible) consciousness” (p. 57). According to Silverstein, a wider interpretation of “contextualization” is necessary for accounting for how meaningfully relevant knowledge about culture and language ideology is understood and reproduced. In this wider view, indexicals may point to “figures” or “tropes,” understood as schemata of culture and language ideology, influenced and constituted in turn by earlier interactions and, at times, by non-textual features of the communication. A case in point, discussed by Blommaert (2015a), are internet “memes,” images or other signs that spread fast and widely online, aided by quasi random algorithmic processes. These memes may broadly index a culturally specified “coolness,” while a specific “meaning” is often lacking. Furthermore, this “coolness” schema depends on, and further constitutes, recognizable earlier occurrences of this type of sharing. In this sense, “[w]e are confronted, in every actual example of discourse, by a complex construction of multiple historicities compressed into one synchronized act of performance” (Blommaert 2015b: 113–114), and, as such, “every instance of micro contextualization would at once be an instance of macro contextualization” (p. 108). This renders the erstwhile macro–micro distinction problematic if not obsolete, though at times it may be difficult to avoid the metaphorical description of “layers” of context (see, e.g., the discussion in Collins 2011). Being relatively recent, however, this multidimensional, historicized, and broader view of context cannot yet be said to be uniformly accepted throughout discourse studies. In Silverstein’s view, indexed ideologies are essentially a social byproduct of linguistic action, analyzable through the interface of metapragmatics. In this view, indexical ideological links take precedence over the “conscious reconstruction of some agentively-centered ‘illocutionary event’” (Silverstein 1992: 69), and, as such, over individual cognition involved in language use. Other frameworks, however, have argued that broader, sociocultural frames should precisely be located in individual cognition, rather than being conceptualized as a byproduct of social action. We may compare Silverstein’s approach to, for example, van Dijk’s (2006) socio-cognitive approach to discourse. There, context is defined as “subjective participant interpretations, constructions or definitions of such aspects of the social environment” (p. 163), which include communicative roles as well as social aspects such as gender and class, and, indeed, any ascribable feature of social groups. This socio-cognitive model is not in opposition to constructionist approaches: It does not deny institutionally influenced participant roles, for example, but highlights their potential cognitive construction. The upshot is

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

3 Context in Discourse Analysis

a centrality of the individual in experiencing, assessing, and constructing a dynamic context. Van Dijk (2006: 163) suggests that contexts are not “objective” or “deterministic” constraints of society or culture at all, but subjective participant interpretations, constructions or definitions of such aspects of the social environment. . . . Thus, it is not “objective” gender, class, ethnicity or power that control the production or comprehension of text and talk, but whether and how participants interpret, represent and make use of such “external” constraints, and especially how they do so in situated interaction.

Much like in Silverstein’s analysis, these schemata or constraints are not assumed to be universally valid; only the processes (cognitive or semiotic) by which participants are able to retrieve relevant contextual aspects are. As both Silverstein and van Dijk also acknowledge, the ideologies at play, then, inform the metapragmatic representations of context. The question that remains, however, is how analysts would be best equipped to access these ideologies, a question to which we return in Section 3.4.

3.3 Illustration: Context in Small Stories Research To illustrate the key features of the analysis of context as described above, the current section will discuss a case-in-point framework, developed by one of this chapter’s authors: that of small stories research, a paradigm for analyzing everyday life stories as contextualized activities. The paradigm synthesizes (i) a broad social interactional position on what context is, and how it can be analyzed, and (ii) interactional sociolinguistic and linguistic anthropological insights, so as to produce multi-layered approaches to context. This synthesis attempts both to go beyond the micro/macro dichotomous conceptualization of context and to introduce emic, ethnographically attested insights into the relationships between text and context. The deployment of certain affiliated concepts as analytical means for establishing and accounting for links between linguistic and other semiotic choices in a text and (wider) contexts is a common hallmark of such synthesizing approaches. Indexicality, timescales (Wortham and Reyes 2015), stance (Jaffe 2009), chronotopes (Blommaert and De Fina 2017), and frames (Tannen 1993), among others, are such affiliated concepts for investigating discourse-in-context. These also feature in small stories research. The rationale for developing small stories research was that a range of narrative activities, commonly occurring in everyday conversational contexts, had not been sufficiently studied within narrative studies, nor had their importance for the interlocutors’ identity work been recognized. These involved stories with fragmentation and open-endedness of tellings, exceeding the confines of a single speech event, and resisting a neat categorization of beginning-middle-end. They were heavily co-constructed (see Section 3.2.2),

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

79

80

CEDRIC DESCHRIJVER AND ALEXANDRA GEORGAKOPOULOU

rendering the sole teller’s story-ownership problematic. Small stories served as an umbrella term for “tellings of ongoing events, future or hypothetical events, shared (known) events, but also allusions to tellings, deferrals of tellings, and refusals to tell” (Georgakopoulou 2006: 130). By synthesizing interactional (e.g., Schegloff 1998) with practice-based approaches to context (Hanks 1996; Scollon and Scollon 2004), small stories research has put forward a heuristic for identifying and analyzing the links between the stories as communicative activities, the social worlds in which they are told and which they invoke, and the tellers’ identities. The heuristic, which explores the connections of three separable but interrelated levels of analysis (see also Section 3.2.4), is a flexible way of analyzing stories-in-context: ways of telling-sites-tellers (Georgakopoulou 2007). “Ways of telling” refer to the communicative “how”: the socioculturally shaped and more or less conventionalized semiotic and, in particular, verbal choices of a story. Part of the analysis in this respect involves uncovering iterativity in the types of stories told as recurrent ways of (inter)acting, embedded in recurrent social practices and engendering expectations about the ongoing activity (Georgakopoulou 2013). The stories’ plots, the types of events and experience that they narrate, and the ways in which they are interactionally managed during the telling are all important in this respect. So are the intertextual links of the current story with other, previous and anticipated, stories, as well as the participants’ metapragmatic orientations to such links. “Sites” in the analytical heuristic of “ways of telling-sites-tellers” refer to the social spaces in which narrative activities take place. These include situational context factors, that include physical (e.g., seating arrangements), mediational, and other tools that the participants may employ. This view of sites is informed by practice-based approaches to language and place that argue for a dialectic relationship between them and that view place not as a monolithic, static entity but as a resource for constituting context (see Baynham, 2015). Stories perform actions in situ and they are integrally connected with social practices, with what gets done, how, and by whom in specific places. Sites thus allow us to explore the significance of social spaces not just for the here and now of the telling activities but also for the tale-worlds invoked in the participants’ stories. Finally, “tellers” comprise the participants of a storytelling activity as complex entities: here-and-now communicators with particular roles of participation; characters in their tales; members of social and cultural groups; last but not least, as individuals with specific biographies, including habits, beliefs, hopes, desires, fears, etc. It is apparent then that the “who” and the “what” of storytelling are seen as being fashioned into and shaped by the “how” of storytelling. Stories emerge as situated, dialogic activities, consisting of multiple tellers and orientations to the world, encompassing a variety of discourse activities and fleeting moments, and being transportable in other contexts (Georgakopoulou 2007: 86). Narrative meaning-making is thus seen as contextualized but also as having the potential to be lifted from its original context

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

3 Context in Discourse Analysis

and to be recontextualized, that is, to acquire new meanings in new contexts (cf. Bauman and Briggs 1990). The historicity and circulation of stories become part of the analysis. A multi-layered approach to context instructs us to see the ways in which the three levels above are connected as presenting contingency but also durability: Some relations become conventional and hold above individual contexts of storytelling. Furthermore, ways of telling-sites-tellers are expected to be reconfigured differently in the different recontextualizations of a story across time and space as well as in different environments. As no contextualized analysis of any discourse activity, in this case, stories, can be exhaustive, there is an acceptance that different analyses will have different priorities and that certain questions may require more focus on any one of the three analytical levels. To forge links between the above interrelated levels of storiesin-context, positioning analysis is used (Bamberg and Georgakopoulou 2008; Georgakopoulou 2013). This aids in the identification of contingency as well as iterativity (i.e., systematic recurrence) in how characters are presented in the tale-world (Level 1 in positioning analysis); how a story is locally occasioned and co-constructed by storytelling participants (Level 2); finally, what aspects of the teller(s), events, and narrated experience are presented as holding above and beyond the specific story and as presenting continuity over time and across spaces (Level 3). Ethnographic understandings and tracking of iterative choices have been proposed as primary ways for uncovering the iterativity of such choices, particularly in relation to Level 3 (De Fina 2013; Georgakopoulou 2013, and cf. the DHA in Section 3.2.4).

3.4 A Key Issue: The Delineation of “Context” As the discussion above demonstrates, in broad terms, the varied branches of discourse analysis have converged on certain key approaches in the study of context. The dynamic and interactional nature of context, as well as its potential division into several analytic layers, are by now commonly accepted conceptualizations. Nevertheless, there is still divergence on other issues, the most complicated of which is the delineation of context. There are two main subordinate problems in this respect: (i) how to a priori or a posteriori decide which part of context is relevant to an interaction, and (ii) how to delimit the notion of “context” since, strictly speaking, any facet of human existence may in some way influence situated meaning-making. The first problem has commonly been cited in criticism toward earlier work in Critical Discourse Studies. It was alleged that by preemptively and theoretically characterizing power relations and dominant ideologies, any piece of text or discourse could be made “suspect,” evidencing power relations merely by virtue of the social theory adopted, while at times omitting extralinguistic contextual factors that would affect discourse (see, e.g., Billig 1999; Breeze, 2011). On the opposite side, Conversation Analysis has been criticized for

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

81

82

CEDRIC DESCHRIJVER AND ALEXANDRA GEORGAKOPOULOU

omitting relevant social and institutional features in their interpretation of a single interaction (see Schegloff 1997; Wetherell 1998, and associated debates). Yet, as discussed in Section 3.2.5, the macro/micro distinction that has for long constituted the topic of debate is itself increasingly brought into question, an evolution that may occur in other branches of discourse analysis. Indeed, Blommaert’s chronotopic view of context (Blommaert et al. 2020) builds on that of Silverstein (cf. Section 3.2.5) and lends itself to a comparison with “intertextuality” as a key notion within CDA. In its original formulation, the notion that texts contained multiple voices and historicities was described by Bakhtin (1981) as “heteroglossia.” Commenting on the “macro”/“micro” distinction regarding language and power in society, Fairclough argues that “the former [macro] are both conditions and resources for the latter [micro], and constituted by the latter” (1992: 211). This interrelation is intertextual: Since texts refer to, re-voice, and echo other texts, they inherently refer to specific sociopolitical environments and become relative to others. Rereading Fairclough in light of Blommaert’s diagnosis of the micro–macro issue (and considering Blommaert et al.’s [2020] concepts of “text trajectories” and “data histories”), the concept of “intertextuality” here seems to foreshadow a similar conclusion, though Fairclough does not explicitly reach it. Much like matrices of indexicality would endow localized and situated interactions with macroscopic relevance (in Silverstein’s/Blommaert’s terms), it would, according to Fairclough (1992), be intertextual properties of texts that mediate language and macroscopic context. In this sense, Fairclough (1992) anticipates the more recent views of context discussed in Section 3.2.5. When this line of reasoning is taken to its logical conclusion, context and its shifting, complex intertextual/indexical links may indeed come to saturate any text. An analysis of collected texts and their multiple, complex textual links within a certain discourse would then render a macroscopic description of context. This approach has recently been suggested by Hamann et al. (2019), who complement discourse analysis with the notion of “dispositif.” This Foucauldian concept “link[s] the heterogeneous textual and non-textual context conditions” of discourse production. Its operationalization “pay[s] attention to hierarchies and power structures in institutions, architectural order, regulations and laws, and scientific statements, and so on that together form a network of dispositional elements” (pp. 52–53). The possibility of meaning-making, it is argued, largely depends on these networks, that is, on the dispositif. It should be noted that this context-centered discourse analysis is necessarily interdisciplinary. In Hamann et al.’s (2019) case study of academic discourses, the authors go beyond textual analysis to rely on sociological study and interpretation, as well as on a statistical analysis of indicators of academic career trajectories. Given the preceding discussion, it is no surprise that “context,” defined in this way, is not entirely retrievable in each individual textual instance itself. No single text, after all, could possibly contain all varyingly explicit indexical links pointing to all relevant sociopolitical and institutional features of context.

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

3 Context in Discourse Analysis

Nevertheless, this type of analysis is valuable in terms of assessing recurring linguistic features in a particular institutional context, as it illuminates the social bases, and social consequences, of discourse. In this sense, it is also tightly bound up with the second abovementioned issue: that of where to delimit context. The question is whether a “‘free’ contextual account [that] has no obvious boundaries . . . might include a vast sociological, political and anthropological study” of any aspects deemed to be tangentially relevant (van Dijk 2006: 162). Additionally, and keeping in mind the recursivity of context (see Section 3.2.3), any broader (textual and discursive) description of indexically reproduced ideologies is vulnerable to recontextualization. Any recontextualization of these ideologies may then become part of the situated encounter, rather than the ideology primarily functioning as the historical/ indexical background. This recontextualization may hinder strong empirical grounds for claims of an overarching (language) ideology or indexical order. Yet, continued developments in communication may present new opportunities for further research on this issue, as will be explained in the next section.

3.5 Future Avenues for Research: “Mediatization” and Context in Discourse Analysis Of particular importance to future research on context within discourse analysis is the increasingly fragmented nature of communication, especially in online communication. Agha terms this phenomenon “mediatization”: a “commoditization of communication,” where language users are presented with greatly expanded uptake possibilities (including for contextual fragments) (Agha 2011; Deschrijver 2020a). This expansion may also cause a fragmentation of possible contextual features, or even “contexts,” to which participants orient, which in turns makes these less visible to the non-informed (or differently informed) participant or analyst. While earlier work has often characterized online communication as operating in a state of context collapse, with fewer cues being available to communicators to interpret context (Davis and Jurgenson 2014), it may just as well be argued that there is a great expansion of context, due to the fragmentation of the indexical links engendered by different interpretations of contextual cues. Indeed, as Blommaert et al. (2020) convincingly argue, context will continue to be engendered dynamically through participants’ situated encounters with text, containing highly adaptable and fragmented cues. In this respect, the non-retrievability of meaning from textual indicators alone also points to the irreducible role of multi-semiosis and multimodality, which is currently prompting discourse analysts to cross-fertilize with studies of multimodality, as well as expanding their notions of text and context. Jones talks about “polyfocality” (2004), the intricate layering and expansion of multiple co-occurring contexts in online discourse. Arguing that we “pay less attention to texts” (p. 29), Jones proposes a focus on a more widely conceived,

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

83

84

CEDRIC DESCHRIJVER AND ALEXANDRA GEORGAKOPOULOU

socially grounded view of interaction that instead pays attention to highly multimodal and semiotic intervention in meaningful displays of identity, common to computer-mediated communication. Similar concerns have been voiced by Schmitz, who argues that ‘the radical prototypical notion of text (as a self-contained, co-written document dealing with one particular subject)’ is no longer sustainable in an age of electronic communication (2014: 296). In the absence of a clearly defined notion of “text,” it is no surprise that the “text”–“context” interrelation also becomes problematic. Similarly affected will be the process of entextualization, which in its original conceptualization involves “making a stretch of linguistic production into a unit, a text, that can be lifted out of its interactional setting” (Bauman and Briggs, 1990). Through modern communication technologies, this process may become “elongated” (Jones 2009: 297), as texts may continuously garner new and varied uptakes far removed from the original entextualization. Here the issue of delineation of context also acquires new dimensions and prompts a need to clarify the role of each modality in constructing context and their interconnections, a project that would, in many ways, need to build on foundational works in linguistics, philosophy, and semiotics (see Tan et al. 2020: 264ff.). These modalities include embodiment and affect, which are currently being (re-)integrated into discourse analysis of context, picking up on Goffman’s earlier focus on these features of context (see Busch 2020). Though context online is fragmented, scattered, and heterogeneously shared, it has been convincingly shown that this does not present insurmountable problems for online interaction, as interaction only necessitates participants’ innate meaning-making capacities. Participants may strategize to circumvent and navigate context collapse constraints by, for example, doing audience selection, that is, tailoring their messages to address different subsets of their (intended) audiences, including ratifying some more than others (Tagg and Seargeant 2015; Georgakopoulou, 2017a). Similarly, recognizability of social actions (including requests, online trolling, responding to questions, etc.) may develop online among users who may not necessarily share much in the way of prior assumptions or understandings about the discourse, through repeated exposure to specific actions taking place in specific environments (Blommaert et al. 2020). From an emic perspective, in other words, fragmented communication does not present insurmountable problems. Of particular potential for further research is the logical consequence that it will become more difficult, from an etic perspective, to accurately identify the text trajectories and data histories evoked in a particular interaction (Blommaert 2015b). The fact that we may reach an analysis of the interpretation of online, fragmented discourse does not imply that we may reach a (correct) analysis of the historicized (language) ideologies that influence the communication, precisely due to the irretrievability of relevant features of context to one or more participants. This brings into focus the diffuse distinction between “context” (as in discourse-analytic settings above) and

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

3 Context in Discourse Analysis

the extra-textual knowledge analysts rely on to make sense of the interaction. Indeed, researchers themselves may become less immune to their own biases, ideologies, and practices of social media engagement: The algorithmic processes of distribution and amplification that characterize much online communication (Section 3.2.5) may cause the research to inadvertently become part of these same processes. Within discourse analysis, this warrants a deeper conversation about researcher reflexivity, which may aid in clarifying the assemblages that create meaning online (Georgakopoulou 2017b). In general, the increased difficulty of assessing context in online communication lends itself to future research to strengthen discourse-analytic claims about context, as exemplified in the following. It is by now increasingly acknowledged that the early twenty-first-century geopolitical international battlefield is partially waged online, as rapid changes in geopolitical circumstances occur concurrent to evolutions in mass communication, and various state actors may seek to spread their narratives or covertly infiltrate debate through new technologies. An important example here is the Chinese government and associated actors, who seek to influence global debate both “above board” (in newspaper opinion pieces, organizational contacts, etc.) as well as covertly, by, for instance, waging Twitter campaigns with multiple fake accounts, or overwhelming news outlet comment sections with specific discussion points (see, e.g., King et al. 2017; Zhang and Wallis 2021). Crucially, these operations are covert, algorithmically driven, and highly nontransparent to the non-informed observer or participant. In early 2021, for example, a campaign was waged against the British Broadcasting Corporation (BBC), based on their reporting on certain issues inconvenient to the Chinese state (Zhang and Wallis 2021). Yet the attack did not focus on the reported issues, as such, but rather wished to recast the BBC as spreader of “fake news,” thus tying in with recent liberal-democratic discourses on “fake news” as a danger for democracy. The use of a similar trope obfuscates the vastly different ideological-contextual background that influences the communication of participants such as Chinese state actors. As such, following Blommaert et al. (2020), while participants (and analysts) will, despite a lack of transparency, navigate through dense indexical networks and historical allusions and voices, they may still misidentify the ideological schemata underlying use of “fake news”. Where AI is involved, we may be dealing with synthetic approximations and reworkings of ideological schemata. This example highlights the fact that, for discourse analysts to make claims about (language) ideologies or indexical orders, it may be necessary to rely on knowledge that is outside of the interactions, and not represented by textual traces of indexicality alone. As additional ethnographic research would not be available in this example either, some a priori knowledge about the potential utterer and geopolitical context would be required to reach with reasonable certainty a conclusion about various aspects of context (including institutional and participant context). It is these “grey” or liminal areas that offer abundant opportunities for theoretical research, areas where entirely different (language-)ideological

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

85

86

CEDRIC DESCHRIJVER AND ALEXANDRA GEORGAKOPOULOU

backgrounds collide, where miscommunication and misinterpretation are quasi-inevitable, and where language as a means for communication is stretched beyond a single “communicative event” frame. Despite the massive scale of disinformation operations and the spread of AI, their effects on public opinion are by no means clear. Disinformation may provoke a backlash rather than successfully shift or divide public opinion, while specific language features of AI-generated texts may engender doubts about the text’s provenance and veracity. In this context, and echoing Gumperz’s research on miscommunications (see Section 3.2.3), there is ample scope to investigate the role of individual cognition, as well as the role of various modalities, in recognizing or resisting the manipulative use of tropes, or to investigate the contextual basis of potential resistance of prototypical text trajectories and indexical links. This may in turn illuminate their “regular” functioning. Similarly, this type of research may update analysts’ models of macroscopic context (i.e., dominant worldviews or (language) ideologies). Notions such as “fake news,” “democracy,” “(neo)liberalism,” “socialism,” “human rights,” “freedom,” and so on, are notoriously difficult to define, even within one specific context (see, e.g., “conspiracy theory” in Deschrijver 2021). The fact that their standard “denotation” is not homogeneously shared (Agha 2007) also makes them vulnerable to referential manipulation (see also Deschrijver 2020b, ‘Unstable Terms’), a commonly attested tactic in propaganda (e.g., Kampf 1987). For discourse analysis, the issue raised is one of the empirical bases to make claims about “macro” context (or indexical orders (Silverstein 2003)). For, if a particular description (such as “fake news”) is recognized and resisted by participants as a description that does not accurately reflect what “goes on” in interaction, it raises the question of the basis on which the researcher may apply this description. Claims about macroscopic context, in other words, should be attested by the data, rather than being imputed onto them.

3.6 Conclusion Various branches of discourse analysis have, as described in Section 3.2, reached key insights on context: It is (i) dynamic and interactive, (ii) analyzable through metapragmatic (para)linguistic features (e.g., contextualization cues and discourse markers), which may be recursively embedded, and (iii) analyzable in “layers.” More recent work has emphasized the historicized nature of context, its transportability across different communicative events, and the interconnections of ways of telling, sites of engagement and communicators with different types of positionality, as we showed with the example of the paradigm of small stories research. Finally, the fragmentation and difficulties in delineation of context have been highlighted vis-à-vis new forms of communication, especially on social media. Considering the future tasks sketched in Section 3.5, as well as the theoretical history of context in discourse analysis, the metapragmatic dimension of language use is a good candidate for future research. As described in Sections 3.2.2 and

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

3 Context in Discourse Analysis

3.2.3, metapragmatics has to a large extent defined analyses of context, from earlier 1950s text-context investigations, through to Gumperz’s contextualization cues and Schiffrin’s discourse markers, up to the most recent approaches to context. Its presence in a variety of different theoretical frameworks makes it a key candidate for further theoretical research. Metapragmatic awareness, by the same token, is present in any linguistic communication: Without a metapragmatic function there would be “no framework of structure” that would “regiment indexicals into interpretable event(s) of such-and-such type that the use of language in interaction constitutes” (Silverstein 1993: 37; and see Verschueren 2000). It can also serve as a point of entry into accessing context through discourse in more complex communicative events and situations. As explained in Section 3.5, important contextual aspects may remain covert. But despite this relative obscurity of what is going on, the analyst can rely on the knowledge that metapragmatic awareness underlies the production and reception of discourse, and any discursively mediated aspects of context. This can provide a first key step toward a grounded analysis. The metapragmatic function is not the only candidate for future research on context that will work on the social interactional–cognitive interface. Auer, for example, suggests that many contextualization cues “have a natural inherent meaning base . . . in some universal requirement of human interaction, of the working of the human mind, or of the articulatory and/or auditory mechanisms involved in speech production and perception” (1996: 24). Indeed, a universalist basis from which to investigate dynamically constructed context would be amenable to several theoretical orientations, including cognitive, sociolinguistic, and ethnographic ones. In resolving how the text–context interface functions, especially in highly complex forms of communication, finding what unifies human communication will be an essential step.

References Agha, A. (2007). Language and Social Relations. Cambridge: Cambridge University Press. Agha, A. (2011). Meet mediatization. Language and Communication, 31, 163–170. Auer, P. (1996). From context to contextualization. Links and Letters, 3, 11–28. Bakhtin, M. M. (1981). The Dialogic Imagination: Four Essays by M. M. Bakhtin (ed. M. Holquist). Austin: University of Texas Press. Bamberg, M., and Georgakopoulou, A. (2008). Small stories as a new perspective in narrative and identity analysis. Text and Talk, 28(3), 377–396. Bar-Hillel, Y. (1954). Indexical expressions. Mind, 63(251), 359–379. Bauman, R., and Briggs, C. L. (1990). Poetics and performance as critical perspectives on language and social life. Annual Review of Anthropology, 19, 59–88. Baynham, M. (2015). Narrative and space/time. In A. De Fina and A. Georgakopoulou (eds.), The Handbook of Narrative Analysis (pp. 117–139). Chichester: Wiley. Berry, R. (2005). Making the most of metalanguage. Language Awareness, 14 (1), 3–20.

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

87

88

CEDRIC DESCHRIJVER AND ALEXANDRA GEORGAKOPOULOU

Billig, M. (1999). Conversation Analysis and the claims of naivety. Discourse and Society, 10, 572–576. Blommaert, J. (2015a). Meaning as a nonlinear effect: The birth of cool. AILA Review, 28, 7–27. Blommaert, J. (2015b). Chronotopes, scales, and complexity in the study of language in society. Annual Review of Anthropology, 44, 105–116. Blommaert, J., and De Fina, A. (2017). Chronotopic identities: On the spacetime organization of who we are. In D. Ikizoglu, J. Wegner, and A. De Fina (eds.), Diversity and Superdiversity: Sociocultural Linguistic Perspectives (pp. 1–15). Washington, DC: Georgetown University Press. Blommaert, J., Smits, L., and Yacoubi, N. (2020). Context and its complications. In A. De Fina and A. Georgakopoulou (eds.), The Cambridge Handbook of Discourse Studies (pp. 52–69). Cambridge: Cambridge University Press. Breeze, R. (2011). Critical discourse analysis and its critics. Pragmatics, 21(4), 493–525. Bridges, J. (2017). Gendering metapragmatics in online discourse: “Mansplaining man gonna mansplain . . .” Discourse, Context and Media, 20, 94–102. Busch, B. (2020). Discourse, Emotions and Embodiment. In A. De Fina and A. Georgakopoulou (eds.), The Cambridge Handbook of Discourse Studies (pp. 327–349). Cambridge: Cambridge University Press. Clarke, I., Kwon, W., and Wodak, R. (2012). A context-sensitive approach to analysing talk in strategy meetings. British Journal of Management, 23, 455–473. Collins, J. (2011). Indexicalities of language contact in an era of globalization: Engaging with John Gumperz’s legacy. Text and Talk, 31(4), 407–428. Cook-Gumperz, J., and Gumperz, J. J. (2011). Commentary: Frames and contexts – Another look at the macro-micro link. Pragmatics, 21(2), 283–286. Davis, J. L., and Jurgenson, N. (2014). Context collapse: Theorizing context collusions and collisions. Information, Communication and Society, 17(4), 476–485. De Fina, A. (2013). Positioning level 3: Connecting local identity displays to macro social processes. Narrative Inquiry, 23(1), 40–61. Deschrijver, C. (2020a). Mediatized communication and linguistic reflexivity in contemporary public and political life. In A. De Fina and A. Georgakopoulou (eds.), The Cambridge Handbook of Discourse Analysis (pp. 687–707). Cambridge: Cambridge University Press. Deschrijver, C. (2020b). Metalinguistic density as an indicator of sharedness: The case of economic and financial terms in online interaction. Language and Communication, 71, 123–135. Deschrijver, C. (2021). On the metapragmatics of “conspiracy theory”: Scepticism and epistemological debates in online conspiracy comments. Journal of Pragmatics, 182, 310–321. Duranti, A., and Goodwin, C., eds. (1992). Rethinking Context: Language as an Interactive Phenomenon (pp. 1–42). Cambridge: Cambridge University Press. Fairclough, N. (1992). Discourse and text: Linguistic and intertextual analysis within discourse analysis. Discourse and Society, 3(2), 193–217. Fairclough, N. (1995). Media Discourse. London: Edward Arnold. Fetzer, A. (2004). Recontextualizing Context: Grammaticality Meets Appropriateness. Amsterdam: John Benjamins. Georgakopoulou, A. (2006). Thinking big with small stories in narrative and identity analysis. Narrative Inquiry, 16(1), 122–130.

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

3 Context in Discourse Analysis

Georgakopoulou, A. (2007). Small Stories, Interaction and Identities. Amsterdam: John Benjamins. Georgakopoulou, A. (2013). Building iterativity into positioning analysis: A practice-based approach to small stories and self. Narrative Inquiry, 23(1), 89–110. Georgakopoulou, A. (2017a). Friends and followers “in the know”: A narrative interactional approach to social media participation. In J. Mildorf and T. Bronwen (eds.), Dialogue across Media (pp. 155–178). Amsterdam: John Benjamins. Georgakopoulou, A. (2017b). “Whose context collapse?”: Ethical clashes in the study of language and social media in context. Applied Linguistics Review, 8(2–3), 169–189. Goffman, E. (1964). The neglected situation. American Anthropologist, 66, 133–136. Goffman, E. (1981). Forms of Talk. Philadelphia: University of Pennsylvania Press. Goffman, E. [1974] (1986). Frame Analysis: An Essay on the Organization of Experience. New York: Harper and Row. Goodwin, C., and Duranti, A. (1992). Rethinking context: An introduction. In A. Duranti and C. Goodwin (eds.), Rethinking Context: Language as an Interactive Phenomenon (pp. 1–42). Cambridge: Cambridge University Press. Gumperz, J. J. (1982). Discourse Strategies. Cambridge: Cambridge University Press. Hacking, I. (1999). The Social Construction of What? Cambridge, MA: Harvard University Press. Halliday, M. A. K., and Hasan, R. (1976). Cohesion in English. Hong Kong: Longman. Hamann, J., Maesse, J., Scholz, R., and Angermuller, J. (2019). The academic dispositif: Towards a context-centred discourse analysis. In R. Scholz (ed.), Quantifying Approaches to Discourse for Social Sciences (pp. 51–87). Cham: Palgrave Macmillan. Hanks, W. F. (1992). The indexical ground of deictic reference. In A. Duranti and C. Goodwin (eds.), Rethinking Context: Language as an Interactive Phenomenon (pp. 43–76). Cambridge: Cambridge University Press. Hanks, W. F. (1996). Language form and communicative practices. In J. J. Gumperz and S. Levinson (eds.), Rethinking Linguistic Relativity (pp. 232–270). Cambridge: Cambridge University Press. Hanks, W. F. (2006). Context, communicative. In K. Brown (ed.), Encyclopedia of Language and Linguistics, 2nd ed. (pp. 115–128). Amsterdam: Elsevier. Heritage, J. (1984). Garfinkel and Ethnomethodology. Cambridge: Polity Press. Hymes, D. (1974). Foundations in Sociolinguistics: An Ethnographic Approach. Philadelphia: University of Pennsylvania Press. Jaffe, A., ed. (2009). Stance: Sociolinguistic Perspectives. Oxford: Oxford University Press. Jones, R. H. (2004). The problem of context in computer mediated communication. In P. LeVine and R. Scollon (eds.), Discourse and Technology: Multimodal Discourse Analysis (pp. 20–33). Washington, DC: Georgetown University Press. Jones, R. H. (2009). Dancing, skating and sex: Action and text in the digital age. Journal of Applied Linguistics, 6(3), 283–302. Kampf, H. A. (1987). The challenge of Marxist-Leninist propaganda. Political Communication, 4(2), 103–122. King, G., Pan, J., and Roberts, M., E. (2017). How the Chinese government fabricates social media posts for distraction, not engaged argument. American Political Science Review, 111(3), 484–501.

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

89

90

CEDRIC DESCHRIJVER AND ALEXANDRA GEORGAKOPOULOU

Malinowski, B. (1923). The problem of meaning in primitive languages. In C. K. Ogden and I. A. Richards (eds.), The Meaning of Meaning (pp. 296–336). New York: Harvest. Martin, J. R. (2003). Cohesion and texture. In D. Schiffrin, D. Tannen, and H. E. Hamilton (eds.), The Handbook of Discourse Analysis (pp. 35–53). Malden, MA: Blackwell. Maschler, Y., and Schiffrin, D. (2015). Discourse markers: Language, meaning, and context. In D. Tannen, H. E. Hamilton, and D. Schiffrin (eds.), The Handbook of Discourse Analysis (pp. 189–221). Chichester: John Wiley and Sons. Ochs, E. (1992). Indexing gender. In A. Duranti and C. Goodwin (eds.), Rethinking Context: Language as an Interactive Phenomenon (pp. 335–358). Cambridge: Cambridge University Press. Reisigl, M., and Wodak, R. (2001). Discourse and Discrimination: Rhetorics of Racism and Antisemitism. London: Routledge. Schegloff, E. (1997). Whose text? Whose context? Discourse and Society, 8, 165–187. Schegloff, E. A. (1998). Reflections on studying prosody in talk-in-interaction. Language and Speech, 41, 235–263. Schiffrin, D. (1987). Discourse Markers. Cambridge: Cambridge University Press. Schmitz, U. (2014). Semiotic economy, growth of mass media discourse, and change of written language through multimodal techniques: The case of newspapers (printed and online) and web services. In J. Androutsopoulos (ed.), Mediatization and Sociolinguistic Change (pp. 279–304). Berlin: De Gruyter. Schneider, H. J. (1993). Ausprägungen pragmatischen Denkens in der zeitgenössischen Sprachphilosophie. In H. Stachowiak (ed.), Pragmatik: Handbuch pragmatischen Denkens (Vol. IV, pp. 1–37). Hamburg: Felix Meiner Verlag. Scollon, R., and Scollon, S. W. (2004). Nexus Analysis: Discourse and the Emerging Internet. London: Routledge. Silverstein, M. (1992). The indeterminacy of contextualization: When is enough enough? In P. Auer and A. Di Luzio (eds.), The Contextualization of Language (pp. 55–76). Amsterdam: John Benjamins. Silverstein, M. (1993). Metapragmatic discourse and metapragmatic function. In J. A. Lucy (ed.), Reflexive Language: Reported Speech and Metapragmatics (pp. 33– 58). Cambridge: Cambridge University Press. Silverstein, M. (2003). Indexical order and the dialectics of sociolinguistic life. Language and Communication, 23(3–4), 193–229. Tagg, C., and Seargeant, P. (2015). Facebook and the discursive construction of the social network. In A. Georgakopoulou and T. Spilioti (eds.), The Routledge Handbook of Language and Digital Communication (pp. 353–367). Abingdon: Routledge. Tan, S., O’Halloran, K., and Wignell, P. (2020). Multimodality. In A. De Fina and A. Georgakopoulou (eds.), The Cambridge Handbook of Discourse Studies (pp. 263– 281). Cambridge: Cambridge University Press. Tannen, D., ed. (1993). Framing in Discourse. Oxford: Oxford University Press. van Dijk, T. A. (2006). Discourse, context and cognition. Discourse Studies, 8, 159–177. Verschueren, J. (1999). Understanding Pragmatics. London: Edward Arnold. Verschueren, J. (2000). Notes on the role of metapragmatic awareness in language use. Pragmatics, 10(4), 439–456.

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

3 Context in Discourse Analysis

Wetherell, M. (1998). Positioning and interpretative repertoires: Conversation Analysis and post-structuralism in dialogue. Discourse and Society, 9, 387–412. Wodak, R. (2007). Pragmatics and Critical Discourse Analysis: A cross-disciplinary study. Pragmatics and Cognition, 15(1), 203–225. Wodak, R. (2011). Complex texts: Analysing, understanding, explaining and interpreting meanings. Discourse Studies, 13(5), 623–633. Wortham, S., and Reyes, A. (2015). Discourse Analysis beyond the Speech Event. Abingdon: Routledge. Zhang, A., and Wallis, J. (2021). Trigger Warning: The CCP’s Coordinated Information Effort to Discredit the BBC. Canberra: Australian Strategic Policy Institute.

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

91

https://doi.org/10.1017/9781108989275.004 Published online by Cambridge University Press

Part II

Philosophical, Semantic, and Grammatical Approaches to Context

Published online by Cambridge University Press

Published online by Cambridge University Press

4 Philosophy of Language and Action Theory María de Ponte, Kepa Korta, and John Perry

4.1 Introduction Language is paradigmatically a human activity, largely consisting of speakers saying things in order to inform, warn, misinform, threat, sell, and so on. Usually, the plans that motivate such utterances include being understood by others and having an effect on their behavior as a result. The use of language belongs squarely in what Frege called the causal realms, the physical and the mental. Language is important because it is a system for doing things. This suggests that the philosophy of action should be a part – a very important part – of the philosophy of language. To a certain extent it is. Charles Morris, in Signs, Language, and Behavior (1946), divided “semiotics,” the study of signs, into three parts: syntax, semantics, and pragmatics.1 Taking language as a system of signs, pragmatics is the study of how language is used, “how to do things with words,” as Austin says. Austin developed his theory of speech acts in the 1950s, and Searle and others have further criticized and developed it (Searle 1985). Speech act theory categorizes and describes actions in terms of what we do to perform them, and what we do in and by performing them. Perhaps by uttering a sentence I say something about a politician (a locutionary act); in saying that, I insult the politician (an illocutionary act); by insulting the politician I annoy his powerful friends (a perlocutionary act), and so on. All of this depends on syntax and semantics to get started, but it is basically pragmatics. A bit later, H. P. Grice (1975) introduced the concept of implicature, which is something conveyed by an utterance beyond the proposition it expresses, We are grateful to the members of the Zoom group of philosophical discussion for their constant inspiration. The first two authors acknowledge two grants, one by the Spanish Government (Grant PID2019-106078GB-I00 funded by MCIN/ AEI/ 10.13039/501100011033) and another from the Basque Government (IT1612-22). We are also grateful to an anonymous referee for their helpful comments. But most of all, the three of us thank the editor Jesús Romero-Trillo for his kind invitation and immense patience. 1 See also Morris (1938). The term “semiotics” is due to Ferdinand de Saussure (1916 [2011]) and was adopted by Charles Peirce (1931–1958); neither used it in exactly the same way Morris did.

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

96

MARÍA DE PONTE, KEPA KORTA, AND JOHN PERRY

beyond what the speaker says in uttering a sentence in a certain occasion. You ask me if I am going to a talk by X; I say “X is a complete idiot.” I have not said that I am not going but have implicated that I won’t. This, however, is not a matter of logic; I can consistently “cancel” the implicature by saying, “but I promised Y I would introduce X, so I’ll be there.” Grice intended his influential theory to be a contribution to semantics. He thought that by taking what is implicated to be what is said, we make mistakes about what words and sentences mean. His original targets were claims, “common-place of philosophical logic” (Grice 1975: 22),2 that the truthconditional analyses logic provides for words like “or,” “and,” “if” do not give their meaning in natural language. But the theory of implicatures, regardless of what Grice intended, is a pragmatic theory, a theory of conversation taken as a “collaborative effort” (p. 26), “a special case or variety of purposive, indeed rational, behavior” (p. 28) where participants observe, or can be expected to observe, rational principles and maxims for (cooperative) action. By and large, semantics itself was “utterance-free” until about the middle of the last century. Not that philosophers and logicians were unaware of utterances until then. But the working assumption was that semantics, the study of meaning and reference, should focus on what all utterances of an expression or sentence have in common, due to meaning, and not on how they differ, due to the particular facts of the utterance. In this chapter we first consider how this assumption was challenged when the philosophy of language began to think about natural languages, as opposed to what Frege called “perfect languages.” We discuss how, in the face of these challenges, attempts were made to finesse utterances, in the analyses of names, and indexicals, by creating what one might call “expression-hybrids,” particularly by David Kaplan. While appreciating the brilliance behind these attempts, we express reservations. Then we turn to our own theory, the reflexive-referential theory, which takes utterances as basic to the semantics and pragmatics of natural language.

4.2 Utterances Lost Discussion of utterances is not common in the philosophy of language developed in the first half of the twentieth century, at least in the analytic tradition inaugurated by Gottlob Frege and Bertrand Russell. The focus was the truthconditions of sentences, and how they were based on the meanings and references of its parts. It was not the utterances of sentences and their parts that were studied, but the expressions themselves, the meanings they have in virtue of the conventions of the language to which they belong, the references they have in virtue of the world, and the truth-values the sentence of which they are parts have in virtue of all of these things. 2

In earlier papers, the targets were claims by Ryle, Malcolm, Wittgenstein, Austin, Strawson, and others, about the meanings of expressions like “X looks φ to A” or “A tried to do x” (see Grice’s “Prolegomena,” in Grice 1989).

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

4 Philosophy of Language and Action Theory

In their early works, neither Frege nor Russell were particularly interested in natural languages except as the inspiration for better, more perfect, languages, like Frege’s Begriffschrift, or the language of Whitehead and Russell’s Principia Mathematica were intended to be. They hoped to use these languages to show that arithmetic could be reduced to logic, a doctrine Russell labeled “logicism.”3 While their efforts did not fulfill their logicist ambitions, they produced a revolution in logic that had a profound effect on the next generation of logicians and philosophers of language. Logic was the main interest of Frege and Russell, at least in his early years. Logic is centrally the study of valid inference. This suggests the activity of saying things that are true or are not and of drawing conclusions. But it was not these activities that were the focus, but rather the properties of sentences, considered as abstract objects, independently of their being spoken or written. How can the truth of one sentence, the conclusion, be guaranteed by the truth of others, the premises? Whether the premises are asserted, and if so by whom, when, and where, and whether the asserter is a reliable person with honorable intentions, is not of any interest. The question is whether what the truth of the premises requires of the world guarantees that the requirements for the truth of the conclusion will be met. So, although actual inferences are actions, with all sorts of causes and effects, it is validity as a property determined by structures of an ordered set of sentences that mainly fascinates logicians. And it was topics connected to this issue that dominated the philosophy of language for the first half of the twentieth century, following Frege’s writings on these topics and Russell and Whitehead’s watershed work, Principia Mathematica. Frege and Russell’s ideas about meaning, reference, truth, and the nature of the propositions gave philosophers of language a lot to think about. It was these ideas, often labeled the “Frege–Russell” theory, that dominated the philosophy of language for the first half of the twentieth century. They are still very much present in contemporary discussions, although their effects are somehow mitigated, or at least combined, with the effects of another “trend” that originated a few decades later, which placed the focus on the study of natural language. This trend, championed by philosophers like Strawson, Austin, and Grice, among others, was not concerned with logic (or not merely) or with the development of any formal language, but rather with understanding the ways in which ordinary languages work and the ways speakers use expressions and sentences to do different things. When the attention of many analytic philosophers turned to natural language, the philosophy of action began to intrude into the citadel of the philosophy of language.

3

The origins of the term “logicism” are not clear. Logicism has its origins in Dedekind’s work and, most importantly, on Frege’s. But neither Dedekind nor Frege, as far as we know, used the term. Russell started using it around 1919, and so did Carnap and others.

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

97

98

MARÍA DE PONTE, KEPA KORTA, AND JOHN PERRY

4.3 Reference: Natural Language Intrudes From the end of World War II on, philosophers of language paid increasing attention to natural language, and to the extent to which the language of logic can mislead us as to how the corresponding expressions are used in natural language. The later Wittgenstein at Cambridge; Urmson, Austin, Grice, Anscombe, and others at Oxford; Peter Geach at Leeds; Max Black, Norman Malcolm, and Keith Donnellan at Cornell; and many others around the world pursued this theme in various ways, seldom in complete agreement with one another. In common to all was the increased attention to action, “how to do things with words,” in Austin’s phrase. We start by describing the effects of this change of focus to the analysis of singular reference and, specifically, to the classic debates about definite descriptions (which are thought by some to refer, but not by all) and about names and indexicals (which are widely thought to refer to things). We pay special attention to indexicals, which might seem the natural candidates for the philosophy of action to intrude into the philosophy of language.

4.3.1 Definite Descriptions Definite descriptions were an early target of philosophers concerned with natural language. Peter Geach in “Russell’s Theory of Descriptions” (1950) and Peter Strawson in “On Referring” (1950) raised a number of objections to Russell’s theory. Geach focused on Russell’s example, “The King of France is bald,” Strawson on a slight variation, “The king of France is wise.” They raised a number of problems, but both focused on presupposition, a key factor in understanding descriptions in natural language. They both were quite clear that they were doing the philosophy of natural language, a somewhat different project from Russell’s. Geach said: “The incorrectness of Russell’s theory as an account of ordinary language in no way goes against it as a proposed convention of symbolism” (1950: 86). And Strawson also makes it clear that his target is natural language: “I think it is true to say that Russell’s Theory of Descriptions . . . is still widely accepted among logicians as giving a correct account of the use of such expressions in ordinary language. I want to show, in the first place, that this theory, so regarded, embodies some fundamental mistakes” (1950: 321). Geach says that Russell’s theory commits “the fallacy of many questions”: the question “Is the present King of France bald” involves two other questions . . .

(4) Is anybody at the moment a King of France? (5) Are there at the moment different people each of whom is a King of France? The question does not arise unless the answer to (4) is positive and the answer to (5) is negative. (1950: 84–85)

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

4 Philosophy of Language and Action Theory

Geach uses the term “presuppose” for what is going on. And this seems an apt term for the phenomenon Strawson describes: to use such an expression as “The king of France” at the beginning of a sentence was, in some sense of “imply”, to imply that there was a king of France. When a man uses such an expression, he does not assert, nor does what he says entail, a uniquely existential proposition. But one of the conventional functions of the definite article is to act as a signal that a unique reference is being made – a signal, not a disguised assertion. When we begin a sentence with “the suchand-such” the use of “the” shows, but does not state, that we are, or intend to be, referring to one particular individual of the species “such-and-such.” (1950: 331)

Strawson forcefully argued for a distinction between the properties that belong to sentences and expressions – meaning – and properties belonging to utterances or uses of sentences or expressions. It’s clear that it is the speaker who does the presupposing, by uttering a sentence at a certain time in a certain situation.4 If, in Oslo, I tell you “The King of Norway is wise,” I am presupposing that there is such a king. I also presuppose, or at least assume, that you know that Norway is a monarchy. Both presuppositions are acts; I perform these acts by uttering a sentence.

4.3.2

Names Proper names are a nightmare for semantics. If it weren’t for their use in calling the kids for dinner, I’d just as soon junk the whole category. (David Kaplan)5

A natural approach to ordinary proper names is to suppose that they are just words assigned to people and things, and it is those things that they contribute to the truth-conditions of the sentences in which they occur. “Francisco Franco was a dictator” is true because the person “Francisco Franco” stands for a dictator. Sometimes nobles have names that contain more information, but we can just call those “titles” and ignore them for the purpose of logic, a democratic enterprise. That seems to have been Frege’s attitude in the Begriffsschrift. The semantics of a sentence of the form “F(a)” is that it stands for the circumstance that the individual “a” refers to falls under the concept referred to by “F”. But then he saw a problem. Suppose “a” and “b” are two different names for the same object. Then “a = b” and “a = a” will refer to the same circumstance. This didn’t seem right. In the Begriffsschrift, he thought this was a problem with the identity sign “=”, which he jettisoned in favor of “≡”, which stood for a relation between names that had the same referent. But before long he realized that the problem was not with the identity sign, but with identity itself and with what 4

5

To be sure, Strawson’s concept of presupposition, characterized in terms of the conditions for the truth of a statement, is semantic. A pragmatic definition of presupposition, in terms of what the speaker does, is due to Stalnaker (1970). John Perry remembers Kaplan saying this, or words to this effect, on several occasions. We have not been able to find a textual reference.

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

99

100

MARÍA DE PONTE, KEPA KORTA, AND JOHN PERRY

he called “cognitive content” (de Ponte et al. 2021). Consider “Mark Twain wrote Huckleberry Finn” and “Samuel Clemens wrote Huckleberry Finn.” Both sentences are true, for “Mark Twain” and “Samuel Clemens” are two names for the same person. But a rational and semantically competent speaker might accept “Mark Twain wrote Huckleberry Finn” while rejecting, or at least having doubts about “Samuel Clemens wrote Huckleberry Finn.” At this point Frege made his distinction between sense and reference ([1892] 1948). Expressions have a sense assigned to them by the conventions of the language of which they are a part. The sense is a set of conditions that an object must fulfill to be the reference of the expression. Expressions with different senses can have the same reference. So “the first planet to appear in the evening sky” and “the last planet to disappear from the morning sky” both refer to Venus, which satisfies both conditions. Names are assigned senses which, at least in all the examples Frege ever gave us, can be captured with definite descriptions. So, in ancient Babylonia, the name “Phosphorus” might be assigned to the condition of being the last planet to disappear in the morning sky, while “Hesperus” is assigned to being the first planet to appear in the evening sky.6 The two names have different senses, but the same reference. As a result, sentences containing them have different cognitive content, or, as modern philosophers of language want to put it, different cognitive significance. So Frege’s distinction between sense and reference seems capable of dealing with differences in cognitive significance. Russell’s theory, which takes names to be abbreviated descriptions, would not have problems with it either. Until the 1950s and 1960s, there was a general consensus on how singular terms, like names and definite descriptions, refer. Definite descriptions refer by describing objects, and names refer by having descriptions associated to them, or by abbreviating those descriptions. But that consensus was broken by ideas championed by “referentialists,” as we will call them; philosophers like Ruth Barcan Marcus (1961), Dagfinn Føllesdal ([1966] 2004), Keith Donnellan (1966, 1970), Kaplan (1989), and, perhaps most importantly, Saul Kripke (1980), who argued that proper names were “rigid designators”: expressions that pick up an object independently of any description, and which pick up the same object in every possible world.7 The label “referentialism” encodes ideas about two different issues (Martí 1995): the mechanism of reference and the contribution of reference. According to referentialists, a name refers to an object by picking it out through an assignment of the expression to the individual: a tagging of sorts. There is no associated sense, description, or identifying condition. The mechanism of reference is a bit more complex and usually involves a causalhistorical process that links the name with the object. 6

7

We are cheating a little to make our life simpler, since the Babylonians did not know that Hesperus and Phosphorus were planets. See Wettstein (1986) who famously called these ideas “the new theory of reference.” As early as 1961, Føllesdal used the phrase “genuine singular terms” for rigid designators ([1966] 2004).

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

4 Philosophy of Language and Action Theory

Referentialism, at least with regard to the contribution of reference, is quite similar to the view defended by Frege in the Begriffsschrift. And, as Frege’s early view, it faces the problem of accounting for cognitive significance. For the referentialist view, “Mark Twain wrote Huckleberry Finn” and “Samuel Clemens wrote Huckleberry Finn” express the same singular proposition, one involving a person who happened to have two names, and that person having written Huckleberry Finn. But then, if they both express the same singular proposition, how is it that a person who knows that “Mark Twain wrote Huckleberry Finn” learns something when she is told that “Mark Twain” is “Samuel Clemens”? And what is it that she learns? Discussions of names, or singular reference in general, are usually regarded as discussions in semantics. But the problem of cognitive significance is very closely connected with pragmatics. The things we do with words, the speech acts we perform, and the effects these acts have on others, depends greatly on the singular terms we use to refer. Even if “Mark Twain” and “Samuel Clemens” refer to the same man (as does “the author of Huckleberry Finn”), the use of one or other name (or definite description) will have a significant impact on the situation in which they are used. Whether the speaker uses one name or the other depends on her communicative intention, and which name she uses will have a quite different effect on the hearer.

4.3.3 Indexicals In “On Denoting” Russell does not start with the example, “The King of France is bald” but rather “The present King of France is bald” (our italics.) There were many Kings of France, some no doubt bald. But there was no King of France in 1905. In the parts of the essay where Russell explains how the meaning of the expressions contribute to the meanings of the sentence in which the description is contained, he leaves out the word “present.” If we say “The present President of the US was born in Pennsylvania,” we identify Joe Biden by the job he has as we write, and predicate something true of him. “Present” seems to be an indexical, referring to the period of time we write, late in 2021. Similarly, if we say “In 2011 the present president was vicepresident,” our description refers to Joe Biden and says something true about him. In a later paper, “Mr. Strawson on Referring” (1957), Russell, answering Strawson, claims that he did not discuss “present” because he was concerned with definite descriptions and not with the “problem of egocentricity,” a problem, he claims, he had already discussed in previous works. To use his own self-citation, in Human Knowledge, he claims: When a word is not egocentric, there is no need to distinguish between different occasions when it is used, but we must make this distinction with egocentric words, since what they indicate is something having a different relation to the particular use of the word. (1948: 107)

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

101

102

MARÍA DE PONTE, KEPA KORTA, AND JOHN PERRY

Russell clearly acknowledged the role of indexicals, and their importance. And so did Frege, who admitted that indexical expressions pose a special problem for his theory of the sense and reference of sentences; that sentences containing indexicals require “the knowledge of certain accompanying conditions of utterance”; without those conditions, “the mere wording . . . is not the complete expression of the thought” ([1918] 1967: 24). Neither Russell nor Frege, however, developed their insights, and they are mainly of interest with respect to Frege or Russell interpretation. The fact remains that they did not develop the required semantic tools to deal with them. There seems to be no way that indexicals such as “I,” “now,” and “present” can be handled with the sparse semantical apparatus Russell provides. We have an expression, “present.” It means something, as Russell uses the terms “means,” in the sense that there is a period of time, the period in which we are writing this essay, that contributes to the truth-conditions of the sentence in which it occurs. But referring to this period of time is not a property of the word “present.” It is a property our utterance of the word “present,” the time at which that event occurred. Strawson points out that his distinctions among sentences/expressions, utterances of sentences/expressions, and uses of sentences/expressions are undeniable when we consider sentences containing indexicals. The sentence “I am hungry now” has a meaning that any competent speaker of English is attuned to. But the sentence, qua sentence, is neither true nor false, since the words “I” and “now,” qua words, do not have a reference. We need to consider utterances of the sentence – that is, particular acts8 of uttering the sentence – a speaker and its time of occurrence, to assign references and, eventually, a truth-value to the utterance. Indexicals – or, “egocentric particulars,” as Russell called them – were promptly admitted to be clear cases of content-sensitivity. In the 1970s Kaplan (1989) formulated an elegant formal theory of indexicals, based on intuitions about the meaning and truth of sentences containing them that are pretty undeniable. Basically, the meaning of indexicals – their “characters” – are functions from “contexts” to references of the sort their grammatical type requires. For Kaplan, the context of an expression is a quadruple of a (potential) speaker, location, time, and possible world. Suppose that Ethel says to Fred, in the afternoon of November 1, 2021: “You must fix dinner today. I have a paper to finish. And I fixed dinner yesterday. So it’s your turn. You’ll find the groceries here, in this bag.” And then suppose that the next day Fred uses the very same sentence to tell Ethel that it is her turn to fix dinner. In the first case the uses of “I” refer to Ethel. The uses of “you” refer to Fred. Uses of “today” refer to November 1, those of “yesterday” refer to October 31. “Here” refers to the kitchen, where Ethel is standing, and “this bag” refers to 8

As we use the terms, acts are particular events, actions are properties or types of such events. Agents perform actions at times, thereby producing acts. The act produced is often designated with a gerund phrase that specifies the agent, time, and action: “Fred’s swimming across the lake earlier this morning.” We basically adopt Goldman’s A Theory of Human Action (1970) account of action, with some changes in terminology. For more about the theory of action, see Anscombe ([1957] 2000), Bratman (1987), Davidson ([1980] 2001), and Grice (1971).

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

4 Philosophy of Language and Action Theory

the bag she is indicating at that moment. If, on the other hand, Fred had said the exact same thing to Ethel, while he was standing in the living room holding a different bag, the truth-conditions of his utterance would be quite different, even though the meanings of the words were the same. Indexicals might seem a natural place for the philosophy of action to stick its nose under the tent occupied by the philosophy of language. When we consider indexicals, it becomes clear that in many cases meaningful sentences with their meaning will not get us all the way to reference and truthconditions. It seems that, as Strawson suggested, we need to consider utterances of sentences and expressions. But Kaplan did not present his account as a move toward integrating the philosophy of language and the philosophy of action. His aim was to develop a logic of indexicals and demonstratives. He wanted to show the flexibility of logic, rather than its limits. The present authors think he succeeded in this endeavor. But in doing so, he raised issues that help us to understand the referential role, not only of indexicals and demonstratives in natural language, but also of names and definite descriptions, in the context of action. Any utterance has countless properties. But for the basic indexicals “I,” “here,” and “now,” three parameters suffice to determine reference: the speaker, the time, and the location. Call these the “bare context” of an utterance. An utterance of “I” refers to the speaker; and utterance of “now” to the time, and an utterance of “here” to the location. For “you” we need to bring in the audience, and perhaps those in the audience that speaker takes himself to be addressing. That is, the speaker intention is relevant. We need to add more and more facts to our utterance. But they all derive, it seems, from properties of the utterance. Indexicals and demonstratives determine reference and truth-condition relative to such contexts. These don’t have to be the speaker, time, and location and circumstances of any actual utterance. But if the speaker is in the location at the time, the context is proper. Consider the following sentence: (1)

You are right about everything we have ever disagreed about.

Let the speaker be John Searle, and the time be November 1, 2021. The location isn’t that relevant, but for the record, let us say it is a pub in Berkeley. The circumstances include John Perry being the person to whom the remark is directed. The sentence is true if, as of November 1, John Perry is right about everything he and Searle have ever disagreed about up to that point. As we write, it is late in the evening November 1, and John Perry is not in a pub in Berkeley talking to John Searle, pleasant as that would be. No utterance of (1), with Searle as speaker and Perry as audience has ever occurred, or is likely to ever occur. Still, we can assess the proposition that would be expressed with the sentence given these parameters. (It would no doubt be false.) So, if we take our subject matter to be sentences in such contexts, as Kaplan did, we can have a sensible concept of the logic of demonstratives and indexicals. A sentence is a logical truth, in the logic of demonstratives, if it is

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

103

104

MARÍA DE PONTE, KEPA KORTA, AND JOHN PERRY

true in any proper context. Kaplan’s example of a theorem in the logic of demonstratives: “I am here now.” An argument will be valid if relative to the same context; truth of the premises guarantees truth of the conclusion. So “You live in Palo Alto, Palo Alto is in California, so you live in California” is valid. And validity may be achieved by additional premises, relating parameters of the contexts. Kaplan took sentences-in-context as the subject matter, which sounds like the formal equivalent of possible utterances. But he did not put utterances at the center. Perry used Kaplan’s work to argue that there were profound difficulties in treating indexicals and demonstratives within a Fregean framework. Eventually he concluded that in order to understand the role of sentences with demonstratives and indexical in action, in particular the role of “I” in expressing self-knowledge, we needed to take a step further than Kaplan had and take utterances to be the primary subject matter. The result of this was the reflexive-referential theory.9 The main departure from Kaplan’s proposal is that, rather than modeling utterances in terms of expressions and contexts, the reflexive-referential theory brings utterances to the fore: it is, actually, a theory about utterances or, more generally, about cognitive episodes. Cognitive episodes are particular types of acts that happen during an interval of time that can be pretty short – in the case of utterances or perceptions – or relatively long – in the case of beliefs, desires, or intentions. We call them “cognitive” because they are intentional and all have contents, which encode the conditions that make them true, satisfied or fulfilled. We will be mostly concerned here with utterances, which are acts of using language, with different contents and properties, all of them causally relevant. By focusing on utterances, rather than modeling them away, we somehow divert from the tradition started by Frege and Russell and continued by Kaplan, and follow instead a tradition originated in the philosophy of language that took language as action, championed by John Austin and Paul Grice. This tradition, however, focused on the study of speech acts and implicatures, and left pretty much aside the study of issues regarding reference and cognitive significance. In other words, pragmatics started as far-side pragmatics – the study of issues that goes beyond saying, such as illocutionary acts and implicatures – leaving near-side pragmatics – the study of the facts that determine what the speaker says, such as reference – to semanticists.10 Austin–Grice is where philosophy of language meets the philosophy of action, producing pragmatics, and Perry is where pragmatically oriented philosophy of language comes to look back at issues of reference and cognitive significance or, if you prefer, where Austin–Grice turned to look back to Frege–Russell.

9

10

The reflexive-referential was developed in Perry’s Reference and Reflexivity (2001 [2012]), in Korta and Perry’s Critical Pragmatics (2011), and in Korta and Perry’s (2013). This chapter heavily relies on the ideas presented in Critical Pragmatics, where the idea of language as action and pragmatic aspects of the theory are extensively explored and discussed. See also de Ponte et al. (2020). On the distinction between near-side and far-side pragmatics, see Korta and Perry (2020).

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

4 Philosophy of Language and Action Theory

4.4 The Reflexive-Referential Account The reflexive-referential account focuses on episodes: utterances, thoughts, perceptions, acts, beliefs, desires, intentions, etc. These things occur in space and time and have causes and effects. We take the relevant utterances to be intentional acts and so call all such episodes “cognitive.” Cognitive episodes have contents. The truth-conditions of a statement or a belief are the paradigm case, but promises and intentions have fulfillment conditions, orders and desires have satisfaction conditions, and so forth. We use “contents” for all such conditions. Contents can be thought of as propositions that encode the conditions of truth for statements, conditions of satisfaction for expressions of hopes and desires, conditions of compliance for orders, and so forth. We take propositions to be abstract objects that are useful for classifying episodes in terms of their conditions of truth, satisfaction, and compliance. We agree with Frege that propositions (“Thoughts”) are not mental or physical episodes or states and so have no causes or effects. We do not, however, accept everything Frege says about propositions. We cannot, for instance, regard mental states, such as belief, desire, and the like, merely as relations to abstract objects, for beliefs and desires have causes and effects. We think that propositions are abstract objects that have been designed to be useful in classifying events in the causal realm. To be useful, such propositions need to be well-behaved and well-understood abstract objects, usually sets. The language that is used for propositions needs to be correlated with the elements of utterances and thoughts that determine their truth-conditions. Sentences play a major role in identifying these elements in the case of utterances and thoughts that can be expressed with sentences. On the reflexive-referential account cognitive episodes have multiple consistent layers of content that can be classified with propositions.11 Consider the follow example. Robyn and Fred are old friends, but up to now they’ve always lived in different cities. Today, however, Robyn has just moved into Fred’s neighborhood in San Francisco. Suppose this is all happening on June 12, 2021, in a particular neighborhood in San Francisco, say, in the Mission. Robyn knocks on Fred’s door and says to Fred, “Today you and I are neighbors.” Call that utterance u. Then on many accounts, Robyn said the proposition P: that Robyn is Fred’s neighbor on June 12, 2021. On a simple account this proposition is a set consisting of the relation of x being a neighbor of y at t; and the sequence of Robyn, Fred, and June 12, 2021. One might take it to be the set of possible worlds in which Robyn be Fred’s neighbor on that day. Or perhaps the set of worlds w in which Robyn’s counterpart in w be a neighbor of Fred’s counterpart in w on that day, or perhaps its counterpart. Indeed, there are many kinds of abstract objects philosophers have taken to be propositions. 11

Somehow confusingly, the reflexive-referential explanation has been labeled as “multi-” or “pluri-propositionalism.” But, since in our account propositions are not the (main) constituents of cognitive episodes, but rather abstract tools to classify the variety of contents, we prefer avoiding this label.

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

105

106

MARÍA DE PONTE, KEPA KORTA, AND JOHN PERRY

We’ll stick with the simple account and assume that the reader gets the general idea. We agree that the proposition P is a truth-condition of Robyn’s utterance. But we don’t agree that it is the truth-condition. On our account, attributions of truth-conditions are ascribed to an utterance relative to what is taken as given. The truth-condition of an utterance, relative to a set of facts g, is what else has to be the case for the utterance to be true, given g. With regard to Robyn’s utterance, we gave you the facts that determined the references of the parts of his utterance. “I” referred to Robyn, “you” referred to Fred, “today” referred to June 12, 2021, the language was English, so “are neighbors” refers to the relation, between two or more people, of living near to. Given all these facts, what else has to be the case for Robyn’s utterance to be true is P: that Robyn is Fred’s neighbor on June 12, 2021.

Now suppose we didn’t bother to tell you when this all happened. Do you know conditions of truth of Robyn’s utterance? Of course you do. Robyn’s utterance is true iff P′: there is some day d during which the utterance occurred, and Robyn and Fred were neighbors on d.

Suppose we also didn’t tell you who answered the door. Then you know that the utterance is true iff P″: there is a day d and a person y such that Robyn was talking to y at the time of the utterance, and Robyn and y were neighbors on that day.

Note that P′ and P″ put conditions on the utterance itself: we call them reflexive or utterance-bound truth-conditions. In contrast, P puts conditions only on the items referred to by the utterance. P gives the referential truthconditions of the utterance. P could be true if the utterance had never been made. But not P′ and P″. These three propositions are different and independent. But they are consistent. P′ follows from P″ when we add the fact that the door answerer was Fred. P follows from P′ when we give the date. Having different propositions, classifying different and incremental truthconditions, is useful for many things. To begin with, of course, it gives a comprehensive account of what needs to be the case for a particular utterance to be true and of the several layers included in what is said by it. With this explanation at hand, it is quite easy to deal with puzzles in philosophy of language, such as the puzzles mentioned above involving names. Let us look at a couple more examples, to see how it works. Bob puts a big platter of kale before his young children and says: “You will love this kale.” He knows his children have never seen this ugly vegetable before. But he thinks they will learn the name for this vegetable from his utterance. They will realize that his utterance is true iff there is some stuff Bob is referring to with “this kale” and they will love it. They realize that “kale,” in

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

4 Philosophy of Language and Action Theory

the singular phrase “this kale,” is most likely a name of the kind of stuff he is pointing at. That is, they know how English works. They do not know this in the way people who have taken a linguistics or a philosophy of language class know it. They may not know what an utterance is, or what referring is, or what a singular term is. But they are attuned to these facts about English and are capable of making inferences from them, without being able to articulate the premises of their reasoning. What explains their inference is an utterancebound content of Bob’s utterance, given the meaning of the sentence and the fact, conveyed by glance and gesture, that he was referring to the green stuff on the platter. Given all that what else had to be the case for Bob’s utterance to be true? That “kale” is a name for the ugly green stuff. Note that in this case Bob succeeds in teaching his children something, that “kale” names kale, even though what he says is almost certainly false. John tells Frenchie he is going to San Sebastián in the Basque Country for a week. She looks at the map of the Basque Country in John’s office for “San Sebastián” but cannot find it. He says, “San Sebastián is Donostia.” Frenchie finds “Donostia” on the map, points to it, and says, “That’s where you are going?” and John says “yes.” John has conveyed to Frenchie that “Donostia” and “San Sebastián” are names for the same city. But what he said was simply that San Sebastián was Donostia, a necessary truth, which is the same thing he would have said with “San Sebastián is San Sebastián” and “Donostia is Donostia”, if what is said is always to be identified with referential content. But those utterances would not have been helpful. A philosophical rule of thumb should be that utterances of identity sentences can almost never be understood simply in terms of their referential content. If you insist on doing that, you will be lucky to get much of anything right. The reflexive-referential theory not only accepts and accommodates the intrusion of the theory of action into the philosophy of language but also takes the theory of action as fundamental to understand how language works. We claim that a theory like this, with utterances at the center and considering them as episodes, that is, as cognitive acts with a variety of contents, is better equipped to deal with the context-sensitivity of language than theories that take (one single) content to be a property of sentences plus contexts.

4.5 Utterance versus Sentence-in-Context Kaplan discusses utterances to motivate his theory, but he does not bring them into his theory as such; they are replaced by, or perhaps modeled as, pairs of expressions and contexts: “expressions-in-context.” Kaplan’s main interest was developing a logic of demonstratives and indexicals. For this purpose he regarded utterances as an unnecessary complication. For one thing, a pair of context and expression can have a content, even if the speaker of the context does not utter the expression at the time and place of the context in the world under consideration. More importantly, utterances take time; the validity of an

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

107

108

MARÍA DE PONTE, KEPA KORTA, AND JOHN PERRY

argument with one hundred steps might depend on the context being the same for all of them, but we can’t talk or write that fast. So, for logical purposes, utterances can get in the way. From the point of view of understanding the relations between the contents of states and their causal roles, however, it is very helpful to have episodes that have contents, have causes, and have effects. So, in a theory of language framed in a theory of action, episodes and, in particular, utterances are too important to ignore, in spite of the complications they pose for logic. For this purpose, we consider the elements of Kaplan’s contexts to simply be properties of utterances, which objects fill the roles of speaker-of, timeof, and location-of. The fact that utterances have speakers and occur in locations at times clearly inspired Kaplan’s concepts of context and character. We promote these inspiring episodes to first-class status. The chief advantage of this view is simply that it accounts for – and makes use of – the fact that utterances have many other properties, in addition to having speakers, locations, and times, that can be relevant to understanding their cognitive significance. Consider Gricean intentions, which are basically intentions to produce concrete utterances in order to achieve certain effects on the hearer by the recognition of those intentions. The meaning of the sentence and expressions uttered, the content uttered, and the elements of Kaplan’s context all contribute to that goal. But they are not enough. Conversations, linguistic transactions, fundamentally involve a speaker, who intends to do something and achieve certain effects by uttering a sentence or an expression, and a hearer, who perceives an event (or a token of such event) and needs to reason about the intentions behind it. Properties associated with meaning and content are key to guessing intentions, but not only. On many occasions, the hearer has to interpret what the speaker is trying to do by producing an utterance, and not only what she says. The hearer, that is, needs to know if the speaker intends to command something, amuse, impress, mislead, etc. One way to handle these, while sticking with Kaplan’s approach, is to add more members to the context set, or to introduce additional sets. The latter is more or less the approach of Jon Barwise and John Perry in Situations and Attitudes (1983). On their “relational theory of meaning,” the meaning of a sentence is taken to be a relation among various situations connected to an utterance, although the utterance itself is, as in Kaplan’s theory, only modeled and not introduced directly into the theory. The basic relation is between the utterance situation, which determines the speaker, location, and time, and, in lieu of propositions, described situations. But various other situations are added to the range of the relation, to deal with names, descriptions, ambiguity, and other phenomena. On the reflexive-referential account, however, the treatment of such factors is simpler and more straightforward. They are all properties of the utterance, recognized as necessary to handle various phenomena. The reflexivereferential theory inherits a key insight of Kaplan’s theory – and Perry’s earlier

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

4 Philosophy of Language and Action Theory

views – but the inclusion of utterances gives it, at least, one important advantage.12 The inherited key insight is the distinction between different ways in which information can be discovered, believed, and asserted. The advantage is, briefly, that the cognitive significance of an utterance for different hearers can depend on the perceptual and causal relations the hearer has to the utterance, which reference-determining facts they know, and how they think of them. To account for this, we need to bring in additional properties of the utterance and, in particular, causal properties. These are not modeled by Kaplan’s expressions-in-context. This advantage is particularly relevant for us here, since it offers a solution to the cognitive significance problem without having to renounce to the referentialist insights.

4.6 Cognitive Significance Consider a classical puzzle in semantics. In his important essay “Has Semantics Rested on a Mistake?” (1986), Howard Wettstein pointed out that whatever virtues of what he called the “Kaplan–Perry” account has in explaining cognitive significance of cases involving indexicals, it does not handle Frege’s Begriffsschrift ([1879] 1967) problem, the origin of worries about cognitive significance, which involves proper names. “Hesperus is Hesperus” and “Hesperus is Phosphorus” clearly have different cognitive significance; one learns from the second that the names co-refer, but not from the first. On a directly referential account of proper names, which is more or less what Frege had in the Begriffsschrift, this is hard to account for. On Kaplan’s account, the character of a proper name is a constant function, from any context to the bearer of the name. The two sentences have the same content, a singular proposition to the effect that Venus is Venus. So, whatever the virtues of the Kaplan–Perry thesis for cases involving indexicals, it does not help with proper names. In response to Wettstein’s essay, Perry (1988) introduced the concept of the proposition created by an utterance, in contrast with the proposition expressed by an utterance, which was basically the distinction between reflexive and referential content. On the reflexive-referential account, the two utterances have different reflexive truth-conditions. The first is true if and only if there is an object named by “Hesperus” which is self-identical. The second is true if and only if there is such an object, and there is also an object named by “Phosphorus,” and the objects are identical. So, even though the referential truth-conditions of the two utterances are the same, they differ in cognitive significance, in virtue of their different reflexive truth-conditions.13 12

13

There is second important advantage, which concerns Prior’s well-known root-canal example and his utterance of “Thank goodness that’s over!” See de Ponte et al. (2020) for a discussion of both advantages. See also Corazza and Korta (2015).

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

109

110

MARÍA DE PONTE, KEPA KORTA, AND JOHN PERRY

Consider now a puzzle involving reference to time. This puzzle involves dates, which are akin to names, and temporal indexicals, such as “today” or “tomorrow.” It is quite common, at least for some of us, not to know the day one lives in. In this case, however, we certainly know when we live; we know that what is happening is happening now or today, even though we do not know what time is “now” or what day is “today.” Also, it is relatively common to be wrong about the date one lives in. One can be in a dateless haze without knowing it; that is, one can be quite certain about the date, but be wrong. In such a case, it is natural to think, in retrospect, that one might have been correct; even that one’s false belief was justified. This provides another advantage for the reflexive-referential theory with respect to Kaplan’s. Suppose John and Dan are planning to go to the Giants game on August 22, 2021. The day before, John types out a reminder: “The Giants game is tomorrow. Don’t forget.” But he forgets to hit the “send” button. He notices that the message has not been sent just before retiring and hits the button. But he doesn’t notice that it is already after midnight. Dan, a dateless-hazer, sees the reminder, “The Giants game is tomorrow,” when he wakes up on August 22, and sees that date on the email heading. He knows that the game is on August 22, and reasons, given John’s notorious reliability, “Today must be August 21.” He then immerses himself in linguistic esoterica until late in the evening, when John calls and says, “You missed the game!” Dan thought, on August 22, that it was August 21. He had good reason for this belief. The game was scheduled for August 22. John said, in an email this morning, that the game was tomorrow. John is very reliable. Therefore, today must be August 21. Could Dan have been right? It seems not, because for his belief to be true, August 22 would have to be August 21, which is not possible. Surely had Dan said, (4) August 22 is August 21. we could diagnose some kind of irrationality (or assume he is making some subtle linguistic point). But if he just says, as he did: (5) Today is August 21. on August 22, this does not seem like the right diagnosis. There is a way in which way Dan’s utterance could have been true: if it had occurred on August 21 rather than August 22. In other words, if the role of time-of for the utterance of “today” had been filled by some moment occurring on August 21, the utterance would have been true. On the expression-in-context approach, this does not seem like an option. Since the pair of expression and context is individuated by its members, we would not have the same pair if the time of the context was August 21. So, we have another advantage for the reflexive-referential theory and for granting utterances first-class status in the semantics of tense and indexicals.

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

4 Philosophy of Language and Action Theory

There is a possible objection to our strategy, as Richard Vallée has pointed out.14 One theory of events is that they are individuated by the time at which they occur, by where they occur, and by which object and properties involved. If we accept this account of event individuation, the reflexive-referential account is no better off than the expressions-in-context account. But we reject this account of the individuation of events. In our opinion, it is a plausible account of the individuation of facts, but not of events. Being a fact is a property of whatever one takes to serve as possibilities, be it circumstances, state of affairs, or whatever. We will not here develop an account of events, which we regard as very basic elements of reality. But we think an adequate account must allow for counterfactuals of the form, “if the election had occurred two weeks earlier, Clinton would have won,” or, to follow with our example, “if Dan’s utterance had occurred on August 21, he would not had missed the Giant’s game.” It is a fact that he missed it and that he made his utterance on August 22. Nothing can change that. But the episodes involved, Dan’s belief and Dan’s utterance “Today is August 21” could have been true, had they occurred on August 21. In his paper “Frege on Demonstratives” (1977), Perry introduced the example of Heimson, who thought he was David Hume. Let’s suppose instead that Heimson thought he was Bob Dylan, which will make it easier to make a case for his rationality. Here is the background story. Heimson falls, hits his head, and has amnesia as a result. He does not know who he is. He carries no identification. He awakes in a hospital where no one has any idea who he is. Heimson decides to figure out who he is. Heimson’s amnesia is of a rather peculiar sort; he retains “third-person” memories about lots of people, he simply does not remember which of them he is. He assumes he is one of the people about whom he knows a great deal. He notices that he knows all of Bob Dylan’s songs by heart, the date of every concert where he performed, and loads of other things. He also knows a lot about a fellow named “Heimson,” but not nearly as much as he knows about Dylan. He decides he is Dylan, and thinks, with some confidence, “I am Bob Dylan.” His thought cannot be true. Indeed, it seems necessarily false. But it might be rational. And we think it does get at a possibility, even if a rather remote one. The possibility is found at the reflexive level. Call his thought – the event of thinking, the episode, not its content – T. If Bob Dylan had the thought T, rather than Heimson, T would be true. This is the possibility that Heimson’s sifting of the evidence available to him led him to think this was the case. David Lewis (1979), considering Perry’s original example, comes to the opposite conclusion. In “Frege on Demonstratives,” Perry advocated a version of the Kaplan–Perry view, that he called the two-tiered view. In a nutshell, to deal with the attitudes we need to recognize two levels of content for beliefs. What is believed is a proposition, often a singular proposition. How it is believed corresponds to character (Kaplan) or role (Perry). 14

Personal communication.

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

111

112

MARÍA DE PONTE, KEPA KORTA, AND JOHN PERRY

Lewis had nice things to say about this account but thought that the level corresponding to character or role could serve as what is believed, and the upper tier could be jettisoned. Lewis noted that a character, a function from contexts to propositions, could be regarded as a property: The property an agent has at a time iff the character, applied to that agent and time, yields a true proposition. So, when JP says, “I am sitting,” he “self-ascribes” the property Psitting: Psitting = the property x has iff the character of “I am sitting,” with arguments x and t, yields a true proposition, that is, the property someone x has at time t, iff x is sitting at t.

On Lewis’s view, properties, rather than propositions, are the true “objects” of beliefs. A belief consists of an agent at a time self-ascribing a property. Lewis’s view, like the Kaplan–Perry approach, does not involve episodes. We have agents, times, the relation of self-ascription, and properties. Lewis (1979, footnote 16) regards singular propositions and “de re” beliefs as unnecessary intrusions into the theory of the attitudes based on preoccupation with the analysis of our customs for reporting beliefs. Lewis’s view incorporates much of the traditional picture that belief is a relation between an agent at a time and a proposition. His innovation is to replace propositions with properties. But we think the idea of objects of beliefs in this sense is a mistake. A belief consists in an agent at a time being in a mental state, an episode. This episode has truth-conditions, which can be characterized by propositions, many different propositions, depending on what is taken as given. But neither propositions, nor characters, nor characters construed as properties, are objects of belief in the sense that belief consists of a relation to them. Propositions, in our sense, are tools we use to characterize and keep track of the truth-conditions of the episode. Lewis’s account, like the Kaplan–Perry account, does not have episodes, utterances or beliefs, as elements. So, on Lewis’s view, Heimson’s belief, whether he thinks he is Hume or thinks he is Dylan, cannot be true in the strong sense that there is no possible way it could be true. On his view, the belief consists of Heimson, the relevant time, and the attitude of self-ascription to the property of being Hume/being Bob Dylan. Since there is no possibility that Heimson at the time has that property, there is no way the belief can be true. Heimson is irrational in all possible circumstances.

4.7 Conclusion In this chapter, we have offered a brief and undoubtedly partial account of the ways an account of language as a human activity intruded in the philosophy of language during the past century, focusing, in particular, on the discussion of singular reference. As a way of confronting the need to embrace this intrusion, we presented and defended the reflexive-referential account, and we discussed

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

4 Philosophy of Language and Action Theory

an example of the beneficial impact of action theory on the philosophy of language. The conception of language as action inaugurated by Austin, Grice, Strawson, and others not only brought new pragmatic theories of language and novel theoretical tools, such as the distinction between locutionary, illocutionary, and perlocutionary aspects of utterances or the systematic study of nonliteral speech via the notion of implicature, but inspired new philosophical theories, like the reflexive-referential theory or critical pragmatics, to deal with classic puzzles in the philosophy of language like reference and cognitive significance.

References Anscombe, G. E. M. [1957] (2000). Intention. Cambridge, MA: Harvard University Press. Barcan Marcus, R. (1961). Modalities and intensional languages. Synthese, 13(4), 303–322. Barwise, J., and Perry, J. (1983). Situations and Attitudes. Cambridge, MA: The MIT Press. Bratman, M. (1987). Intention, Plans, and Practical Reason. Cambridge, MA: Harvard University Press. Corazza, E., and Korta, K. (2015). Frege on subject matter and identity statements. Analysis, 75(4), 562–565. Davidson, D. [1980] (2001). Essays on Actions and Events. Oxford: Clarendon Press. de Ponte, M., Korta, K., and Perry, J. (2020). Utterance and context. In T. Ciecierski and P. Grabarczyk (eds.) The Architecture of Context and Context-Sensitivity (pp. 15–29). Cham: Springer. de Ponte, M., Korta, K., and Perry, J. (2021). Four puzzling paragraphs: Frege on “≡” and “=.” Semiotica, 240, 75–95. de Saussure, F. [1916] (2011). Course in General Linguistics. New York: Columbia University Press. Donnellan, K. S. (1966). Reference and definite descriptions. The Philosophical Review, 75(3), 281–304. Donnellan, K. S. (1970). Proper names and identifying descriptions. Synthese, 21, 335–358. Føllesdal, D. [1966] (2004). Referential Opacity and Modal Logic. New York and London: Routledge. Frege, G. [1879] (1967). Begriffsschrift, Eine der Arithmetischen Nachgebildete Formelsprache des Reinen Denkens. Trans. S. Bauer-Mengelberg, Begriffsschrift, a Formula Language, Modeled upon that of Arithmetic, for Pure Thought. Halle a. S.: Louis Nebert. Repr. in Van Heijenoort (1967), pp. 1–82. Frege, G. [1892] (1948). Sense and reference. Trans. Max Black of Frege (1892) Über Sinn und Bedeutung. The Philosophical Review, 57(3), 209–230. Frege, G. [1918] (1967). Der gedanke: Eine logische untersuchung. Trans. A. Quinton and M. Quinton, The thought: A logical inquiry. Beitra¨ge zur Philosophie des deutschen Idealismus, I (1918–1919), 58–77. Repr. in Strawson (1967), pp. 17–38.

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

113

114

MARÍA DE PONTE, KEPA KORTA, AND JOHN PERRY

Geach, P. T. (1950). Russell’s theory of descriptions. Analysis, 10(4), 84–88. Goldman, A. I. (1970). A Theory of Human Action. Englewood Cliffs, NJ: Prentice-Hall. Grice, H. P. (1971). Intention and uncertainty. Proceedings of the British Academy, 57, 263–279. Grice, H. P. (1975). Logic and conversation. In P. Cole and J. L. Morgan (eds.), Syntax and Semantics. Vol. III: Speech Acts (pp. 41–58). New York: Academic Press. Repr. in Grice (1989), pp. 22–40. Grice, H. P. (1989). Studies in the Way of Words. Cambridge, MA: Harvard University Press. Kaplan, D. (1989). Demonstratives. In J. Almog, J. Perry, and H. Wettstein (eds.) Themes from Kaplan (pp. 481–563). Oxford/New York: Oxford University Press. Korta, K., and Perry, J. (2011). Critical Pragmatics: An Inquiry into Reference and Communication. Cambridge: Cambridge University Press. Korta, K., and Perry, J. (2013). Highlights of critical pragmatics: Reference and the contents of the utterance. Intercultural Pragmatics, 10(1), 161–182. Korta, K., and Perry, J. (2020). Pragmatics. In E. N. Zalta (ed.) The Stanford Encyclopedia of Philosophy. Metaphysics Research Lab, Stanford University, Spring edition. Kripke, S. A. (1980). Naming and Necessity. Cambridge, MA: Harvard University Press. Lewis, D. (1979). Attitudes de dicto and de se. The Philosophical Review, 88(4), 513–543. Martí, G. (1995). The essence of genuine reference. Journal of Philosophical Logic, 24 (3), 275–289. Morris, C. (1946). Signs, Language and Behavior. Englewood Cliffs, NJ: PrenticeHall. Morris, C. W. (1938). Foundations of the theory of signs. In O. Neurath, R. Carnap, and C. Morris (eds.), International Encyclopedia of Unified Science (Vol. I, pp. 1–59). Chicago: Chicago University Press. Peirce, C. S. (1931–1958). Collected Papers of Charles Sanders Peirce. Cambridge, MA: Harvard University Press. Perry, J. (1977). Frege on demonstratives. The Philosophical Review, 86(4), 474–497. Perry, J. (1988). Cognitive significance and new theories of reference. Noûs, 22(1), 1–18. Repr. in Perry (2000), pp. 189–206. Perry, J. (2000). The Problem of the Essential Indexical and other Essays. Stanford: CSLI. Perry, J. [2001] (2012). Reference and Reflexivity, 2nd ed. Stanford: CSLI Publications. Russell, B. (1948). Human Knowledge: Its Scope and Limits. New York: Simon and Schuster. Russell, B. (1957). Mr. Strawson on referring. Mind, 66(263), 385–389. Searle, J. R. (1985). Expression and Meaning: Studies in the Theory of Speech Acts. Cambridge: Cambridge University Press. Stalnaker, R. (1970). Pragmatics. Synthese, 22, 272–289. Repr. in Stalnaker (1999), pp. 31–47. Stalnaker, R. (1999). Context and Content. London: Routledge. Strawson, P. F. (1950). On referring. Mind, 59(235), 320–344.

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

4 Philosophy of Language and Action Theory

Strawson, P. F. (1967). Philosophical Logic. London: Oxford University Press. Van Heijenoort, J. (1967). From Frege to Gödel: A Source Book in Mathematical Logic, 1879–1931. Cambridge, MA: Harvard University Press. Wettstein, H. (1986). Has semantics rested on a mistake? The Journal of Philosophy, 83 (4), 185–209.

https://doi.org/10.1017/9781108989275.005 Published online by Cambridge University Press

115

5 A Functional Approach to Context Rebekah Wegener and Lise Fontaine

5.1 Introduction The aim of this chapter is to highlight and explain the principal ways in which context is situated within the framework of Systemic Functional Linguistics (SFL) and to explore the main themes and parameters which situate context within an integrated theory of language as a semiotic resource. SFL has viewed context as integral to linguistic theory, rather than an optional extra or a complementary add on. In Butler’s review of three structural-functional theories, he states (2004: 164) that SFL is “the only functional theory to have built in a specific model of social context, through the parameters of field, tenor and mode.” However, not only has context been a core concept in the development of the theory, it also has explanatory power through the SFL concepts of realization, instantiation, and stratification. Within the SFL framework, “context is regarded as explaining the form of the language system as a whole. The clusterings of choices in the language system are hypothesised to derive from clusterings of contextual features” (Berry 2017: 43). As we will show below, the relationship of realization more accurately captures the relationship between context and language, since one is not, strictly speaking, derived from the other in any directional sense. For Halliday, “every act of meaning has a context of situation, an environment within which it is performed and interpreted” (1979/2002: 201, emphasis in original). Not only is language use shaped by context but also context is shaped by language. Within SFL, context refers to all aspects of a social (communicative) process that have an impact on meaning and shape the outcome of that social process, including the material situational setting, the topic, the modality, and the relationships between the social actors. This view of context is situational and is defined primarily in terms of three parameters (sometimes referred to as variables): the Field of discourse, the Tenor of discourse and the Mode of We are very grateful to Anne McCabe and to our anonymous reviewer for comments on earlier drafts.

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

5 A Functional Approach to Context

discourse. Each is described within SFL as follows. Field represents the social context, seen as “the nature of the social process in which the text is embedded – ‘what is going on’” (Halliday 1981/2002: 227); Tenor includes social relationships and roles, where a text is viewed as “the interpersonal relationships among the participants – ‘who are taking part’” (p. 227); and Mode concerns the symbolic organization of language, including “the role assigned to the text, including both medium and rhetorical function – ‘what part the language is playing’” (p. 227). How these parameters relate to semantics and the lexicogrammar within Halliday’s framework will be discussed below. It is important to note at this point that the field of SFL scholarship is broad and while context will be represented in this chapter following Halliday, his is not the only framework within SFL. For example, Fawcett (1980) developed a different place for context in his model. While the concepts are all very similar in many ways, they do not take up the same place in the framework. For Fawcett, context is a type of knowledge (or information) held by the speaker, and in this sense there are different contexts. For example, Fawcett’s model has a built-in distinction between “context of situation” (recoverable from the observable), “context of co-text” (recoverable from preceding text), and “context of culture” (recoverable from long-term memory). This representation brings Fawcett’s model much closer to the models of context put forward by Philipp Wegener or even Breal (see Nerlich 1990 or Nerlich and Clarke 1996). Fawcett’s approach to context is perhaps more cognitive and individualistic than Halliday’s more semantic and social one. It is, however, true that the concept of context in Fawcett’s model has not been developed as explicitly as that of Halliday’s. See Fawcett (2015) for an overview of his account of context. While it may be the case that the SFL framework brought context into its theory of language from a very early stage, it was not independent of strong influences both from other linguistic theories and from other disciplines such as anthropology and sociology. This chapter will begin in Section 5.2 with an overview of the historical development of context as a concept within SFL theory. Following this, Section 5.3 examines its place in the SFL theory of linguistics, including how the model accounts for its interaction with language. In Section 5.4 we consider some of the challenges with respect to studies of context: (i) how we can reconcile the place of context within a theory of linguistics and (ii) whether the theory predicts different contexts for different modalities when working with multimodality. Finally, we conclude the chapter with a brief summary and what we feel are key directions for future research in this area.

5.2 Historical Development of Context within SFL As a functional approach to language, SFL has its theoretical roots very firmly in the European Functionalist tradition (Halliday 1978), which many argue differs significantly from the American Functionalist tradition including in key

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

117

118

REBEKAH WEGENER AND LISE FONTAINE

assumptions about the nature of language and linguistic evidence (for more on this debate, see Dirven and Fried 1987 or Newmeyer 2001). While the European tradition is typically traced back to the Prague School,1 clearly the crucial assumptions of European functionalism go back much further and did not arise out of nowhere with the Prague School (Daneš 1987). What is important for understanding a functional approach to context is the assumptions that these approaches make about language, the nature of linguistic data, and the fundamental role of linguistics as a discipline, since this shapes how theories and models are built and how they are applied in research. Halliday’s functionalism takes its systematicity from Saussure and Hjelmslev, but with a distinct cybernetics flavor that results in system networks as a means of representation that model both open and closed systems. Halliday (1995) is quite clear in relating this to the debate about language evolution: Since language is a fourth order system, at once physical and biological and social, as well as having its own special character as semiotic, the total context must be in terms of theories about systems of all these types. It will be clear that the ideological antecedents of my discourse lie not in the formal grammars and truth-conditional semantics of the latter part of the century, but in a more functionally-oriented linguistics: that of Sapir and Whorf, Malinowski and Firth, Bühler, Mathesius and Trubetzkoy, Hjelmslev, Benveniste and Martinet, among many others.

Functional perspectives on language generally assume that language is a tool used as an instrument to facilitate communication or social interaction. The primary concern is thus on language in use and on better understanding language process and structure, which presupposes modeling both the speaker and the listener.2 In this view, how language is actually used in communicative contexts becomes crucial. Modeling context, therefore, must be integral to functional approaches to language and all functional models of language must account for context to some extent. However, as Butler and Gonzálvez-García (2014) argue very few functional models have a detailed model of context, with SFL being a notable exception in this regard. Perhaps the most significant influence on SFL in terms of context has been Malinowski, who has typically been considered to belong to the macrosociological or anthropology tradition. However, Malinowski’s account should not be underestimated. In his preface to Malinowski’s book, Frazer (1922: ix) explains that his method “takes full account of the complexity of human nature. He sees [human nature], so to say in the round, and not in the flat.” While many at that time considered that “pure sociology should confine itself 1

2

According to François (2018: 7), its beginnings can be traced to 1929 with the Prague group and Frei’s La grammaire des fautes: Introduction à la linguistique fonctionnelle, assimilation et différenciation, brièveté et invariabilité, expressivité (our translation: “The grammar of errors: Introduction to functional linguistics, assimilation and differentiation, brevity, invariability and expressivity.” See Polguère (2015) for an argument in favor of modeling primarily the speaker in a meaning-driven functional model of language (Meaning-Text Theory, MTT).

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

5 A Functional Approach to Context

to the description of acts” (Frazer 1922: ix), it is clear that Malinowski saw the purpose of sociology as being centrally concerned with meaning. In viewing the object of study “in the round,” we see that Malinowski draws from both the micro and the macro in his establishment of the scope of modern anthropology. The SFL approach has tended to relate context very closely to the grammar and, as will be discussed below, the concept of metafunction (Butt and Wegener 2008), but also through a kind of distributed semantics that allows for a fuzzy boundary between grammar and semantics. Coming to functionalism through his mentor Firth and proto-pragmatists like Wegener, Whittney, and Breal (Nerlich and Clarke 1996: 248) and directly influenced by Malinowski’s modeling of situated interaction, Halliday’s “context of situation” has always had a distinct place alongside other kinds of linguistic statements. As Halliday (1991/2007) argues, when we arrive at the level of a given context of situation, we are already in the culture – hence, there is no need to proceed to culture. Rather, we have the task of elucidating what we find there in the typical-actual, as Firth referred to it. Halliday’s own practice in this regard appears to be cautious: While he arranges his investigations with respect to field, tenor, and mode, the variables seen as relevant for any given account of context/text stay in close proximity to the register under description. Firth’s work with Malinowski in London, and Firth’s own fieldwork from Kenya to Afghanistan (Rebori, 2002), confirmed the significance of context and incorporated the notion into a polysystemic, relational theory of choices at many levels. Halliday (1973) integrated “context of situation” into a linguistic theory that included a separate semantic stratum and a more abstract notion of function – “metafunction.” Context of situation became the interface between language and the socio-material order (or, more correctly, between language and the dimensions of the socio-material order that are of importance to the processes of meaning in a given instance). Firth’s ideas on context were extended by Ellis (1966), who provided some clarifications, and Mitchell (1957) who applied it to an actual community. Ure (1969) brought a typological order to the linguist’s work with semantic varieties, or registers. Work by Hasan (1978, 1984, 1999), in particular, has elaborated and extended the Hallidayan approach by conceptualizing the stratum with explicit motivations for the contrasts within the three major systems of field/tenor/mode, and by developing the systematization of the semantic stratum (crucial to activating the descriptive power of context, through realization and hence, inevitably, in terms of delimitation in specific descriptions). Halliday’s work represents an important extension to Malinowski’s legacy. Everything in a cultural context may be functional and therefore – in that sense – meaning bearing. Halliday is also, however, demonstrating, through the polysystemic mapping of semantic choice, that function (albeit of an abstract kind) provides the optimum way to understand the internal relations

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

119

120

REBEKAH WEGENER AND LISE FONTAINE

of a language system, not just the externalized tasks to which it is employed. A realizational model is a mapping of patterns of patterns, all of which occur together on different levels of abstraction – from social configuration to modalities of manifestation (e.g., in sound or writing). This evolved “happening together” is a key to the power of languages for extending their potential as systems (or for their speakers to so extend them). Languages are not encodings, but encodings of encodings of encodings (see Wegener 2011 and Butt and Wegener 2008). While the story of context within SFL developed from the historical roots as just described, it has not remained static. Other theorists, in particular Martin (1985, 1993, 1997), have extended the stratification above context to genre, and on to ideology, in order to treat variation through Gleason’s (1965) notion of “agnation” (albeit here, an issue of context and of the levels “above” context, not of grammatical variants). There have been numerous other significant contributors to the discussion of context in the SFL paradigm, for example Gregory and Carroll (1978) and Ventola (1983), who emphasized the flowing and dynamic nature of context, a notion that we see in Berry’s early work but also recently (2016a), O’Donnell (1999, 2021), Bartlett (2016) and Moore (2020). Others have picked up Hasan’s early representations of context as system networks and developed system networks for field, tenor, and mode (Butt 1999/2004; Bowcher 1999; Hasan 1999). Berry (2016b) and Wegener (2011, 2016) have worked on context for situated interaction and in particular have focused on new divisions for context. In the next section, we turn our attention to the relationship between context and language, focusing on the way in which context is situated within the SFL theory of linguistics.

5.3 Context and Language We might want to ask, as does Hasan (2014: 47), whether “the study of context deserves the attention and care being demanded here3” since, as she explains, “[i]n modern linguistics until quite recently context had been viewed as an optional extra.” For SFL, the answer is clearly “yes,” since it is one of the few functional theories to have both centralized the concept of context and made of it “an analytical level of description” (Bowcher 2018: 2). The concept of context is theoretically motivated within the SFL framework. Context, as Halliday (2013: 34) argues, is considered to belong “within the domain of a linguistic theory.” Although central to the theoretical framework, context is extralinguistic for Halliday. Critically, for SFL, context is semiotic and its relation(s) to language is explicit, as will be shown below.

3

Of course Hasan’s use of “here” (there in the context of her writing) is deictically different from our use of “here,” but we feel it is relevant given this volume dedicated to context.

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

5 A Functional Approach to Context

However, we must be clear that context itself, as a semiotic construct, is not “out there,” as in outside the speaker, in any material sense as we might imagine would be the case of the physical surroundings for a speech act. A discussion of semiotic context, as concerns our purposes here as compared to material context, is beyond the scope of what we will consider (but see Bowcher 2018 for a detailed discussion of these two types of context within the SFL framework). While it might be considered outside the language system, context is not independent of language use; it is connected to the language system. Halliday (1991/2007: 281) explains the interaction as follows: How do you set about “creating” a context for language? You cannot do it by means of legislation, like decreeing that poems are to be written in praise of a national leader. The only way is for the text itself to create its own context of situation.

This sense of context being created from text makes clear the inherent relationship between context and language use; language construes context. This perspective on context is summarized very nicely by Berry (2017: 43): SFL is a theory of language as choice; the view is that it is impossible to make sense of linguistic choices without reference to context and also that it is impossible to make sense of context without reference to linguistic choices. Linguistic choices are strongly influenced by the contexts of the situation in which they occur, but in certain circumstances linguistic choices may bring about a change in the context, for instance leading it to become more formal or more informal.

In this sense, we might gloss the relationship between context and language as dialectical in the SFL framework. The definition and place of context within the theory is interpreted somewhat differently by different scholars (see, e.g., Martin 1992; Wegener 2011; Matthiessen 2015; Bartlett 2017; Tam 2017; and various chapters in Fontaine et al. 2013). However, there is general agreement concerning the significance of four theoretical concepts which contribute to defining, describing, and parametrizing context. These concepts, which we will now explore, are stratification, realization, instantiation, and construal. Within SFL, stratification is a key pillar within the theory, one of five main dimensions (Wegener 2011). A stratified language system is one which “consists of different coding levels” (Taverniers 2011: 1102), where the relationship between the strata is accounted for by realization (see below). It is through the stratified system that language becomes a semiotic system, i.e., one which creates meaning. Stratification is sometimes discussed in terms of abstraction, i.e., higher levels are more abstract than lower ones; for example, context is more abstract than lexicogrammar, although this notion of abstraction is not entirely problem free (for a useful discussion of the concept of abstraction within the theory, see Williams et al. 2017). The strata are labeled (from more abstract to more concrete) as follows: Context , Semantics , Lexicogrammar , Expression. This organization is

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

121

122

REBEKAH WEGENER AND LISE FONTAINE

Extralinguistic Interface context (eco-social environment)

semantics lexicogrammar

Form of the content

Content plane Figure 5.1 A Functional approach to context within a theory of linguistics (adapted from Halliday 2013)

typically represented in SFL literature by a series of embedded circles, each representing a particular stratum.4 Context is always illustrated as the highest level in order to represent “how the stratified linguistic system is ‘embedded’ in context” (Halliday and Matthiessen 2014: 25). We illustrate the stratification of the content plane in Figure 5.1, which is adapted from Halliday (2013). In the SFL model, semantics is the stratum which serves as an interface between context and lexicogrammar. Halliday (2013: 34) explains this stratification as follows: “Somewhere between the lexicogrammar and the context – the significant eco-social environment – we recognise an interface stratum, that of semantics, Hjelmslev’s ‘content substance.’” Figure 5.1 represents only a partial view, we have excluded the expression plane (e.g., phonology or graphology) as it does not directly concern our current discussion; however, see Taverniers (2011) for more detail about the full representation. As a stratified layer in the theory of linguistics within SFL, Figure 5.1 can be explained as follows: Context is realized as semantics, semantics is realized as lexicogrammar.5 Equally though, as noted above, text construes context. The relationship between the strata is therefore bidirectional, rather than deterministic. The concepts of realization and construal are discussed in more detail below, as is the dynamic nature of context, but for now we continue with the consequences of stratified context within a theory of linguistics. As mentioned above, each stratum constitutes a different coding level. Context is organized by the parameters of Field, Tenor, and Mode of discourse, commonly referred to as the variables of register (see Lukin et al. 2011 and Bowcher 2017). Semantics is organized metafunctionally, i.e., the broad set of functions that language serves. Taverniers (2019) offers an in-depth discussion of semantics as an interface stratum, providing considerable insight into how it contributes to language as a meaning-making system. In brief, there are three 4

5

While the view represented here can be considered standard in the sense of most commonly used, there are some considerable differences as concerns the strata of context and the place of genre, indeed whether or not genre as a concept is needed. For a very useful overview of this issue, see Matthiessen et al. (2019). Although not shown here, we could continue and add that lexicogrammar is realized as phonology (see Taverniers 2011 and/or Lukin 2017).

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

5 A Functional Approach to Context

main metafunctions: Experiential, Interpersonal and Textual. The metafunctional organization of language at the semantic stratum as expressed in the clause is well discussed throughout the SFL literature, perhaps most effectively and most recently by Berry (2019). Lexicogrammar represents a “construct of wording” (Halliday 1981/2002: 221); in other words, the form of the content in wordings where lexis and grammar are continuous, inseparable. The organizing principle here is that of rank, with the unit of the clause as the central or core unit. While the concept of rank is largely unproblematic in its broadest sense, i.e. that smaller units make up larger ones and vice versa, the concept of the rank scale as used in SFL is open for some debate (see Fontaine and Schönthal 2019; Berry 2017; and Fawcett 2000). For our purposes here, it suffices to view rank as an organizing principle of the lexicogrammar stratum where the clause, the main grammatical unit, is said to consist of phrase units, phrase units consist of word units, and word units consist of morpheme units. As mentioned above, the relationship between context and language, as different strata, is that of realization. For Halliday (1992: 352) realization is an “interstratal relationship, meanings are realized as wordings, wordings realized as sound (or soundings).” Realization is the principle that accounts for how “patterns on one level both construe and construct patterns on another level” (Cassens and Wegener 2008: 208). While realization may, at first glance, seem somewhat similar to the relations described above for the rank scale, it is not, as Wegener et al. (2008) explain, a “consists of” type of relation. Attempts have been made within the SFL literature to account for the relation between context and metafunction by means of a so-called context-metafunction hookup hypothesis, or CMHH (see Hasan 1995). However, as Berry (2017: 47) points out, despite a lack of evaluation, this hook-up hypothesis is generally accepted as fact. The main problem according to Berry (p. 47) is as follows: “if contextual features are established via semantic features and semantic features are established via contextual features, the reasoning is circular and the hypothesis becomes untestable.” While there is no space to consider these issues here, Berry (2017) lists several works for the interested reader, including Thompson (1999), Clarke (2013), and Berry (2016b). In Figure 5.2, which situates context and language in the SFL framework, we can see that realization is the principle accounting for the relationship between context and language along the stratification dimension. This means that context is “realised in / construed by” the stratum below (Halliday 1991/ 2007: 275), which, in the stratified model given in Figure 5.1, is semantics. The same can then be said for semantics, i.e., that semantics is realized in/construed by lexicogrammar, and so on. Therefore, an inherent aspect of realization is the SFL concept of construal. Construal,6 as a theoretical term in SFL, is what gives meaning to the instances of language. For example, an imperative clause such as eat your 6

Construal as used in SFL is conceptually and operationally different from the use of this term in other functional frameworks. See, for example, Langacker (2019).

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

123

124

REBEKAH WEGENER AND LISE FONTAINE

instantiation

SYSTEM

realization

CONTEXT

context of culture (cultural domain)

LANGUAGE language as system

(register)

INSTANCE

context of (situation situation type)

(text type)

language as text

Figure 5.2 Language and context, system and instance (adapted from Halliday 1991/2007: 275)

vegetables is said to construe a command. This command, in turn, would be said to construe a feature of the context parameter of Tenor, for example construing an unequal power relationship (e.g., social hierarchy). Lukin et al. 2011 provide a detailed discussion of construal as it relates to realization between the strata of context and semantics. Construal is, then, the process of meaning-making; we might think of it as a kind of inference, or interpretation. According to Taverniers (2011: 1122), construal refers to “the relationship between language and extralinguistic reality” and is therefore an essential component of inter-stratal realization. Hasan (2010: 12) explains this relationship as follows: “In the reception of the utterance, realisation is construal of the relevant choice at the higher level: thus, in decoding an utterance, the choice in wording construes meaning, the choice in meaning construes context.” Here we can see that realization and construal are nearly equated in terms of, in the one sense, relating the wording to the semantics, but we also find simultaneously that the wording construes meaning, i.e., it is meaning-making. For further discussion of construal and how it relates to other concepts touched upon briefly here, see Berry (2017). Having considered realization and construal in relation to the stratification of context, we also note in Figure 5.2 a relation between context as culture and context as situation, which are differentiated by the principle of instantiation. In contrast to the inter-stratal relationship of realization, instantiation is an intra-stratal relationship of the move between the system and the instance (Halliday 1994/2002: 352). Taverniers (2011: 1120) explains that this principle “concerns the relationship between a schema and an instance (a particular usage/an instantiation) of this schema.” Berry (2017: 53) describes instantiation as “a relationship between the theory and the data.” In this view, context of culture as a theoretical instantiation holds a significant place in the SFL model of context in that, as Matthiessen (2015: 34) explains, the two perspectives on context are differentiated by instantiation: “context of culture is located at the potential pole, context of situation at the instance pole.” For Halliday (1991/2007: 282) “the culture is construed by systems of language choice; the situation is construed by patterns of language use.”

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

5 A Functional Approach to Context

However, he warns that in talking about these two aspects of context, we must not slip into thinking that these are two different things. The representation in Figure 5.2 of the relationship between system and instance is meant to show one and the same object of study from two different perspectives. The system perspective offers one of meaning potential, in other words “the potential that lies behind all the instances” (Halliday 2002: 10). Space does not permit a detailed exploration of culture here but see Lukin (2017) for a more detailed discussion of context of culture together with its relation to ideology. For our purposes here, we can think of instantiation as a type of abstraction, where instance (here context of situation) is more concrete, and potential (here context of culture) is more abstract. As Williams et al. (2017: 14) explain, “The notion of potentials is fundamental in systemic theorising, because it allows the linguistic model to apply not just to previously encountered linguistic utterances, but also to additionally try to explain how speakers produce and interpret novel instances of language.” The view we now have of context in the SFL framework is that context is realized in language, and also construed by it. As O’Donnell (2020: 73) explains, “Context of situation is thus defined in relation to language, rather than being defined only in relation to the social system itself.” As a stratified layer in the framework, the relationship between context and language is mediated by semantics. Halliday’s (1991/2007: 282) claim that “text and situation come into being together” can be interpreted as text and context co-constructing. Context is not deterministic, nor is it static or stable. This perspective on context is captured very well by O’Donnell (2021) as follows: I argue that at each point of a text/interaction, we as interactants have the choice to affirm the contextual expectation, or to vary from it, either using novel means to achieve some situational goal, or by shifting to a distinct Context of Situation (as when the speaker in a conference talk makes smalltalk with someone in the audience). Rather than focusing on the context-forms -text-with-dynamic-exceptions approach, I argue we should be taking the approach such that every act creates its own context, which sometimes is coherent to the context created by prior acts of the interactants.

We have focused on the strong connection that the SFL notion of context makes with language. The theory views the language system as stratified. In this sense, we find interacting layers or strata which account for how meanings are construed in contexts of language use. Context, while viewed as extralinguistic, is nevertheless implicated in the theoretical linguistic framework by the realization principle which provides the account of how context relates to lexicogrammar and at this interface is where we find the stratum of semantics. In the next section, we examine this framework for context in more detail by considering two aspects which pose theoretical challenges. In doing so, we also provide a sketch of important directions for future research and development.

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

125

126

REBEKAH WEGENER AND LISE FONTAINE

5.4 Two Challenges in Working with Context While in this section we give our attention specifically to the perspective of SFL, the two areas we discuss are in some ways common to all functional approaches and applications, at least to some extent. Of particular interest for us is (i) where context sits in relation to language and (ii) what the relationship is between context and different modes of meaning-making behavior.

5.4.1 Where in the Model is Context? As we have just discussed, language is modeled in SFL as a stratificational model, where the different strata are related through the principle of realization, rather than a causal relationship, since for Halliday (1994/2002b: 358) realization is best treated as a dialectic relation. A text is created by its context, the semiotic environment of people and their activities that we have construed via the concepts of field, tenor and mode; it also creates that context. The relationship that we refer to as realisation between levels of semiosis – situation (doing) realized in semantics (meaning), semantics realized in lexicogrammar (wording), and so on – is a dialectic one involving what Lemke (1984) interprets as n-order metaredundancies. A semiotic event is an event on many levels. (Halliday 1994/2002a: 254)

Realization is theoretically abstract in nature, i.e., the strata in the model do not exist in any concrete sense, and realization is typically seen as the dialectal relationship between these strata, i.e., an inter-stratal relationship. But there is also an intra-stratal use of the term to “refer to any move which constitutes a link in the realisational chain, even one that does not by itself cross a stratal boundary (for example, features realized as structures)” (Halliday 1994/2002b: 352). Hence realization is a shorthand for the entire chain or sequence of relations involved in realization (both inter- and intra-stratal), which is crucial for a stratified representation of language. SFL’s stratified model incorporates both the levels of the expression plane (briefly mentioned above) and the content plane, which includes lexicogrammar (wording and structure), semantics (the meaning system), and context (culture and situation – elements of the social structure as they pertain to meaning). Context is not language, so it is not treated as part of the stratal arrangement of language; however, it does stand in a realizational relationship to semantics, which is part of the stratal arrangement of language. For SFL, it is part of the development of the adult language system that both the content and the expression plane become stratified (Halliday and Matthiessen 2004). Thus, as adults we have access to a meaning potential which is able to “expand, more or less indefinitely” (Halliday and Matthiessen 2004: 24), allowing us to variously construe our experience and our social relations. In this way, uniqueness is built in, although it is a very socially oriented

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

5 A Functional Approach to Context

uniqueness (hence original rather than individual), since to be shared it must be coded (Hasan 1999). What this means is that in order to share our internal world with others we must encode it in a form that others can recognize and understand, at least to some extent. Halliday and Matthiessen (2004) suggest that it is not only the content plane which is stratified, but the expression plane as well. Thus, where content is split into semantics and lexicogrammar, the expression plane divides into phonology and phonetics, “separating the organizing function from the function of interfacing with the environment” (p. 25). One consequence of this thinking is that if we recognize that stratification takes place for both the content and the expression planes, we might consider whether further stratification is possible at the context plane. Do we find an organizing function which is separate from the function of interfacing with the socio-material environment? If there is stratification of the contextual plane, then we would have a stratum with an interfacing function which makes contact with the socio-material world (the socio-material setting), much as is the case with the semantics stratum, and another one with an organizing function, i.e., organizes context. In many respects this is how context of culture and context of situation are often used, although the relationship between culture and situation is one of instantiation, as we see in Figures 5.1 and 5.2. For those who model context as systems, the relation of instantiation (as set out in Figures 5.1 and 5.2) becomes more problematic as Hasan (2004: 175) argues: If context of situation is to context of culture as text is to the system of language, then, by analogy, so far as context theory in SFL is concerned it is like having the theory of text, but without an ability to show its relation to the theory of language system.

For Hasan, then, by explicitly relating context to lexicogrammar, Halliday had made the realizational chain very complete at one end, but with a lot of missing links at the other end. While Hasan is clearly correct in this observation, it could be argued that this is in keeping with the deliberately fuzzy notion that Halliday has of context and language. Halliday views context as everything that is necessary to fully elucidate meaning, but also as the causal explanation of variation in language behavior. Therefore language is embedded in context, and context is an integral part of language, which largely follows from Firth, who, according to Halliday (2003: 243): consistently maintained that all study of language was a study of meaning; and meaning was function in a context, where the “context” was located both in the various strata of language itself and in its situational and cultural environment – all of which came within the compass of linguistic theory.

For purely practical purposes, this fuzzy representation can be difficult to actually work with, because it forces us to ask: How do you model context as both distributed among different strata and contained within a specific stratum? If we consider Figure 5.2 again, we can see that Halliday is positing

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

127

128

REBEKAH WEGENER AND LISE FONTAINE

a realization relation between the abstract system of language and the abstract system of culture, and the instantial relation to situated language. He proposes that if there is a dialectal relation between language and culture, then we can expect that there will be realization statements. But there is no reason to expect that these realization statements will be of the same form or nature as realization statements between other strata, since the system is not symmetrical. As long as you concentrate your attention on the core of the linguistic system, on linguistic form (grammar and vocabulary), then the interrelationships that you are studying are – or can be treated as if they were – wholly bounded within language, since their immediate points of reference are also within language: on the one hand the semantic system, and on the other hand the phonological system. But once you become concerned with the linguistic system as a whole, including the semantic system, then you have to look outside language for your criteria of idealization. (Halliday 1974: 82)

What we can glean from Halliday’s claim here is that to fully understand meaning, we have to go outside of language, but going outside language is not the job of the linguist. The implications of this claim are twofold: Firstly, it seems likely that the linguist alone will not be able to fully “solve” meaning. Secondly, by extension, meaning is best modeled as a transdisciplinary problem that needs to be viewed from multiple perspectives. To address the questions about meaning that he was interested in, Halliday employed Bernstein’s (1971) theorization for the following two reasons: 1. Bernstein’s work is “a theory of the social system with language embedded in it” (Halliday 1974: 83), and 2. It examines the function of language in the social system. Halliday’s idealization is motivated by a view of language embedded in and motivated by the social system: “the semantic system, which is the meaning potential embodied in language, is itself the realisation of a higher level semiotic which we may define as a behavioural system or more generally as a social semiotic” (Halliday 1974: 86). The exact nature and representation of this higher semiotic and its relation to language remains rather elusive and underspecified although it is a key area of ongoing research (see, e.g., Bateman 2019, or Maiorani 2017). This then raises the question of what counts as language if our meaning potential is distributed across different modes of meaning-making behavior or multimodality, which is the second challenge for studies of context.

5.4.2 Multimodality and Context As with all functional approaches, SFL sees language as a tool and language use as a form of behavior. In line with his view of the importance of looking at everything from the perspective of how it makes meaning, Halliday (2003)

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

5 A Functional Approach to Context

follows Ellis (1966) in arguing that language “has to create a universe of its own, a parallel reality that is as it were made of meaning.” If we take this perspective seriously, we are in the position of viewing all behavior as potentially meaning bearing, and indeed, all artifacts and even the environment itself (Fawcett 1990). These different modes of meaning-making are all used by humans in their interaction and to the extent that they are useful for us, we can organize them in various ways and to different extents as distinct codes.7 The broad-scale use of different modes by humans does not mean that the different modes have the same meaning potential, since, as Hasan suggests, “despite overlaps, what can be said through the verbal code is not coextensive with what can be said through the gazing code or the gesture code or the code of dress” (Hasan 1980: 107). Each mode carries distinct representational capacities which relate to the means of interaction and are codified in different ways and to different extents. Not having equal access to the full range of meanings in a given mode is distinct from the mode itself having a limited potential. Individuals may not have the same access to the mode, but the mode has the same potential whether we access it or not. Here the situation is that the modes themselves do not have the same meaning potential. The implication of this limit on the meaning-making potential for different modalities is that they will have a different semantic stratum and different contexts, although they share the same material-situational and temporal setting and are part of the same “text.” They will also have different reactances in the contextual stratum and may not even have a distinct organizing stratum. That is to say that what counts as context for one mode may be quite different to what counts as context for a different mode and even in the one shared situation, different aspects of the context will have different effects on different modalities. Indeed, there is no reason to suspect that each mode should conform to the same dimensional arrangement at all, and we cannot expect that there is the same metafunctional arrangement. So with multimodal texts (and this includes all forms of meaning-making behavior) we are in the situation of having very fuzzy boundaries between text and context. In order to make sense of and manipulate our socio-material world, we typically make use of all of the modalities we have at our disposal. Some of these modalities we organize, and they become codified to a greater or lesser extent, but this is not static. Technological changes alter the potential of modalities and change what we can do with them. They change in their permanence, their shareability, their editability and adaptability, and in turn, this changes how we can use them and what we can use them for, and a contextual theory of language has to be able to manage these changes. Access to multiple modalities gives us certain advantages for meaningmaking in that we can use it for redundancy, or expansion. For example, we 7

While some use the terms code and mode interchangeably, we prefer to keep them distinct, especially as concerns multimodality.

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

129

130

REBEKAH WEGENER AND LISE FONTAINE

can use different available modalities to encode the same meaning (redundancy), or we can use these different modalities to create quite distinct meanings (expansion). Here we might think of an exit sign that uses color, light, words, and images to mean the same thing (redundancy) or a parent engaging multiple children at once through different modalities (expansion). We can make use of redundancy for emphasis or as a reminder, and we can make use of expansion as a way to split the semiotic load between modalities. This splitting of the semiotic load is incredibly useful particularly in some domains such as surgery or air traffic control, or indeed any situation where participants are engaged in a shared task with common goals (see Wegener 2016). But if we want to consider context, whether that is for the purposes of fully understanding meaning or for mapping variation or similarity and difference, then we need to be able to define what it means to have a multimodal text before we can consider this in context. Most human communication is multimodal, no matter how you define multimodality (see Bateman et al. 2020). The value of being able to access multimodal data more fully is something that contextual and multimodal AI is only just beginning to realize and address (Wegener and Cassens 2019). However, to make use of this, we have to be able to model multimodal meaning-making in context, which means being able to define boundaries, to work out segmentation and modal alignment, and to look at this in context. Despite being theoretically open to modeling language as inherently multimodal, SFL historically has some problems that Hasan (2004: 21) highlights in her claim that: unlike the “cultural activity theory” associated with the Russian, especially (neo-)Vygotskian literature (Engestrom et al. 1999), “context theory” was not intended to apply to all kinds of social action, being designed specifically with discourse in mind.

Although this distinction varies across the body of her work, Hasan does typically focus her attention on those aspects of social action which relate specifically to discourse or are “construed by discourse.” By discourse, Hasan usually means spoken or written language, with the multimodal aspects of these interactions being covered by context in the form of either material situational setting, situation, or culture. So in Hasan’s model, context of situation includes other modalities as part of the context rather than as part of the discourse. In the wider context of SFL, this means that modalities are potentially present on the expression plane as separate modes of meaning and represented in the contextual plane as part of the context for discourse. Although it would seem to present a theoretical and modeling difficulty, depending on how you go about modeling context, this dual representation does not have to be problematic in application. For example, working with applications is already messy (and/or noisy), and it may appear at first glance

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

5 A Functional Approach to Context

that it is reasonable for practical application purposes to gloss over the wrinkles in theory. But what does this mean for theory if practitioner after practitioner glosses over those wrinkles for the purposes of their specific project? Working with theory is generally quite abstract and developing a model from theory reduces complexity and abstraction as the model becomes more concrete. In turn, applying the model extends this even further such that theoretical concerns are often hidden and ignored. Given the work that is currently being done (O’Halloran et al. 2019; Stöckl 2019) and the importance of contextual and multimodal AI for human-cenetred technology design and development, work on this area is bound to be an important area of future development for SFL and for studies of context more broadly.

5.5 Concluding Remarks and Future Directions SFL has always had a transdisciplinary and applied focus, something which has shaped and will continue to shape the theory. As we see models of context applied across more varied domains and engaging with different disciplines, we see the theory taking new paths in new directions. For example, applications in domains such as health and medicine (Moore 2019), education and learning (Byrnes 2019; Mickan 2019), and computing and artificial intelligence (Bateman et al. 2019; Wegener and Cassens 2019; Zappavigna 2019) have forged new pathways for the development of context modeling in SFL. It is fair to say that research drawing on the SFL framework has generally focused to a large extent on the social aspect of language and context. More recently we have seen a return to some consideration of the individual. This more cognitive aspect was always available (consider, for example, the work of Philipp Wegener and even Firth and Ellis), but it was much less prominent and underdeveloped to the extent that one might argue it was completely ignored. Recent work from Martin on individuation (Martin 2019) and Wegener on a person-centric approach to context for human-centered AI (Wegener 2016; Cassens and Wegener 2019) has renewed this focus on cognition and the person, in different ways and for different motivations. It may be perceived by some that after what may seem like an extended period of focus on the social aspects of context, the cognitive aspects have still not made an impact on the development of theory. The question of having “an overall cognitive theory of context as a type of mental model” (van Dijk 2005; see also van Dijk 2006) is a worthwhile one. Far from treating mental aspects as alien, SFL provides a venue for examining the features, which combine to produce the structural nature of mental aspects such as purpose. No theory has all the answers to all the questions, but if, as Butler and Gonzálvez-García (2014) suggest, there is fertile ground to explore, it would best be done collaboratively, since each functional theory brings its experience and developments from their own extended periods of focus whether that concerns cognitive processes, typological comparisons, or text in context.

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

131

132

REBEKAH WEGENER AND LISE FONTAINE

If we accept that context cuts across all strata of language, and as such aids the understanding of meaning at all levels, then we can define it simply as everything that is around the text. Context can, and should, be used to understand the linguistic process; the design of research questions, the collection, storage, and analysis of data, and the interpretation and representation of results. In this sense it is both theory and method. As textual analysis, multimodal or otherwise, is carried out, it is context that makes any findings meaningful. Context does not need to be mysterious or idealized, but by the same principle, neither is it helpful to reduce or oversimplify. Perhaps what is most needed in future work is a clear setting out of a process, one that supports the shift from theory to practice, while enabling and supporting collaboration and transdisciplinary work on context. There are two key areas for such a development. Firstly, there is a need for greater transparency around context. While there is a relatively clear process for carrying out grammatical analysis or morphological analysis, and even to a certain extent semantic analysis, analysis at the level of context has no unified or clear procedure. There are no established units, no guidance as to how one might set about analyzing a text in context with the result that we usually end up with a very discursive description of context. Secondly, without a clear procedure it is almost impossible to combine analysis from different projects for comparison or for complementation. At the moment, if one wishes to compare two contexts, it is necessary for the analysis to have been done using the same approach and preferably by the same researcher. Ideally it should be possible to compare across research projects no matter who the researcher and no matter what approach was taken. If our comparison is made on a functional basis, this should be achievable. Just as Halliday and Matthiessen (2004) compare the traditional grammatical terms with the functional ones, allowing comparison of analysis, so it should be possible to compare different contextual analyses if we know the functional motivation of the distinctions. This process is an exercise in ontology matching, which, as Wegener (2011) explains, is useful since there are definite limits to what can be considered context. Different theoretical approaches and different models within theories differ not so much on what they consider context to be, but on how they organize their description. It is possible to see how these different approaches have arranged information in different ways, foregrounding some aspects and hiding others but in many respects covering the same ground. Grimshaw’s (1994) complementary studies of professional discourse show this very clearly. Ultimately, the biggest challenges for the future of context in SFL seem to be modeling and representational challenges. With the exception of some small difficulties, the theory itself provides a useful space for context, and applications of SFL across varied domains make use of this, but it is still a rather case-specific application. Some solutions will undoubtedly be technical, particularly with respect to corpus work, particularly large-scale

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

5 A Functional Approach to Context

multimodal and multiparticipant corpora. But much of the future work lies with modeling context to make that connection between theory and practice. New tools change theory and practice, just as tools change us as humans and individuals. We would like to end this chapter on context with the following words from Halliday (2003: 3): Meaning needs matter to realize it; at the same time, matter needs meaning to organize it. Human history is a continuing interplay of the material and the semiotic, as modes of action – ways of doing and of being. . . . This is the context in which language needs to be understood.

References Bartlett, T. (2016). Phasal dynamism and the unfolding of meaning as text. English Text Construction, 9(1), 143–164. Bartlett, T. (2017). Context in systemic functional linguistics: Towards scalar supervenience? In T. Bartlett and G. O’Grady (eds.), The Routledge Handbook of Systemic Functional Linguistics (pp. 375–390). London/New York: Routledge. Bateman, J. (2019). Transmediality and the end of disembodied semiotics. International Journal of Semiotics and Visual Rhetoric, 3(2), 1–23. Bateman, J., McDonald, D., Hiippala, T., Couto-Vale, D., and Costetchi, E. (2019). Systemic functional linguistics and computation. In G. Thompson, W. L. Bowcher, L. Fontaine, and D. Schonthal (eds.), The Cambridge Handbook of Systemic Functional Linguistics (pp. 561–586). Cambridge: Cambridge University Press. Bateman, J. A., Wildfeuer, J., and Hiippala, T. (2020). A question of definitions: Foundations for multimodality. A response to Charles Forceville’s review. Visual Communication, 19, 317–320. Bernstein, B. (1971). Class, Codes, and Control, Vol. I: Theoretical Studies Towards a Sociology of Language. London: Routledge and Kegan Paul. Berry, M. (2016a). Dynamism in exchange structure. English Text Construction, 9(1), 33–55. Berry, M. (2016b). On describing contexts of situation. In Society in Language, Language in society (pp. 184–205). London: Palgrave Macmillan. Berry, M. (2017). Stratum, delicacy, realisation and rank. In T. Bartlett and G. O’Grady (eds.), The Routledge Handbook of Systemic Functional Linguistics (pp. 42–55). London/New York: Routledge. Berry, M. (2019). The clause: An overview of the lexicogrammar. In G. Thompson, W. Bowcher, L. Fontaine, and D. Schönthal (eds.), The Cambridge Handbook of Systemic Functional Linguistics (pp. 92–117). Cambridge: Cambridge University Press. Bowcher, W. (1999). Investigating institutionalization in context. In M. Ghadessy (ed.), Text and Context in Functional Linguistics (pp. 141–176). Amsterdam: John Benjamins. Bowcher, W. (2017). Field, tenor and mode. In T. Bartlett and G. O’Grady (eds.), The Routledge Handbook of Systemic Functional Linguistics (pp. 391–403). London/ New York: Routledge.

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

133

134

REBEKAH WEGENER AND LISE FONTAINE

Bowcher. W. (2018). The semiotic sense of context vs the material sense of context. Functional Linguistics, 5(5), 1–19. Butler, C. (2003). Structure and Function: A Guide to Three Major Structural-Functional Theories, Part I: Approaches to the Simplex Clause (Studies in Language Companion Series 63). Philadelphia, PA/Amsterdam: John Benjamins. Butler, C. (2004). Corpus studies and functional linguistic theories. Functions of Language, 11(2), 147–186. Butler, C., and Gonzálvez-García, F. (2014). Exploring Functional-Cognitive Space. Amsterdam: John Benjamins. Butt, D. (1999/2004). Parameters of Context. Mimeo Sydney: Department of Linguistics, Macquarie University. Butt, D., and Wegener, R. (2008). The work of concepts: Context and metafunction in the systemic functional model. In R. Hasan, C. M. I. M. Matthiessen, and J. Webster (eds.), Continuing Discourse on Language, Vol. II: A Functional Perspective (pp. 589–618). London: Equinox. Byrnes, H. (2019). Applying SFL for understanding and fostering instructed second language development. In G. Thompson, W. L. Bowcher, L. Fontaine, and D. Schonthal (eds.), The Cambridge Handbook of Systemic Functional Linguistics (pp. 512–536). Cambridge: Cambridge University Press. Cassens, J. and Wegener, R. (2008). Making use of abstract concepts: Systemic functional linguistics and ambient intelligence. In M. Bramer (ed.), Artificial Intelligence in Theory and Practice II, IFIP 20th World Computer Congress, IFIP AI Stream, vol. 276 of IFIP (pp. 205–214). Milan: Springer. Clarke, B. P. (2013). The differential patterned occurrence of ellipsis in texts varied for contextual mode: Some support for the “mode of discourse-textual metafunction” hook-up? In G. O’Grady, T. Bartlett, and L. Fontaine (eds.), Choice in Language: Applications in Text Analysis (pp. 269–297). Sheffield: Equinox. Daneš, F. (1987). On Prague school functionalism in linguistics. In R. Dirven and V. Fried (eds.), Functionalism in Linguistics (pp. 3–38). Amsterdam: John Benjamins. Dirven, R., and Fried, V. (1987). Functionalism in Linguistics. Amsterdam: John Benjamins. Ellis, J. (1966). On contextual meaning. In C. E. Bazell, J. C. Catford, M. A. K. Halliday, and R. H. Robins (eds.), In Memory of Firth (pp. 79–95). London: Longman. Fawcett, R. (1980). Cognitive Linguistics and Social Interaction: Towards an Integrated Model of a Systemic Functional Grammar and the Other Components of an Interacting Mind. Heidelberg: Julius Groos. Fawcett, R. (1990). The COMMUNAL project: Two years old and going well. Network: News, Views and Reviews in Systemic Linguistics and Related Areas, 13, 35–39. Fawcett, R. (2000). A Theory of Syntax for Systemic Functional Linguistics. Amsterdam/ Philadelphia: John Benjamins. Fawcett, R. (2015). System networks, codes, and knowledge of the universe. In R. Fawcett, M. Halliday, S. Lamb, and A. Makkai (eds.), Semiotics of Culture and Language, Vol. II: Language and Other Semiotic Systems of Culture (Linguistics: Bloomsbury Academic Collections) (pp. 135–180). London: Bloomsbury Academic.

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

5 A Functional Approach to Context

Fontaine, L., Bartlett, T., and O’Grady, G. (2013). Systemic Functional Linguistics: Exploring Choice. Cambridge: Cambridge University Press. Fontaine, L., and Schönthal, D. (2019). The rooms of the house: Grammar at group rank. In G. Thompson, W. Bowcher, L. Fontaine, and D. Schönthal (eds.), The Cambridge Handbook of Systemic Functional Linguistics (pp. 118–141). Cambridge: Cambridge University Press. François, J. (2018). The stance of Systemic Functional Linguistics amongst functional(ist) theories of language and its “systemic” purpose. In A. Sellami-Baklouti and L. Fontaine (eds.), Perspectives from Systemic Functional Linguistics (pp. 6–25). London: Routledge. Frazer, J. G. (1922). The Golden Bough. New York: Macmillan. Gleason, H. A. (1965). Linguistics and English Grammar. New York: Holt, Reinhart and Winston. Gregory, M., and Carroll, S. (1978). Language and Situation: Language Varieties and Their Social Contexts. London: Routledge and Kegan Paul. Grimshaw, A., ed. (1994). What’s Going On Here? Complementary Studies of Professional Talk (Advances in Discourse Processes, Vol. 43). Norwood, NJ: Ablex. Halliday, M. A. K. (1973). Language in a Social Perspective: Explorations in the Functions of Language. London: Edward Arnold. Halliday, M. A. K. (1974). Interview with M. A. K. Halliday. In H. Parret (ed.), Discussing Language (pp. 81–120). The Hague: Mouton. Halliday, M. A. K. (1978). Language as Social Semiotic: The Social Interpretation of Language and Meaning. London: Edward Arnold. Halliday, M. A. K. (1979/2002). Modes of meaning and modes of expression: Types of grammatical structure and their determination by different semantic functions. In Jonathan Webster (ed.), The Collected Works of M. A. K. Halliday, Vol. I: On Grammar (pp. 196–218). London/New York: Continuum. Halliday, M. A. K. (1981/2002). Text semantics and clause grammar: How is a text like a clause? In Jonathan Webster (ed.), The Collected Works of M. A. K. Halliday, Vol. I: On Grammar (pp. 219–260). London/New York: Continuum. Halliday, M. A. K. (1991). Corpus linguistics and probabilistic grammar. In K. Aijmer and B. Altenberg (eds.), English Corpus Linguistics: Studies in Honour of Jan Svartvik. (pp. 30–44). New York/London: Longman. Halliday, M. A. K. (1991/2007). The notion of “context” in language education. In Jonathan Webster (ed.), The Collected Works of M. A. K. Halliday, Vol. IX: Language and Education (pp. 269–290). London: Equinox. Halliday, M. A. K. (1992). New ways of meaning: The challenge to applied linguistics. In M. Putz (ed.), Thirty Years of Linguistic Evolution (pp. 59–97). Philadelphia/ Amsterdam: John Benjamins. Halliday, M. A. K. (1994/2002a). So you say “pass” . . . thank you three muchly. In Jonathan Webster (ed.), The Collected Works of M. A. K. Halliday, Vol. II: Linguistic Studies of Text and Discourse (pp. 228–254). London: Continuum. Halliday, M. A. K. (1994/2002b). How do you mean? In Jonathan Webster (ed.), The Collected Works of M. A. K. Halliday, Vol. I: On Grammar (pp. 352–368). London/ New York: Continuum. Halliday, M. A. K. (1995). On language in relation to the evolution of human consciousness. In S. Allen (ed.), Of Thoughts and Words: Proceedings of Nobel Symposium 92:

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

135

136

REBEKAH WEGENER AND LISE FONTAINE

“The Relation between Language and Mind” (pp. 45–84). London: Imperial College Press. Halliday, M. A. K. (1997/2002). Text as semantic choice in social contexts (1977). In Jonathan Webster (ed.), The Collected Works of M. A. K. Halliday, Vol. II: Linguistic Studies of Text and Discourse (pp. 23–84). London: Continuum. Halliday, M. A. K. (2002). Introduction: A Personal Perspective. In J. Webster (ed.), The Collected Works of M. A. K. Halliday, Vol. I: On Grammar (pp. 1–16). London/ New York: Continuum. Halliday, M. A. K. 2003. On Language and Linguistics, Vol. III, The Collected Works of M. A. K. Halliday, ed. J. Webster. Continuum: London. Halliday, M. A. K. (2013). Meaning as choice. In L. Fontaine, T. Bartlett, and G. O’Grady (eds.), Systemic Functional Linguistics: Exploring Choice (pp. 15–36). Cambridge: Cambridge University Press. Halliday, M. A. K. and Matthiessen, C. M. I. M. (2004/2014). An Introduction to Functional Grammar, 3rd ed. London: Edward Arnold. Hasan, R. (1978). Text in the systemic functional model. In W. Dressler (ed.), Current Trends in Text Linguistics (pp. 228–246). Berlin: Walter Gruyter. Hasan, R. (1980). What’s going on: A dynamic view of context. In The Seventh LACUS Forum (pp. 106–121). Columbia, SC: Hornbeam Press. Hasan, R. (1984). What kind of resource is language? Australian Review of Applied Linguistics, 7(1), 57–85. Hasan, R. (1995). The conception of context in text. In P. H. Fries and M. Gregory (eds.), Discourse in Society: Systemic Functional Perspectives. Meaning and Choice in Language: Studies for Michael Halliday (pp. 183–283). Norwood, NJ: Ablex. Hasan, R. (1999). Speaking with reference to context. In M. Ghadessy (ed.), Text and Context in Functional Linguistics: Systemic Perspectives (pp. 219–328). Amsterdam: John Benjamins. Hasan, R. (2004). Analysing discursive variation. In L. Young and C. Harrison (eds.), Systemic Functional Linguistics and Critical Discourse Analysis: Studies in Social Change (pp. 15–52). London: Continuum. Hasan, R. (2010). The meaning of “not” is not in “not.” In A. Mahboob and N. Knight (eds.), Appliable Linguistics (pp. 267–306). London/New York: Continuum. Hasan, R. (2014). Towards a paradigmatic description of context: Systems, metafunctions, and semantics. Functional Linguistics, 1(9), 1–54. Langacker, R. (2019). Construal. In E. Dąbrowska and D. Divjak (eds.), Cognitive Linguistics: Foundations of Language (pp. 140–166). Berlin/Boston: De Gruyter Mouton. Lukin, A. (2017). Ideology and the text-in-context relation. Functional Linguistics, 4 (16), 1–17. https://doi.org/10.1186/s40554-017-0050-8. Lukin, A., Moore, A., Herke, M., Wegener, R., and Wu, C. (2011). Halliday’s model of register revisited and explored. Linguistics and the Human Sciences, 187–213. Maiorani, A. (2017). Making meaning through movement: A functional grammar of dance. In M. G. Sindoni, J. Wildfeuer, and K. L. O’Halloran (eds.), Mapping Multimodal Performance Studies (pp. 39–60). London/New York: Routledge.

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

5 A Functional Approach to Context

Martin, J. (1985). Process and text: Two aspects of human semiosis. In J. D. Benson, and W. S. Greaves (eds.), Systemic Perspectives on Discourse, Vol. I (pp. 248–274). Norwood, NJ: Ablex. Martin, J. (1992). English Text: System and Structure. Amsterdam: John Benjamins. Martin, J. (1993). A contextual theory of language. In W. Cope and M. Kalantzis (eds.), The Powers of Literacy: A Genre Approach to Teaching Literacy (Critical Perspectives on Literacy and Education) (pp. 116–136). London: Falmer. Martin, J. (1997). Analysing genre: Functional parameters. In F. Christie and J. Martin (eds.), Genre and Institutions: Social Processes in the Workplace and School (pp. 3–39). London: Cassell. Martin, J. (2019). Discourse semantics. In G. Thompson, W. L. Bowcher, L. Fontaine, and D. Schonthal (eds.), The Cambridge Handbook of Systemic Functional Linguistics (pp. 358–381). Cambridge: Cambridge University Press. Matthiessen, C. M. (2015). Register in the round: Registerial cartography. Functional Linguist, 2(9), 1–48. Matthiessen, C. Wang, B., and Ma, Y. (2019). Expounding register and registerial cartography in systemic functional linguistics: An interview with Christian M. I. M. Matthiessen, WORD, 65(2), 93–106. https://doi.org/10.1080/00437956 .2019.1599544. Mickan, P. (2019). Language and education. In G. Thompson, W. L. Bowcher, L. Fontaine, and D. Schonthal (eds.), The Cambridge Handbook of Systemic Functional Linguistics (pp. 537–560). Cambridge: Cambridge University Press. Mitchell, T. F. (1957/1975). The language of buying and selling in Cyrenaica: A situational statement. In T. F. Mitchell (ed.), Principles of Firthian Linguistics (pp. 1–41). London: Longman. Moore, A. (2019). Language and medicine. In G. Thompson, W. L. Bowcher, L. Fontaine, and D. Schönthal (eds.), The Cambridge Handbook of Systemic Functional Linguistics (pp. 651–695). Cambridge: Cambridge University Press. Moore, A. R. (2020). Progress and tensions in modelling register as a semantic configuration. Language, Context and Text: The Social Semiotics Forum, 2(1), 22–58. Nerlich, B. (1990). Change in Language: Whitney, Breal and Wegener. London: Routledge. Nerlich, B., and Clarke, D. (1996). Language, Action and Context: The Early History of Pragmatics in Europe and America. Amsterdam: John Benjamins. Newmeyer, F. (2001). The Prague School and North American functionalist approaches to syntax. Journal of Linguistics, 37(1), 101–126. O’Donnell, M. (1999). Context in dynamic modelling. In M. Ghadessy (ed.), Text and Context in Functional Linguistics (pp. 63–99). Amsterdam: John Benjamins. O’Donnell, M. (2020). On the abstractness of levels of description in Systemic Functional Linguistics. In G. Tucker, G. Huang, L. Fontaine, and E. McDonald (eds.), Approaches to Systemic Functional Grammar: Convergence and Divergence (pp. 57–75). London: Equinox.

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

137

138

REBEKAH WEGENER AND LISE FONTAINE

O’Donnell, M. (2021). Dynamic modelling of context: Field, tenor and mode revisited. Lingua, 261, 102952. https://doi.org/10.1016/j.lingua.2020.102952. O’Halloran, K. L., Tan, S., and Wignell, P. (2019). SFL and multimodal discourse analysis. In G. Thompson, W. L. Bowcher, L. Fontaine, and D. Schonthal (eds.), The Cambridge Handbook of Systemic Functional Linguistics (pp. 433–461). Cambridge: Cambridge University Press. Polguère, A. (2015). Lexical contextualism: The Abélard Syndrome. In N. Gala, R. Rapp, and G. Bel-Enguix (eds.), Language Production, Cognition, and the Lexicon, 48 (pp. 53–73). London: Springer. Rebori, V. (2002). The legacy of J. R. Firth: A report on recent research. Historiographia Linguistica, 29(1/2), 165–190. Stöckl, H. (2019). Linguistic multimodality – multimodal linguistics: A state-of-the-art sketch. In J. Wildfeuer, J. Pflaeging, J. Bateman, O. Seizoy, and C.-I. Tseng (eds.), Multimodality: Disciplinary Thoughts and the Challenge of Diversity (pp. 41–68). Berlin/Boston: De Gruyter. Tam, K. (2017). Context and meaning in the Sydney architecture of systemic functional linguistics? In T. Bartlett and G. O’Grady (eds.), The Routledge Handbook of Systemic Functional Linguistics (pp. 438–456). London/New York: Routledge. Taverniers, M. (2011). The syntax–semantics interface in Systemic Functional Grammar: Halliday’s interpretation of the Hjelmslevian model of stratification. Journal of Pragmatics, 43(4), 1100–1126. Taverniers, M. (2019). Semantics. In G. Thompson, W. Bowcher, L. Fontaine, and D. Schönthal, D. (eds.), The Cambridge Handbook of Systemic Functional Linguistics (pp. 55–91). Cambridge: Cambridge University Press. Thompson, G. (1999). Acting the part: Lexico-grammatical choices and contextual factors. In M. Ghadessy (ed.), Text and Context in Functional Linguistics (pp. 101–124). Philadelphia, PA/Amsterdam: John Benjamins. Ure, J. (1969). Practical register: Parts I and II. English Language Teaching, 23(1/2). van Dijk, T. (2005). Contextual knowledge management in discourse production. A CDA perspective. In R. Wodak and P. Chilton (eds.), A New Agenda in (Critical) Discourse Analysis. (pp. 71–100) Amsterdam: John Benjamins. van Dijk, T. (2006). Discourse, context and cognition. Discourse Studies, 8(1), 159–177. Ventola, E. (1983). Contrasting schematic structures in service encounter. Applied Linguistics, 4, 242–258. Wegener, R. (2011). Parameters of context: From theory to model and application. Ph.D. thesis, Macquarie University, Sydney. Wegener, R. (2016). Studying language in society and society through language: Context and multimodal communication. In W. Bowcher and J. Liang (eds.), Society in Language, Language in Society: Essays in Honour of Ruqaiya Hasan (pp. 227–248). London: Palgrave Macmillan. Wegener, R., and Cassens, J. (2019). Blending SFL and Activity Theory to model communication and artefact use: Examples from human–computer interaction. In M. Kaltenbacher and H. Stöckl (eds.), Analyzing the Media: A Systemic Functional Approach (pp. 167–188). London: Equinox. Wegener, R., Cassens, J., and Butt, D. (2008). Start making sense: Systemic-functional linguistics and ambient intelligence. Revue d’Intelligence Artificielle 22(5), 629–645.

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

5 A Functional Approach to Context

Williams, J., Russell, N., and Irwin, D. (2017). On the notion of abstraction in systemic functional linguistics. Functional Linguistics, 4(13). https://doi.org/10.1186/s40554017-0047-3. Zappavigna, M. 2019. Language and social media. In G. Thompson, W. Bowcher, L. Fontaine, and D. Schonthal (eds.), The Cambridge Handbook of Systemic Functional Linguistics (pp. 715–738). Cambridge: Cambridge University Press.

https://doi.org/10.1017/9781108989275.006 Published online by Cambridge University Press

139

6 The Grammar of Incremental Language Production in Context J. Lachlan Mackenzie

6.1 Introduction There are two basic perspectives to be taken on a clause. It can be seen principally as a hierarchical structure or as a succession of units. Hierarchical syntactic analysis is most saliently exemplified in contemporary linguistics by phrase-structure trees. These serve to bring out language users’ tacit knowledge of relationships among constituents of the clause. They show which elements are closely related and which are more distantly connected. Hierarchical analysis can take the form either of constituency diagrams that display relations of sisterhood between co-constituents of a hierarchically higher constituent or of dependency diagrams that show the hierarchy of headmodifier relations between elements. Such hierarchical representations are particularly prominent in approaches to syntax that focus on linguistic competence: Their aim is to lay bare language users’ underlying knowledge of the syntactic potential of their language. In some such approaches, the syntactic analysis is limited to indicating relationships of constituency or dependency, with the left-to-right (or earlier-to-later) order of elements being regarded as a matter of phonology. By contrast, an emphasis on the succession of syntactic units is more typical of grammatical approaches that foreground performance and language processing. Here the focus is on representing the timecourse of the utterance, the gradual, incremental build-up of the clause as one component unit follows the other in real time. While an emphasis on hierarchy tends to be associated with grammars that abstract away from contexts of use and accentuate the grammaticality of sentences, the incrementalist approach generally views language use as situated action and stresses the achievement of situation-appropriateness. This involves a consideration of how an utterance reflects a language user’s communicative goals and needs and the context in which they are conversing with their interlocutor(s). Whereas the perspective that privileges the attribution of

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

6 The Grammar of Incremental Language Production

a hierarchical structure understandably focuses chiefly on complete clauses and relies heavily on intuitive judgments of well-formedness, the alternative incrementalist view is more open to including incomplete or fragmentary clauses in the purview of syntax and to drawing its data from transcriptions of verbal interaction. The present chapter aims to bring out how such incrementalist approaches to grammar have developed in tandem with psycholinguistic modeling of language production and comprehension. An important emphasis of the chapter will be on how aspects of the multisensory situational context are rapidly integrated into language users’ processing strategies. Against this background, we will see that research has been moving in the direction of an integration of the two perspectives distinguished in the opening paragraphs. In terms of the structure of the chapter, Section 6.2 will sketch how work on contextually embedded language processing has had implications for the well-established hierarchical analysis of syntactic units, while Section 6.3 will consider various approaches that privilege incremental processing. Section 6.4 will turn to findings relating to the emergence of hierarchy from sequence and thus the integration of the two perspectives, and Section 6.5 will close the chapter by considering the matter from the viewpoint of dialogue.

6.2 Hierarchical Structure Defended, Curbed, and Challenged The strongest measure of hierarchization in a constituent structure tree is achieved by uniformly imposing binary branching, i.e., limiting the number of co-constituents (“sisters”) to two. This is notably characteristic of Chomsky’s Minimalist Program (Boeckx 2006) within generative grammar, as a result of the generalized application of the operation Merge, which combines exactly two elements, a and b, to form a set b {a, b}, where a and b are immediate constituents of a more embracing unit b. As Kornmesser (2019: 505; italics in original) points out, “postulating merge to take at least and at most two syntactic objects and hence to produce binary branching structures is considered to satisfy the requirement of economy.”1 Several assumptions of generative grammar, minimalist or not, notably the presence of various “empty” elements in the tree structure and the analysis of certain morphological phenomena as “syntactic” and as binary-branching phrases, as well as the absorption of semantic phenomena such as Tense and pragmatic phenomena such as Topic, Focus, Contrast, and even (illocutionary) Force into syntax, again in the form of branching phrases, all contribute to a very high degree of hierarchization. Extreme hierarchization is particularly evident in so-called cartographic approaches to generative syntax (e.g., 1

This institutionalizes what has been the norm in generative grammar since the eighties; earlier work did embrace ternary branching (cf. Luraghi and Prodi 2013: 206).

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

141

142

J. LACHLAN MACKENZIE

Cinque 1999; Rizzi 2004), which entail abundantly branching analyses, with over thirty “projections” just for adverbial modifiers.2 Ray Jackendoff, once a major figure in generative syntax (1977), has come to reject such “syntactocentric” approaches, proposing (Culicover and Jackendoff 2005; Jackendoff and Audring 2020) alternative models (Simpler Syntax and Relational Morphology respectively), in which constituent structure is retained but no longer restricted to binary branching and is freed from non-syntactic notions. In this approach, syntax is one of three modules of a Parallel Architecture (phonological structures; syntactic structures; semantic/conceptual structures) that coexist in the grammar. For comparable but distinct architectures of this kind, see Functional Discourse Grammar (which has four modules, one of them devoted to pragmatics; Hengeveld and Mackenzie 2008) and the Autonomous Modular approach of Sadock (2012; with six modules). Culicover and Jackendoff (2005) defend what they call their “relatively flat” (p. 108), multiple-branching structures, and as they point out, these structures are analogous to what has been proposed in other offshoots of the generative paradigm such as Lexical Functional Grammar (LFG; Bresnan 2001),3 HeadDriven Phrase Structure Grammar (HPSG; Pollard and Sag 1994)4 as well as the more practical grammar of English edited by Huddleston and Pullum (2002); to quote an observation by Culicover and Jackendoff (2005: 107) that will be significant for later sections of this chapter, an emphasis on sequence rather than hierarchy reflects “the view of syntax generally adopted in psycholinguistic research.” This line of thought is followed up to an extreme degree by Jackendoff and Wittenberg (2014), who argue that there are certain forms of linguistic organization that dispense with hierarchical structure altogether. They defend the view that in pidgins, late L2 acquisition, “home signs” (i.e., spontaneously arising gestural languages), agrammatic aphasia, and even certain fully functional languages such as Riau Malay, the sequence of syntactic units is purely linear, with no hierarchy being detectable. This is a situation which they model by having semantic/conceptual structure map directly onto phonological structure without any syntactic structure coming into play (i.e. an architecture with only two modules). The ultimate position along this trajectory of thought – the claim that syntax is not needed for any language – is defended by Nordström (2017), who argues that Merge is not implementable and proposes what she calls a “flat structure” (p. 697), which, again significantly, she qualifies as being “much more in line with language production and comprehension” (p. 702). Steedman (2022: 22) observes in addition that “[m] 2

3

4

Bošković (2022: 7) interprets Cinque (1999) as espousing the view that “all these projections are present even in both clauses of I said that he left, a sentence with no adverbials.” Representing LFG, Dalrymple and Findlay (2019: 132–133) write that “where there is no strong evidence for the existence of a particular hierarchical structure, we do not feel constrained to nonetheless posit it . . . This avoids the requirements of heavily articulated trees with unpronounced functional elements which a theory restricted to purely binary branching necessitates.” Note, however, that Müller and Machicao y Priemer (2019: 330) do assume binary branching for HPSG.

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

6 The Grammar of Incremental Language Production

any languages have freer word order than English, and do not support any clear notion of surface constituency. Even for English, there is no clear consensus on whether the surface structure of the ditransitive VP or the subject auxiliary-inverted question is flat and ternary-branching, leftbranching, or right-branching.” Everett (2015: 361) has written in this regard that Merge “is potentially falsified by any exocentric or non-binary- (ternary-, quaternary-, etc.) branching structure, e.g. a structure with flat syntax.” Although these various proposals to restrain or even abandon hierarchical structure have largely come from critics who share generative grammar’s orientation to competence, it was curiously the appearance of a short paper by Frank et al. (2012) that caused the greatest consternation in generative circles. The introduction to that article makes it clear that the authors do not doubt the importance of hierarchical structure for the description of languages and are merely questioning its relevance for the study of use, which is a matter of performance, i.e., the second approach distinguished here. Nevertheless, the paper’s impact was considerable, and the leading generative grammarian, David Pesetzky, confronted it directly in his plenary lecture to the 2013 Linguistic Society of America’s annual meeting, speaking of Frank et al.’s paper as a “nightmare” and citing Chernyshevsky’s (and Lenin’s) not entirely rhetorical question Что дѣлать? (What Is to Be Done?) in response. It is beyond doubt that grammarians who see themselves as mentalists, as unveiling cognitive structures through their study of language, are sensitive to findings from psycho- and neurolinguistics, and these findings are that binary-branching hierarchical constituent structure is not a given: “[r]ather, considerations of simplicity and evolutionary continuity force us to take sequential structure as fundamental to human language processing. Indeed, this position gains support from a growing body of empirical evidence from cognitive neuroscience, psycholinguistics and computational linguistics” (Frank et al. 2012: 4530). As we shall see in the following section, there are scholars out there, a notable example being Christiansen and Chater (2016), who are willing and able to shake off what they call “Chomsky’s hidden legacy” (p. 6) of focusing on linguistic competence and to explore the implications of not separating the study of grammar from the study of language processing in context.

6.3 Elements of Incremental Syntax Although, as we have seen, mainstream currents in syntax have emphasized constituency (or dependency) and hierarchical structure, there has always been a strong undercurrent of work that has privileged incrementality. This has tended to go hand in hand with the assumption that each increment leads to a reduction in information entropy, understood as the information that the recipient lacks at any point in the construction of a discourse world. Just as, in Groenendijk and Stokhof’s words (2006: 29), “the interpretation of a discourse – a sequence of sentences – takes the form of an incremental

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

143

144

J. LACHLAN MACKENZIE

construction of a discourse representation structure,” so within a sentence, too, the production and comprehension of the various components have been seen as similarly incremental. This notion has found expression in a range of different theoretical initiatives, for example in the Grammar of Speech of Brazil (1995), which reflects an actional view of syntax, seeing the syntactic unit (chiefly the clause) as “something that begins, continues, and ends in time: it happens” (p. 11). In this view, syntactic structures are viewed as the result of the gradual, piecemeal construction of utterances in time. The initial state (i.e., the first increment) sets up a framework of expectation that ultimately leads to the target state, i.e., the final increment. Brazil’s work was continued and developed by Sinclair and Mauranen (2006) in their Linear Unit Grammar, which, as its name suggests, focuses on syntagmatic relations within clauses but goes further in examining how certain successions of units are linked together to form “chunks.” They stress (pp. 130, 134) that chunking is a “pre-theoretical term” and not an exact science in the way that constituent structure analysis seeks to be exact. In fact, they see the very flexibility of their conceptual framework as advantageous in permitting cross-speaker variation within an overall speech community. This emphasis on the practical benefits of approximation will return in various psycholinguistic approaches that encompass the notion of “good enough” processing. This perspective on syntax engenders a particular view of minimal or holophrastic utterances, those that consist of one chunk only. Whereas these have been seen from the viewpoint of hierarchical syntax as reductions of full clauses, the incrementalist position views them as complete within the context in which they occur. For example, in one-chunk answers to questions, the question itself provides a large amount of the context in which the brief answer will be understood. Similarly, holophrastic greetings will typically elicit an equally holophrastic response. The ubiquitous occurrence of minimal utterances in conversation follows, given a sufficiently rich context, from the Gricean Maxim of Quantity, which enjoins the language user to invest a linguistic expression with no more (and obviously no less) information than the interlocutor requires for satisfactory interpretation. Levinson (1987: 68), too, proposes a Maxim of Minimization “Say as little as necessary,” that is, “produce the minimal linguistic clues to achieve your communicational ends.” Many utterances in adult language are of course not minimal in this sense, and the logic of the incrementalist approach is that these are sophistications of the holophrase rather than the latter being an elliptical version of the former. As will be immediately clear, this direction of argument recapitulates the familiar series of stages in child language acquisition, where the child initially produces single-word utterances (“the holophrastic stage”) and then progresses gradually to more complex utterances. Greenfield et al. (1985) apply a Principle of Informativeness to the child’s holophrastic stage, which states that the child will verbalize new rather than given information. At the later two-word stage, the child expresses information that, at the preceding stage, could only be inferred by the addressee from gestural or intonational means

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

6 The Grammar of Incremental Language Production

and/or by drawing on the context. The underlying assumption is that we communicate in order to draw attention to variability in the environment, that is, what is novel, changing, or alternative. For example, if a mother rolls a ball to her infant, the latter will utter ball rather than Mummy, because it is the ball that is in movement and undergoing transfer, while the mother is relatively static and her agency is presupposed rather than asserted. This insight applies equally well to the adult use of holophrases: for a minimal utterance to function as complete within its context, it must contain the one linguistic clue from which the hearer can read off or infer the crucial information that the speaker wishes to impart. For conversation, this notion is present in Chafe’s (1994: 119, 159) vision of each unit making a “single point” (in both English and the Indigenous language Seneca) and, for writing, in Givón’s (1995: 358) “one-chunk-per-clause” constraint, which states – again using the above-mentioned term of art “chunk” – that “[c]lauses in natural text tend to have only one chunk (usually a word) of new information per clause.” It is worth stressing at this point that it is not just holophrastic utterances that rely on context for full comprehensibility. Full clauses, too, can only be understood in their entirety if they are submitted to inferential interpretive processes: The hearer does not merely incrementally decode (or “parse”) the incoming utterance but also interprets it incrementally in its context in line with his understanding of the ongoing and dynamic discourse situation. This suggests a fundamental continuity between minimal and maximal utterances but also raises the question of how more sophisticated utterances differ from holophrases in their communicative content. The connection lies in the distinction between assertion and presupposition: While a (declarative) holophrase offers only assertion, leaving the presupposed information to be inferred from the context, the fuller utterance contains a combination of asserted and presupposed information. The traditional term “assertion” will, however, not be retained here because it is so strongly associated with declarative illocutions (“statements”). A preferable term is the information-structural notion “focus,” which is neutral with respect to illocutionary distinctions (Hengeveld and Mackenzie 2008: 89 ff.). In a wh-interrogative, for example, be it holophrastic or not, it is the wh-word that functions as its focus, in the sense of identifying the particular piece of information to be elicited from the addressee. Consider (1), cf. Mackenzie (2020: 193): (1)

[Who]focus [were you talking to just now]presupposition?

Here the speaker fills out the utterance by reiterating the interactants’ shared knowledge that the hearer was talking to someone in the immediate past and thereby creates a context for the interaction. She5 could also have invoked the context with greater reliance on inference, as in (2), with the anaphoric pronoun that: (2) 5

[Who]focus [was that]presupposition?

In this chapter, the speaker will be referred to as she and the hearer as he.

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

145

146

J. LACHLAN MACKENZIE

Or she could even have formulated it holophrastically, perhaps with a supportive gesture, gaze, or head movement, for example, if the question is whispered surreptitiously to the hearer: (3)

[Who]focus?

He similarly has various answering options: (4)

a. b. c. d.

[I was talking to]presupposition [Bill]focus [just now]presupposition [I was talking to]presupposition [Bill]focus [To]presupposition [Bill]focus [Bill]focus

Since the information about the conversation with Bill not only is shared but has also been explicitly mentioned in the question, the holophrastic answer in (4d) or the near-holophrastic one in (4c) is sufficient. The presupposed information can be repeated in part as in (4b) or in whole as in (4a). But to do this, especially in full, is to opt for the marked option in conversation, which may spark off implicatures (i.e., indirectly communicated meanings), for example that the speaker of (4a) is irritated by the question. It is useful at this stage to bring in a crucial notion from Functional Discourse Grammar (FDG; Hengeveld and Mackenzie 2008), that of the discourse act. A discourse act is a unit of linguistic action that may take the outward form of a full clause, a sequence of phrases that is less than a clause, or a holophrase. Each discourse act is made up of a series of subacts, which together constitute the communicated content of the discourse act. One of these subacts bears the focus of the communicated content (cf. Chafe’s 1994 “single point” or Givón’s 1995 “one chunk” referred to above), while the other subacts, if any, are backgrounded. FDG’s more action-oriented approach allows us to move away from counting only single words as holophrases: They are not identified in morphosyntax but rather at the level of the pragmatic activities of reference and ascription (Searle 1969). Just as children at the holophrastic stage may use certain word combinations as unanalyzed wholes (allgone, allfalldown), so formulaic or prefabricated units (of different degrees of rigidity) also play an important role in adult usage, being interwoven with more compositional elements. Wray (2002: 8) points out that “[o]nly with the new generation of grammatical theories, based on performance rather than competence . . ., has the idea of holistically managed chunks of language been slowly reinstated, and its implications recognized.” FDG adopts the following compromise with regard to the hierarchy‒ sequence debate (Mackenzie 2014a): Syntax is seen as primarily linear, while pragmatic and semantic organization are predominantly hierarchical, with the hierarchies in question representing scope relations (e.g., subacts scoping under discourse acts, or states-of-affairs scoping under propositional contents). Now, when it comes to the syntactic description of incrementally composed utterances, FDG recognizes various salient (so-called “absolute”) positions in the linear order of the various morphosyntactic units, first and

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

6 The Grammar of Incremental Language Production

foremost among which is the initial position (PI). Morphosyntactic units that correspond to the upper reaches of the pragmatic and semantic scopal hierarchies that describe the formulation of the discourse act have prior access to those absolute positions. For example, discourse markers, which are analyzed as modifiers of the discourse act, tend to be placed in initial position by virtue of their high position in the pragmatic hierarchy, which represents wide scope. Returning now to holophrases (or one-chunk utterances), these correspond to discourse acts whose communicated content contains only one subact, necessarily (or “trivially”) marked as focus. They thus rightfully claim the absolute initial position, even in the absence of further material. Thus, all the following examples will be assigned to the initial position of the syntactic unit: (5)

a. b. c. d. e.

Brilliant. Goal! Arthur! Ball. (as in holophrastic-stage child speech) Allgone. (as in holophrastic-stage child speech)

Things become more interesting where the communicated content contains more subacts, as in the following examples: (6)

a. b. c. d. e.

That was brilliant. They scored a goal! Hey Arthur! Mummy ball. (as in two-word-stage child speech) Banana allgone. (as in two-word-stage child speech)

What we see in (6), as compared to (5), is that units expressing other (nonfocus) subacts have usurped PI, pushing the focus unit to a later position in the sequence. Here the “minimization” advantage of the smallest possible expression in (5) has been abandoned in favor of providing sufficient contextual grounding for the hearer: in (6a), for example, the initial that, possibly accompanied by a gesture, provides the hearer with some indication of what the speaker finds to be brilliant. However, there are advantages for the speaker, too, in formulating a discourse act non-holophrastically. As Wagner (2016) points out, “[s]tudies from the production planning literature [in psycholinguistics, JLM] have found that one crucial factor in such word order choices is that speakers tend to realize constituents encoding more active and accessible referents earlier in the sentence.” In other words, speakers can in practice gain precious milliseconds from not encoding the focus immediately, but anchoring it in knowledge that is already “active” (i.e., energetically present in the common ground) and accessible (i.e., relatively easily retrieved, for both parties). In FDG terms, such subacts are known as topics, defined technically as “signalling how the Communicated Content relates to the gradually constructed record in the Contextual Component” (Hengeveld and Mackenzie 2008: 92), where the Contextual Component is an

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

147

148

J. LACHLAN MACKENZIE

extra-grammatical module that implements the notion of a shared “common ground” of contextual assumptions. Since topic, like focus, is assigned to a subact, the two are found at the same hierarchical level of pragmatic organization, and therefore if syntactic order reflects hierarchical status, languages may be expected to differ in which “wins out” in terms of access to the initial position PI. In English, generally, it is the topic unit – which is often also the clausal subject – that goes to initial position, pushing the focus unit further to the end. By contrast, in Nootka (Fortescue 2004: 155), the element considered by the speaker to be the most “newsworthy” (a notion that correlates with focus status) remains in initial position, in keeping with a “strategy . . . found in many North American languages with pragmatically controlled word order.” In English, too, of course, there are constructions in which the focus retains its initial positioning, most notably in wh-interrogatives, where the wh-word – cf. (1)–(3) above – is the focus of the question. This also applies in what Hannay (1991) calls the “reaction mode” of declarative discourse acts, as in (7): (7)

Brilliant that was!

where accessible information (that) is mentioned but is denied the status of topic, since (7) is not presented as being about “that.” Since the only pragmatic function present in (7) is the focus, it has no rival for initial position and that is where it appears. The phonological phrase that was in (7) can feel almost like an afterthought to a holophrase and is accordingly pronounced on a low, flat contour. The approach being sketched here applies without further ado to declaratives, be they exclamative or not. However, with other illocutions (interrogative, imperative), in English it is the marker of illocutionary status that appears in PI, reflecting the fact that in pragmatic organization these are hierarchically higher than the subacts that carry the functions topic or focus. The result is that any topic unit is pushed to a later position and the focus unit even later: (8)

a. Does (PI) that (Top > PI+1) matter (Foc > PI+2) ?[Interrogative] b. Give (PI) me (PI+1) a pen (Foc > PI+2) ! [Imperative]

The effect of postponing the focus unit can lead to an analytical decision to associate focus status (at least in certain configurations) with clause-final position (PF), where the language user’s strategy is geared to exploiting the very prominence of that final position (a strategy recognized in the literature as “end focus”).

6.4 Incrementality and Prediction in Production and Comprehension Finding the balance between accessibility (with such associated notions as grounding, topicality, and processing ease) and the need to make one’s point

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

6 The Grammar of Incremental Language Production

quickly in the rapid to and fro of dialogue (with associated notions like focality and attention) has been the subject of increasing interest in more recent psycholinguistic work. Both in theoretical linguistics and in psycholinguistics, however, there has been a tendency to concentrate on the single language user, doubtless because of the long-established emphasis on language competence and the associated individual cognitive abilities – as well as the logistical difficulties tied to experimentation with multiple subjects in conversation. Chomsky (1965: 3) famously stated that “linguistic theory is concerned with an ideal speaker-listener in a completely homogeneous speech community,” and despite taking issue with this characterization, many functional approaches have adopted a similar orientation to the single language user, as when Hengeveld and Mackenzie (2008: 1–2) present FDG as “a model of grammar [that] will be more effective the more its organization resembles language processing in the individual” (my emphasis). Pickering and Garrod (2021: 11) point out that the very same applies to the cognitive sciences, including psycholinguistics in general, which “have focused on individual rather than joint activities. Specifically, there has been little concern with the mental representations and processes that take place within individuals as they participate in joint activities . . . This is the case both for joint activity in general and for dialogue in particular.” Nevertheless, there have been repeated calls for a reorientation to dialogue, and specifically to context-embedded dialogue, in both psycholinguistic and more narrowly linguistic work. This has led to such initiatives in linguistics as Dynamic Syntax (Gregoromichelaki and Kempson 2019: 201), which adopts a view of syntactic structure “where utterances are taken as goal-directed physical actions coordinating with equally goal-directed mental actions within and across speakers . . . such actions are aimed to locally and incrementally alter the affordances of the internal/external context for both one’s self and one’s interlocutors.” The essence of Gregoromichelaki and Kempson’s approach is “a grammar architecture whose core notion is incremental interpretation of word sequences (comprehension) or linearisation of contents (production) relative to context” (p. 192), and their work features such “compound contributions” as (9): (9) A [seeing B emerge from a smoke-filled kitchen]: Did you burn B [interrupting]: myself ? No, fortunately not.

where only the dialogic co-construction of the shared (strictly speaking ungrammatical) utterance Did you burn myself? can explain the occurrence of myself rather than yourself (p. 190). A comparable position is adopted from a more psycholinguistic perspective by Momma and Philips (2018), who argue for a strong measure of similarity between language production and comprehension, specifically with regard to the granularity of incremental processing, that is, the extent to which speakers plan ahead as they utter each unit and the extent to which comprehenders predict upcoming material. The particularities of the language being spoken

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

149

150

J. LACHLAN MACKENZIE

clearly can interfere with what they call the “left-to-right” planning of lexical items in production (Momma and Philips 2018: 246). In a language with gender like German, for example, it seems likely that the (later) head noun will be planned before any associated definite determiner, the form of which is dependent upon the gender of the noun, whereas this is not a relevant factor in English where the definite determiner is invariable: cf. der Aufsatz (‘article,’ masculine) / das Buch (‘book,’ neuter) vs. the article / the book. Similarly, in comprehension, the gender of the definite article in German narrows the expectations of the hearer and, especially in a context that already limits plausible predictions (say, a discussion of academic publications), contributes to efficient processing. In incrementally decoding the initial words of a question Darf ich deinen . . . ‘May I [VERB] your . . .,’ the hearer can discount Buch and entertain Aufsatz as a possible continuation, because of the masculine-accusative marking of deinen, restricting his search space. This greater efficiency is bought, of course, at the expense of remembering the gender of each noun, a long-term memory requirement not imposed on the speaker of English. Much relevant work in psycholinguistics, understood as the study of how language is produced and comprehended, is oriented to laying bare what is involved in those processes. Accordingly, a focus on the timecourse of verbal interaction entails for many psycholinguists (e.g., Chater, McCauley, and Christiansen 2016) that there is no need to posit static and complete knowledge representations (forming a structured and autonomous language competence) that are independent of procedures for language processing, since these are seen as skills that are in large measure comparable with other timebound skills. The challenge is then to understand the mechanisms of those skills as against the limited capacity of working memory. The solution again lies, in ways that are reminiscent of the division of labor proposed in theoretical linguistics, in striking a balance between sequence and hierarchy, whereby successive units are recoded in comprehension into hierarchically higher units through a process, to use a term familiar from Section 6.3, of “chunking.” The resultant chunks are stored in an inventory that Christiansen and Chater (2016: 115) have dubbed a “chunkatory,” using an impromptu term that mimics the informal nature of the chunks that are assembled in this way. These are induced from the processes of comprehension and then immediately become available for production processes, that is, implementable as higher-order units available for incremental sequencing. The resultant sequences themselves form ever larger chunks, ultimately yielding a hierarchy or layered network of chunks of different degrees of inner complexity. There are strong parallels between this so-called Chunkand-Pass approach to production and comprehension (also often referred to as “cascaded processing”) and the proposals of the usage-based work in Construction Grammar (Goldberg 1995), where the structured network representing all of the linguistic units we know – of whatever complexity – is referred to as the “constructicon.”

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

6 The Grammar of Incremental Language Production

Whereas the Construction Grammar tradition focuses on the unification of chunks to form ever more complex signs, i.e., pairings of form and meaning, the psycholinguistic approach is less concerned with building up semantic representations of formal units and more with how they are interpreted, i.e., how they impact on the hearer’s general understanding of what is being said. These processes of interpretation are triggered by the words said but are only very partially linguistic and involve multiple pragmatic considerations such as the identification of referents or the differential status of information units as “new” or “given” and crucially also a range of contextual factors, which can include world knowledge of various degrees of generality or specificity, the visual context, the social context, and a wealth of prior experience of language chunks and of how they have been interpreted in the past. In interpreting linguistic input, the hearer is involved in a very rapid process of updating, using all imaginable aspects of context to fit the incoming information into a consistent and relevant understanding of the ongoing discourse. It is becoming increasingly clear that these interpretation processes have to contend with a massive amount of ambiguity in the linguistic input. Piantadosi et al. (2012) have not only demonstrated the ubiquity of multi-interpretable chunks of language but also argued convincingly that the lack of bijective relations between form and meaning is actually “a desirable property of communication systems” (p. 281) because it keeps utterances short and simple. The apparent shortcomings of the signal are more than adequately compensated for by the fact that context is so richly informative about meaning. Hearers draw on context to infer the intended meaning: Any extra cost that this may entail (typically in the form of abductive reasoning) is fully offset by the benefits of having an efficient system that makes copious reuse of the same resources. As Piantadosi et al. (2012: 283) remark, “when the individual units of [such] an efficient communication system are viewed out of their typical contexts, they will look ambiguous”; however, it is only in linguists’ abstract examples that the appearance of ambiguity is troubling. One famous case of such ambiguous-when-out-of-context sentences is the kind of “garden-path” sentences created by researchers with the intention of leading experimental subjects down a syntactic garden path (the reference here is to an idiom that means “cause to take the wrong route”). The prediction is that because of the constructed ambiguity the subjects, confronted with such a sentence out of context, will parse it in the most obvious way, then discover that their parse leads to a dead-end, backtrack, and finally parse the sentence successfully at the second attempt. A familiar example is The horse raced past the barn fell, where raced is initially parsed as a finite verb in the past tense only for the reader to encounter another finite verb in the past tense (fell): [[The horse]NP [raced past the barn]VP [fell]VP]. The reader then tries again, with raced now as a past participle, and the result is a successful parse: [[The horse raced past the barn]NP [fell]VP]. However, examination of the use of such garden-path sentences in context (Ferreira 2008) has shown that they rarely cause any difficulties because the reader who has already encountered “the horse” (definite and

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

151

152

J. LACHLAN MACKENZIE

specific) will then typically also know whether the horse was being ridden, which would favor the second reading, or running free, which would favor the first. Even the notion of the “most obvious” way to parse a sentence makes an implicit reference to imagined contexts: Since general knowledge tells us that horses are swift animals, it is natural to think of one “racing” and to try that interpretation first. It has been argued that insights into how the pieces of incoming information are integrated into the hearer’s understanding of the ongoing discourse (the Chunk-and-Pass account) are insufficient to account for the astounding rapidity with which speech is processed and with which speakers initiate their turns. Those insights have been supplemented in recent work by a realization that hearers additionally use probabilistic cues to predict future words and chunks, and even how these will be interpreted, massively augmenting the efficiency of language processing. Rappe (2019) refers to this perspective as the Predictive Processing framework. It shares the notion that the hearer converts a linear signal through chunking into a hierarchical representation but differs in making crucial reference to context in the sense of prioritizing higher-level representations, i.e., those that contextualize the ongoing discourse. These relatively general representations become the basis for the hearer’s predictions of what is to come; as Rappe (2019: 362) observes, “[l]inguistic processing is at least minimally predictive in that linguistic stimuli come with some sort of context, even if what we call context is only the previously processed linguistic information. This information reaches and affects the system before the input becomes available and facilitates further processing.” If the context is already activated before any linguistic input arrives, it will only be further enriched once the speech begins to flow. One of the options considered by Rappe is that since the aim of understanding is to achieve a satisfactory interpretation of the discourse as a whole, lower-level representations may not be fully constructed in practice (however much linguists may squabble about the details). This is the notion of “good-enough processing” as first presented by Ferreira et al. (2002), who found that experimental subjects, confronted with out-of-context sentences, regularly come up with interpretations that are at odds with the complete semantic representation of those sentences. Rather than analyzing the meaning in the manner of a semanticist, the language user employs expectations about the discourse context to predict what a sentence will mean and may then also persist with that interpretation despite linguistic evidence to the contrary. In other words, hearers are found to sidestep strict one-by-one incremental build-up of the meaning of sentences and to commit early to higher-level representations. The human brain as a predictive organ is central to Clark’s (2015) proposals for the processing of perceptual information. Rather than a one-way flow of sensory information from the external world to the brain, Clark posits sensory information as a potential corrective to top-down predictions: Incoming stimuli then either confirm the predictions or adjust them for future occasions. The top-down predictions can be so strong as to cause the brain to ignore or

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

6 The Grammar of Incremental Language Production

“explain away” contradictory information. Rappe (2019: 374) received a personal communication from Clark in 2018, speculating that “language has evolved specifically to ‘fit’ our multi-level predictive brains.”

6.5 Dialogue The assumption behind much of the work summarized in the preceding sections is that the human being is inherently social and interactive and that the primary arena for all communication, whether by speech or not, is dialogic interaction in context (Weigand 2017). The grammatical framework to which most attention has been devoted in this chapter, Functional Discourse Grammar, admittedly espouses an architecture that, in line with much grammatical tradition, portrays the linguistic activity of the individual speaker, but that grammar is only part of an overall model of verbal interaction, which assumes the co-presence of an addressee and a “contextual component” to which both dialogists have access and which represents the common ground that is the basis for dialogue. Certain aspects of this context are relatively timestable (e.g., institutional status and roles, or identities in terms of seniority, gender, sexual orientation, ethnicity, physical ability), while others are in continual flux, reacting to the vicissitudes of the moment, to the continual updates arising from the ongoing interchange (Hengeveld and Mackenzie 2014; Mackenzie 2014b). For a thorough-going consideration of how common ground is constructed in the process of interaction as a “shared workspace,” see Pickering and Garrod (2021). Where the context is relatively stable and mutually understood, communication can in principle take place without any use of language, even between individuals capable of language use: Pointing, other gestures, eye movements, etc. may function as discourse acts and be enough to guarantee successful interaction. There are also specific situations in which silent communication is indeed favored, for example, in hunting to avoid being heard by the prey or in secret gestural behind-the-back communication between doubles partners in a tennis competition, or where the background noise is so intense that gestures are the only way to communicate information. The amount of information that can be imparted in this manner is limited, of course, in the first instance to indexical identification of some aspect of the physical context and human movement within that context. But the dependence upon context is extreme in such communication and it is prone to misunderstandings. The challenge may lie in distinguishing between gestures that arise from a communicative intention and other chance bodily movements that are not designed to convey information. However, there are many situations in which it is perfectly possible for adults who are capable of speech to cooperate without talking. As the Viennese linguist Bühler (1990: 47) pointed out, “Productive and creative man, working in cooperation with his fellows, often remains silent as long as

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

153

154

J. LACHLAN MACKENZIE

each fully understands what the other is doing and acts appropriately.” Such situations can be said to exploit the context of communication to the maximum: Where there is a high measure of common ground, speech is redundant and indeed superfluous. Bühler (1990: 176) asks, “To what end should one speak if one gets on just as well or better in practical life without speaking?” He answers by invoking “islands of language” emerging “from within the sea of silent but unequivocal communication at the places where a differentiation, a diacrisis, a decision between several possibilities has to be made, and easily can be made by interspersing a word,” exemplifying this statement with his memories of earlier times in Vienna when there was a single kind of tram ticket and silent interaction with the conductor sufficed; only with the introduction of simple and transfer tickets did the need for the “empractical” (i.e., standalone) utterances Geradeaus “straight through” and Umsteigen “change” arise. The holophrastic utterances in question are, it should be noted, no more complex than is required in the highly restricted framework of two public transport options. One is an adverb, and the other is a verb in the infinitive form, but each is complete in its context; no misunderstanding is possible. Of course, the major emphasis in studies of dialogic interaction has been on less constrained situations, where the syntax is more sophisticated and the full multimodal mix is available (notably prosody and gesture). The dialogists have a sufficiently shared awareness of the context and sufficient experience of how it interacts with the linguistic cues on offer to make communication in most situations satisfactory for both participants. As has been mentioned in previous sections, language is both an imperfect tool, in that the relation between communicative intentions and linguistic forms is complex and insecure, and also a close-to-perfect tool in making creative use of a restricted set of syntactic and lexical instruments. The result is dialogue as a form of tacit cooperation (even in the case of a furious row between the interlocutors), in which throughout each utterance the operations of chunk-and-pass and of predictive processing actively contribute to the inferences that each is drawing about the other’s communicative intentions. Understanding is never guaranteed, however, and conversations can jump to the meta-level, with requests being made and granted for reformulations, before the conversation returns to the question under discussion. The dialogical perspective upon language use draws attention away from the criterion of grammaticality toward that of situation-appropriateness (Anward and Linell 2016). From the viewpoint of the participants in the dialogue, the ultimate questions are: How much does the utterance contribute to the achievement of our goals as interactants? And how well does it fit into the context in which we find ourselves and which we are co-creating? Prioritizing answers to these questions leads Anward and Linell, in their study of dialogues in Swedish, to make observations that are unattainable with a methodology that considers only sentences in isolation. One is the ubiquity of “responsiveness,” i.e., structures designed to occur in reactive moves, as in (10):

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

6 The Grammar of Incremental Language Production

(10)

a.

b.

å vilken hjälp har du and what help have you ‘And what help did you get?’ inte jättemycke hjälp har NEG awfully.much help have ‘Not awfully much help did I get.’

fått? received ja I

inte NEG

fått. received

The structure with the double negative shown in (10b) only occurs in responses: The initial focus “not awfully much” morphs into a topic “awfully much” with its comment “I did not receive.” The initially placed focus serves the dialogic purpose of answering the question promptly (see the discussion of Hannay’s “reaction mode” in Section 6.3 above), while the topic structure with its reuse of hjälp and fått contributes to the alignment of question and answer. Another property found in the Swedish dialogues is “incrementality,” understood as we have seen as the stepwise build-up of an utterance, with the speaker not knowing exactly how it is going to end; it may even “end up in an incoherent syntactic structure, as judged from conventional sentence grammar” (Anward and Linell 2016: 17; emphasis in original), but without triggering any reformulations or requests for repair. A final property is “syntactic ambiguity,” understood here as referring to multiple internal relations that arise from changes in pragmatic properties in the course of the utterance, as in (11), which is not “grammatical” as it stands but shows evidence of its piecemeal construction in real time: (11)

de va bra (.) att it was good P A U S E that ‘It was good that he did it.’

han he

gjorde. did

In this case, the pronoun de is initially understood as anaphoric; then the delayed incrementation of the “extraposed” subordinate clause renders it cataphoric; and finally, it comes to be interpreted as the object of gjorde. This example shows the dynamic interaction between linguistic form and context: The speaker initially assumes contextual clarification of the implicit reference of de; then after a pause makes this explicit but in the meantime has reinterpreted the de as the topic of the utterance and yet semantically integral to the subordinate clause, which consequently lacks its full complement of arguments. These examples manifest a usage-based view of linguistic data, presented in the form of transcriptions of speech, as protocols of joint action that took place between certain individuals at a certain time in the past. The analyst can discern, in the left-to-right sequencing of data, the effects of those individuals’ incremental co-creation of meaning. However suggestive such linguistic data may be, the need for objective criteria has encouraged the use of various new technologies designed to detect the cognitive processing underlying dialogical speech. It has long been understood, relatively informally, that such matters as speech-accompanying gestures, changes in

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

155

156

J. LACHLAN MACKENZIE

interactants’ direction of gaze, interruptions, and requests for speech repairs can all be treated as clues to what is going on, but it is only fairly recently that equipment for tracking saccades (rapid movements of the eye between fixation points) has become available, allowing precise conclusions to be drawn about how dialogists interpret each other’s disfluencies and prosody. In particular, use is made in laboratory settings of “confederates,” that is, members of the team who enter into dialogue with experimental subjects who are fitted up with the gaze-tracking equipment. The confederates work with scripts that maneuver the naïve subjects toward certain ambiguities, allowing the researchers to infer relevant disambiguating cognitive processes from the subjects’ eye movements (see Brennan and Hanna 2017 for details). It is expected that future developments will reinforce our understanding of the sequences of cognitive processes that underlie the incremental build-up of discourse acts and longer stretches of discourse.

6.6 Conclusion This chapter has offered an overview of developments that reflect a convergence of work in linguistics and psycholinguistics around the implications of the piecemeal composition of units of speech for an understanding of grammar and the cognitive processing that underlies the production, comprehension, and interpretation of utterances. Notions from Functional Discourse Grammar have been employed to present a view of syntactic structure as arising from the incremental expansion of the minimal utterance, the holophrase. By prioritizing the timecourse of language processing, we have interpreted syntactic hierarchy as arising from chunk-and-pass operations, arguably supported by predictive processing. Finally, our attention turned briefly to dialogue as a general framework for the understanding of interaction, communication, and language. All of these notions make essential reference to context as a necessary component of the operations and processes that we have reviewed.

References Anward, J., and Linell, P. (2016). On the grammar of utterances: Putting the form vs. substance distinction back on its feet. Acta Linguistica Hafniensia, 48(1), 35–58. Boeckx, C. (2006). Linguistic Minimalism: Origins, Concepts, Methods, and Aims. Oxford: Oxford University Press. Bošković, Ž. (2022). Formalism and, not vs, functionalism. Lingbuzz. https://ling .auf.net/lingbuzz/006337. Brazil, D. (1995). A Grammar of Speech. Oxford: Oxford University Press. Brennan, S. E., and Hanna, J. E. (2017). Psycholinguistic approaches: Meaning and understanding. In E. Weigand (ed.), The Routledge Handbook of Language and Dialogue (pp. 93–108). New York/Abingdon: Routledge.

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

6 The Grammar of Incremental Language Production

Bresnan, J. (2001). Lexical-Functional Syntax. Oxford: Blackwell. Bühler, K. (1990). Theory of Language (trans. of Sprachtheorie, 1934). Amsterdam/ Philadelphia, PA: John Benjamins. Chafe, W. (1994). Discourse, Consciousness and Time. Chicago/London: Chicago University Press. Chater, N., McCauley, S. M., and Christiansen, M. H. (2016). Language as skill: Intertwining comprehension and production. Journal of Memory and Language, 89, 244–254. Chomsky, N. (1965). Aspects of the Theory of Syntax. Cambridge, MA: MIT Press. Christiansen, M. H., and Chater, N. (2016). Creating Language: Integrating Evolution, Acquisition, and Processing. Cambridge, MA: MIT Press. Cinque, G. (1999). Adverbs and Functional Heads: A Cross-linguistic Perspective. New York: Oxford University Press. Clark, A. (2015). Surfing Uncertainty: Prediction, Action, and the Embodied Mind. Oxford: Oxford University Press. Culicover, P. W., and Jackendoff, R. (2005). Simpler Syntax. Oxford: Oxford University Press. Dalrymple, M., and Findlay, J. Y. (2019). Lexical functional grammar. In A. Kertész, E. Moravcsik, and C. Rákosi (eds.), Current Approaches to Syntax: A Comparative Handbook (pp. 123–154). Berlin/New York: De Gruyter Mouton. Everett, D. L. (2015). The role of culture in the emergence of language. In B. MacWhinney and W. O’Grady (eds.), The Handbook of Language Emergence (pp. 354–376). Malden, MA/Oxford/Chichester: Wiley Blackwell. Ferreira, F., Bailey, K. G. D., and Ferraro, V. (2002). Good-enough representations in language comprehension. Current Directions in Psychological Science, 11(1), 11–15. Ferreira, V. (2008). Ambiguity, accessibility, and a division of labor for communicative success. Learning and Motivation, 49, 209–246. Fortescue, M. (2004). The complementarity of the process and pattern interpretations of Functional Grammar. In J. L. Mackenzie and M. Á. Gómez González (eds.), A New Architecture for Functional Grammar (pp. 151–178). Berlin/New York: Mouton de Gruyter. Frank, S. L., Bod, R., and Christiansen, M. H. (2012). How hierarchical is language use? Proceedings of the Royal Society B: Biological Sciences, 279 (1747), 4522–4531. Givón, T. (1995). Functionalism and Grammar. Amsterdam/Philadelphia, PA: John Benjamins. Goldberg, A. (1995). A Construction Grammar Approach to Argument Structure. Chicago: University of Chicago Press. Greenfield, P., Reilly, J., Leaper, C., and Baker, N. (1985). The structural and functional status of single-word utterances and their relationship to early multi-word speech. In M. Barrett (ed.), Children’s Single-Word Speech (pp. 233–267). Chichester: Wiley. Gregoromichelaki, E., and Kempson, R. (2019). Procedural syntax. In K. Scott, B. Clark, and R. Carston (eds.), Relevance, Pragmatics and Interpretation: Essays in Honour of Deirdre Wilson (pp. 209–246). Cambridge: Cambridge University Press. Groenendijk, J., and Stokhof, M. (2006). Dynamic semantics. In K. Brown (ed.), Encyclopedia of Language and Linguistics, 2nd ed. (Vol. IV, pp. 28–33). Oxford: Elsevier. Hannay, M. (1991). Pragmatic function assignment and word order variation in a functional grammar of English. Journal of Pragmatics, 16, 131–155.

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

157

158

J. LACHLAN MACKENZIE

Hengeveld, K., and Mackenzie, J. L. (2008). Functional Discourse Grammar: A Typologically-Based Theory of Language Structure. Oxford: Oxford University Press. Hengeveld, K., and Mackenzie, J. L. (2014). Grammar and context in Functional Discourse Grammar. Pragmatics, 24(2), 203–227. Huddleston, R., and Pullum, G. K., eds. (2002). The Cambridge Grammar of the English Language. Cambridge: Cambridge University Press. Jackendoff, R. (1977). X’ Syntax: A Study of Phrase Structure. Cambridge, MA: MIT Press. Jackendoff, R., and Audring, J. (2020). The Texture of the Lexicon: Relational Morphology and the Parallel Architecture. Oxford: Oxford University Press. Jackendoff, R., and Wittenberg, E. (2014). What you can say without syntax: A hierarchy of grammatical complexity. In F. J. Newmeyer and L. B. Preston (eds.), Measuring Linguistic Complexity (pp. 65–82). Oxford: Oxford University Press. Kornmesser, S. (2019). The multiparadigmatic structure of science and generative grammar. In A. Kertész, E. Moravcsik, and C. Rákosi (eds.), Current Approaches to Syntax: A Comparative Handbook (pp. 493–520). Berlin/New York: De Gruyter Mouton. Levinson, S. C. (1987). Minimization and conversational inference. In M. Bertuccelli Papi and J. Verschueren (eds.), The Pragmatic Perspective: Selected Papers from the 1985 International Pragmatics Conference (pp. 61–129). Amsterdam/Philadelphia, PA: John Benjamins. Luraghi, S., and Parodi, C., eds. (2013). The Bloomsbury Companion to Syntax. London/New York: Bloomsbury Academic. Mackenzie, J. L. (2014a). Morphosyntax in Functional Discourse Grammar. In A. Carnie, D. Siddiqi, and Y. Sato (eds.), Routledge Handbook of Syntax (pp. 627–646). London: Routledge. Mackenzie, J. L. (2014b). The Contextual Component in a dialogic FDG. Pragmatics, 24(2), 249–273. Mackenzie, J. L. (2020). Functional approaches. In B. Aarts, J. Bowie, and G. Popova (eds.), Oxford Handbook of English Grammar (pp. 180–200). Oxford: Oxford University Press. Momma, S., and Phillips, C. (2018). The relationship between parsing and generation. Annual Review of Linguistics, 4, 233–254. Müller, S., and Machicao y Priemer, A. (2019). Head-Driven Phrase Structure Grammar. In A. Kertész, E. Moravcsik, and C. Rákosi (eds.), Current Approaches to Syntax: A Comparative Handbook (pp. 317–359). Berlin/New York: De Gruyter Mouton. Nordström, J. (2017). Language without narrow syntax. The Linguistic Review, 34(4), 687–740. Pesetsky, D. (2013). Что дѣлать? What is to be done? Plenary talk to the annual meeting of the Linguistics Society of America, January 4, 2013. Piantadosi, S. T., Tily, H., and Gibson, E. (2012). The communicative function of ambiguity in language. Cognition, 122 (3), 280–291. Pickering, M. J., and Garrod, S. (2021). Understanding Dialogue: Language Use and Social Interaction. Cambridge: Cambridge University Press. Pollard, C., and Sag, I. (1994). Head-Driven Phrase Structure Grammar. Chicago: University of Chicago Press.

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

6 The Grammar of Incremental Language Production

Rappe, S. (2019). Now, never, or coming soon? Prediction and efficient language processing. Pragmatics & Cognition, 26(2/3), 357–385. Rizzi, L., ed. (2004). The Structure of CP and IP: The Cartography of Syntactic Structures, Vol. II. New York: Oxford University Press. Sadock, J. M. (2012). The Modular Architecture of Grammar. Cambridge: Cambridge University Press. Searle, J. (1969). Speech Acts: An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. Sinclair, J., and Mauranen, A. (2006). Linear Unit Grammar: Integrating Speech and Writing. Amsterdam/Philadelphia, PA: John Benjamins. Steedman, M. (2022). Combinatory Categorial Grammar. https://homepages .inf.ed.ac.uk/steedman/papers/ccg/moravcsik2.pdf. Wagner, M. (2016). Information structure and production planning. In C. Féry and S. Ishihara (eds.), The Oxford Handbook of Information Structure (pp. 541–561). Oxford: Oxford University Press. Weigand, E. (2017). The concept of language in an utterance grammar. In E. Weigand (ed.), The Routledge Handbook of Language and Dialogue (pp. 214–233). New York/ Abingdon: Routledge. Wray, A. (2002). Formulaic Sequences and the Lexicon. Cambridge: Cambridge University Press.

https://doi.org/10.1017/9781108989275.007 Published online by Cambridge University Press

159

7 Cognitive Linguistics and Context Dirk Geeraerts

7.1 Introduction Cognitive Linguistics, as meant in this chapter, is the linguistic framework represented by handbooks like Evans and Green (2006), Geeraerts and Cuyckens (2007), Littlemore and Taylor (2014), Dąbrowska and Divjak (2015), Dancygier (2017), or Wen and Taylor (2021), readers like Evans et al. (2007) or Geeraerts (2006), and shorter introductory texts like Winters and Nathan (2020). The very fact that I am identifying the approach by means of encompassing reference works and not by referring to a single dominant name is significant. Generative grammar is predominantly and perhaps overwhelmingly Chomskyan grammar, just like Systemic Functional Grammar is Hallidayan linguistics, but Cognitive Linguistics is not the linguistic approach developed by a single Big Name. Rather, like perhaps American structuralism or the Prague School, it is best characterized as a movement fed by the ideas of a number of inspirational figures. In the first generation of cognitive linguists, these would include people like Ronald Langacker, George Lakoff, Leonard Talmy, Gilles Fauconnier, with other key figures emerging in later generations. The recognition that Cognitive Linguistics is a collective enterprise (and perhaps a rather loose one at that) raises the question of what holds the framework together: There will surely be an affinity between the various ideas introduced by the central members, but how can that affinity be best described? A network of thinkers produces a network of analytical concepts and descriptive practices, but what are the cohesive forces within that network? In this chapter, I will develop the idea that context is a crucial factor not only for the internal cohesion of Cognitive Linguistics but also for the position of Cognitive Linguistics in the evolution of linguistics in the twentieth and twenty-first centuries. In particular, I will argue that Cognitive Linguistics is a recontextualizing framework par excellence, i.e., one that reintroduces the contextual embeddedness of language in linguistic theory formation and that in doing so it counterbalances the decontextualizing tendencies that are at work in formal grammar and formal semantics.

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

7 Cognitive Linguistics and context

Three caveats will be useful before we proceed. First, emphasizing the importance of context for Cognitive Linguistics does not imply that it is the only contemporary or recent framework in linguistics in which context plays a central role. In particular, much of functional linguistics is about contextual factors and language as a social semiotic with a communicative function. However, a systematic comparison between functional and Cognitive Linguistics (as in Nuyts 2007) is beyond the scope of the present chapter. Second, as implied by the previous paragraph, I will try to do justice to Cognitive Linguistics as a diverse framework that brings together different, largely complementary approaches to language, without privileging any one in particular. Specifically, although some readers may tend to associate Cognitive Linguistics predominantly with the Lakoffian tradition of research into conceptual metaphors and embodied cognition, they should be aware that this line of investigation, however impressive, is but one strand within the larger texture of Cognitive Linguistics. Third, I have made the point about the recontextualizing nature of Cognitive Linguistics before (Geeraerts 2010, 2015), and some readers might remark that the following pages are basically a recombination of ideas that I expressed earlier. They would not be wrong, but even if the present text will repeat, to some extent literally, my earlier treatments of the issue, it will also offer a more analytical and systematic view of the recontextualizing core of Cognitive Linguistics.

7.2 Cognitive Linguistics in the History of Modern Linguistics To understand the specific position of Cognitive Linguistics in the history of modern linguistics, we may focus on two aspects of generative grammar: its changing view on the role of semantics, and its overall decontextualization of grammar as autonomous syntax.

7.2.1 Meaning in Generative Grammar As a first step, we should go back to the beginning of generative grammar, and to the changing position of semantics in the internal development of the theory. In the framework defined by Noam Chomsky’s Syntactic Structures (1957), meaning does not play a role in the conception of grammaticality (and a fortiori, in the grammar as the rule system governing that grammaticality). The iconic sentence Colorless green ideas sleep furiously is considered meaningless, but at the same time it is taken to be grammatical, because its syntactic structure corresponds entirely to a fully grammatical sentence like Bright young linguists talk endlessly. Meaningfulness, in other words, is not a criterion for grammaticality, but syntactic well-formedness is. In the model defined by Aspects of the Theory of Syntax, Chomsky (1965) switches position. The description of meaning is incorporated into grammar, and while Colorless green ideas sleep furiously is still considered to be meaningless, it acquires the

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

161

162

DIRK GEERAERTS

status of an ungrammatical sentence that needs to be excluded by the formal grammar: Ungrammaticality may be due to semantic ill-formedness. The incorporation of meaning and semantic well-formedness into the heart of the grammar did, however, create a problem with regard to the notion of “transformation,” which was at that point another crucial aspect of the formal framework of generative grammar. The algorithmic description of the grammatical structures of a language went in two steps. First, phrase structure rules produce an initial syntactic tree, the so-called “deep structure,” which may then, second, be transformed into a different type of tree, the “surface structure.” The question of whether transformations are meaning-preserving then became a hotly debated topic in generative theory, ultimately leading to a rift between two virulently opposed camps. On the one hand, if you believe that transformations are meaningpreserving, all the semantic information you need is already available in the deep structure, and the deep structure as such becomes equivalent to the semantic description of the sentence. Semantics accordingly takes precedence in the linguistic description; the first and basic step in the process of building syntactic structures is a semantic one. This was the position taken by the so-called Generative Semantics movement, which in works like McCawley (1968) developed its program by creating a hybrid and rather ill-fitting fusion of linguistic syntax with descriptive notions taken from formal logic. On the other hand, if you believe that transformations can change meaning, the primacy of syntax can be maintained: Semantic interpretation will be placed at the end of the process of building grammatical tree structures. This was the position that was ultimately favored by Chomsky, for reasons that can be easily understood in light of one of Chomsky’s basic targets for linguistic theorizing, namely, to explain the process of language acquisition. If, like Chomsky, you believe in a genetic endowment for language, then it is highly unlikely that that genetic module will involve something as ephemeral and diverse and variable as meaning. If there is an ingrained linguistic knowledge (and as is well known, Chomsky believes there has to be to explain language acquisition), it is more likely to involve the formal, structural aspects of the language, i.e., syntax. This is all the more so because syntax underlies that other main feature of language emphasized by Chomsky: the capacity of human beings to produce an infinite number of different sentences. Therefore, within the generative tradition, Generative Semantics lost out against so-called Interpretive Semantics, which then became the first of a series of successive models within generative grammar ultimately leading to the current notion of Universal Grammar. What all of these models (and a number of other theories of formal grammar) have in common is the idea of an autonomous syntax, that is, the notion that the syntax of the language is a module of the grammar that stands on its own and that can be described largely independently of considerations of meaning and function. Such a module will surely interact with semantics and pragmatics, but essentially works according to its own set of principles. But importantly from a historical point of view, the

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

7 Cognitive Linguistics and context

demise of Generative Semantics within the generative grammar tradition became an important stimulus for the development of semantics in two different traditions outside of generative grammar. First, some of the linguists who were active in the broad circle of Generative Semantics became founding figures of what we now know as Cognitive Linguistics. This applies to George Lakoff and Ronald Langacker, and to some extent also to Charles Fillmore, who was a major inspiration for Cognitive Linguistics but who never self-identified as a cognitive linguist. Second, the rapprochement with logical semantics which was rather clumsily executed by Generative Semantics motivated scholars with a background in logic, like Richard Montague (1974) and Barbara Partee (1976), to develop a formal kind of natural language semantics that was firmly and unambiguously rooted in logical semantics, and that became the basis for the very rich tradition of formal semantics as we currently know it. So overall, if we look at the role and position of meaning, we may distinguish three broad strands of thought in current linguistic theory: formal syntax, formal semantics, and cognitive semantics. The first two share the interest in a formal, symbolically representational approach to linguistic description that was introduced by generative grammar, and while the first approach minimizes the role of meaning, the last two frameworks endorse the primacy of semantics that was originally (but unsuccessfully) promoted within generative grammar by the Generative Semantics movement. Characteristically though, the logical inspiration of formal semantics leads to an entirely different type of meaning description from what is common in Cognitive Linguistics. Not only is formal semantics couched in the formal language of logic, but it focuses on the truth-functional compositionality of linguistic utterances. Given a strictly referential conception of meaning, it examines how the truth-functional properties of an entire proposition are compositionally derived from the properties of its components. Cognitive semantics, by contrast, does not restrict linguistic meaning to its referential and truth-functional aspects but starts from a broad conception of meaning in which imagery, affect, experience may all take pride of place, and in which rigid formalization becomes less of a desideratum. For instance, to come back to the sentence with which we started this section, Cognitive Linguistics will readily accept the possibility of a metaphorical reading of Colorless green ideas sleep furiously (in the sense, say, in which not very inspiring ecological ideas impatiently remain hidden and inactive) – a figurative interpretation that lies beyond the range of most types of formal semantics. So, as a first characterization of the theoretical position of Cognitive Linguistics, we can think of it historically as an approach in which, in the area of meaning description, demands of descriptive scope take precedence over the pursuit of formalizability.

7.2.2 Systemic Decontextualization We need to take a further step backwards to enrich the picture painted so far. Generative grammar separated grammar, reduced to autonomous syntax, from a thick conception of meaning, but it also separated it from other contextual

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

163

164

DIRK GEERAERTS

features. To see how this happened, we need to specify the relationship between generative grammar and structuralism. In its Saussurean form, structuralism emphasizes the distinction between language system (langue, a social phenomenon) and language use (parole, an individual, psychological phenomenon). By favoring langue over parole, and by thinking of langue as a homogeneous “system,” this distinction introduces an element of decontextualization: Whereas language use is by definition contextualized because it happens in a specific situation and environment, focusing on the linguistic system, by contrast, means abstracting away from these specific speech situations and looking for what is invariant at the level of the common system in contrast with the variability at the level of individual language use. But generative grammar takes decontextualization even further. The Saussurean framework makes a distinction between the system of language as a social phenomenon and language use as an individual phenomenon, but it doesn’t have much to say about the psychological bridge between system and use. To explain how language users can use the linguistic system to begin with, a third notion needs to be introduced that is as such absent from the Saussurean model, namely, the individual language user’s knowledge of the linguistic system. It is that knowledge that mediates between the individual activity of the language user and the existence of the linguistic system as a social phenomenon, and it is that knowledge, equally obviously, that is captured by Chomsky’s notion of “competence.” While Chomskyan “performance” and Saussurean parole are essentially the same (and in both cases, equally secondary in the hierarchy of things linguistic), Chomskyan competence fills the psychological gap that exists between langue as system and parole as activity: It introduces the individual’s knowledge of the language system as a crucial element of linguistic description. At the same time, however, Chomskyan generativism leaves a new gap by ignoring or neglecting the social nature of language. The lack of relevance (for his research program, that is) that Chomsky has generally attributed to the social aspects of language involves not just sociolinguistics and language-internal language variation, but more importantly also the social aspects of the process of language acquisition. If language is a social semiotic in the Saussurean sense, the acquisition of language is, at least to some extent, a process of cultural transmission: It is part of a socialization and acculturation process. Conversely, in the Chomskyan mindset, the acquisition of language is not primarily grounded in social interaction but based on man’s genetic endowment for language. Language acquisition in the Chomskyan view may require some input from the environment, but it is essentially a process of biological growth, not one of social integration. In this sense, there is a correlation between Chomsky’s limited attention for the social dimension of language and his genetic view of language acquisition. Let us summarize. Generative grammar, by defining grammar in terms of an autonomous, genetically entrenched universal syntax, dissociates what it considers to be the core discipline of linguistics from meaning, from the social

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

7 Cognitive Linguistics and context

function of language, from language usage. When Cognitive Linguistics emerged in the 1980s, it defined itself in contrast with the then dominant generative paradigm on two dimensions. First, against the minimization of semantics in generative grammar (and also against the reduction of meaning in formal semantics), it put a broad conception of meaning at the heart of linguistic description. And second, it reacted against the dissociation of grammar from social variation and actual usage. More precisely, we can describe Cognitive Linguistics as a systematic attempt to recontextualize linguistics on both relevant dimensions: It recontextualizes semantics, and it recontextualizes the grammar at large. In the following two sections, we will explore what this recontextualization implies by examining each of both dimensions in more detail. More precisely, I will argue that there are two main forms in which Cognitive Linguistics recontextualizes the linguistic conception of meaning, and then again, that there are two main forms in which it recontextualizes the linguistic conception of language. These two pairs of recontextualizing tendencies are not unrelated and describing the links between them will be the subject matter of a brief final section.

7.3 Recontextualizing Meaning To appreciate how Cognitive Linguistics recontextualizes meaning, we will first go over the most prominent models for meaning description suggested by Cognitive Linguistics, and then identify the two major ways in which these approaches embody a contextualized perspective on meaning.

7.3.1 Models of Meaning in Cognitive Linguistics The innovations brought to semantics by Cognitive Linguistics fall roughly into three groups. First, a number of important ideas put forward by cognitive semantics focus on the internal semantic structure of natural language expressions, i.e., the relationship between the various senses of the expressions. Thus, the “radial network” model (Lakoff 1987) describes a category structure in which a central case of the category radiates toward novel instances: Less central category uses are extended from the center. Brugman’s seminal study (1988) of over, for instance, takes the “above and across” reading of over (as in the plane flew over) as central, and then shows how less central readings extend from the central case. These can be concrete extensions, as in a “coverage” reading (the board is over the hole), but also metaphorical ones, as in temporal uses (over a period of time). Radial categories constitute but one type of a broader set of models that fall under the heading of “prototype theory” (Rosch 1978; Geeraerts 1989; Taylor 1989). In general, these look at the interaction between the extensional aspects of the category (specifically, the salience of category members) and their intensional aspects (the definitional demarcation between senses and the possibility of formulating a definition of

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

165

166

DIRK GEERAERTS

the category in terms of necessary and sufficient conditions). This interaction can take various forms, to the extent that it has been argued that the notion of “prototype” is itself a prototypically structured one, i.e., that there is no single definition that captures all and only the diverse forms of “prototypicality” that linguists have been talking about. Second, other branches of cognitive semantics concentrate on the conceptual mechanisms that realize the creation of new meanings, specifically, metaphor and metonymy. The notion of “conceptual metaphor” (introduced by George Lakoff’s and Mark Johnson’s book Metaphors We Live By of 1980) captures the recognition that a given metaphor need not be restricted to a single lexical item but may generalize over different expressions. Such general patterns may then be summarized in an overall statement like L O V E I S W A R , a pattern that ranges over expressions like He is known for his many rapid conquests, He fled from her advances, She is besieged by suitors. The attention for metaphor sparked off a number of related concepts. Image schemas (Johnson 1987; Hampe 2005) are regular patterns of sensory or motor experiences that recur as a source domain (or a structuring part of a source domain) for different target domains. Typical image schemas include containment, path, scales, verticality, and center-periphery. If metaphor is analyzed as a mapping from one domain to another, the question arises as to how such mappings take place: How does the structure of the source domain get mapped onto the target domain? The notion of “conceptual integration” (aka blending) developed by Gilles Fauconnier and Mark Turner (Fauconnier 1985; Fauconnier and Turner 2002) provides a descriptive framework to answer that question; it describes the derived reading as a merger of the source space and the target space. Although this “mental spaces” approach clearly links up with the analysis of metaphor as mapping across domains, the conceptual integration framework is more general, in the sense that it can be applied to a variety of phenomena, many of which (like counterfactuals) are not related to metaphor. Although less dominant than conceptual metaphor and its offshoots, metonymy research (Kövecses and Radden 1998; Barcelona 2000; Dirven and Pörings 2002) also received an important impetus in cognitive semantics, more specifically from the suggestion that metonymy could receive a definition that mirrors that of metaphor: If metaphor is seen as a mapping from one domain to the other, metonymy can be seen as a mapping within a single domain or a domain matrix. This suggestion is not without issues (the relevant definition of “domain” is not self-evident), but regardless, the cognitive linguistic study of metonymy is thriving. A third group of models examines the mechanisms of grammatical construal in languages: How do the grammatical resources of a language contribute to the conceptualization of reality? This approach, epitomized by the work of Ron Langacker (1987, 1991) and Len Talmy (2000), analyzes the semantic import of syntactic, morphological, constructional mechanisms. Thus, for instance, Langacker suggests a semantic definition of word classes. On conceptual grounds, he distinguishes between a number of basic classes of

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

7 Cognitive Linguistics and context

predications: entities and things versus relations, and within the relational predicates, stative relations, complex atemporal relations, and processes. The formal word classes of a language, he then argues, typically express a basic type of predication. For instance, while nouns express the notion of “thing” (a bounded entity in some domain), adjectives will typically be stative relational predicates, and verbs processes. Langacker’s grammatical model is a systematic exploration of such dimensions of construal, which further include, among others, perspective (described in terms of figure/ground alignment) and prominence (described in terms of base/profile configurations). Similarly, Talmy describes various conceptual subsystems underlying grammatical constructs, like force dynamics and causation, attention phenomena, plexity, event structures and their lexicalization. Importantly, construal does not just involve the semantics of separate forms of expression but crucially presupposes the presence of alternatives, of different ways of portraying reality. Accordingly, studies of grammatical construal tend to look at the systems lying behind functionally equivalent or near-equivalent alternatives. In Talmy’s work on lexicalization patterns of motion events, for instance, the variation shows up typologically: “Verb-framed” languages encode the path of motion into the verb, whereas “satellite-framed” languages encode it in a particle, a prepositional phrase, or similar. But the alternatives can obviously also exist within a single language. Frame semantics (Fillmore 1977, 1985) is an influential model describing intralinguistic alternative argument structure choices. One essential starting point is the idea that one cannot understand the meaning of a word (or a linguistic expression in general) without access to all the encyclopedic knowledge that relates to that expression. For example, in order to understand the word sell, you need to have world knowledge about the situation of commercial transfer. This comprises, besides the act of selling, a person who sells, a person who buys, goods to be sold, money or another form of payment, and so on. A semantic frame of this type is a coherent structure of related concepts where the relations have to do with the way the concepts co-occur in real-world situations. Specific patterns of expressions evoke the frame and highlight individual concepts within the frame. In the standard commercial transaction example, for instance, sell construes the situation from the perspective of the seller and buy from the perspective of the buyer.

7.3.2 Experiential Grounding and Contextual Dialectics Given this overview of various types of cognitive semantics, let us now identify the two characteristics that allow us to say that cognitive semantics is indeed a recontextualizing way of looking at linguistic meaning. In the first place, the approaches mentioned above embody different ways in which linguistic meaning is grounded in experience. The meanings expressed by linguistic forms do not exist in a vacuum, but in various ways they reflect and express the experience of the language user. We may discern two basic ways in which this is the case.

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

167

168

DIRK GEERAERTS

First, language users are embodied beings, not pure minds. Our organic, physiological, anatomic nature influences our experience of the world, and this experience is reflected in the language we use. Image schemas are a clear case in point: They embody patterns of spatial, and more broadly sensorimotor interaction with the world. Verticality, for instance, corresponds with the posture of the human body, and a containment schema draws on experiences of visibility and haptic accessibility. In addition, there is a link with cognitive development. Grady (1997) argues for a distinction between sensory and non-sensory schemas, with image schemas restricted to fundamental units of sensory perception. He further distinguishes a class of primary metaphors that essentially map basic sensory schemas onto non-sensory ones, as in M O R E I S U P , where height functions as a source domain for quantity: “more” can be seen as “up” because in both cases, gradation plays a crucial role. Primary metaphors are based on correlations experienced in childhood: The child building a tower with blocks learns by experience that more blocks build a higher tower. The bodily experiential nature of language is not restricted to image schemas and preconceptual experience. It extends to the idea that conceptual construal in language may rely on cognitive capacities and processes that are not of a specifically linguistic kind. This is the case, for instance, with Langacker’s analysis of perspective, or with Talmy’s attentional subsystem. As an illustration, consider spatial perspectives in linguistic expressions, and how they can construe the same objective situation linguistically in different ways. Standing in the back garden and describing the position of a bicycle, someone could say either It’s behind the house or It’s in front of the house. These would seem to be contradictory statements, except that they embody different perspectives. In the first expression, the perspective is determined by the way the speaker looks: The object that is situated in the direction of the gaze is in front of the speaker, but if there is an obstacle along that direction, the thing is behind that obstacle. In this case, looking in the direction of the bicycle from the back garden, the house blocks the view, and so the bike is behind the house. In the second expression, however, the point of view is that of the house: A house has a canonical direction, with a front that is similar to the face of a person. The way a house is facing, then, is determined by its front, and the second expression takes the point of view of the house rather than the speaker, as if the house were a person looking in a certain direction. Seeing things in this way is not a primarily or typically linguistic skill, but approaches like Langacker’s and Talmy’s show how it shapes specific grammatical resources of the language. But second, we are not just biological entities: We also have a cultural and social identity, and our language may reveal that identity, i.e., languages may embody the historical and cultural experience of groups of speakers (and individuals). Think of birds (as a typically prototypical category). The encyclopedic nature of language implies that we have to take into account the actual familiarity that people have with birds: It is not just the general definition of

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

7 Cognitive Linguistics and context

“bird” that counts, but also what we know about sparrows and penguins and ostriches, etc. But these experiences will differ from culture to culture: The typical, most familiar birds in one culture will be different from those in another, and that will affect the knowledge people associate with a category like “bird.” Linguistic categories, then, may express cultural models, i.e., the more or less coherent sets of concepts that cultures use to structure experience and make sense of the world. These are not reinvented afresh with every new period in the culture’s development. Rather, it is by definition part of their cultural nature that they have a historical dimension. It is only by investigating their historical origins and their gradual transformation that their contemporary form can be properly understood. In recent years, the importance of culture for metaphor research, for instance, has received an increasing recognition, also among the major spokesmen of Conceptual Metaphor Theory: see Kövecses (2005). A second form of contextualization is slightly more abstract than the experiential grounding of meaning: Next to its experiential grounding, linguistic meaning has an interactive relationship with regard to its contexts of usage. Language is used for talking about certain things, situations, events, actions. In the simplest case, a concept like “bird” refers to a specific animal in a given situation, and then to another in another situation, or perhaps to all members of the class “bird” as a whole. The specifics of these usage situations are part of the context of language use, and Cognitive Linguistics explores how linguistic meaning takes into account that context, and more specifically, it takes that context as interacting with the linguistic meaning applied to it. The latter specification is necessary, because no theory of meaning would simply deny the existence of that situational context. A simple view of semantics merely holds that meanings are applied to contexts, while a more sophisticated view analyzes the interaction of meanings and contexts. That interaction goes in two directions, in fact, and both directions are conspicuously ingrained in the models for meaning description developed within Cognitive Linguistics. On the one hand, linguistic meaning is dynamic and flexible. Meanings change, and there is a good reason for that: meaning has to do with shaping our world, but we have to deal with a changing world. New experiences and changes in the environment require that people adapt their semantic categories to transformations of the circumstances and that we leave room for nuances and slightly deviant cases. For a theory of language, this means that we cannot just think of language as a more or less rigid and stable structure: If meaning is the hallmark of linguistic structure, then we should think of those structures as flexible. Again, this is firmly embodied in the approaches that we presented: A radial structure of senses (whether it is built up through prototypicality, metaphor, metonymy or other mechanisms) is precisely the outcome of a flexible usage of existing concepts. Importantly, this involves not just the development of new senses but also the modulation of existing ones. When applying the concept “bird” to ostriches, for instance, another set of

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

169

170

DIRK GEERAERTS

characteristics is activated than when you talk about sparrows, if only because ostriches do not fly. But we would not say that the word bird has two different meanings, one that includes sparrows and another that includes ostriches. On the other hand, linguistic meaning is perspectival. Meaning, according to cognitive semantics, is not just an objective reflection of the outside world, it is a way of shaping that world, it is a form of conceptual construal, and the various types of conceptual construal underlie the diverse descriptive models that are developed in cognitive semantics. Metaphor, metonymy, mental spaces, grammatical mechanisms in a Langackerian or Talmyan vein, alternatives within a frame: These are all ways of looking at one thing in terms of another and construing a perspective on reality. To come back to the bird example, classifying an ostrich as a bird means tweaking the common concept of “bird” a little bit, but at the same time, it is also a way of looking at the ostrich from the point of view of what you already know about birds. In this way, we can say that there is an interactive, dialectic relationship between concept and context: The concept provides a perspective from which to conceptualize the context, but there is a feedback loop from the context to the concept.

7.4 Recontextualizing Grammar In the previous section, we saw not only how Cognitive Linguistics places semantics at the heart of linguistic description, but how it specifically develops a contextualized conception of meaning. The basic impulse of Cognitive Linguistics is to put the description of meaning in a central position in linguistics, but just as importantly, that meaning is not studied as something isolated but rather as something that is embedded in a wider context – and that context takes different forms: the body, society and culture, and actual language use. In this section, we will see how that recontextualizing tendency also applies to the way in which Cognitive Linguistics looks at linguistic systems at large. As in the previous section, there are two relevant features to pay attention to, and they are in fact related to the features discussed there. First, similar to the interactive relationship between concepts and their contexts of usage, we will describe how usage-based linguistics generalizes that idea to language at large. Second, similar to the experiential grounding of meaning, we will introduce the intersubjective grounding of linguistic systems as the basis for a “social turn” in Cognitive Linguistics.

7.4.1 Usage-based Linguistics Prototypicality, as we have seen, illustrates the interactive relationship between concepts and contexts. However, it not only plays a role with regard to the description of meaning as such but also has an influence on the way in which linguistic categories in the broader sense are described: If grammatical

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

7 Cognitive Linguistics and context

categories are conceptual entities (and that is one of the main perspectives driving Cognitive Linguistics), then the characteristics of categorization unearthed by prototype theory should also play a role in the description of grammatical categories and linguistic systems at large. This can be specified with regard to both the nature of grammatical descriptions and the nature of the linguistic system as a whole. In the first case, we look at the characteristics of the categories that are needed in a grammatical description. Two features of current thinking about grammatical categories link up with prototype theory: the inclusion of an extensional level of description, and the acceptation of fuzziness. (For a richly illustrated introduction to prototype effects in grammar, see Taylor 2015.) The first feature relates to the fact that prototype theory has made it clear that it is necessary not only to look at the definition of a category, but also to have a look at the members of that category, and the exemplars instantiating it. The information that we have about a category like “fruit” may not only be enshrined in an abstract definition, but it may also be carried by the knowledge and experience that we have of actual, specific types of fruit. If you apply that to syntactic categories and grammatical rules, you will need to have a look at the way in which those entities are realized by actual words and word groups. An adequate description of a category like “passive” cannot stop short at an abstract characterization of passivization but needs to have a look at typical passive constructions, just in the way in which the description of a category like “fruit” needs to take into account both typical and atypical types of fruit. Langacker (1987) has referred to this recognition as a rejection of the “rule/ list fallacy.” A traditional approach would have it that when you can describe something as a grammatical rule it is no longer necessary to look at the list of elements that instantiate the rule. But from a prototype-theoretical perspective, that is an error: An adequate description of a category requires looking at the intensional level (the abstract level of definitions) as well as the extensional level (the level of the actual elements filling in a category). As a consequence, a second feature of prototypicality emerges: The categories of grammatical description may be fuzzy at the edges, and there will be clines of applicability of categories to the entities that instantiate them (as when some passive constructions are more typical than others, and the possibility of passivization may in some cases be in doubt). On a higher level, the distinction between grammar and lexicon itself stops being categorical. This is most conspicuous in construction grammar (Hoffmann and Trousdale 2013), in the sense that the notion of “construction” covers fully lexicalized idioms as well as syntagms consisting of abstract entities like word classes, and hybrid entities that combine open slots with fully or partially specified lexical classes. The prototype-theoretical effect of breaking down categorical barriers also applies to the distinction between the linguistic system and the usage of that system. That distinction has dominated twentieth-century linguistics both in the structuralist, Saussurean form of the distinction between langue and parole, and in the generative form of the distinction between “competence”

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

171

172

DIRK GEERAERTS

and “performance.” From a prototype-theoretical point of view, this is again an example of the distinction between an intensional level and an extensional level, that is to say, if you think of a linguistic system as a category, then that category is instantiated by the communicative events making use of that system. Accordingly, a good understanding of the system (the overarching category, the language) requires taking into account the actual usage events realizing the system. It is, for instance, only by looking at actual usage events that changes in the linguistic system can be explained. Systemic changes occur when marginal or new ways of speaking become more and more frequent, and this points to another feature of prototypicality: salience effects and entrenchment, as reflected by differences of frequency. In the course of the first two decades of the twenty-first century, this dynamic, dialectic relationship between linguistic system and linguistic usage has become more and more prominent within Cognitive Linguistics (see among many other references, Langacker 2000 and Bybee 2010 for foundational statements, and Diessel 2019 and Divjak 2019 for forceful recent texts). One might even say that we are currently witnessing a trend toward usagebased linguistics that is broader than Cognitive Linguistics, in the sense that an interest in the frequency-based, probabilistic description of corpus data and experimental results takes precedence over the semantic interests of Cognitive Linguistics as it originally emerged. That shift toward usage-based linguistics is to some extent the result of a technological pull exerted by the increasing availability of corpus data embodying spontaneous language use, and computational tools for dealing with such large amounts of data. But at the same time, it is the result of a theoretical push in which the prototype-theoretical insistence on the importance of an extensional level of description – of looking not just at the abstract characterization of concepts but also at the exemplars that they apply to – plays a major role. The current emergence of usage-based linguistics is the ultimate consequence of a process that began half a century ago, when the incorporation of prototype theory into lexical semantics and nascent Cognitive Linguistics constituted a first step toward breaking down the theoretical barrier between meaning and world knowledge, langue and parole, system and usage.

7.4.2 Intersubjectivity and Sociocultural Variation Cognitive Linguistics in the new millennium is characterized by a growing attention to the social and cultural aspects of language on three levels of analysis. The first level considers variation within languages: To what extent do the phenomena that are typically focused on in Cognitive Linguistics exhibit variation within the same linguistic community? The research conducted within this approach links up with the research traditions of sociolinguistics, dialectology, and stylistic analysis, using the same meticulous empirical methods as these traditions. Focusing on language-internal variation in all its dimensions, this research field may be illustrated by several

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

7 Cognitive Linguistics and context

publications referring to “Cognitive Sociolinguistics” or “social cognitive linguistics” as the study of lectal variation in the context of Cognitive Linguistics: Kristiansen and Dirven (2008), Croft (2009), Geeraerts et al. (2010), Pütz et al. (2012), and Kristiansen and Geeraerts (2013). Cognitive Sociolinguistics, in the way demarcated by these publications, strives toward a convergence of the usage-based traditions of language studies, as represented by pragmatics and sociolinguistics, and the post-generative theories of grammar illustrated by Cognitive Linguistics. Variationist studies of this kind involve issues such as the social distribution of prototype-based meaning extensions (Robinson 2010), the lectal productivity of metonymical patterns (Zhang et al. 2015), the variable use of metaphor in discourse (Semino 2008), lexical variation in pluricentric languages (Soares da Silva 2014), usage-based approaches to borrowing (Zenner et al. 2012), spatial semantics at dialect level (Berthele 2006), and lectal variation of constructions and constructional semantics (Colleman 2010, among many others). The next level is that of variation among languages and cultures, taking the form of cultural and anthropological comparisons, or of historical investigations into changing conceptualizations across time periods. An interest in cultural effects at the level of interlinguistic variation existed from an early date in the history of Cognitive Linguistics. For instance, Rosch’s (1978) research on prototype categorization, which had a major influence on theory formation in Cognitive Linguistics, is characterized by an anthropological background, just like Berlin’s research on color terms and ethnobiological classification from which it derived (Berlin and Kay 1969). Questions of cultural relativity play a natural role in this kind of investigation, together with the notion of “cultural model”: see Holland and Quinn (1987) for an influential early volume. Cross-cultural studies of metaphorical patterns and conceptual metaphors are by now an established line of research: see Kövecses (2020). A broadly anthropological view on cultural linguistics has been developed by, among others, Palmer (1996) and Sharifian (2017). The third level, beyond intralinguistic and interlinguistic variation, is that of language as such: Here we can situate foundational studies that emphasize and analyze the way in which the emergence of language as such and the presence of specific features in a language can only be adequately understood if one takes into account the socially interactive nature of linguistic communication. Representatives of this strand of research include Croft (2009) on a socioevolutionary view of language, Sinha (2009) on language as an epigenetic system, Zlatev (2008) on situated embodiment, Itkonen (2003) on the social nature of the linguistic system, Verhagen (2005) on the central role of intersubjectivity in language, Harder (2010) on the socio-functional background of language, Schmid (2020) on the interplay of psychological and social entrenchment of linguistic patterns, and in the wake of Tommasello (2003), the rich literature on socio-communicative, usage-based theories of language acquisition. Regardless of their differences, these approaches share a foundational

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

173

174

DIRK GEERAERTS

perspective: They present high-level models of the principled role of social factors and intersubjective phenomena in language and linguistic evolution. Taken together, these three areas of research constitute a “social turn” in Cognitive Linguistics that completes the recontextualization tendencies. The social aspects of language that were relegated to the periphery by approaches that consider “autonomous syntax” to be the core of the language are now seen as naturally and inescapably intertwined with grammatical description. Also, this social turn is obviously related to the other forms of recontextualization mentioned in previous sections. An interest in the experiential grounding of meaning easily translates into an interest in interlinguistic variation: To what extent do different cultures express a different construal of the world in their language use? And a usage-based model almost inevitably yields a concern with intralinguistic variation. When we say that common linguistic behavior derives from the interaction between language users, it needs to be established just how common that behavior actually is, and how the existing variation is structured by social factors – precisely the kind of questions that are central to dialectology and sociolinguistics: “usage-based implies variational” (Geeraerts 2005, 2016). In the context of the history of linguistics, the third, theoretically inclined line of research mentioned above is particularly telling: It invests the socio-communicative and intersubjective features of language with a foundational importance, or in other words, the social turn in Cognitive Linguistics is a return to a Saussurean conception of language as a social semiotic (though, given the acceptance of usage-based variation, without the mainstream structuralist assumption of the internal homogeneity of language systems).

7.5 Conclusion Let’s wrap up. We have positioned Cognitive Linguistics against the background of twentieth-century developments in linguistics and described how Cognitive Linguistics constitutes a recontextualization of linguistics in contrast with formal semantics and formal grammar. In contrast with the former, Cognitive Linguistics works with a model of meaning that emphasizes the experiential grounding of meaning and the dialectic interaction of senses and context of use. In contrast with the latter, Cognitive Linguistics develops a usage-based model of linguistics and reintroduces the social dimension as an inherent dimension of language, and social variation as an inherent part of grammar. Both pairs of recontextualizing trends are related, as can be readily appreciated if we interpret them against the background of the semiotic triangle as classically described by Ogden and Richards (1923). The semiotic triangle relates the linguistic sign, on the one hand, to its reference, and on the other, to the language user. These two poles may precisely provide a link between the two pairs of recontextualizations, if we double the triangle by not only applying it in its original form to individual signs, but by also applying it to

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

7 Cognitive Linguistics and context

Table 7.1 Fundamental recontextualizing characteristics of Cognitive Linguistics contextualizing meaning

contextualizing grammar

relating to the user pole experiential grounding of intersubjective and social of the semiotic triangle meaning grounding of language relating to the reference pole dialectic relationship between dialectic relationship between of the semiotic triangle concept and context system and usage

language at large. In the latter case, we consider how the linguistic system (the large-scale equivalent of the individual sign in the basic Ogden and Richards conception) relates to users and groups of users, and then to its instantiation as usage events. “Instantiation” is the crucial notion here: Just like the referents or exemplars of a single sign instantiate the associated category, a linguistic system is extensionally represented by all the occasions on which, and all the specific forms in which, it is put into practice. If we accept this doubling of the semiotic triangle (corresponding, needless to say, with the distinction between Sections 7.3 and 7.4 in the pages above), we can define the essential similarity between the two forms of contextualization that we identified within each of the two perspectives. First, both the experiential grounding of meaning and the intrinsically social, intersubjective nature of language involves the user: users sharing common physiological and anatomical embodiment; users with their individual experiences and background; users as part of intersubjective and communicative communities. Second, a dialectic relation between language and its instantiation in a specific situational context exists both for individual linguistic categories and for language as a whole. The resulting picture of the fundamental recontextualizing characteristics of Cognitive Linguistics in their mutual relations can be summarized in Table 7.1. It appears that (re)contextualization indeed provides a unifying perspective on the contribution of Cognitive Linguistics to the development of linguistic theory in the late twentieth and the early twenty-first century. But to round off, it should be remarked that this unifying perspective should not be taken to imply a complete homogeneity of Cognitive Linguistics. The systematic and pervasive relevance of recontextualization within Cognitive Linguistics should not obscure the fact that tensions may exist within the framework: Even if recontextualization is a defining overarching characteristic of Cognitive Linguistics, various forms of recontextualization do not necessarily point in the same direction. Specifically, an experiential grounding of meaning as embodiment has a universalist tendency (assuming that the human body is the same for everybody), while a sociocultural conception of language will tend to emphasize the diversity and historicity of the human experience. If it is correct to characterize Cognitive Linguistics in terms of recontextualization, determining the proper relationship between what is universal and what is historical and specific may well be a major focus for the further evolution of the framework.

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

175

176

DIRK GEERAERTS

References Barcelona, A., ed. (2000). Metaphor and Metonymy at the Crossroads: A Cognitive Perspective. Berlin: Mouton de Gruyter. Berlin, B., and Kay, P. (1969). Basic Color Terms: Their Universality and Evolution. Berkeley: University of California Press. Berthele, R. (2006). Ort und Weg: Die sprachliche Raumreferenz in Varietäten des Deutschen, Rätoromanischen und Französischen. Berlin: Walter de Gruyter. Brugman, C. (1988). The Story of “Over”: Polysemy, Semantics and the Structure of the Lexicon. New York: Garland. Bybee, J. (2010). Language, Usage and Cognition. Cambridge: Cambridge University Press. Chomsky, N. (1957). Syntactic Structures. The Hague: Mouton. Chomsky, N. (1965). Aspects of the Theory of Syntax. Cambridge, MA: MIT Press. Colleman, T. (2010). Lectal variation in constructional semantics: “Benefactive” ditransitives in Dutch. In D. Geeraerts, G. Kristiansen, and Y. Peirsman (eds.), Advances in Cognitive Sociolinguistics (pp. 191–221). Berlin: De Gruyter Mouton. Croft, W. (2009). Towards a social cognitive linguistics. In V. Evans and S. Pourcel (eds.), New Directions in Cognitive Linguistics (pp. 395–420). Amsterdam: John Benjamins. Dąbrowska, E., and Divjak, D., eds. (2015). Handbook of Cognitive Linguistics. Berlin: De Gruyter Mouton. Dancygier, B., ed. (2017). The Cambridge Handbook of Cognitive Linguistics. Cambridge: Cambridge University Press. Diessel, H.. (2019). The Grammar Network: How Linguistic Structure Is Shaped by Language Use. Cambridge: Cambridge University Press. Divjak, D. (2019). Frequency in Language: Memory, Attention and Learning. Cambridge: Cambridge University Press. Dirven, R., and Pörings, R., eds. (2002). Metaphor and Metonymy in Comparison and Contrast. Berlin: Mouton de Gruyter. Evans, V., and Green, M. (2006). Cognitive Linguistics: An Introduction. Mahwah, NJ: Lawrence Erlbaum. Evans, V., Bergen, B., and Zinken, J., eds. (2007). The Cognitive Linguistics Reader. London: Equinox. Fauconnier, G. (1985). Mental Spaces: Aspects of Meaning Construction in Natural Language. Cambridge, MA: MIT Press. Fauconnier, G., and Turner, M. (2002). The Way We Think: Conceptual Blending and the Mind’s Hidden Complexities. New York: Basic Books. Fillmore, C. J. (1977). Scenes-and-frames semantics. In A. Zampolli (ed.), Linguistic Structures Processing (pp. 55–81). Amsterdam: North-Holland Publishing. Fillmore, C. J. (1985). Frames and the semantics of understanding. Quaderni di Semantica, 6, 222–254. Geeraerts, D. (1989). Prospects and problems of prototype theory. Linguistics, 27, 587–612. Geeraerts, D. (2005). Lectal variation and empirical data in Cognitive Linguistics. In F. Ruiz de Mendoza Ibáñez and S. Peña Cervel (eds.), Cognitive Linguistics: Internal Dynamics and Interdisciplinary Interactions (pp. 163–189). Berlin: Mouton de Gruyter.

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

7 Cognitive Linguistics and context

Geeraerts, D., ed. (2006). Cognitive Linguistics: Basic Readings. Berlin: Mouton de Gruyter. Geeraerts, D., and Cuyckens, H., eds. (2007). The Oxford Handbook of Cognitive Linguistics. New York: Oxford University Press. Geeraerts, D. (2010). Recontextualizing grammar: Underlying trends in thirty years of Cognitive Linguistics. In E. Tabakowska, M. Choinski, and L. Wiraszka (eds.), Cognitive Linguistics in Action: From Theory to Application and Back (pp. 71–102). Berlin: De Gruyter Mouton. Geeraerts, D. (2015). From structure to context: Modern linguistics from a distance. Μελέτες για την Ελληνική Γλώσσα. Studies in Greek Linguistics, 35, 35–51. Geeraerts, D. (2016). The sociosemiotic commitment. Cognitive Linguistics, 27, 527–542. Geeraerts, D., Kristiansen, G., and Peirsman, Y., eds. (2010). Advances in Cognitive Sociolinguistics. Berlin: De Gruyter Mouton. Grady, J. (1997). “Theories are buildings” revisited. Cognitive Linguistics, 8, 267–290. Hampe, B., ed. (2005). From Perception to Meaning: Image Schemas in Cognitive Linguistics. Berlin: Walter de Gruyter. Harder, P. (2010). Meaning in Mind and Society: A Functional Contribution to the Social Turn in Cognitive Linguistics. Berlin/New York: De Gruyter Mouton. Hoffmann, T., and Trousdale, G., eds. (2013). The Oxford Handbook of Construction Grammar. New York: Oxford University Press. Holland, D., and Quinn, N., eds. (1987). Cultural Models in Language and Thought. Cambridge: Cambridge University Press. Itkonen, E. (2003). What Is Language? A Study in the Philosophy of Linguistics. Turku: Åbo Akademis tryckeri. Johnson, M. (1987). The Body in the Mind: The Bodily Basis of Meaning, Imagination, and Reason. Chicago: University of Chicago Press. Kövecses, Z., and Radden, G. (1998). Metonymy: Developing a cognitive linguistic view. Cognitive Linguistics, 9, 37–77. Kövecses, Z. (2005). Metaphor in Culture: Universality and Variation. Oxford: Oxford University Press. Kövecses, Z. (2020). An extended view of conceptual metaphor theory. Review of Cognitive Linguistics, 18, 112–130. Kristiansen, G., and Dirven, R., eds. (2008). Cognitive Sociolinguistics: Language Variation, Cultural Models, Social Systems. Berlin: Mouton de Gruyter. Kristiansen, G., and Geeraerts, D., eds. (2013). Contexts of Use in Cognitive Sociolinguistics (Thematic issue of Journal of Pragmatics). Amsterdam: Elsevier. Lakoff, G., and Johnson, M. (1980). Metaphors We Live By. Chicago: University of Chicago Press. Lakoff, G. (1987). Women, Fire and Dangerous Things: What Categories Reveal about the Mind. Chicago: University of Chicago Press. Langacker, R. W. (1987). Foundations of Cognitive Grammar, Vol. I: Theoretical Prerequisites. Stanford: Stanford University Press. Langacker, R. W. (1991). Foundations of Cognitive Grammar, Vol. II: Descriptive Application. Stanford: Stanford University Press. Langacker, R. W. (2000). A dynamic usage-based model. In M. Barlow and S. Kemmer (eds.), Usage-based Models of Language (pp. 1–63). Stanford: CSLI Publications. Littlemore, J., and Taylor, J. R., eds. (2014). The Bloomsbury Companion to Cognitive Linguistics. London: Bloomsbury.

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

177

178

DIRK GEERAERTS

McCawley, J. D. (1968). Lexical insertion in a transformational grammar without deep structure. In B. J. Darden, C.-J. Bailey, and A. Davidson (eds.), Fourth Regional Meeting of the Chicago Linguistic Society (pp. 71–80). Chicago: Chicago Linguistic Society. Montague, R. (1974). Formal Philosophy: Selected Papers of Richard Montague. New Haven: Yale University Press. Nuyts, J. (2007). Cognitive linguistics and functional linguistics. In Dirk Geeraerts and Hubert Cuyckens (eds.), The Oxford Handbook of Cognitive Linguistics (pp. 543–565). New York: Oxford University Press. Ogden, C. K., and Richards, I. A. (1923). The Meaning of Meaning: A Study of the Influence of Language upon Thought and of the Science of Symbolism. London: Routledge and Kegan Paul. Palmer, G. B. (1996). Toward a Theory of Cultural Linguistics. Austin: University of Texas Press. Partee, B. H., ed. (1976). Montague Grammar. New York: Academic Press. Pütz, M., Robinson, J. A., and Reif, M., eds. (2012). Cognitive Sociolinguistics: Social and Cultural Variation in Cognition and Language Use. Amsterdam: John Benjamins. Robinson, J. A. (2010). Awesome insights into semantic variation. In D. Geeraerts, G. Kristiansen, and Y. Peirsman (eds.), Advances in Cognitive Sociolinguistics (pp. 85–110). Berlin: De Gruyter Mouton. Rosch, E. 1978. Principles of categorization. In E. Rosch and B. B. Lloyd (eds.), Cognition and Categorization (pp. 27–48). Hillsdale, NJ: Lawrence Erlbaum. Semino, E. (2008). Metaphor in Discourse. Cambridge: Cambridge University Press. Schmid, H.-J. (2020). The Dynamics of the Linguistic System: Usage, Conventionalization, and Entrenchment. Oxford: Oxford University Press. Sharifian, F. (2017). Cultural Linguistics: Cultural Conceptualisations and Language. Amsterdam: John Benjamins. Sinha, C. (2009). Language as a biocultural niche and social institution. In V. Evans and S. Pourcel (eds.), New Directions in Cognitive Linguistics (pp. 289–309). Amsterdam: John Benjamins. Soares da Silva, A. (2014). The pluricentricity of Portuguese: A sociolectometrical approach to divergence between European and Brazilian Portuguese. In A. Soares da Silva (ed.), Pluricentricity: Language Variation and Sociocognitive Dimensions (pp. 143–188). Berlin: De Gruyter Mouton. Talmy, L. (2000). Toward a Cognitive Semantics. Cambridge, MA: MIT Press. Taylor, J. R. (1989). Linguistic Categorization. Prototypes in Linguistic Theory. Oxford: Clarendon Press. Taylor, J. R. (2015). Prototype effects in grammar. In E. Dąbrowska and D. Divjak (eds.), Handbook of Cognitive Linguistics (pp. 562–579). Berlin: De Gruyter Mouton. Tomasello, M. (2003). Constructing a Language: A Usage-Based Theory of Language Acquisition. Harvard: Harvard University Press. Verhagen, A. (2005). Constructions of Intersubjectivity: Discourse, Syntax, and Cognition. Oxford: Oxford University Press. Wen, X., and Taylor, J. R. eds. (2021). The Routledge Handbook of Cognitive Linguistics. New York/London: Routledge. Winters, M. E., and Nathan, G. S. (2020). Cognitive Linguistics for Linguists. New York: Springer.

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

7 Cognitive Linguistics and context

Zenner, E., Speelman, D., and Geeraerts, D. (2012). Cognitive Sociolinguistics meets loanword research: Measuring variation in the success of anglicisms in Dutch. Cognitive Linguistics, 23, 749–792. Zlatev, J. (2008). The co-evolution of intersubjectivity and bodily mimesis. In J. Zlatev, T. Racine, C. Sinha, and E. Itkonen (eds.), The Shared Mind: Perspectives on Intersubjectivity (pp. 215–244). Amsterdam: John Benjamins. Zhang, W., Geeraerts, D., and Speelman, D. (2015). Visualizing onomasiological change: Diachronic variation in metonymic patterns for “woman” in Chinese. Cognitive Linguistics, 26: 289–330.

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

179

https://doi.org/10.1017/9781108989275.008 Published online by Cambridge University Press

Part III

Pragmatic Approaches to Context

Published online by Cambridge University Press

Published online by Cambridge University Press

8 The Role of Context in Gricean and Neo-Gricean Pragmatics Jacques Moeschler

8.1 Introduction It is generally assumed that pragmatics is “the study of language use, as opposed to language structure. . . . In a more focused sense . . . pragmatics contrasts with semantics, the study of linguistic meaning, and is the study of how contextual factors interact with linguistic meaning in the interpretation of utterances” (Wilson and Sperber 2012: 1). According to this presentation of pragmatics, it is obvious both that pragmatic meaning is outside the scope of linguistic meaning and that it is largely contextual; that is, it is determined in accordance with contextual factors. This vision of pragmatics is consistent with recent developments in pragmatics such as Relevance Theory (see Chapter 11 on Relevance Theory and context), but is not particularly compatible with early developments in pragmatics, which were influenced by the philosophy of language (see Chapter 4): Speech act theory, as initiated by John Austin in the 1950s and John Searle in the 1970s (see respectively Austin 1962; Searle 1969, 1979), is more a semantic, or conventional, theory of meaning; in a similar way, Grice’s theory of nonnatural meaning (Grice 1989), developed in his theory of conversational implicature (Grice 1975), hardly addresses context (see Section 8.3). The above-mentioned background shows up a historical paradox: Whereas pragmatics has become a trend in research on context, as shown by recent experimental developments (see Noveck and Sperber 2004; Noveck and Reboul 2008; Noveck 2018), its early advances were primarily noncontextual. This paradox must be explained, or at least called into question. The first aim of this chapter is to explain how and why the substantial developments in pragmatics launched in Grice’s article “Logic and Conversation” were noncontextual. I will also show how the Gricean conversational machinery (Section 8.3) is anticontextualist (Recanati 1994), and how the neo-Gricean paradigm, chiefly developed by Gazdar (1979), Horn (1984, 1989), and Levinson (2000), has given

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

184

JACQUES MOESCHLER

context a double-check function which verifies whether implicatures are compatible with the common ground (Sections 8.5 and 8.6). The grammatical approach to implicature, mainly represented by Fox (2007) and Chierchia (2013), has on the other hand developed a complex reasoning process for deriving scalar implicatures, the type of generalized conversational implicature that is most often studied (Section 8.7). However, before introducing the starting point of all the research that has been done on conversational implicatures, and for which context is the crucial issue, I must focus on two additional topics. Firstly, Grice’s theory of conversational implicature is mainly addressed in the literature as a complement to utterance meaning as opposed to “what is said”; that is, the content or proposition expressed by an utterance. The issue at stake is what “what is said” consists of. It will be argued that most philosophical approaches (Kaplan 1989; Stalnaker 1978; Korta and Perry 2012) give a contextual flavor to “what is said” (Section 8.2). This paradox will be extended by introducing a strongly disputed concept, conventional implicature (Section 8.4). This concept was introduced by Grice in his definition of implicatures and of the criteria that distinguish between different types of implicatures (Sadock 1978). Although conventional implicatures are semantically encoded within the linguistic meaning of words or expressions, they are contextually constrained in their usages by their relationship to the common ground, or the set of propositions supposedly shared by the participants in a conversation (Karttunen and Peters 1979). The general conclusion of this chapter will be surprising, and certainly controversial: Whereas literal meaning, or “what is said,” is strongly contextually constrained – context is defined by Kaplan as the sequence of an agent, a time, and a world – implicatures, or pragmatically inferred nonliteral meanings reached through maxims of conversation and general principles, are also contextual. This conclusion could be the starting point of a very important debate in pragmatics on the role of contextual factors in the comprehension process. These factors impact not only linguistic and pragmatic competence development, and second-language learning, but also clinical linguistics, artificial intelligence, and social media, which are all research fields developed in this handbook.

8.2 Context in Content It is a common-sense assumption that the propositional content of a sentence, that is, its linguistic meaning, is not computed throughout contextual information, but simply as a result of compositional rules that are described by semantics. However, several arguments can be raised against this anticontextualist vision of “what is said.” The first argument was clearly stated by Searle (1979: 117): For him, “the notion of literal meaning of a sentence only has application relative to a set of contextual or background assumptions.” For

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

8 Context in Gricean and Neo-Gricean Pragmatics

instance, Searle contrasts the truth-conditions of a sentence like (1) when uttered in ordinary earthly conditions where gravity is at work, with situations in which gravity is not present, for example, in a space station. In the latter case, the sentence would describe a situation in which the cat is only touching the edge of the mat without sitting on its entire surface: (1) The cat is on the mat.

The second argument against the noncontextual vision of “what is said” is given by all types of classic semantic formal language:. A sentence like (2) is certainly not true of all times and all locations: it is a contingent proposition which receives a truth-value for a determinate time and a specific location. (2) It is raining.

For instance, if I utter (3), after returning home, I certainly do not wish to say that at every moment and on every part of the motorway from Geneva to Mâcon it is true that it was snowing: (3) It was snowing on the motorway.

What I am suggesting is that there is a set of pairs of times and places for which the proposition it is snowing is true. For example, if I left the university parking lot at 4:00 p.m., it might be possible that it snowed at kilometer 60 at 4:30 p.m., but not at 4:45 at km. 75, and that it snowed again at 5:00 p.m. at km. 100.1 The third argument stems from an analysis of indexicals, or referential expressions that are only interpreted relative to the context of utterance, such as I, here, now. Korta and Perry (2012: 20) present Kaplan’s classical analysis of indexicals, which introduces the technical concepts of character, content, and context. In a nutshell, On Kaplan’s theory a sentence like “I am sitting” has a certain sort of meaning, which he calls “character,” that is modeled as a function from contexts to contents. “Content” is used by Kaplan as a theoretical term, that gets at the ordinary concept of “what is said” or “the proposition expressed.” Following the tradition that came from Carnap and had been developed in modal intensional logic, Kaplan took propositions to be intensions, that is, functions from possible words (or pairs of times and worlds – Kaplan’s own preference) to truth-values for sentences, and other appropriate values for other expressions. Contexts were sequences of an agent, a time, a location, and a world. So the character of the sentence “I am tired now,” as spoken by a at t gives us the proposition that a is tired at time t. (Korta and Perry 2012: 20)

In other words, a context consists of a speaker, a time, a location, and a world. What is crucial is that the proposition expressed, or intension, depends on the possible world in which it is indexed: In an alternative world, it is false 1

For more technical arguments, see Barwise and Perry (1983), the standard textbook on situation semantics.

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

185

186

JACQUES MOESCHLER

that it was snowing during my drive from Geneva to Mâcon, that is, in an alternative pair of times and places. We should bear in mind that intensional semantics is based on the intuition that “a proposition – the content of an assertion or belief – is a representation of the world as being a certain way” (Stalnaker 1978: 316). In other words, “a proposition is a rule determining a truth value as a function of the facts – the way the world is. More roughly and intuitively, a proposition is a way – any way – of picking a set of possible states of affairs – all those for which the proposition takes the value true” (p. 316). This approach to “what is said” (content in Kaplan’s terms, proposition in most other semantics approaches) is a function from a possible world to a truth-value. The most important element for our purposes is that meaning is described by Kaplan as a function from context to content, which Kaplan refers to as character.2 In other words, you choose a context, defined by a speaker a, a time t, and a place l, you pick a possible world defined by these indices, and you map your possible world onto a truth-value, which results in the character of the assertion, or its meaning. So, in intensional semantics, a proposition is contextually defined, mainly by a speaker and the time and place in which it is true, that is, the world in which its content has a truth-value. The agenda of intensional semantics might appear to be completely disconnected from pragmatic issues, but this is not the case. In their 2012 book, Korta and Perry suggest a version of pragmatics, termed critical pragmatics, that is based on intensional semantics. On the other hand, Wilson and Sperber (2012, chapter 7), in their seminal article “Linguistic Form and Relevance,” explicitly refer to Kaplan to show that indexicals encode truth-conditional procedural meanings, which are responsible for the computation of what they call explicature – the pragmatic enrichment of what is said. In order to disentangle these approaches, as well as the role of context in utterance interpretation, I will now return to Grice and his theory of implicature.

8.3 Gricean Pragmatics: Conversational Implicatures and Context One of the commonplaces about Grice and his theory of implicatures and meaning is to interpret them in contextualist terms. In fact, Grice’s theory is based on two strong assumptions: first, that meaning in language use is nonnatural; and second, that access to speaker meaning is based on cooperative communication. 2

Content, context, and character can be formally described in these ways: a. content = {truth-values}possible worlds = {1,0}W b. context = {agent, time, location} = {a, t, l} c. character = {content}context = ({1,0}W)a,t,l

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

8 Context in Gricean and Neo-Gricean Pragmatics

The first claim is certainly the most important for a theory of meaning because it departs from the traditional view of linguistic meaning, based, from Saussure to lexical semantics, on the assumption that meaning is the result of conventions.3 On the contrary, for Grice, meaning in natural language is nonnatural, which implies that what is conveyed by a speaker goes beyond what lexical units conventionally mean: (4) Nonnatural meaning (Grice 1989: 219) “A meantNN something by x” is roughly equivalent to “A uttered x with the intention of inducing a belief by means of the recognition of this intention.”

This definition of non-natural meaning is crucial because it implies a double intention on the speaker’s part, which is known in Relevance Theory (Sperber and Wilson 1986: 29) as the informative intention (“the intention of inducing a belief”) and the communicative intention (“by means of the recognition of this intention”). This conception of meaning is not unrelated to what has been termed an implicature: Grice gives the example of a conversation between A and B about C, who works in a bank. A asks B how C is getting on in his job, and B replies (5): (5) Oh quite well, I think; he likes his colleagues, and he hasn’t been to prison yet.

Grice (1975: 43) comments that “it is clear that whatever B implied, suggested, meant, etc., in this example, is distinct from what B said, which was simply that C had not been to prison yet.” This intended meaning is called an implicature and is distinct from “what is said,” which is, “closely related to the conventional meaning of the words (the sentences) he has uttered” (p. 44). Whereas “in some cases the conventional meaning of the words used will determine what is implicated, besides helping to determine what is said” (Grice 1975: 44), most of the cases examined by Grice are cases of nonconventional implicatures, known as conversational implicatures, which are “essentially connected to certain general feature of discourse” (p. 45). According to Grice, our conversational exchanges are mainly characterized by cooperative efforts, leading to the participants’ acceptance of a common purpose or “at least a mutually accepted direction” (p. 45). This assumption is stated in the Cooperative Principle: (6) Cooperative Principle (Grice 1975: 45) Make your conversational contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged.

In order to be cooperative, the speaker must respect, or exploit – that is, ostensively violate – nine maxims of conversations, which are grouped into four Kantian categories: Quantity, Quality, Relation, and Manner (Grice 1975: 3

An example is given by one basic assumption of cognitive grammar, defining a language as a set of form–meaning pairs: “The grammar of a language is defined as a structured inventory of conventional linguistic units” (Langacker 1991: 263–264).

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

187

188

JACQUES MOESCHLER

45–46). The maxims of Quantity state that the contribution must be “as informative as required” (Quantity1), but “not more informative than is required” (Quantity2). The supermaxim of Quality states that the contribution must be true: “Do not say what you believe to be false” (Quality1), and “Do not say that for which you lack adequate evidence” (Quality2). The maxim of Relation requires the speaker to be relevant, that is, she must contribute to the topic of the conversation. Finally, the supermaxim of Manner, “Be perspicuous,” requires avoiding obscurity (Manner1) and ambiguity (Manner2), as well as being brief (Manner3) and ordered (Manner4). Because conversational implicatures are distinct from “what is said” (the content of the proposition expressed),4 and nonconventional, they must be calculated, or “worked out.” Here is how Grice describes this working-out process (1975: 50): To work out that a particular conversational implicature is present, the hearer will rely on the following data: (1) the conventional meaning of the words used, together with the identity of any references that may be involved; (2) the CP and its maxims; (3) the context, linguistic or otherwise, of the utterance; (4) other items of background knowledge; and (5) the fact (or supposed fact) that all relevant items falling under the previous headings are available to both participants and both participants know or assume this to be the case. (my emphasis)

In other words, what is implied in the working-out process is the linguistic meaning of the words uttered, the CP, and the maxims, as well as the context and background knowledge (known as the common ground, see Section 8.4). Context is, at first glance, a crucial feature for the calculation of a conversational implicature. However, the procedure of reaching a conversational implicature does not rely on contextual information. Here is how a hearer will reason in order to draw a conversational implicature (Grice 1975: 50): He [the speaker] has said that p; there is no reason to suppose that he is not observing the maxims, or at least the CP; he could not be doing this unless he thought that q; he knows (and knows that I know that he knows) that I can see that the supposition that he thinks that q I S required; he has done nothing to stop me thinking that q; he intends me to think, or is at least willing to allow me to think, that q; and so he has implicated that q.

In this working-out process, no contextual information is at stake: only the CP and the maxims are concerned. This general template is the case for what is known as generalized conversational implicatures (GCIs): GCIs are implicatures that are normally triggered by a linguistic form, “in the absence of special 4

“What is said” is more imprecise in Grice’s theory of implicature than in Kaplan’s intensional semantics. It can, however, be more easily approached from the standpoint of the classic philosophy of language, as demonstrated by Austin (1962: 94), who defines “‘meaning’ in the favorite philosophical sense of that word, i.e. with a certain sense and with a certain reference.” In other words, “what is said” corresponds to sense and reference, in the Fregean sense of the term (Frege [1892] 1948).

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

8 Context in Gricean and Neo-Gricean Pragmatics

circumstances” such as contextual or linguistic information, which cancels the implicature. When a conversational implicature is obtained without any linguistic contribution, the implicature is a particularized conversational implicature, defined as “an implicature [that] is carried by saying that p on a particular occasion in virtue of special features of the context” (Grice 1975: 56). The property that gives sense to the concept of conversational implicature, apart from its calculability, is its cancellability. If cancellation occurs, it must occur explicitly, “by the addition of a clause that states or implies that the speaker has opted out.” For instance, a classic case of cancellation results from an in fact clause, as in (7), where the scalar implicature “not all the students passed” is suspended: (7) Some, in fact all, the students passed.

But Grice imagines cases in which the implicature is contextually cancelled “if the form of utterance that usually carries it is used in a context that makes it clear that the speaker I S opting out” (Grice 1975: 56). For instance, in cases when meiosis occurs, such as in (8), the speaker intends to minimize and thus to cancel the stronger reading (9) in order to convince a policeman not to administer a blood alcohol test: (8) He drank a little. (9) He drank a lot.

In other words, generalized conversational implicatures can be cancelled either linguistically or contextually. The role of context therefore appears to be limited to one specific function: to allow for implicature cancellation. This raises the question of the status of implicatures as regards context, or in more general terms, the common ground. I will investigate this relationship in the following sections, first in terms of the compatibility between conventional implicatures and the common ground (Section 8.4), and second in terms of generalized conversational implicatures and context (Section 8.5).

8.4 Conventional Implicatures and the Common Ground Grice addressed the concept of conventional implicature (CI) in an extremely sketchy manner. His examples of connectives such as therefore suggest that a consequence relationship is added on to the assertions of the connected discourse segments, as in (10): (10) He is an Englishman; he is, therefore, brave.

For Grice, being brave is a consequence of being an Englishman, but he suggests that this consequence relationship is not “said” by the speaker, even if the speaker has “certainly indicated, and so implicated, that this is so” (1975:

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

189

190

JACQUES MOESCHLER

45). Grice’s main argument to support this claim is that the discourse connective therefore does not contribute to the truth-conditions of the utterance: in a segment of the form p, therefore q, the truth-values of p and q are defined independently of the consequence-relation introduced by therefore. This relationship can also be accessed without the connective, as in (11), which clearly shows that therefore does not impact truth-conditions:5 (11) He is an Englishman; he is brave.

So, CI is an additional non-truth-conditional content to “what is said.” This non-truth-conditional content resembles a presupposition, which is also insensitive to the truth-value of the proposition expressed. Suppose that I tell you (12): (12) My daughter is in Japan.

In this case, the referential expression my daughter presupposes that “the speaker has a daughter.” But the same is true when I utter a negative sentence such as (13): (13) My daughter is not in Japan.

In this case, it is still true that “the speaker has a daughter.” Now, suppose that I want to show off by uttering that my daughter is in Japan, although I do not have a daughter. Is my assertion true or false? Some philosophers, including Strawson (1950), have argued that such assertions simply do not have truth-values. For instance, if we answer (14) to a previous statement such as (15), “we would certainly not say we were contradicting the statement that the king of France is wise. We are, rather, giving a reason for saying that the question whether it’s true or false simply doesn’t arise” (Strawson 1950: 330): (14) There is no king of France. (15) The king of France is wise.

Speakers’ presuppositions, therefore, have often been defined as background information, as belonging to the common ground (Stalnaker 1977). If this is true, two types of implied content crucially interact with context: conventional implicatures on the one hand, and presuppositions on the other. Whereas presuppositions are evaluated in terms of their belonging to the common ground,6 the relationship between conventional implicatures and the common ground is an open issue. Two key answers have been given, respectively, by Karttunen and Peters (1979) and Potts (2005). 5

6

When there is no connective, the implicature is conversational, more specifically a particularized conversational implicature. A presupposition is said to be accommodated when it gives new as opposed to old information, for instance, in the case that you do not know I have a daughter when you process my utterance My daughter is in Japan. See Beaver and Zeevat (2007) on accommodation.

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

8 Context in Gricean and Neo-Gricean Pragmatics

In one of the earliest approaches to conventional implicatures, Karttunen and Peters (1979: 14) define CIs as non-truth-conditional meanings which should contribute to the common ground:7 It has often been observed that what we here call conventional implicatures play a role in determining the felicity or appropriateness of utterances in conversational settings. As a general rule, in cooperative conversation a sentence ought to be uttered only if it does not conventionally implicate anything that is subject to controversy at that point in the conversation. Since the least controversial propositions of all are those in the common ground, which all participants already accept, ideally every conventional implicature ought to belong to the common set of presuppositions that the utterance of the sentence is intended to increment.

In other words, CIs are part of the common ground, defined as “the set of propositions that any participant is rationally justified in taking for granted,” or “the common ground or the common set of presumptions” (Karttunen and Peters 1979: 13). This implies that a second type of content (apart from the proposition or character in Kaplan’s terms) is sensitive to context: A CI must be compatible with the common ground. Intuitively, this makes sense: When a CI is triggered, because of its conventional but non-truth-conditional meaning, it is not supposed to add new information to the common ground. An example is the case of even. In uttering (16), a rational speaker cannot use even in an appropriate way if it is shared in the common ground that nobody except Bill likes Mary: (16) Even Bill likes Mary.

The proposition “others than Bill like Mary” is therefore still part of the common ground. On the other hand, the proposition “Bill is the least likely to like Mary” could be new information, because it might be unexpected for the hearer that Bill likes Mary. As this example shows, what is pertinent here is that none of this information is required to be contradictory with the common ground. If the knowledge that Bill likes Mary is shared, mutual, and unsurprising, the use of even would be inappropriate, because its CI would contradict a proposition in the common ground. The second type of approach to CIs is that of Potts (2005), which makes a different diagnosis: In his view Cis are speaker-oriented and antibackgrounding. Potts’ arguments are based on two types of empirical evidence, expressives on the one hand and supplements on the other. Expressives, like damn, idiot, fuck, etc., do not contribute to the truth-conditions of the proposition expressed; they only give information about the emotional state of the speaker: If I say (17), to 7

Conventional implicature has a troubled history: This concept was not developed in neo-Gricean approaches to implicatures, which primarily focused on generalized conversational implicatures (Horn 1989; Levinson 2000 among others). Relevance Theory completely ignored it, which has been reinterpreted as procedural meaning (Blakemore 1987; Cartons 2002). More controversially, CIs were strongly refuted by Bach (1999), who considered them an unnecessary concept.

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

191

192

JACQUES MOESCHLER

no one in particular after lunch, what damn adds to the proposition “I have to mow the lawn” is not information about the action I am going to undertake, but the fact that I am displeased about or unwilling to mow the lawn: (17) I have to mow the damn lawn.

This type of information is not at issue: Even though it does not contribute to the truth-conditions of the utterance, it makes a substantial contribution to speaker meaning in terms of her propositional attitude and emotions. This is not an anecdotal meaning, because ignoring its contribution to utterance meaning would have dramatic consequences as regards what the speaker means. A shallow processing, ignoring the CI, could mislead the addressee in determining the speaker meaning. The second piece of empirical evidence mentioned by Potts is supplements, which are linguistically realized by parentheticals and explicative relative clauses. Potts’ example (18) shows that the main function of the supplement is to add new information which is, however, not put in the foreground: (18) Sweden may export synthetic wolf urine – sprayed along roads to keep elk away – to Kuwait for use against camels.

The information carried by the supplement is secondary information, whose contribution is to make sense of the proposition expressed: In (18), the at-issue proposition deals with Swedish exportation of wolf urine to Kuwait, while the supplement explains why it is relevant. Potts terms this type of information antibackgrounding. In other words, the example of CIs shows that a second type of conventional meaning, parallel to propositional content, is context sensitive, but in a different way from conversational implicatures: Whereas CIs make contributions to the common ground based on a consistency criterion – a CI which is contradictory to the common ground would point at the inappropriateness of the at-issue entailment of the utterance – conversational implicatures must be checked against the common ground. When they are contradictory with certain propositions in the common ground, they must be explicitly cancelled, in the same way that false presuppositions which do not match the common ground must be explicitly rejected by the speaker. The double-check test for conversational implicatures will be described in Section 8.5. In Section 8.6, I will examine the question of whether generalized conversational implictures are default inferences, or whether on the contrary they are contextually triggered.

8.5 The Contextual Cancellation of Generalized Conversational Implicatures The main property of a conversational implicature is its cancellability. Among all the properties of conversational implicatures proposed by Grice (1975: 57–58) – cancellability, nondetachability, nonconventionality, carried

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

8 Context in Gricean and Neo-Gricean Pragmatics

by the saying of what is said, and indeterminacy – cancellability is described by Sadock (1978: 292) as the best test to define conversational implicatures: “The [cancellability] test is based on the notion that conversation implicatures are not part of the conventional force of utterances but are figured out in context.” In other words, when a conversational implicature is cancelled, the cause is the context. How does this occur? The issue here is contextualization. Now, the simplest version of contextualization would say that context is that which differentiates between sentence meaning and utterance meaning. However, this straightforward explanation is incorrect. We have already observed that the character of an assertion is contextually determined, even though the context is defined as a set of indices, including a speaker, a time, and a place of utterance. So, a sentence is firstly and secondly contextualized, as stated by Gazdar (1979: 5): “we can treat the meaning of the utterance as the difference between the original context and the context arrived at by uttering the sentence. A reasonable goal for pragmatics is thus the provision of a function that will map E [the set of sentences of some natural language D mapping the set of contexts M] into M.” Meaning, in other words, is the result of context change. This assumption explains why Gazdar points out two steps in the computation of an implicature. First a potential implicature, or implicature, is computed, and second, following a second contextualization, the actual implicature is derived, but only in those cases in which the context does not cancel the potential implicature. Both (19) and (20) im-plicate (21), but only (19) implicates (21), or makes the potential implicature actual. In (20), the potential implicature is explicitly cancelled (Gazdar 1979: 135): (19) Some of the boys were there. (20) Some of the boys, in fact all of them, were there. (21) Not all of the boys were there.

The way a potential implicature can be triggered depends on quantitative scales, which differentiate between a strong scalar expression and a weaker one. In the above examples, define a quantitative scale in which all N VP entails some N VP, and some N VP im-plicates not all N VP. The same two-step computation process can also be applied to presuppositions, which explains how and why presuppositions can be cancelled (Gazdar 1979: 124): I will define the potential presupposition of sentences in terms of their components and constructions, as if potential presuppositions were something given to us by the lexicon and the syntax. . . . Sets of them [potential presuppositions or pre-suppositions] are assigned to sentences in a completely mechanical way. They are what the presuppositions would be if there was no “projection problem,” no “ambiguity” in negative sentences, and no contextsensitivity.

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

193

194

JACQUES MOESCHLER

For instance, (22) pre-supposes (23), but in (24) the pre-supposition is cancelled because of the new context: (22) John does not regret having failed. (23) John failed. (24) John does not regret having failed, because he passed.

As the above-mentioned examples show, context appears twice: first as a set of propositions, and second as the result of the utterance interpretation, which contributes to the modification of the context or its confirmation (when no new information is added). It must be observed that this second step does not behave in the same way for conversational implicatures and presuppositions. In terms of presuppositions, the difference between (22) and (24) is negation: Negation is descriptive or internal in (22) – behaving like a hole – whereas it is metalinguistic or external in (24), and behaves a plug (Gazdar 1979: 109).8 On the other hand, when an im-plicature (potential implicature) is cancelled no negation is required, as shown in (20). Only an explicit cancelation is required, and it can be introduced, for example, by in fact.9 However, metalinguistic negation can also cancel an implicature: In this case the cancelled implicature is the im-plicature of its positive counterpart which contains a scalar predicate. For instance, the predicate excellent can be metalinguistically negated to mean outstanding, as in (25): (25) Your paper is not excellent, it is outstanding.

Outstanding entails excellent, which excludes the ordinary or descriptive use of negation, because this would give rise to a contradiction: (26) # It is not the case that your paper is excellent, it is outstanding.

While the positive counterpart (27) quantitative–im-plicates (28), which corresponds to the negation of an upper-bound predicate, the cancelled proposition in (25) is the im-plicature (28): (27) Your paper is excellent. (28) Your paper is no more than excellent.

The implicature (28) is obtained through the presumption that the speaker respects the first maxim of quantity (Quantity1), which states that the contribution must be as informative as required. In uttering (27), a cooperative speaker implicates that she cannot say more: in other words, the addressee’s

8

9

On metalinguistic negation, see the classic works by Horn (1985, 1989). Other in-depth works are Carston (2002), Noh et al. (2013), Blochowiak and Grisot (2018), and Moeschler (2018, 2019). The usual test, introduced by Grice (1989: 214), is the but-test: “I can use the first of these [‘Those three rings on the bell (of the bus) mean that the bus is full’] and go on to say, ‘But it isn’t in fact full – the conductor has made a mistake.’”

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

8 Context in Gricean and Neo-Gricean Pragmatics

paper is not outstanding. This meaning is an implicature, since it can be cancelled, as (29) demonstrates: (29) Your paper is excellent, and even outstanding.

To sum up, a conversational implicature can be cancelled when it does not fit the common ground. When the implicature is triggered by a linguistic expression – when it is a generalized conversational implicature – the cancellation must be explicit and cannot be left to the context. On the other hand, when the implicature is particularized, its cancellation can be left implicit. This is exactly what happens in the following example (Reboul 2019: 77): (30) Q: Do you have any bank accounts in Swiss banks, Mr. Bronston? A: No, sir. Q: Have you ever? A: The company had an account there for about six months, in Zurich.

The questioner (Q) is an American judge and the answerer (A) an American businessman accused of having opened an account in a Swiss bank. As explained by Anne Reboul, The relevant utterance is the second answer . . . . It implicitly communicates the false information that Bronston did not have a (personal) account in Swiss banks, while, as a matter of fact, he did. Bronston clearly did not communicate this information explicitly in his second answer. In his first answer, he communicated explicitly that, at the moment of utterance, he did not have such an account, which was true. However, he had had one before. And he communicated implicitly through his second answer that he had not. (p. 77)

If Bronston had been cooperative, he would have respected Quantity1, and therefore the judge could have inferred, as an implicature, that Bronston had not had a bank account in a Swiss bank before. However, the judge understood that the implicature was false; in other words that Bronston was not being cooperative. The judge requested a retrial for perjury, but he lost the case, because Bronston did not lie: the judge assumed that Bronston did not comply with Quality, whereas in fact he did not comply with Quantity. In other words, the cancellation of a particularized conversational implicature is made possible by its compatibility with the common ground. The issue, therefore, is the addressee’s capacity to address the common ground and to check whether the implicature is true or false. The relevant points here are the reliability of the speaker and the addressee’s confidence in the speaker. A more general question now arises: Are generalized conversational implicatures triggered by default or not? The neo-Gricean approach answers in the affirmative, as I will demonstrate in Section 8.6.

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

195

196

JACQUES MOESCHLER

8.6 Are Generalized Conversational Implicatures Default Inferences? One is tempted to view GCIs as default interpretation. This view is possible only if, firstly, implicatures are the result of general principles or rules which apply by default, based on some heuristics, and secondly, if they are based on a precise hierarchy between types of inference. This model, which was developed in Levinson (2000), has recently been challenged because of two types of facts: The first is the experimental testing of the predictions of the default interpretation; the second is the difficulty of obtaining, in an automatic procedure, the correct implicature. Let us start with the principled approach to speaker meaning, which was first put forward by Horn (1984) and developed by Levinson (2000). Their idea was to reduce the nine Gricean maxims of conversation to two principles, which are generalizations of the maxims of Quantity on the one hand, and the maxim of Relation of the other.10 According to Horn (1984: 13), these two principles are formulated as follows: (31) The Q-Principle Make your contribution sufficient (cf. Quantity). Say as much you can (given Relation). (32) The R-Principle Make your contribution necessary (cf. Relation, Quantity, Manner). Say no more than you must (given Quantity).

Levinson’s formulation is slightly different: The R-Principle is termed the I(informativity)-Principle (Levinson 2000: 76, 114): (33) Q-Principle Do not provide a statement that is informationally weaker than your knowledge allows. (34) I-Principle Say as little as necessary.

The idea behind both formulations is straightforward: Either the speaker gives the strongest information (Quantity1) or he says as little as necessary (Relevance). The two principles are therefore balanced and operate when specific lexical items or linguistic constructions are addressed. These are respectively described as Q- and I-heuristics (Levinson 2000: 35, 37): (35) Q-Heuristics What isn’t said, isn’t. 10

I do not have enough space here to develop a third principle, based on the Gricean maxims of Manner, the M-Principle: “Indicate an abnormal, nonstereotypical situation by using a marked expression that contrasts with those you would use to describe the corresponding normal, stereotypical situation” (Levinson 2000: 136). This principle triggers M-implicatures, which indicate abnormal situations. The classic example is the contrast between stopping the car and getting the car to stop, which M-implicates an abnormal way of stopping the car.

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

8 Context in Gricean and Neo-Gricean Pragmatics

(36) I-Heuristics What is expressed simply is stereotypically exemplified.

Scalar implicatures that are triggered by weak expressions in quantitative scales, for instance, are examples of Q-Heuristics, whereas pragmatic enrichments illustrate I-Heuristics: (37) Quantitative scales , , , , , etc. (38) Pragmatic enrichments a. conditional perfection: if for if and only if b. conjunction buttressing: and for and then c. bridging: the X . . . the Y for Y is a part of X d. negative strengthening: do not like for dislike etc.11

In other words, when an expression like those mentioned above occurs in an utterance, the default inference is triggered. The implicature is cancelled only if there is a conflict between the implicature and the common ground or between the implicature and a stronger inference. Levinson (2000: 49) describes the process in this way: The process of updating the common ground with the communicative content of a new utterance can then be treated as the ordered incrementation of that background with the entailments, implicatures, and presuppositions of the utterance. . . . Informally and metaphorically, we can think of the common ground as a bucket, holding all the facts mutually assumed, either because they’re common knowledge or because they have been asserted and accepted. (my emphasis)

The order of the incrementation contains the condition that “each incrementation is consistent with the content of the bucket” (Levinson 2000: 50): (39) Order of incremented information a. Entailments b. Q GCIs (. . .) c. M GCIs d. I GCIs

In other words, entailments are stronger than GICs. Among GICs, Q-implicatures (Q GCIs) are stronger than M-implicatures (M GCIs), which are stronger than I-implicatures (I GCIs). To give the simplest possible illustration, in the case of a scalar implicature whose implicature is cancelled, this cancellation occurs because the stronger expression has been made explicit, and particularly because this content is entailed and as such cancels the 11

Experimental works on pragmatic enrichment are for conditional perfection, Blochowiak et al. (2022); for conjunction buttressing, Noveck et al. (2009) and Blochowiak and Castelain (2018); for negative strengthening, Gotzner et al. (2018).

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

197

198

JACQUES MOESCHLER

implicature. The following incremented information explains why example (40) is not a contradiction: (40) Some students, in fact all, were at the party. a. Implicature: not all students were at the party. b. Entailment: all students were at the party. c. Cancellation of the implicature: it is not the case that not all of the students were at the party. d. Incrementation of the common ground: all students were at the party.

Therefore, as the neo-Gricean approach predicts, only the explicitation of the negation of the Q-implicature makes entailment (40b) possible. This explains why (41) is generally not interpreted as (40b): (41) Some students were at the party. Implicature: not all students were at the party

This prediction is very strong. But it raises three major points. First, there are contexts in which a Q-implicature is not the preferred interpretation; that is, there are contexts in which the default pragmatic interpretation is not appropriate. Second, the results of experiments on Q-implicatures do not confirm the predictions of the default interpretation. Third, scalar implicatures cannot be explained by referring to the Gricean maxims alone. I will develop these issues in Section 8.7.

8.7 The Role of Context in Scalar Implicature Computation Let us start with the third issue. Fox (2007) has shown that, in the case of or, which belongs to the quantitative scale , it is not possible to derive the scalar implicature not and (43b). The only possible derivation is the ignorance implicature (43c): (42) Sue talked to John or Fred. (43) a. Basic inference Sue talked to John or Fred or both. b. Scalar inference Sue didn’t talk to both John and Fred. c. Ignorance inference The speaker doesn’t know that Sue talked to John. The speaker doesn’t know that Sue talked to Fred.

The reason why (43b) is not inferable is as follows: First, it is evident that each disjunct is more informative than the disjunction; second, by referring to Quality1 (“Do not say what you believe to be false”), we can infer that the speaker believes that her utterance, which is a disjunction, is true. But the speaker does not know that both the first and the second disjunct are true because of Fox’s reformulation of the maxim of Quantity:

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

8 Context in Gricean and Neo-Gricean Pragmatics

(44) Maxim of Quantity (Fox 2007: 73) If S1 and S2 are both relevant to the topic of the conversation and S1 is more informative than S2, if the speaker believes that both are true, the speaker should utter S1 rather than S2.

Therefore, in Fox’s view, we conclude that the speaker is ignorant of the truth or the falsehood of p and q (the stronger expression), which is as relevant in terms of the topic of the conversation as the disjunction p or q. In other words, with Quantity and Quality, only the ignorance implicature can be driven.12 How can we now derive the scalar implicature “it is not the case that p and q” of p or q? According to Chierchia (2013: 102), the generation of a scalar implicature requires an exhaustification of the assertion: “The hearer, running through relevance, quantity, and opinionatedness, concludes that the assertion is the only member of the alternative set that is, in fact, true according to the speaker.” In other words, the generation of a scalar implicature requires, along with the maxims of Quantity and Quality, something that Chierchia calls opinionatedness: the fact that the speaker has an opinion about the truth or falsity of the utterance, and also about the truth or falsity of the stronger expression p and q. So, in the case of p or q, the speaker’s opinion is that the conjunction p and q is false. For this opinion to be true, the speaker must access a set of alternatives. This is exactly what Horn’s scales provide. But instead of being the starting point of a default inference, accessing the scalemate is required throughout an entire reasoning process. This type of reasoning, which targets or-scalar implicature, is illustrated in (45) (Chierchia 2013: 101): (45) a. b. i. ii. iii. iv. v. vi. vii.

John or Bill will show up. John and Bill will show up. The speaker has said [a] and not [b], which, presumably, would have been also relevant. [relevance] [b] entails [a], and hence [b] is more informative than [a]. If the speaker believes that [b] holds, she would have said so. [quantity] It is not the case that the speaker believes that [b] holds. The speaker has an opinion as to whether [b] holds. [opinionatedness] The speaker takes [b] to be false. The speaker is conveying that John or Bill but not John and Bill show up. [scalar implicature]

As this example shows, a scalar implicature is the result of a reasoning, and not a default inference. Experimental data challenge the default approach, too. First, in the large number of experimental works on scalar implicatures,13 the general trend is 12

13

However, Zhan (2018) has given experimental arguments showing that both scalar and ignorance implicatures are computed immediately when encountering the connective or. See among others Papafragou and Musolino (2003) and Breheny et al. (2006). For a synthesis, see Noveck (2018, chapters 6 and 7) and Zufferey et al. (2019, chapter 7).

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

199

200

JACQUES MOESCHLER

that scalar implicatures are not default inferences. Breheny et al. (2006: 434) claim for instance that “our results suggest that these scalar implicatures are dependent on the conversational context and that they show none of the autonomy predicted by the Default view.” In other words, the default inference approach advocated by neo-Gricean pragmatists is not backed up by experimental data. These empirical results are not surprising. It was demonstrated in the earliest days of experimental trends in pragmatics that implicatures are costly: they require more cognitive efforts than semantic or logical readings, at least in terms of the processing of logical words (Noveck 2001; Noveck and Reboul 2008 for a synthesis).14 Second, experimental research has proven15 that scalar implicatures are not generated in some specific contexts, such as a context of politeness. For example, the scalar implicature of some, in other words “not all,” is not triggered in face-threatening contexts (Brown and Levinson 1987), as opposed to its logical reading, which would be “some if not all.” Bonnefon et al. (2009: 251, 255) presented the following results: Imagine that you have joined a poetry club, which consists of five members in addition to you. Each week one member writes a poem, and the five other members discuss the poem in the absence of its author. This week, it is your turn to write a poem and to let the others discuss it. After the discussion, one fellow member confides to you that “Some people hated your poem.” . . . When X in “some X-ed” threatens the face of the listener, individuals are less likely to infer that the speaker meant or knew that not all X-ed.

How can we explain this very crucial observation? Here is a possible explanation (Moeschler 2021): “It is obvious that face-work, which implies minimisation to preserve the hearer’s positive face, or his self-image, wins over the process of the working out of the implicature triggered by some. This is a situation in which two types of logic clash: the logic of rationality and reasoning is ruled out by the logic of politeness.” In other words, the logical reading has a minimization effect in terms of politeness. This is not surprising: The logical reading is weaker than the pragmatic one and is less specific. In a face-threatening context, therefore, the logical reading of some is compatible with its maximal reading (“all”), which would be strongly face-threatening if expressed explicitly. By contrast, the scalar implicature reading would be difficult for the speaker, because it might be the case that that implicature (“not all”) is false, while the logical reading (“all”) is optimally relevant. From the perspective of the I-Principle it could be argued, however, that the speaker says less to convey more: According to this interpretation, face-threatening contexts would favor an I-implicature over a Q-implicature. But context is crucial, even in this interpretation, and no default interpretation is at stake. 14

15

More recent experimental research has shown that scalar implicature is not a uniform phenomenon (Doran et al. 2009), and that the processing cost depends on the semantic property of the scalar word (van Tiel et al. 2019). See among others Bonnefon et al. (2009), Mazzarella (2015), Terkourafi et al. (2020).

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

8 Context in Gricean and Neo-Gricean Pragmatics

8.8 Conclusion This chapter has investigated the contribution of context in the triggering of meaning and has covered issues ranging from propositional context, including presuppositions, to conventional implicatures and conversational implicatures. A first nontrivial finding was that expressed propositions are context-sensitive, as are conventional implicit meanings including conventional implicatures and presuppositions. Second, the focus on generalized conversational implicatures confirmed a limited contribution to context, which most often becomes apparent when implicatures are explicitly cancelled. However, as demonstrated through theoretical and experimental arguments, the prototypical examples of scalar implicatures, mainly triggered by logical words like some and or, are in fact context sensitive. We can therefore draw the conclusion that context is pervasive in the computation of implicatures. In this case it is possible to assert that the use of Grice’s concepts, such as the Cooperative Principles and the maxims of conversation, give rise to a contextual mechanism. This conclusion goes far beyond the classical anticontextualist interpretations of Gricean and neo-Gricean pragmatics.

References Austin, J. L. (1962). How to Do Things with Words. Cambridge, MA: Harvard University Press. Bach, K. (1999). The myth of conventional implicatures. Linguistics and Philosophy, 22(4), 327–366. Barwise, J., and Perry, J. (1983). Situations and Attitudes. Cambridge, MA: MIT Press. Beaver, D., and Zeevat, H. (2007). Accommodation. In G. Ramchand and C. Reiss (eds.), Oxford Handbook of Linguistic Interfaces (pp. 503–538). Oxford: Oxford University Press. Blakemore, D. (1987). Semantic Constraints on Relevance. Oxford: Basil Blackwell. Blochowiak, J., and Castelain, T. (2018). How logical is natural language conjunction? An experimental investigation of the French conjunction et. In P. Saint-Germier (ed.), Language, Evolution and Mind: Essays in Honour of Anne Reboul (pp. 97–125). London: College Publications Blochowiak, J., and Grisot, C. (2018). The pragmatics of descriptive and metalinguistic negation: experimental data from French. Glossa: A Journal of General Linguistics, 3(1), 50. https://doi.org/10.5334/gjgl.440. Blochowiak, J., Castelain, T., Rodriguez-Villagra, O. A., and Musolino, J. (2022). If and only if people were logical! The effect of pragmatic enrichment on reasoning with abstract and realistic materials. Journal of Pragmatics, 197, 137–158. Bonnefon, J.-F., Feeney, A., and Villejoubert, G. (2009). When some is actually all: Scalar inference in face-threatening contexts. Cognition, 112, 249–258. Breheny, R., Katsos, N., and Williams, J. (2006). Are generalized conversational implicatures generated by default? An on-line investigation into the role of context in generating pragmatic inferences. Cognition, 100(3), 434–363.

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

201

202

JACQUES MOESCHLER

Brown, P., and Levinson, S. C. (1987). Politeness: Some Universals in Language Use. Cambridge: Cambridge University Press. Carston, R. (2002). Thoughts and Utterances: The Pragmatics of Explicit Communication. Oxford: Basil Blackwell. Chierchia, G. (2013). Logic in Grammar: Polarity, Free Choice, and Interpretation. Oxford: Oxford University Press. Doran, R., Baker, R., McNabb, Y., Larson, M., and Ward, G. (2009). On the non-unified nature of scalar implicature: An empirical investigation. International Review of Pragmatics, 1, 211–248. Fox, D. (2007). Free choice and the theory of scalar implicatures. In U. Sauerland and P. Stateva (eds.), Presupposition and Implicature in Compositional Semantics (pp. 71–120). Basingstoke: Palgrave. Frege, G. [1892] (1948). Sense and reference. The Philosophical Review, 57(3): 209–230. Gazdar, G. (1979). Pragmatics: Implicatures, Presupposition, and Logical Form. New York: Academic Press. Gotzner, N., Solt, S., and Benz, A. (2018). Scalar diversity, negative strengthening, and adjectival semantics. Frontiers in Psychology, 9, 1659. doi: 10.3389/fpsyg.2018.01659. Grice, H. P. (1975). Logic and conversation. In P. Cole and J. L. Morgan (eds.), Syntax and Semantics, Vol. III: Speech Acts (pp. 41–58). New York: Academic Press. Grice, H. P. (1989). Meaning. In Studies in the Way of Words. (pp. 213–223). Cambridge, MA: Harvard University Press. Horn, L. R. (1984). Toward a new taxonomy for pragmatic inference. In D. Schiffrin (ed.), Form and Use in Context: Linguistic Applications (GURT’84) (pp. 11–42). Washington, DC: Georgetown University Press. Horn, L. R. (1985). Metalinguistic negation and pragmatic ambiguity. Language, 61(1), 121–174. Horn, L. R. (1989). A Natural History of Negation. Chicago: The University of Chicago Press. Horn, L. R. (2004). Implicature. In L. R. Horn and G. Ward (eds.), The Handbook of Pragmatics (pp. 3–28). Oxford: Blackwell. Kaplan, D. 1989. Demonstratives. In J. Almog, J. Perry, and H. Wettstein (eds.), Themes from Kaplan (pp. 481–563). Oxford: Oxford University Press. Karttunen, L. and Peters, S. (1979). Conventional implicatures. In C.-K. Oh and D. A. Dinneen (eds.), Syntax and Semantics, Vol. XI: Presuppositions (pp. 1–56). New York: Academic Press. Korta, K., and Perry, J. (2012). Critical Pragmatics: An Inquiry into Reference and Communication. Cambridge: Cambridge University Press. Langacker, R. (1991). Concept, Image, and Symbol: The Cognitive Basis of Grammar. Berlin: Mouton de Gruyter. Levinson, S. C. (2000). Presumptive Meanings: A Theory of Generalized Conversational Implicatures. Cambridge, MA: MIT Press. Mazzarella, D. (2015). Politeness, relevance and scalar inferences. Journal of Pragmatics, 79, 93–106. Moeschler, J. (2018). A set of semantic and pragmatic criteria for descriptive vs. metalinguistic negation. Glossa: A Journal of General Linguistics, 3(1), 58. 1–30. https://doi.org/10.5334/gjgl.439. Moeschler, J. (2019). Non-Lexical Pragmatics: Time, Causality and Logical Words. Berlin: Mouton de Gruyter.

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

8 Context in Gricean and Neo-Gricean Pragmatics

Moeschler, J. (2021). Why Language? What Pragmatics Tells Us about Language and Communication. Berlin: Mouton de Gruyter. Noh, E. J., Choo, H., and Koh, S. (2013). Processing metalinguistic negation: Evidence from eye-tracking experiments. Journal of Pragmatics, 57, 1–18. Noveck, I. (2001). When children are more logical than adults: Investigations of scalar implicatures. Cognition, 78(2), 165–188. Noveck, I. (2018). Experimental Pragmatics: The Making of a Cognitive Science. Cambridge: Cambridge University Press. Noveck, I., and Sperber, D., eds. (2004). Experimental Pragmatics. Basingstoke: Palgrave. Noveck, I., and Reboul, A. (2008). Experimental pragmatics: A Gricean turn in the study of language. Trends in Cognitive Sciences, 12(11), 425–431. Noveck I., Chevallier, C., Chevaux, F., Musolino J., and Bott L. (2009). Children enrichment of conjunctive sentences in context. In P. De Brabanter M. and M. Kissine (eds.), Current Research in the Semantics/ Pragmatics Interface (pp. 211–234) Bingley: Emerald. Papafragou, A., and Musolino, J. (2003). Scalar implicatures: Experiments at the semantics/pragmatics interface. Cognition, 86, 253–282. Potts, C. (2005). The Logic of Conventional Implicatures. Oxford: Oxford University Press. Reboul, A. (2019). Particularized Conversational Implicatures: Why there are Conversational Implicatures. In S. Zufferey, J. Moeschler, and A. Reboul, Implicatures (pp. 67–87). Cambridge: Cambridge University Press. Recanati, F. (1994). Contextualism and anti-contextualism in the philosophy of language. In S. L. Tsohatzidis (ed.), Foundations of Speech Act Theory (pp. 156–166). London: Routledge. Sadock, J. (1978). On testing for conversational implicature. In P. Cole (ed.), Syntax and Semantics, Vol. IX: Pragmatics (pp. 281–297). New York: Academic Press. Searle, J. R. (1969). Speech Acts: An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. Searle, J. R. (1979). Expression and Meaning: Studies in the Theory of Speech Acts. Cambridge: Cambridge University Press. Sperber, D., and Wilson, D. (1986). Relevance: Communication and Cognition. Oxford: Basil Blackwell. Stalnaker, R. (1977). Pragmatic Presuppositions. In A. Rogers, B. Wall, and J. P. Murphy (eds.), Proceedings of the Texas Conference on Performatives, Presuppositions and Implicatures (pp. 135–147). Arlington: Center for Applied Linguistics. Stalnaker, R. (1978). Assertion. In P. Cole (ed.), Syntax and Semantics, Vol. IX: Pragmatics (pp. 315–332). New York: Academic Press. Strawson, P. (1950). On referring. Mind, 59(235), 320–344. Terkourafi, M., Weissman B., and Roy, J. (2020). Different scalar terms affected by face differently. International Review of Pragmatics, 12, 1–43. van Tiel, B., Pankratz, E., and Sun, C. (2019). Scales and scalarity: Processing scalar inferences. Journal of Memory and Language, 105, 93–107. Wilson, D., and Sperber, D. (2012). Meaning and Relevance. Cambridge: Cambridge University Press.

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

203

204

JACQUES MOESCHLER

Zhan, L. (2018). Scalar and ignorance inferences are both computed immediately upon encountering the sentential connective: The online processing of sentences with disjunction using the visual world paradigm. Frontiers in Psychology, 9, 61. doi: 10.3389/fpsyg.2018.00061. Zufferey, S., Moeschler, J., and Reboul, A. (2019). Implicatures. Cambridge: Cambridge University Press.

https://doi.org/10.1017/9781108989275.009 Published online by Cambridge University Press

9 Sociopragmatics and Context Sophia Marmaridou

9.1 Introduction Arising in the domain of the philosophy of language, pragmatics has been broadly defined as the study of language in use, or “the general conditions of the communicative use of language” (Leech 1983: 10). Within this framework, a distinction has been drawn between sociopragmatics, relating pragmatic meaning to an assessment of participants’ social distance, the language community’s social rules and appropriateness norms, discourse practices and accepted behaviors, and pragmalinguistics, referring to the study of the particular resources that a given language provides for conveying illocutionary and interpersonal meaning (Leech 1983; Thomas 1983). For example, the T/V distinction is available in several languages, including French and Greek, to mark social proximity or distance between interlocutors and, as such, is an object of pragmalinguistic study. How this distinction is deployed to create, or reflect, social context at the time of the utterance is addressed within the area of sociopragmatics. The impact of this distinction has been particularly felt in the conceptual organization of some theoretical constructs in the field of applied linguistics, such as pragmatic failure, pragmatic transfer, and pragmatic development (see Marmaridou 2011). Moreover, this distinction has also informed historical pragmatics, or pragmaphilology, describing the contextual aspects of historical texts (Jucker 2006). In this framework, historical sociopragmatics focuses on the interaction between specific aspects of social context and particular historical language use that leads to pragmatic meanings. Conversely, it also leads to the reconstruction of contexts on the basis of historical texts, the latter viewed as carrying evidence of, or projecting, their own contexts (Culpeper 2009). Notably, the association between language and social context in sociopragmatic research has not been typically held to be systematically encoded in language. This seems to follow directly from its differentiation from pragmalinguistics, which, as already noted, is the locus of linguistic resources and

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

206

SOPHIA MARMARIDOU

regularity. While the distinction between pragmalinguistics and sociopragmatics seems less relevant in a wider conception of pragmatics as a “general cognitive, social, and cultural perspective on linguistic phenomena in relation to their usage in forms of behaviour” (Verschueren 1999: 7), yet, when maintained, it still reflects the separation of linguistic form from the concept of context as a cognitive, social, and linguistic construct (Fetzer 2011). Admittedly, context is not easy to pin down and, therefore, its association with language in a principled manner is a challenging task. In view of the above, and within a Construction Grammar framework (Fried and Östman 2004), the aim of this chapter is to show that the locus of sociopragmatic analysis, Leech’s “local contexts,” can in fact be viewed as the domain of socioculturally defined genres that are often associated with particular (genre) constructions. The question to be addressed in this framework is to what extent speakers’ understanding of context is systematic, conventional, and, hence, an inherent part of grammar and the description of language. Construction Grammar is well suited to address this issue in that it seeks to account for language in its totality, from highly productive patterns of language to more idiomatic and conventional expressions, and from sentencelevel to discourse-level constructions. The data to be discussed include extracts from recipes (Östman 2005; Ruppenhofer and Michaelis 2010), labelese (Ruppenhofer and Michaelis 2010), couple talk (Antonopoulou et al. 2015), stage directions (Nikiforidou 2021), and TV shows (Marmaridou, forthcoming). On this evidence, it is possible to argue that sociopragmatic context, typically encoded at the meso-level of genre, is often associated with a set of discourse specifications that are routinely incorporated in the description of a language’s grammatical constructions, thereby linking context to grammatical form. To respond to the above-stated aim, it is necessary to trace the relationship between sociopragmatics and context. As will become clear in the following sections, the latter has been variously construed as the situation in which communication is taking place, or as the societal conditions of a speech event in which interlocutors are engaged, or as a cognitive object comprising knowledge schemas available to interlocutors during communication (Fetzer 2011). In a sense, the development of sociopragmatics as a research program within the broader field of pragmatics mirrors corresponding developments in perspectivizing context and, as a result, has informed research areas such as L2 pedagogy, historical pragmatics, and pragmatic approaches to Genre. In what follows, the concept of context will be focused upon as it emerges in the sociopragmatics agenda, the latter associated with what Leech (1983: 10) has termed general pragmatics. Toward this attempt, and in the next section, the origins of sociopragmatics will be sought in Leech’s (1983) theory of pragmatics and Thomas’s (1983) investigation of L2 learners’ pragmatic failure, with a view to highlighting the understanding of context in their joint program. In Sections 9.3 and 9.4 sociopragmatics and co-text will be associated with L2 pedagogy and historical pragmatics respectively, in an attempt to

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

9 Sociopragmatics and Context

bring forward the development of the concept of context within each field of sociopragmatic inquiry. Section 9.5 will focus on genre as a social, functional, and cognitive construct encompassing contextual parameters of linguistic communication. On the basis of examples drawn from different genres, it will be argued that the latter are characterized by specific constructional reflexes of sociopragmatic context which are available to speakers of the language as knowledge schemas, i.e., as part of a speaker’s knowledge of the language. The concluding section summarizes the basic claims of the paper and poses questions for further research.

9.2 A View of the Origins of the Sociopragmatics Program The emergence of the term sociopragmatics is associated with its corresponding term pragmalinguistics, both introduced in Thomas (1983) and Leech (1983) in their joint attempt to define the object of pragmatic study and its role in L2 pedagogy. To appreciate the need to (re)define the object of pragmatics, as these scholars did, one has to resort to the so-called narrow view of this field as it emerged in American pragmatism, represented in the semiotic tradition of Peirce and Morris in the 1930s, and in the English (Anglo-) ordinary language philosophy of the 1950s, starting with Wittgenstein and Austin and developing later into Searle’s theory of speech acts and intentionality, and Grice’s theory of meaning and communication – hence the name AngloAmerican tradition of pragmatics. All the above-mentioned scholars, among others, had an understanding of linguistic meaning as residing in the use of language by its speakers in the social world and were quite explicit about this idea. For example, Morris (1938: 108) defines pragmatics as in the following extract: It is a sufficiently accurate characterization of pragmatics to say that it deals with the biotic aspects of semiosis, that is, with all the psychological, biological, and sociological phenomena which occur in the functioning of signs.

Similarly, Austin delimits the category of performative utterances, and later of illocutionary acts, as part of institutionalized procedures conforming to social convention, as spelt out in felicity conditions, which are in fact conditions of appropriateness of use (1962: 14–15). In further elaborating his theory of speech acts with the concept of the illocutionary force of a speaker’s utterance, he introduces an intentional orientation, to be taken up in Searle’s (1975) theory of speech acts and intentionality and Grice’s (1975) theory of communication and implicature. It should be pointed out that these so-called ordinary language philosophers must be credited with a shift of emphasis from language as description to language as action and the consequent focus on the speaker, or addressee, and context (Marmaridou 2000: 22). Defining pragmatics as the study of language use, complementing syntax and semantics in accounting for linguistic meaning, has been associated with a number of problems extensively

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

207

208

SOPHIA MARMARIDOU

discussed in Levinson (1983: 7–32), who finally points out that “the upper bound of pragmatics is provided by the borders of semantics and the lower bound by sociolinguistics,” while at the same time admitting that “drawing a boundary between sociolinguistic and pragmatic phenomena is likely to be an exceedingly difficult enterprise” (p. 29). It is probably such problems that led Leech (1983) and Thomas (1983) to proceed on the one hand to a definition of general pragmatics as comprising the interaction of the Cooperative and Politeness Principles and, on the other, to the introduction of two subfields of pragmatic study, namely sociopragmatics and pragmalinguistics. Sociopragmatics refers to more specific, “local” conditions on language use that account for the variability of different cultures or language communities, different social situations, or different social classes (Leech 1983: 10). Pragmalinguistics, on the other hand, concerns “the study of the more linguistic end of pragmatics – where we consider the particular resources which a given language provides for conveying particular illocutions” (Leech 1983: 11). It transpires from the above that these two subfields of pragmatics primarily target appropriateness conditions of speech acts in particular sociocultural contexts, typically in instances of face-to-face interaction. Emphasis is placed on a set of parameters that guide interlocutors’ cooperative communication, such as their relative social status, their institutional roles, and the (in)formality of the situation within what is perceived as the cultural norm.

9.3 Sociopragmatics in L2 Pedagogy The contribution of the pragmalinguistics/sociopragmatics distinction to L2 pedagogy needs to be assessed against the overall aim of L2 instruction, which was to enable learners to communicate in a foreign language (mainly English) with native speakers of that language. Therefore, emphasis was placed on the linguistic realization of speech acts in culturally appropriate ways, in a largely comparative and cross-cultural framework that associates linguistic and cultural difference with communication difficulty. In this framework, Thomas (1983) must be credited with the incorporation of the sociopragmatics/pragmalinguistics distinction in L2 pedagogy in her investigation of issues relating to L2 linguistic competence. She views the latter as consisting of grammatical competence (i.e., knowledge of phonology, syntax, and semantics) and pragmatic competence as “the ability to use language effectively in order to achieve a specific purpose and to understand language in context” (Thomas 1983: 94). In this framework, pragmatic failure reflects the L2 learner’s inability to either recognize the illocutionary force of a speaker’s utterance or produce an utterance expressing her intentions in a socially appropriate way. While accepting that there are no strict boundaries between the two fields, the author shows that the inappropriate choice of forms and strategies in performing a speech act leads to pragmalinguistic failure, whereas the inability to recognize the size of imposition of a speech act in a particular culture, the existence of taboos, or

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

9 Sociopragmatics and Context

differing assessments of relative power or social distance between interlocutors in it, leads to sociopragmatic failure. Prominent in this discussion is the issue of politeness which, on the one hand, belongs to the object of general pragmatics and, on the other, is the cornerstone of appropriateness in performing speech acts. It should be noted that in her discussion of pragmalinguistic failure, Thomas (1983: 101) acknowledges the predictability and highly conventionalized form of certain utterances to perform particular speech acts, such as the often-quoted “Can you do X . . .” example, which can be fairly easily taught in an L2 context. At the same time, the identification of social roles and stereotypes, power relations between interlocutors relating to age, social status, or dialect, socioculturally based evaluations of situations and linguistic value judgments constitute social variables that L2 learners must take into account in order to employ the appropriate linguistic forms in their performance of speech acts. At a theoretical level, the study of such variants is typically part of sociolinguistics, which raises the methodological issue of the dividing line between this field of investigation and sociopragmatics. In other words, while pragmalinguistics seems to cover the micro-level of analysis of language use, i.e., the linguistic end of the interaction, sociolinguistics must be seen as the macro-level specifying the parameters that affect the interlocutor’s decisions on what it is appropriate to say on particular occasions of use. This allocates sociopragmatics the task of addressing the interface between the two and specifying this meso-level of analysis, i.e., the local context of language use. The same theoretical concerns emerge in the subsequent focus on developmental issues in L2 learning, which has brought to the fore the issue of pragmatic transfer as an aspect of an L2 learner’s interlanguage (Kasper and Rose 2002). However, the lack of sharp boundaries between pragmalinguistics and sociopragmatics is also testified in studies of pragmatic transfer and L2 development, as it is usually difficult to diagnose the source of transfer in a learner’s inappropriate use of a speech act strategy at any particular stage of L2 development. Regardless of the indeterminacy of the findings of L2 acquisition and developmental research, the subfields of pragmalinguistics and sociopragmatics have informed L2 teaching practices more successfully, as one might expect. Given that the widely accepted aim of L2 teaching is to enable successful communication in a foreign language, the distinction between the above subfields serves to familiarize learners with the appropriate use of linguistic resources in L2 to convey communicative acts and encode interpersonal meaning on the one hand, and, on the other, to enable them to perceive, and respond accordingly to, the sociocultural parameters of a communicative encounter, such as relative social distance between interlocutors, institutional roles, and respective rights and obligations, the degree of imposition involved in a particular speech act, and variables such as the degree of familiarity between interlocutors, sex, and age (Kasper and Rose 2002).

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

209

210

SOPHIA MARMARIDOU

The usefulness of the pragmalinguists/sociopragmatics distinction in L2 teaching contexts is not paralleled by that of L2 testing, given that it is often difficult to evaluate whether a learner’s error reflects pragmalinguistic or sociopragmatic failure (McNamara and Roever 2006: 55). For example, the absence of politeness markers from a learner’s utterance might reflect pragmalinguistic weakness, if the learner does not know the specific markers, or sociopragmatic deficit if she is not aware that they should be used in a particular circumstance. Moreover, L2 tests do not seem to establish context in a way that authentic communication does in the real world, nor do they represent the speaker’s or the addressee’s actual face needs in real terms while they restrict the number of contextual cues that would be available in a real encounter. In view of difficulties in teasing apart L2 learners’ sociopragmatic from pragmalinguistic competence, attempts have shifted to establishing the priority of sociopragmatic development over pragmalinguistic development, its correlations with overall L2 proficiency, and the context of L2 learning itself. Chang’s (2011) study on interlanguage pragmatic development indicates that the relation between pragmalinguistic competence and sociopragmatic competence is a complex and interwoven one rather than a simple, linear ‘‘whichprecedes-which’’ kind of relation. Overall L2 proficiency levels also play a role in pragmatic development, but awareness of sociopragmatic parameters in communication is not always paired with appropriate pragmalinguistic resources. For example, a student may be aware of the social distance with a teacher but may use an inappropriate formula in apologizing (e.g., “Pardon me . . .”). Evidently, insufficient knowledge of the linguistic form–pragmatic force relationship (pragmalinguistics) may bring about an inappropriate demonstration of sociopragmatic competence. Apart from L2 proficiency, duration of study abroad seems to impact the acquisition of sociopragmatic variation patterns. Expectedly, while student mobility was found to be beneficial for the development of learners’ sociopragmatic competence, it was also evident that it is not the context per se that promoted learning, but also the nature of the interactions, the quality of the experiences, and the overall effort made to use the L2 (Devlin 2019). A shift of L2 research away from the appropriate use of speech act forms in face-to-face interaction with native speakers of the language has led to a different perspective on sociopragmatic context as emerging in the field of intercultural pragmatics and the development of learners’ intercultural communicative competence. Intercultural pragmatics is concerned with instances of communication between speakers with different first languages and usually representing different cultures, who communicate in a common language (Kecskes 2012). In this perspective, “individual prior experience and actual situational experience conditioned by socio-cultural factors are equally important in meaning construction and comprehension” (Kecskes 2012: 608). It follows that intercultural pragmatics embraces both a sociocultural view of L2 development and a subjective, experiential understanding of this

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

9 Sociopragmatics and Context

process. On this analysis, context is both a societal construct consisting of the social conditions affecting intercultural communication and therefore mainly static and objective, and a subjective one in that it is perceived by individual speakers who process objective societal factors on the basis of their prior experience, linguistic, nonlinguistic, affective, etc. Investigating the subjective dimensions of intercultural communication, Chang and Haugh (2017) show that L2 learners’ attitudes affect the behavior of L2 learners, while their experiences take precedence over the current situational context in developing knowledge of the L2 and its use. This shift of emphasis away from given sociocultural aspects of context and toward attitudinal effects on interaction has brought forward aspects of meaning associated with metaphor, irony, humor, and interjections or prosodic markers. To the extent that such “non-propositional” aspects of meaning affect interlocutors’ understanding of each other as situated in a social context, or their understanding of societal attitudes promoted in a text, they are said to be sociopragmatic par excellence. Within this framework, Ifantidou (2019) found that in opinion articles from broadsheet British and US newspapers, L2 learners understood metaphor correctly regardless of the number of unknown words they encountered. It was also found that experiment participants relied less on contextual linguistic cues and more on personal experience in inferring the meaning of metaphorical expressions which included unknown words. Given that metaphorical sentences activate brain regions associated with emotion processing more strongly than their literal counterparts (Citron and Goldberg 2014), it is not surprising that L2 learners prefer metaphors over other linguistic stimuli. What is significant in this type of research is that, apart from a shift from a given sociocultural context to attitudinal aspects of communication, there is a parallel shift from face-to-face interaction to aspects of text understanding. Apparently, texts, such as opinion articles in newspapers, are generally experienced as artifacts known to represent, maintain, and create diverse positions and values, while tapping on emotions and triggering reader reactions. As such, they are likely to create expectations in the L2 readers’ minds that may be said to facilitate understanding. In this sense metaphor understanding is both inference-based and socially situated. Knowledge of face-to-face conventions of appropriateness and societal norms appears to be paralleled by knowledge of textual convention of sociocultural artifacts. Such knowledge, comprising speakers’, and writers’, cognitive context, seems to play a significant role in the analysis of texts within the framework of historical sociopragmatics, as will be shown in the next section.

9.4 Historical Sociopragmatics The cognitive view of context as comprising knowledge of textual convention is observed in historical sociopragmatics focusing on language data in written form, even when oral communication is represented. More specifically,

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

211

212

SOPHIA MARMARIDOU

historical sociopragmatics investigates language use in its situational, local context and the ways in which situational contexts generate norms which interlocutors employ or exploit for pragmatic purposes. In fact, Culpeper (2009) identifies three levels of context against which texts may be understood: the most local, immediate text and co-text of interlocutors, the medial level of social situation (e.g., speech events, activity types, frames), par excellence the domain of sociopragmatics in the author’s view, and the most general level referring to national and regional cultures, institutional cultures, etc. From a theoretical point of view, sociopragmatic context seems to be reflected in the textual categories adopted in annotating available historical corpora (e.g., the Helsinki Corpus of English Texts). These categories refer to written genres, including essays, fiction, personal letters, and manuals for good behavior, but also to speech-based genres, including dialogues from plays and fiction, court proceedings, political debates, public speeches, and sermons (Jacobs and Jucker 1995). From an analytic point of view, the examination of the data indicates that interlocutors, in communicating their intentions, are aware of both social and textual convention and exploit such convention to achieve their purposes. For example, Fitzmaurice (2009) focuses on the sociopragmatic construction of implicature and inference in the illicit courtship correspondence between Edward Wortley and Mary Pierrepont. The communicative practices employed by the correspondents are shown to be embedded in the local context comprising the activity type (Levinson (1992) and social ritual they engage in, namely courting through correspondence. On such evidence, it can be argued that in sociopragmatic analysis all aspects of context are necessarily present: linguistic context or co-text, social context comprising social roles, values and norms, and cognitive context comprising knowledge of such norms and values that are followed or challenged in using language toward achieving specific purposes. Prioritizing context in the analysis of historical texts, Archer and Culpeper (2009) introduce the term sociophilology to refer to the study of the ways in which historical contexts, including the co-text, genre, social situation, and/or culture, shape the functions and forms of language as used within these contexts. In this approach, context is the investigative starting point, rather than language form or interactive function, while corpus linguistic techniques are pivotal in this enterprise (Archer 2017). Evidently, “sociophilology shares with sociopragmatics a fundamental interest in ‘local’ conditions of language use” (Archer and Culpeper 2009: 288). Using the Sociopragmatic Corpus, a subsection of the Corpus of English Dialogues: 1560–1760, the authors examine two genres, those of drama and trial proceedings, to analyze the speech of dyads such as master/mistress with servant and examiner with examinee respectively, using key parts of speech and key semantic fields that are statistically characteristic of these genres. They show that their method captures asymmetrical power relations that seem to be stereotypical of these particular genres at the time.

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

9 Sociopragmatics and Context

Apparently, the significance of historical sociopragmatic research in approaching sociopragmatic context lies in the fact that, on the one hand, it broadens its scope by relating it to genre and, on the other, it brings out the significance of language, and lexis in particular, in revealing aspects of context, thereby transgressing Leech’s definition of sociopragmatic context as providing “a link between micro, more linguistically-oriented considerations (the typical focus of pragmalinguistics), and the macro, more sociologically-oriented considerations” (Culpeper 2009: 181). Such considerations necessitate a closer look on the concept of genre in relation to context, and what possibilities genre affords for pragmatic analysis, especially as regards the two strands that have been associated with it, namely pragmalinguistics and sociopragmatics.

9.5 Genre as Context: The Constructional Perspective While, as already mentioned, the sociopragmatics/pragmalinguistics distinction was initially introduced to elaborate on Leech’s (1983) General Pragmatics program, a wider conception of pragmatics as a “general cognitive, social, and cultural perspective on linguistic phenomena in relation to their usage in forms of behaviour” (Verschueren 1999: 7), often referred to as a “broad view” of pragmatics (Culpeper 2021), makes the distinction between sociopragmatics and pragmalinguistics a matter of research focus (Marmaridou 2011; Archer 2017) rather than two distinct subfields of pragmatics. At the same time, this latter view of pragmatics seems to reflect the concept of context as a cognitive, social/sociocultural, and linguistic construct respectively. Linguistic context is functionally equivalent to co-text and constitutes the first overt and verifiable connection between speaker and addressee (Romero-Trillo and Maguire 2011). Social context, and sociocultural context in particular, includes extralinguistic aspects of communication such as the interlocutors, physical surroundings such as time and location, and institutional and noninstitutional domains, while cognitive context frames content and is the foundation on which inference and other forms of reasoning are based (Fetzer 2011: 35). Moreover, contextualization cues, such as intonation patterns, various discourse markers, or forms of code-switching (Verschueren 2021: 127) enable interlocutors to adapt their frame of mind to incoming information toward reshaping their cognitive expectations (Romero-Trillo 2007) so that context can in fact be characterized as flexible and dynamic. While such distinctions have long been maintained in pragmatic research, not enough attention has been paid to the fact that contextual parameters of communication are systematically encoded in language, constitute aspects of the speaker’s knowledge of language, and affect inferencing. As Kecskes (2012) points out, the use of particular linguistic expressions is equally important as sociopragmatic context in inferential communication. Embracing an encyclopedic, and hence sociocultural, view of the lexicon, he points out that focus on

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

213

214

SOPHIA MARMARIDOU

prior experience with language brings the cognitive component of context into focus. For example, if, upon an invitation for a drink, a speaker refuses by saying “Sorry, I can’t. My doctor won’t let me,” his interlocutor’s reply “What’s wrong with you?” implies an interest in the first speaker’s health (Kecskes 2012: 606). If, however, the response to the same invitation is “Sorry, I can’t. My mother-in-law won’t let me,” the same reply by the interlocutor implies a criticism of the speaker’s character, crucially depending on the sociocultural stereotype activated by the word “mother-in-law” in the particular situation, which is deeply conventionalized and cognitively salient. In fact, such observations cast doubt on the generality of inferencing. As Terkourafi (2021: 45) claims, sociopragmatics, especially concerned with discourse and physical context, as well as background knowledge, contributes to the study of inference and implicature by actually questioning their generality. Evidently, sociocultural context is not added to linguistic input to lead to appropriate inferences concerning the speech act performed. Rather, context is already linguistic at the onset of communication. This idea is specifically explored in Terkourafi (2009), who argues that there is strong interdependence between co-text and context in that linguistic cotext already “contains” extralinguistic context. The author takes issue with utterances from Cypriot Greek realizing offers or requests. She notes that while there is a general construction realizing requests, there are other lexically specific constructions that are preferred in specific contexts. For example, in transactional settings, where the speaker is asking the shop assistant for the availability of a particular product, (e.g., “is there . . .,” “do you have . . .”), the verb exo (have) in third-person singular (eçi) is preferred by working-class speakers, while second-person plural (eçete) is preferred by middle-class speakers. In interview settings, though, in addressing a request to the middleclass interviewer, the middle-class interviewee uses the construction θa iθela (I would like to + VP). In spite of the possibility of using the latter construction in transactional settings, too, frequency counts indicate a preponderance of each construction for the respective settings. Among the contextual parameters involved, the social class of speaker and addressee, their relationship, and the physical setting, comprise different parameters whose values identify a minimal context in each case, to the effect that different minimal contexts correlate positively with different linguistic constructions. Terkourafi claims that these parameters are fixed on the basis of background knowledge, or even perceptually, prior to the onset of communication, which then proceeds in parallel with the utterance. In this sense, a minimal context constitutes a convenient shorthand for an accumulated body of sociocultural knowledge against a background that on the one hand “colors” minimal context and, on the other is presupposed by it. So illocutionary force assignment proceeds on the basis of two types of contextual information. One pertaining to sociocultural categories of a minimal context available at speech onset and serving to set up expectations about the intended illocutionary force; and the other comprising paralinguistic and intonational cues while speech unfolds, and

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

9 Sociopragmatics and Context

serving to confirm or revise the projected illocutionary force. Apart from the obvious advantage of Terkourafi’s analysis to associate lexically specific constructions with a more schematic construction of realizing requests, it advances a motivated assumption about the boundedness of sociopragmatic context in utterance understanding, crucially reflected in linguistic structure. As Terkourafi notes, “A potentially all-inclusive notion of context is practically unmanageable and of little empirical usefulness” (p. 36), while a minimal context appears to tap on routine ways of interacting with the social environment. Apparently, the concept of minimal context refers to the conventionalization of a relationship holding between utterances and contexts in one’s experience on the basis of frequency of occurrence. Along the same lines, Moser and Panaretou (2011) examine the speech act of “ruling,” i.e., setting a rule and delineating the consequences of breaking this rule, in two different contexts, i.e., in law texts and in the everyday discourse of a mother to her children. In particular, the authors focus on the grammatical categories of tense, aspect, and modality and find that their use in law texts deviates from everyday use. Using questionnaires and interviews to elicit native speakers’ judgments on law texts and everyday uses, and subjecting their data to qualitative analysis, the authors find that, while certain uses are unanimously judged ungrammatical in everyday contexts, they go completely unnoticed in law texts. On the basis of such evidence, they point out that what licenses the observed “deviant” uses is the genre of law texts rather than the speech act performed therein. They argue that the genre of law, as a form of highly institutionalized meso-context embedded in the wider sociocultural context, sets up a cognitive frame which imposes interpretations to grammatical forms that would be unacceptable in other contexts. They conclude that genre is a bridging notion, connecting the macro-sociocultural context with the micro-context of laws. The above studies make a case for systematically correlating contextual specifications with linguistic form in particular genres. In very broad terms, the concept of genre emerges as a relation of language to context (Corbett 2006) and, as such, it is reminiscent of Leech’s (1983) “middle” level of analysis that Culpeper (2009) identified with genre. Admittedly, linguistic regularities play an important role in genres; what is more, genres have been viewed as cognitive constructs, “recognized, maintained and employed by members of a given discourse community” (Stukker et al. 2016: 9). In exploring the conventional language of Greek folk tales, Nikiforidou (2016) views genre as a frame “creating conventional expectations in the choice and interpretation of lexis and grammar with particular roles and structure in the frame indexed by particular linguistic choices” (p. 345). Similarly, in analyzing poetic discourse, Piata (2016) considers genre as a schema, i.e., as “a mental representation shared among language users and guiding language production and comprehension” (p. 225). The issue of the correlation of linguistic form and meaning with contextual parameters of a speech event or genre has been systematically taken up within the framework of usage-based Construction Grammar. In this

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

215

216

SOPHIA MARMARIDOU

approach, language, and knowledge of language, consists of a structured inventory of constructions, i.e., form–meaning pairings, at various levels of schematicity, and encompassing all linguistic levels, from morphemes through the lexicon (including idiomatic characterizations) to the various patterns of syntactic structure associated with corresponding semantic, pragmatic, and discourse patterns. In this framework the construction, the basic unit of linguistic communication, is also the basic mental unit for genre information and is part of a speaker’s knowledge space. More specifically, constructional approaches to genre model genre knowledge in terms of genre-based constructions, whose pragmatic specifications include their association with a particular sociocultural context (Nikiforidou 2018: 543). While this model seems well suited to the investigation of conventional aspects of discourse, it is relatively recently that specific patterns have been associated with particular genres, viewed as institutionalized contexts and definable in sociocultural terms. Perhaps one of the earliest attempts in this direction is Östman (2005), who claims that much of discourse is conventionalized and that acceptability and conventionality are relative to context (p. 126). Adopting a frame-semantics perspective, he observes that there are discourse-level frames that can be viewed as genres, the latter indicating how we know that what we say is appropriate in a particular setting.1 These frames for understanding are what he calls discourse patterns. Notably, he shows that recipes, for example, constitute such a discourse pattern, including the ingredients section, followed by the cooking instructions, in a specific outline, or graphic display, that is part of this discourse-level categorization. Newspaper headlines are another case in point, characterized by determiner omission, which is not otherwise acceptable. For example, “Mother drowns baby” is a pattern licensed by the particular genre, rather than any of the “core” constructions like the Determination construction, the Subject-Predicate construction and the Verb Phrase construction, which together would license constructs like “A mother drowned a baby.” However, there are restrictions as to the application of the headlines template. For example, “Emelda likes vase” is not licensed in this context. Apparently, Östman (2005: 123) observes, in headlines only a human agent as subject and an animate patient as object are acceptable without articles or other determiners.2 In their constructional account of genre-based argument omissions, Ruppenhofer and Michaelis (2010) identify the dimensions of genre that are conventionally associated with the use of null arguments, as in product label statements, e.g., “Ø Contains alcohol,” in which the null argument is associated with the subject role.3 They argue that such omissions are constructional 1

2

3

In fact, for Östman (2005: 132), the notion of discourse pattern is a filter mediating between genre and text-type descriptions, i.e., discourse patterns are conventionalized associations between text type and genre. In fact, the animate object may be recovered through a part-for-whole metonymy, as in “Sewage plant worker finds arm in plant” (from Ruppenhofer and Michaelis 2010: 166). Other instances of genre-based argument omission discussed by the authors include recipes, as in “Blend all the ingredients in an electric blender. Serve Ø cold,” or diary style, as in “Ø read Michelet; Ø wrote to Desmond . . . ; Ø played gramophone . . .,” among others.

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

9 Sociopragmatics and Context

relative to a genre, rather than predicational/lexical (as in “I understand Ø” [what you’re saying]), or as a result of a syntactic role, such as by-phrase agents, as in “The arrow was found Ø near a tree.” In this analysis, genres enable omissions that the predicator or construction would not otherwise permit, as in “*I avoid cocktails because Ø contain alcohol.” Moreover, genresensitive argument omissions are constrained by particular specifications. For example, anaphoricity is possible only under specific restrictions, as when the relevant genre presupposes the salience of certain entities. This is the case with labelese (product labels), which, as an indexical genre, features inherently selfreferential predications. Omission is also possible in diaries because it is possible to track the “missing” referent, as in “Ø read Michelet; Ø wrote to Desmond . . . ; Ø played gramophone . . . .” The same holds for recipes, as in “Blend all the ingredients . . . Ø Serve cold.” While genre-based omissions are not obligatory, in the sense that non-omission of arguments does not lead to ungrammaticality, they nevertheless invoke the particular text type. In this sense genres are not independent of the omission involved. Rather, they are, at least partly, constituted by the practice of omission, along with other lexicogrammatical choices. Interestingly, genre-based argument omissions are often associated with what is canonical in the genre. For example, in labelese, argument omission is allowed only for statements describing “the provenance, constitution, qualities or efficacy of the product” (Ruppenhofer and Michaelis 2010: 169). So, while “Ø contains no hydrogenated oils” or “Ø creates visibly fuller, thicker hair” are acceptable, “Ø has flourished in cultivation for over 5,000 years” on a product box is not. Apparently, what is canonical in each case is sensitive to socioculturally determined practices, or sociopragmatic context. In a study aiming to establish the conventional links of context to grammatical form, Nikiforidou (2021) focuses on the well-established and clearly delimited genre of stage directions, as, for example “Enter Violet” or “Exit Marianne,” taking the standard form [Enter – NPsubject]. The expressions licensed by this pattern “function as genre markers in the sense that their presence uniquely and unequivocally evokes a theater play setting” (p. 191). Although such expressions are formulaic, they are not totally fixed. Rather, they exhibit some flexibility in the verb slot, as well as the subject. For example, while “enter” can be used for subjects in the singular or plural (“Enter Camilla . . . and King Charles III”), “exit” is typically used for a singular subject. Following the Latin verb paradigm, plural post-posed subjects require the third-person plural verb suffix, i.e., “exeunt,” as in “Exeunt Newman and Mme. De Cintre, right.” While lexical variability for “enter” is limited to the prefixation with “re-,” as in “Re-enter Alan,” the subject slot can be occupied by all kinds of NPs ranging from proper names to heavily modified nouns, as in “Enter Lear in a chair carried by Servants.” However, the post-posed subject construction cannot feature pronouns in the subject position. This accords with the function of the construction, which is to provide hearer(reader)-new focal information, i.e., which characters are to appear in the next scene. While this construction can be described as semi-substantive, it does relate formally

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

217

218

SOPHIA MARMARIDOU

and functionally to other constructions in the language, such as Lakoff’s (1987) central deictic (e.g., “There goes Harry with his red sneakers on”), which draws on the “pointing-out” idealized cognitive model, whereby the speaker aims to focus the hearer’s attention on the referent of the post-posed NP, which is in her visual field. As with the post-posed subject construction of stage directions, the deictic-there construction is incompatible with pronominal subjects for the same functional reason, i.e., both constructions draw the hearer’/reader’s attention to an individual about to appear on the scene of events. Such observations point to the functional motivation of the post-posed subject construction and its embeddedness in the rest of the grammar. Apart from formal and functional motivations for the [Enter – NPsubject] construction, there is a set of semantically characterizable words which appear frequently and consistently in all plays, including modifiers of speech (e.g., narrative tone, with senile quaver), modifiers of expression or posture (e.g., smiling, distraught), and modifiers of actions (e.g., urgently, dramatically). The use of such lexis in stage directions is evidently motivated by genre requirements to the extent that these expressions, conventionally interpreted as instructions on speech and expression/posture, function as conventional genre markers, uniquely evoking the specific context.4 In focusing further on sentence-level constructions evoking particular discourse contexts, Antonopoulou et al. (2015: 303) show that there is a conventional link between specific forms and specific contexts in which these forms appear, to the extent that they become bearers of those discoursal properties, through frequent use and consequent entrenchment. They take issue with a particular instance of interaction, namely couple talk and make reference to the expression we need to talk as a phrase introducing the speaker’s proposal to his/her partner to break up their relationship. Notably, this is a truncated form, i.e., without the prepositional complements “to NP” or “about/of NP,” which qualifies as a separate construction characterized by a pragmatic point (Fillmore et al. 1988). In support of this analysis, the authors observe that the implicature associated with the construction is not readily available to propositionally equivalent expressions, e.g., we must talk; moreover, its form is fully fixed, in that it consists of the first-person plural pronoun subject, the modal need in present tense (only) and the infinitival complement to talk. To test the discourse-specific conventionality of the construction, and in the absence of discourse-specific corpora, the authors resort to the online Urban Dictionary (2011), where definitions are contributed by the users, posted, and approved or not with votes by other users whose number is recorded. The answers for the expression we need to talk are suggestive of a particular contextual meaning, while the fact that it is included as an expression is itself suggestive of its constructional, unit-like status. The above formal, lexical, and pragmatic specifications of the construction, along with speaker 4

As the author observes, regardless of whether the basic meaning of such a term refers more to speech or to expression/ posture, in practice, it is sometimes impossible to distinguish between the two (Nikiforidou 2021: 196).

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

9 Sociopragmatics and Context

judgments and dictionary evidence, point to its status as a substantive idiom in the language which functions as a marker of the particular type of interaction, i.e., couple talk. So far, it has been pointed out that genre, as a complex knowledge schema that individual language users have at their disposal to engage in discourse, and as the “middle” level on which sociopragmatic context resides, is conventionally associated with particular constructions to the extent that such constructions may in fact function as markers for the particular discourse. However, as has been pointed out by several scholars (e.g., Antonopoulou and Nikiforidou 2011; Diewald 2015; Fischer 2015), not all contextual information (e.g., as in genres) should be directly associated with specific constructions. A criterion for the systematicity of associating discourse-sensitive contextual information with a particular construction is transferability, which indicates the degree of entrenchment and conventionalization of a linguistic form for a particular discourse setting. As Antonopoulou et al. (2015) have shown, transferability is often associated with a humorous effect in discourse. Examples are provided from the humorous exploitation of classroom discourse, the discoursal frame of the Alcoholics Anonymous (AA), the scholarly edition, horoscopes, and couple talk. The authors systematically show how conventional discoursal patterns coming from a specific genre are exploited in a different one, so as to create a humorous effect. The humorous, or other, exploitation of a construction associated with particular discourse contexts has also become apparent in Greek TV shows (Marmaridou, forthcoming). What is perhaps significant in this case is that the construction humorously transfers to other TV programs, develops formal variation, and is consistently spreading in other discourses. More specifically, in the last few years, TV shows in Greece feature contests in various domains of experience, such as fashion and styling, modeling, cooking, dancing, singing, and performing, among others. The structure of these contests is very similar. There is a committee of three to four judges who, after signaling out the final list of contestants, require of them to perform several tasks. As a result of their performance in each one of them, the number of contestants weekly diminishes, until the last three or four enter the final phase that will provide the winner. Notably, the judges’ verdict on each contestant’s performance is often expressed by a simple affirmative or negative expression of the form apo mena (ine) ne/oçi (“from me (is) yes/no” – “from me (it is) yes/no”).5 Expectedly, search in two Greek language corpora, also representing oral discourse, does not yield this expression. Direct observation and search on YouTube have shown that this pattern is functionally motivated and has constructional status, in that it qualifies as a substantive idiom, inheriting formal and semantic properties from more general constructions in the language, while also exhibiting its distinctive properties, crucially including its systematic association with the “TV contest show” discourse, as already mentioned. 5

This pattern was initially observed in the 2019 show Greece’s Next Top Model and was subsequently taken up by other contests of a similar orientation.

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

219

220

SOPHIA MARMARIDOU

As with the humorous exploitation of other discourse-specific constructions, the occurrence of this construction in different contexts, albeit featuring formal or lexical variation, evokes the popular TV contest shows genre in a humorous, if not critical, manner. For example, in announcing recent news of the development of a Russian vaccine, the host of a humorous TV show exclaimed “apo mas ine da” (from us is yes), “from us it is yes (in Russian)” to show approval of this achievement. In this case, the variation not only affects the personal pronoun, which now features in the plural, but also the language in which the affirmative is glossed to metonymically stand for the origin of the vaccine. The transferability of this idiom in discourse is further exemplified by its humorous exploitation in an email thread between colleagues making arrangements for the dates of their leave of absence from the office during the summer vacation. When a final plan is reached with each member’s preferences, the Director of Studies indicates agreement by “apo mena ine ne” (from me is yes).6 Notably, the agreement of two further members of staff is expressed by adopting the same idiom, enriched by typical phrases enacting corresponding practices: (in translation) “And from me yes. Congratulations, you pass to the next phase,” and “And from me yes, but I press the golden buzzer, too, and we all go straight to the finals!” respectively. Significantly, the latter two utterances are followed by smiley and winking emoticons underlining the humorous effect. Clearly, the perlocutionary effect of this construction, i.e., the contestants’ remaining – or not – in the contest, does not apply to its humorous uses. What is significant for the purposes of the present chapter is that the sociopragmatic parameters of a communicative event, such as a TV contest show, are systematically associated with a particular construction, so that the construction evokes the particular discourse context.

9.6 Concluding Remarks As already mentioned, in their joint attempt to constrain the field of pragmatics, Leech (1983) and Thomas (1983) introduced the concept of sociopragmatics (as distinct from pragmalinguistics) to refer to more specific, “local” conditions on language use, which were subsequently associated with corresponding parameters of the interactional context of communicative acts. These typically refer to the relative status of interlocutors, institutional roles, and the (in) formality of the situation, within perceptions of a cultural norm. The linguistic means available to interlocutors to respond to sociopragmatic requirements were the object of pragmalinguistic study. In subsequent construals of context as linguistic (co-text), social, and cognitive (commonly frame- or schema-based and inference-creating), sociopragmatics was essentially viewed as a set of 6

I gratefully acknowledge the purposeful initiation of this thread to V. Geka, as well as the recording of the responses of the unsuspecting subjects.

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

9 Sociopragmatics and Context

contextual features readily associated with specific discourse patterns, activity types or genres. Especially in the framework of Construction Grammar, where contextual specifications are routinely incorporated in the description of grammatical constructions, genres are characterized by specific constructional reflexes of sociopragmatic context which are available to speakers of the language as knowledge schemas, i.e., as part of a speaker’s knowledge of the language, or a speaker’s repertoire of genres. In this analysis, the distinction between sociopragmatics and pragmalinguistics is less pronounced than it initially was, while sociopragmatic context is often shown to be systematically encoded in particular discourse constructions. Obviously, this is not to deny the dynamicity of communication. In fact, genres themselves, initially considered static and well entrenched, have been shown themselves to be dynamic, changing, remediated, overlapping, or fluid, as, for example, is the case with instances of computer-mediated communication (Herring et al. 2013: 10). However, there is ample evidence that genres systematically feature conventional and highly entrenched constructions that function as markers for the particular genres, are recognizable as such by speakers of the language, and are hence exploitable for humorous or other effects. In this context, a challenge facing future research on genre constructions, especially within a usage-based model, such as Construction Grammar, is their identification and frequency of occurrence within particular genres. Evidently, this calls for the compilation of specialized corpora that will provide clearer insights into the actual usage of such constructions, which might otherwise go unnoticed in general corpora. To the extent that genres comprise Leech’s local conditions of language use and incorporate aspects of sociopragmatic context as reflected in particular constructions, work toward discourse specific corpora seems well justified.

References Antonopoulou, E., and Nikiforidou, K. (2011). Construction grammar and conventional discourse: A construction-based approach to discoursal incongruity. Journal of Pragmatics, 43(10), 2594–2609. Antonopoulou, E., Nikiforidou, K., and Tsakona, V. (2015). Construction grammar and discoursal incongruity. In G. Brône, K. Feyaerts and T. Veale, eds., Cognitive Linguistics Meets Humor Research: Current trends and new developments. Berlin/ New York: Mouton de Gruyter, pp. 289–314. Archer, D. (2017). Context and historical (socio-)pragmatics 20 years on. Journal of Historical Pragmatics, 18(2), 315–336. Archer, D., and Culpeper, J. (2009). Identifying key sociophilological usage in plays and trial proceedings (1640–1760): An empirical approach via corpus annotation. Journal of Historical Pragmatics, 10(2), 286–309. Austin, J. L. (1962). How to Do Things with Words. Oxford: Clarendon Press. Chang, Y.-F. (2011). Interlanguage pragmatic development: The relation between pragmalinguistic competence and sociopragmatic competence. Language Sciences, 33(5), 786–798.

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

221

222

SOPHIA MARMARIDOU

Chang, W.-L. M., and Haugh, M. (2017). Intercultural communicative competence and emotion among second language learners of Chinese. In I. Kecskes and C. Sun (eds.), Key Issues in Chinese as a Second Language Research (pp. 267–286). New York: Routledge. Citron, F., and Goldberg, A. (2014). Metaphorical sentences are more emotionally engaging than their literal counterparts. Journal of Cognitive Neuroscience, 26(11), 2585–2595. Corbett, J. (2006). Genre and genre analysis. In K. Brown (ed.), Encyclopedia of Language and Linguistics (pp. 26–32). Boston: Elsevier. Devlin, A. M. (2019). The interaction between duration of study abroad, diversity of loci of learning and sociopragmatic variation patterns: A comparative study. Journal of Pragmatics, 146, 121–136. Diewald, G. (2015) Modal particles in different communicative types. Constructions and Frames, 7(2), 218–257. Fetzer, A. (2011). Pragmatics as a linguistic concept. In W. Bublitz and N. R. Norrick (eds.), Foundations of Pragmatics (pp. 23–50). Berlin: De Gruyter Mouton. Fillmore, Ch., Kay, P., and O’Connor, M. C. (1988). Regularity and idiomaticity in grammatical constructions: The case of let alone. Language, 64(3), 501–538. Fischer, K. (2015). Situation in grammar or in frames? Evidence from the so-called baby talk register. Constructions and Frames, 7(2), 258–288. Fitzmaurice, S. M. (2009). The sociopragmatics of a lovers’ spat: The case of the eighteenth-century courtship letters of Mary Pierrepont and Edward Wortley. Journal of Historical Pragmatics, 10(2), 215–237. Fried, M., and Östman, J.-O. (2004). Construction Grammar: A thumbnail sketch. In Μ. Fried and J.-O. Östman (eds.), Construction Grammar in a Cross-Language Perspective (pp. 11–86). Amsterdam/Philadelphia: John Benjamins. Grice, P. (1975). Logic and conversation. In P. Cole and J. Morgan (eds.), Syntax and Semantics, Vol. IX: Pragmatics (pp. 113–128). New York: Academic Press. Herring, S. C., Stein, D., and Virtanen, T. (2013). Introduction to the pragmatics of computer-mediated communication. In S. C. Herring, D. Stein, and T. Virtanen (eds.), Handbook of Pragmatics of Computer-Mediated Communication (pp. 3–31). Berlin: Mouton. Ifantidou, E. (2019). Relevance and metaphor understanding in a second language. In K. Scott, B. Clark, and R. Carston (eds.), Relevance: Pragmatics and Interpretation (pp. 218–30). Cambridge: Cambridge University Press. Jacobs, A., and Jucker, A. H. (1995). The historical perspective in pragmatics. In A. H. Jucker (ed.), Historical Pragmatics: Pragmatic Developments in the History of English (pp. 3–33). Amsterdam: John Benjamins. Jucker, A. H. (2006). Historical pragmatics. In Östman, J.-O. and J. Verschueren (eds.), Handbook of Pragmatics 2006 (pp. 1–14). Amsterdam: John Benjamins. Kasper, G., and Rose, K. (2002). Pragmatic Development in a Second Language. Oxford: Blackwell. Kecskes, I. (2012). Sociopragmatics and cross-cultural and intercultural studies. In K. Allan and K. M. Jaszczolt (eds.), The Cambridge Handbook of Pragmatics (pp. 599–616). Cambridge: Cambridge University Press. Lakoff, G. (1987). Women, Fire and Dangerous Things: What Categories Reveal about the Mind (pp. 462–585). Chicago/London: University of Chicago Press. Leech, G. N. (1983). Principles of Pragmatics. London: Longman.

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

9 Sociopragmatics and Context

Levinson, S. N. (1983). Pragmatics. Cambridge: Cambridge University Press. Levinson, S. N. (1992). Activity types and language. In Drew P. and J. Heritage (eds.), Talk at Work (pp. 66–100). The Hague: Mouton. Marmaridou, S. (2000). Pragmatic Meaning and Cognition. Amsterdam/Philadelphia: John Benjamins. Marmaridou, S. (2011). Pragmalinguistics and sociopragmatics. In W. Bublitz, and N. R. Norrick (eds.), Foundations of Pragmatics (pp. 77–106). Berlin: De Gruyter Mouton. Marmaridou, S. (forthcoming). Re-cycling discourse constructions: The case of [apo NPAcc. (ine3rd p. sing.) ne/oçi] in Greek TV shows. McNamara, T., and Roever, C. (2006). The social dimension of proficiency: How testable is it? Language Learning, 56(2): 43–79. Morris, Ch. W. (1938). Foundations of the Theory of Signs. In N. O. R. Carnap and C. W. Morris (eds.), International Encyclopaedia of Unified Science (Vol. I, pp. 77–138). Chicago: University of Chicago Press. Moser, A., and Panaretou, E. (2011). Why a mother’s rule is not a law: The role of context in the interpretation of Greek laws. In A. Fetzer and E. Oishi (eds.), Context and Contexts (pp. 11–40). Amsterdam/Philadelphia: John Benjamins. Nikiforidou, K. (2016). “Genre knowledge” in a constructional framework: Lexis, grammar and perspective in folk tales. In N. Stukker, W. Spooren, and G. Steen (eds.), Genre in Language, Discourse and Cognition (pp. 331–360). Berlin: Walter de Gruyter. Nikiforidou, K. (2018). Genre and constructional analysis. Pragmatics and Cognition, 25(3), 543–575. Nikiforidou, K. (2021). Grammatical variability and the grammar of genre: Constructions, conventionality, and motivation in “stage directions.” Special issue of Journal of Pragmatics (eds. Y. Matsumoto and S. Iwasaki), Multiplicity in Grammar: Modes, Genres and Speaker’s Knowledge, 173, 189–199. Östman, J.-O. (2005). Construction discourse: A prolegomenon. In J.-O. Östman and M. Fried (eds.), Construction Grammars: Cognitive Grounding and Theoretical Extensions (pp. 121–144). Amsterdam/Philadelphia: John Benjamins. Piata, A. (2016). Genre “out of the box”: A conceptual integration analysis of poetic discourse. In N. Stukker, W. Spooren, and G. Steen (eds.), Genre in Language, Discourse and Cognition (pp. 225–250). Berlin: Walter de Gruyter. Romero-Trillo, J. (2007). Adaptive Management in discourse: The case of involvement discourse markers in English and Spanish conversations. Catalan Journal of Linguistics, 6(1), 81–94. Romero-Trillo, J., and Maguire, L. (2011). Adaptive context: The fourth element of meaning. International Review of Pragmatics, 3(2), 228–241. Rose, K. R. (2000). An exploratory cross-sectional study of interlanguage pragmatic development. Studies in Second Language Acquisition 22(1), 27–67. Ruppenhofer, J., and Michaelis, L. (2010). A constructional account of genre-based argument omissions. Constructions and Frames, 2(2), 158–184. Searle, J. R. (1975). Indirect speech acts. In P. Cole, P. and J. Morgan (eds.), Syntax and Semantics, Vol. III: Speech Acts (pp. 59–82). New York: Academic Press. Stukker, N., Spooren, W., and Steen, G. (2016). Introduction. In N. Stukker, W. Spooren, and G. Steen (eds.), Genre in Language, Discourse and Cognition (pp. 1–14). Berlin: Walter de Gruyter.

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

223

224

SOPHIA MARMARIDOU

Terkourafi, M. (2009). On de-limiting context. In A. Bergs and G. Diewald (eds.), Contexts and Constructions (pp. 17–42). Amsterdam/Philadelphia: John Benjamins. Terkourafi, M. (2021). Inference and Implicature. In M. Haugh, D. Z. Kadar, and M. Terkourafi (eds.), The Cambridge Handbook of Sociopragmatics (pp. 30–47). Cambridge: Cambridge University Press. Thomas, J. A. (1983). Cross-cultural pragmatic failure. Applied Linguistics, 4(2), 91–112. Verschueren, J. (1999). Understanding Pragmatics. London: Hodder Arnold. Verschueren, J. (2021). Reflexivity and meta-awareness. In M. Haugh, D. Z. Kadar, and M. Terkourafi (eds.), The Cambridge Handbook of Sociopragmatics (pp. 117–139). Cambridge: Cambridge University Press.

https://doi.org/10.1017/9781108989275.010 Published online by Cambridge University Press

10 Natural Semantic Metalanguage and Context Anna Gladkova 10.1 Introduction Natural language is a human-specific symbolic system for expressing meaning. Meaning is embodied at the level of each linguistic element, as well as through combinations of elements. When meaning is considered in combination with other meanings, we name it context. The Natural Semantic Metalanguage (NSM) is an approach to studying linguistic meaning in isolation, as well as in context. Context can be considered broadly here and includes the syntagmatic and paradigmatic relations of a word, as well as the social and cultural contexts it functions in. Therefore, when it comes to interpreting linguistic meaning in context, NSM is a unique semantic approach which can analyze and describe meaning and its functions in various types of contexts.1 Anna Wierzbicka, the founder of the NSM approach, famously declared the centrality of meaning to be a linguistic function and emphasized that the study of language should be primarily conducted through the prism of meaning. Wierzbicka (1996: 3) wrote: To study language without reference to meaning is like studying road signs from the point of view of their physical properties (how much they weigh, what kind of paint are they painted with, and so on), or like studying the structure of the eye without any reference to seeing.

Wierzbicka started working on identifying language universals in the 1960s, and this research agenda led her to developing a versatile system of studying meaning. The chapter is structured as follows. Section 10.2 describes theoretical foundations of the NSM approach and its main concepts, such as semantic primitives, universal syntax, and semantic molecules. Section 10.3 demonstrates how NSM can be applied to studying “formal” linguistic context, in particular through studying lexical meaning and conceptual inter-connections I am grateful to Anna Wierzbicka and Cliff Goddard for their comments on this chapter. All errors remain mine. Other approaches which pay particular attention to meaning in context are the Moscow School of Semantics (Apresjan 1992) and Meaning-Text Theory (Mel’č uk 2016). The NSM framework shares many theoretical views and assumptions with these two approaches.

1

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

226

ANNA GLADKOVA

and constructing semantic explications. Section 10.4 focuses on the aspects of cultural context in interpreting meaning, in particular the notions of cultural keywords, cultural influences in conceptualizations, and cultural scripts. Section 10.5 concludes with the description of recent NSM development, such as Minimal Language and its applications.

10.2 NSM: Basic Concepts 10.2.1 Semantic Primitives The NSM approach to studying meaning in natural language originated as a research program by Anna Wierzbicka in the 1960s. She was later joined by Cliff Goddard as well as other Australian and international scholars (Wierzbicka 2021a). The NSM approach relies on the hypothesis put forward by Leibniz (1956, [1678] 1987) that a set of concepts should exist that cannot be reduced further and need to be accepted as undefinable. Wierzbicka’s idea is that such concepts should also be present in every language. Therefore, a key unit of the NSM approach is a semantic primitive (or a prime) – a lexically encoded meaning which can be realized at the level of a word, or a bound lexeme, and cannot be defined further. Wierzbicka’s initial list of primes included fourteen words (1972). In the last decade, it has been proposed that the search for semantic primitives has reached its goal and is set at a total of sixty-five (see Table 10.1) (Goddard and Wierzbicka 2014; Goddard 2018a). The primitives have been identified by trial and error in the process of designing definitions of several lexical domains, primarily emotion terms, speech acts, and concrete vocabulary. The universality of NSM has been tested on a significant number (about three dozen) of related and unrelated languages (Goddard and Wierzbicka 1994, 2002; Peeters 2006; see Gladkova and Larina 2018 and the NSM Homepage for an overview). The primes include several thematic groups: substantives (e.g., I and YOU), evaluators (GOOD and BAD), mental predicates (KNOW, THINK, WANT, DON’T WANT, FEEL, SEE, HEAR), speech (SAY, WORDS, TRUE), actions, events, movement (DO, HAPPEN, MOVE), time (TIME~WHEN, NOW, BEFORE, AFTER, A LONG TIME, A SHORT TIME, FOR SOME TIME, MOMENT), space (PLACE~WHERE, HERE, ABOVE, BELOW, FAR, NEAR, SIDE, INSIDE, TOUCH), among others. The primes are anthropocentric and reflect essential qualities of human cognition and sociality. The exponents of primes can be realized at different levels across languages – words, bound morphemes, or phrasemes. Also, the same universal meaning can sometimes have several representations in a given language which are called allolexes. For example, in English other and else are allolexes of the prime OTHER, I and me are allolexes of the prime I. In some cases, a single word called portmanteau expresses a combination of primes – e.g., tak in Russian for LIKE and THIS.

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

10 Natural Semantic Metalanguage and Context

Table 10.1 Semantic primes, English exponents (after Goddard and Wierzbicka 2014) I, YOU, SOMEONE, SOMETHING~THING, PEOPLE, BODY

substantives

KINDS, HAVE PARTS

relational substantives

THIS, THE SAME, OTHER~ELSE

determiners

ONE, TWO, SOME, ALL, MUCH~MANY, LITTLE~FEW

quantifiers

GOOD, BAD

evaluators

BIG, SMALL

descriptors

KNOW, THINK, WANT, DON’T WANT, FEEL, SEE, HEAR

mental predicates

SAY, WORDS, TRUE

speech

DO, HAPPEN, MOVE

actions, events, movement

BE (SOMEWHERE), THERE IS, BE (SOMEONE/SOMETHING)

location, existence, specification

(IS) MINE

possession

LIVE, DIE

life and death

TIME~WHEN, NOW, BEFORE, AFTER, A LONG TIME, A SHORT TIME, FOR SOME TIME, MOMENT

time

PLACE~WHERE, HERE, ABOVE, BELOW, FAR, NEAR, SIDE, INSIDE, TOUCH

place

NOT, MAYBE, CAN, BECAUSE, IF

logical concepts

VERY, MORE

augmentor, intensifier

LIKE

similarity

Notes: • Exponents of primes can be polysemous, i.e., they can have other, additional meanings • Exponents of primes may be words, bound morphemes, or phrasemes • They can be formally complex • They can have language-specific combinatorial variants (allolexes, indicated with ~) • Each prime has well-specified syntactic (combinatorial) properties.

10.2.2 NSM Universal Syntax Empirical research has identified that certain syntactic properties of semantic universals are also universal. Such combinations are called canonical contexts. For example, the prime FEEL is used in the following combinations: someone feels something someone feels something good/bad someone feels something in a part of the body

At the same time, not all uses of a word with the universal meaning are shown to be universal. For example, the use to feel good/bad is not shown to be a universal use of the prime FEEL. This is so because in some languages (e.g., Russian), this meaning would be realized by a reflexive form – č uvstvovat’ sebja xorošo ‘feel oneself good.’ The use of the reflexive form brings in additional meaning to this syntactic use of to feel and excludes it from the list of universal contexts.

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

227

228

ANNA GLADKOVA

10.2.3 Semantic Molecules Empirical research conducted over time has indicated that certain concepts cannot be decomposed via semantic primitives only and require the use of more complex terms. In NSM, they are called semantic molecules (Goddard 2016). Semantic molecules are universal or near-universal meanings decomposable in terms of semantic primes (and sometimes other molecules) which are required in explications of certain concepts. Semantic molecules are often salient in demonstrating conceptual dependence. Wierzbicka (2015: 5–6) illustrates such a chain of dependency with the explications of legs, long, and hands: the explication of legs contains the molecule long, which in turn is explicated via hands, with the latter being explicated via semantic primes only: Explication [A] legs (someone’s legs) a) b) c) d) e)

two parts of someone’s body these two parts are below all the other parts of the body these two parts are long (m) these two parts of someone’s body can move as this someone wants because people’s bodies have these two parts, people can move in many places as they want

Explication [B] something long (e.g., a cucumber, a tail, a stick) a) when someone sees this thing, this someone can think about it like this: b) “two parts of this thing are not like any other parts because one of these two parts is very far from the other” c) if someone’s hands (m) touch this thing everywhere on all sides, this someone can think about it in the same way

Explication [C] hands (someone’s hands) a) b) c) d) e) f) g)

two parts of someone’s body they are parts of two other parts of two sides of this someone’s body these parts of someone’s body can move as this someone wants these parts of someone’s body have many parts these parts can move in many ways as this someone wants because people’s bodies have these parts, people can do many things with many things as they want because people’s bodies have these parts, people can touch many things as they want

Some molecules are hypothesized to be universal, while others are not. Goddard (2021) distinguishes universal, approximate, and culture-specific semantic molecules. Universal molecules include fundamental concepts from the immediate human environment found across languages and themselves constitute frequent building conceptual blocks e.g., hands, eyes, sun, sky, fire, water, children, men, women, long, round. Approximate molecules do not have full conceptual equivalence across languages. For example, “river” is a type of water flow found in different parts of the globe. “Rivers” differ depending on the environment, climate conditions, and terrain, and consequently, their conceptualization across languages have been shown not to be universal (Bromhead 2018). Lastly, there are

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

10 Natural Semantic Metalanguage and Context

culture-specific molecules, which are significantly culture- and society-dependent (e.g., number, money, doctor, priest). Semantic molecules and semantic primes are salient in semantic conceptual construction, yet they vary in their linguistic status and “flavor.” Semantic primes can have different lexical status – words, bound morphemes, or phrasemes. Semantic molecules are more commonly separate lexemes. Simultaneously, their semantic “flavor” is richer and has broader semantic and cultural associations than semantic primes have.

10.3 NSM in the Analysis of Linguistic Meaning at a Conceptual Level 10.3.1 How to Construct a Semantic Explication The result of semantic analysis of a word is recorded in the form of a semantic explication. A semantic explication is a semantic formula written in semantic primes (or semantic primes and semantic molecules if required) which represents the semantic content of a lexeme. Several issues are at stake here: the issue of polysemy, conditions of creating a good semantic explication, and the issue of a semantic template (Goddard 2011). Goddard (2020: 96) identifies the criteria for a good semantic explication as follows: (a) it is phrased entirely in words and grammar of the NSM metalanguage; (b) it is coherent, i.e., makes sense as a whole; (c) it is compatible with the range of use of the word being explicated and with its relations with other words, entailments, frequent collocations, and so on; and (d) it satisfies native speaker intuitions about interpretation in context. The issue of identifying polysemy is also essential to the NSM-based semantic analysis. Explications worded in semantic primitives serve as a reliable test in identifying and confirming polysemy of a word. A word can be classified as having a single general meaning if its range of use can be covered by a single explication. On the contrary, a word can be classified polysemous (as having two or more meanings) if two or more related explications need to be proposed to cover its uses (Goddard 2000). Both issues of proposing and constructing an NSM explication and distinguishing a polysemy of a word will be demonstrated using the example of analysis of the word virus by Goddard and Wierzbicka (2021). Goddard and Wierzbicka (2021) explore the semantics of virus- and disease-related terminology which has become particularly prominent in the time of coronavirus pandemic. They posit the polysemy of the word with a core meaning and a recently emerging meaning due to the COVID-19 pandemic. Goddard and Wierzbicka emphasize that their definition represents a naïve worldview which does not fully coincide with the scientific understanding of the concept. The explication is constructed based on the linguistic evidence.

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

229

230

ANNA GLADKOVA

Virus1 is used in contexts like the following (Goddard and Wierzbicka 2021: 13): (1a) (1b) (1c) (1d)

I must’ve picked up a virus. Colds are caused by a virus. Howard is out of . . . [the] team after going down with a virus. It’s only a matter of time before such a virus adapts itself to spread more easily from person to person and causes a severe worldwide outbreak.

The core meaning is represented as follows (Goddard and Wierzbicka 2021: 4): Explication [D] virus1 (e.g., a virus) a) something of one kind, there are many kinds of things of this kind b) when there is something of this kind somewhere, there are many very very small things of this kind there, people can’t see them c) there can be many such things in people’s bodies at some times d) when there are many of these very very small things in someone’s body, a very short time after there can be many of them in the bodies of other people in the same place e) when they are in someone’s body, something bad can happen in this someone’s body because of it f) people can think about things of this kind like this: “these things do very bad things to people’s bodies”

The explication contains four main conceptual sections. Section (a) introduces virus1 as “something of one kind” with many subkinds, therefore underlying that there can be different viruses. Part (b) explains that viruses are perceived to contain small parts which are not seen by people. Consequently, it is known that often when a virus spreads, small particles are dispersed in the air. Components (c)–(e) are closely related and explain that these particles can get into a person’s body and they can spread to other people and cause harm. The final section (f) in the explication is anthropocentric and explains the effect of a virus on a human body. Goddard and Wierzbicka also argue that a COVID-19-specific use of the term virus has emerged in the situation of the global coronavirus pandemic. They identify this meaning as virus2 and argue that it refers to COVID-19 virus. The authors observe that this meaning tends to be used with the definite article as in the following examples (p. 10): (2a) Even before the virus, the economy was barely limping along. (2b) And now for the latest news on the virus.

The explication capturing this meaning is the following (p. 10): Explication [E] virus2 (e.g., “even before the virus, . . .”) a) something very bad is happening in many countries on earth now, very bad things are happening to many people this is happening because there is virus-1 of one kind (“coronavirus”) in these people’s bodies b) when there is virus-1 of this kind in the bodies of some people in a place, a short time after, this virus-1 can be in the bodies of many other people in this place c) because of this, after some time this virus-1 can be in the bodies of many people in many places

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

10 Natural Semantic Metalanguage and Context

The explication refers to the first meaning virus1 which is used here as a derivational base. In this explication, part (a) describes the situation of the global coronavirus pandemic, part (b) spells out the characteristic of the coronavirus when it can spread easily from person to person, part (c) explains the results of the spread of the virus and its effect on many people. Goddard and Wierzbicka hypothesize that with the decline of the pandemic this meaning is likely to fade out of use. The proposed explications satisfy the criteria of an NSM semantic explication spelled out by Goddard above; they are worded in the NSM metalanguage, they are coherent, satisfy a native speaker intuition, and reflect a naïve worldview. Also, they are compatible with the range of use of the word and can be substituted in the context of use. Analysis of the use of the term virus also illustrates the way polysemy is treated in the NSM analysis. This analysis identifies an emerging use of the word virus relating to the coronavirus pandemic. The use of NSM allowed the authors to design two different explications satisfying the context of use, thus confirming the polysemy of the word.

10.3.2 Semantic Template: Conceptual Structure and Context Semantic template is a salient notion in the NSM approach. According to Goddard (2020: 97), a semantic template is “an arrangement of component types which is shared by explications for words of a particular class or subclass.” Research has identified the importance of establishing a semantic template for subclasses of physical activity verbs (e.g., run, eat, cut, grind), life-form words (e.g., fish, bird), emotions (e.g., happy, sad), among many others. Semantic templates, on the one hand, allow to achieve greater coherence in an explication and simultaneously they reflect the cognitive ability to process the complexity of information. On the other hand, a semantic template can explain how a lexical unit functions in the linguistic context and how it ‘hooks’ with the neighboring words. We will illustrate the issues of a semantic template and polysemy with the semantic analysis of the verb to dig presented in Goddard (2015). Goddard argues that to dig has two inherent properties – first, it is locational in the sense that digging always happens in a particular PLACE and, second, it implies an effect on the ground in that place. He proposes to identify three main uses of to dig – to dig1 as basic activity in progress, to dig2 + NOUN (something created by a human in the ground) as exemplified in (3a and 3b) and to dig3 for as exemplified in (4): (3) a. He dug a hole (in the ground). b. They dug a well (tunnel, grave, trench). (4) He was digging for potatoes (gold, water, buried treasure).

Therefore, the three uses of to dig can be distinguished by different lexicosyntactic frames of the verb. Goddard proposes the following explications for three uses of to dig:

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

231

232

ANNA GLADKOVA

Explication [F]: Someone is digging1 somewhere. someone is doing something somewhere for some time because of this, something is happening in this place during this time this someone is doing it with something at many times when someone does this in a place, it is like this: – a short time before, this someone thought like this: “I want some of the ground [m] in this place to be after this time not in the place where it is now”

LEXICOSYNTACTIC FRAME PROTOTYPICAL SCENARIO

when someone does this in a place, the same thing happens many times it happens like this: – this someone holds [m] something for some time – during this time, this someone does something to this something with parts of this someone’s body – because of this, part of this thing is inside the ground [m] in this place for a short time – during this time, this thing moves as this someone wants – because of this, something happens to the ground [m] in this place at this time – because of this, after this, some of the ground [m] is not in the place where it was before

MANNER + EFFECT (HOW IT HAPPENS)

if this someone does this for some time, after this, much of the ground [m] in this place is not in the place where it was before

POTENTIAL OUTCOME

Explication [G]: Someone X dug2 a hole (in the ground) in this place. someone X did something in this place at this time because this someone wanted there to be a hole in the ground [m] in this place

LEXICOSYNTACTIC FRAME

because of this, after this, there was a hole in the ground [m] in this place

OUTCOME

it happened like this: this someone was digging [d] in this place for some time

HOW IT HAPPENED

Explication [H]: Someone was digging3 for potatoes (water, gold, etc.) in this place. someone was doing something somewhere for some time because this someone wanted to do something with this something

LEXICOSYNTACTIC FRAME

it happened like this: – some time before, this someone thought like this about this place: “maybe there is something good of one kind (potatoes, water, gold) below the ground [m] in this place I want to know if there is something like this below the ground [m] in this place, I want it not be below the ground [m] anymore I want to do something with it” – because of this, this someone wanted to do something to the ground [m] in this place – because of this, this someone was doing something in this place for some time like someone does when someone is digging [m] in a place for some time

HOW IT HAPPENED

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

10 Natural Semantic Metalanguage and Context

The semantic template of the main use (dig1) consists of a lexicosyntactic frame, prototypical scenario, manner and effect (how it happens), and a potential outcome. The lexicosyntactic frame constitutes the foundation of the explication and reflects the inherent syntactic properties of the analyzed word. Goddard (2015) defines a lexicosyntactic frame as “the top-most set of components, expressed in a very generic way. These identify the core participants, inherent aspect, and other conceptual components that are important in the macro grammar of a language.” He further elaborates that they “account for much of the morphosyntactic behavior of a given verb, as they identify the core participants and their semantic roles, inherent aspect, and, often, components of volitionality and control.” In the case of to dig1, the process is necessarily durational (it happens for some time) and as a result of it, something happens in the place due to a person performing this activity in that place. Using NSM it is formulated as follows: someone is doing something somewhere for some time because of this, something is happening in this place during this time this someone is doing it with something

The prototypical scenario in explication (F) describes the intention of the person to conduct this activity. The manner and effect section contains the description of the actual process and the potential outcome section reaffirms the result of the process. The structure of the extensional uses of to dig2 and to dig3 have a lot in common; however, they are not fully identical and reflect syntactic and semantic properties of these uses. In both cases, these meanings are presented as extensions of the main use of to dig1, and the meaning of to dig1 is incorporated into the explications of to dig2 and to dig3. Broadly speaking, the semantic template of the explication is reflective of the semantic content and the context the word functions in. Thus, since to dig2 applies to nouns, such as well, tunnel, grave, and trench, the final section of the explication, includes an “analogy of action” component: this someone was doing something in this place for some time like someone does when someone is digging [m] in a place for some time

Therefore, the activity described in the process to dig a well (a tunnel, etc.) is depicted not as ordinary digging, but like ordinary digging. This component provides a mechanism of meaning extension. As for the “digging for” construction, it involves knowledge that things of a certain kind that people might want to have can be found in or under the ground. Consequently, explication (H) contains reference to the object of action (e.g., potatoes, gold). This explication includes dig as a molecule, that is, explication of dig1 is incorporated into this explication.

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

233

234

ANNA GLADKOVA

10.4 NSM in the Interpretation of Cultural Context 10.4.1 Culture as a Context for Linguistic Meaning We now turn to the issue of the influence of broader context on linguistic meaning in a particular cultural context. By culture here we mean ideas, beliefs, and understandings that are shared by speakers of a particular language in a certain locality. The locality is considered broadly here. The important view here is that language serves as an embodiment of cultural ideas that are shared by its users. We consider the definition of culture proposed by anthropologist Clifford Geertz to be relevant here. By culture he understood “a historically transmitted pattern of meanings embodied in symbols, a system of inherited conceptions expressed in symbolic forms by means of which men communicate, perpetuate, and develop their knowledge about and attitudes towards life” (Geertz 1973: 89). This interpretation of culture is labeled “semiotic.” Geertz also emphasized the permeability of cultural ideas and beliefs. Geertz (1973: 5) uses a metaphor suggested by Max Weber to explain how an anthropologist can approach the task of studying culture: Believing, with Max Weber, that man is an animal suspended in webs of significance he himself has spun, I take culture to be those webs, and the analysis of it to be therefore not an experimental science in search of law, but an interpretive one in search of meaning.

To develop the ideas of Geertz and Weber in relation to context further, culture can be regarded as a “thick” context in which the use of language needs to be understood and interpreted. A linguist’s task lies in the search for and interpretation of the meanings of cultural “webs” as they are encoded in language – the primary means of human communication. The NSM approach provides a methodological tool which is capable of interpreting meaning in cultural context. We will demonstrate the application of NSM in the interpretation of cultural meaning using examples of cultural keywords, culture-specific meanings, and cultural scripts.

10.4.2 Cultural Keywords Anna Wierzbicka’s search for lexical universals in language went hand in hand with the search for culture-specific elements of meaning. This kind of research led to the understanding that culture manifests itself in language in an array of ways. “Cultural ideas” penetrate language and get encoded in the meanings of lexemes, morphemes, and grammatical constructions, as well as speech practices. Cultural keywords are an illustration of such cultural meanings (Wierzbicka 1997; Levisen and Waters 2018). Wierzbicka (1997: 15–16) defines cultural keywords as “words which are particularly important and revealing in a given culture.” Cultural keywords are prominent in the collective psyche of a society and their meanings resonate with meanings of other linguistic units and cultural practices. Often, these words denote values, attitudes, speech acts, social

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

10 Natural Semantic Metalanguage and Context

categories, among others. Wierzbicka (1997: 16) explains that “there is no finite set of such words in a language, and there is no ‘objective discovery procedure’ for identifying them. To show that a particular word is of special importance in a given culture, one has to make a case for it.” Wierzbicka further claims that a word can be regarded as a keyword if it meets the following criteria: (1) it has a relatively high frequency in the semantic domain it belongs to; (2) it is at the center of a phraseological cluster; and (3) it might be part of common sayings or expressions. Wierzbicka (1997: 16) sums up the procedure of identifying cultural key words as follows: But the question is not how to “prove” whether or not a particular word is one of the culture’s key words, but rather to be able to say something significant and revealing about that culture by undertaking an in-depth study of some of them. If our choice of words to focus on is not “inspired” we will simply not be able to demonstrate anything of interest.

Therefore, no strict rule exists in respect of the part of speech or lexical domain that a cultural keyword can belong to. In fact, it can be represented by almost any part of speech if it is shown to reveal important cultural attitudes. Observations suggest that most commonly such words are from domains of emotions, values, attitudes, social categories, among others. Wierzbicka’s research has identified the following cultural keywords for Anglo English: privacy, personal autonomy, fairness, mind, reasonable, sense, evidence, experience, among others (Wierzbicka 2006, 2014). The main principles of the NSM semantic analysis also apply to the analysis of cultural keywords. A thorough semantic investigation leads to the proposal of a semantic explication of the keyword in simple universal concepts as they are identified in the NSM. The universal properties of the primes guarantee their translatability into all languages. Thus, when culture-specific meanings are formulated using primes, they can be understood by speakers of any languages in the world. We will illustrate the idea of a cultural keyword and its NSM analysis with Levisen’s (2012, 2017) research on hygge “pleasant togetherness, coziness.” Levisen demonstrates that hygge is an extremely linguistically productive word and occurs in a wide range of constructions and derivatives: • Verbs: hygger and a light reflexive hygger sig • Speech formulae: hyg dig! and the variants ka’ du hygge dig! or du må hygge dig!, which fills in the conversational slot as the English parting routine have fun!, and can be represented as variant of ‘have hygge!’ • Adjectives: hyggelig can be used predicatively, e.g., in a det var hyggeligt ‘that was hyggelig’ • Negated forms: uhygge and uhyggelig. Uhygge, however, is not the exact opposite of hygge, and uhyggelig differs in meaning from the analytically

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

235

236

ANNA GLADKOVA

generated and truly negated phrase ikke hyggelig ‘not hyggelig.’ Uhygge, can be translated roughly as ‘sinister atmosphere, eeriness’ and uhyggelig means roughly ‘eery, grim.’ • Fixed compounds: Nouns – hyggespreder ‘someone who spreads hygge,’ hyggeaften ‘hygge-evening,’ julehygge ‘Christmas-hygge,’ adventshygge ‘advent-hygge,’ hyggebelysning ‘hygge-illumination,’ hyggemusik ‘hyggemusic,’ hyggeonkel ‘hygge-oncle,’ hyggekrog ‘hygge-nook,’ hyggepianist ‘hygge-pianist,’ hyggesnak ‘hygge-chat,’ råhygge ‘raw-hygge,’ familiehygge ‘family-hygge,’ hjemmehygge ‘home-hygge’; Verbs – julehygger ‘to Christmashygge’; hyggesnakker ‘to hygge-chat.’ Therefore, hygge well fits the criterion of linguistic productivity and cultural salience as a cultural keyword. It is also a routine- and practice-forming concept. In this regard, Levisen mentions the metapragmatic code of hyggesnak ‘hygge talk,’ which stands for a relaxed, friendly frame of talking. Hyggesnak can be compare to the Anglo concept of “small talk.” Hyggesnak, however, is usually socially grounded. It typically takes place outside working hours, and within an extended time frame. Levisen claims, “Hyggesnak emerges through a collaborative social process where all participants contribute, and no one dominates the conversation. Since the pleasure of talking with others is the sole purpose, hygge-talkers actively avoid serious discussions, difficult issues, or socially divisive topics.” Levisen proposes the following NSM explication for hygge: Explication [I] hygge a) something b) people can say what this something is with the word hygge c) someone can say something about something with this word when this someone thinks like this about a place: d) it can be like this: e) good things are happening in this place now, because people are with other people now in this place for some time f) during this time people don’t want to do many things they want to do some things with the other people in this place they want to say many things to the other people in this place

Levisen (2012) demonstrates an extremely high cultural significance of hygge and relates it to an array of Danish-specific cultural values and understandings, such as the value of a “strong group identity,” the value of a “group inclusivity,” the value against, roughly, being “verbally non-participatory,” the value against, roughly, “being verbally dominant,” the value against, roughly, “not including everyone,” the value against, roughly, “raising sensitive issues,” the value against, roughly, “raising serious issues,” and the value against, roughly, “making gloomy predictions.”

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

10 Natural Semantic Metalanguage and Context

10.4.3 The Influence of Multiple Types of Contexts of the Word Meaning We will illustrate the influence of social, historical, and economic context on language using the example of the concept “beautiful” across three languages – English, Russian and Spanish. Gladkova and Romero-Trillo (2014) analyze the word beautiful in English and its equivalents krasivyj in Russian and bonito in Spanish. They conclude that the three languages share the major prototypical meaning referring to positive cognitive and emotional experience evoked by seeing someone or something – beautiful1, krasivyj1, and bonito1: Explication [J] something/someone is beautiful1 (This woman is beautiful. This vase is beautiful.) (a) (b) (c) (d)

this thing [this someone] is like this: at many times, when someone sees this thing this someone can’t not feel something very good because of this at the same time, this someone can’t not think something very good about it

The analysis also confirms the presence of two other meanings in the three languages, which are regarded as extensions of the first prototypical meaning. The second meaning encodes appreciating something beautiful via hearing. This explication contains reference to the positive experience of appreciating something via seeing, that is the first meaning: Explication [K] beautiful2 singing/voice/tune (a) (b) (c) (d) (e)

this thing is like this: at many times, when someone hears this thing this someone can’t not feel something very good because of this like people can’t not feel something very good at some times when they see some things at the same time, this someone can’t not think something very good about it

The three languages are also shown to share the third use of beautiful where it describes positive experience from someone’s intellectual or physically skillful performance: Explication [L] someone does beautiful3 something (Barcelona play beautiful football, he had a beautiful idea) (a) it can be like this: (b) someone does something at some time (c) when someone else thinks about it, this someone can’t not feel something very good because of this, (d) like people can’t not feel something very good at some times when they see some things (e) at the same time, this someone can’t not think something very good about it

The three languages are shown to diverge when it comes to describing the positive experience achieved via tasting, touching, smelling, or feeling: English

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

237

238

ANNA GLADKOVA

has this use of beautiful, while Russian and Spanish do not. The explication for beautiful4 is the following: Explication [M] something is beautiful4 (It’s beautiful flavor. This tastes beautiful. It was a beautiful experience.) (a) (b) (c) (d) (e)

someone can think like this at some time: ‘‘something is happening to me now, I can’t not know it I can’t not feel something very good because of this, like people can’t not feel something very good at some times when they see some things at the same time, I can’t not think something very good about it’’

The analysis demonstrates the influence of linguistic, historical, and social contexts on the meanings under analysis. From a linguistic perspective, the words sense and senses are important cultural and cognitive concepts in English with a relatively high frequency of use (Wierzbicka 2014). In Russian, however, the vocabulary for talking about “senses” differs from English, as senses are usually represented via different words – č uvstva ‘feelings,’ ošč ušč enie ‘feeling/ perception/sense’ and vosprijatie (roughly, ‘perception’). Iordanskaja (1979) on the basis of the analysis of the semantics of corresponding Russian verbs č uvstvovat’ ‘feel,’ ošč ušč at’ ‘feel/perceive/sense,’ and vosprinimat’ ‘perceive’ argues that in the Russian “naïve” model of perception smelling, touching, and taste are opposed to sight and hearing. In particular, she demonstrates that the verb vosprinimat’ ‘perceive’ is applicable to any mental ability, including reason, sight, and hearing, while the verbs č uvstvovat’ ‘feel’ and ošč ušč at’ ‘feel/perceive/sense’ are applicable to olfaction, touch, taste, kinesthetic, and other internal senses, but not seeing and hearing. Therefore, in the Russian worldview the division between seeing and hearing, on the one hand, and touching, smelling, and tasting on the other is more distinct than in English. This may influence the conceptualization of evaluative aesthetic vocabulary. From a sociological perspective, some economic factors could also play a role in this linguistic development. Howes (2005), for instance, argues that in modern culture consumerism is associated with the involvement of all senses. In particular, he associates the emergence of departmental stores and the related practice of displaying goods approachable and freely touchable for the public in the mid-nineteenth century with the transformation of the nature of capitalism from “industrial” to “consumerist” (Howes 2005: 284). Based on this fact, Howes observes a significant increase in the appeal to the five senses since the mid-nineteenth century, with a particular increase in the past decades. The revitalization of the senses was a typical phenomenon between the 1890s and the first years of the twentieth century. In this period, and in order to maintain the level and cycle of consumerism, firms promoted the appeal to the senses by the socialization of luxury goods, with their aesthetic attraction, no matter how superfluous their function might be in comparison with other more economical objects that realize the same function, e.g., luxury penmanship instruments.

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

10 Natural Semantic Metalanguage and Context

From the point of view of Anglo, Russian, and Spanish cultures discussed in Gladkova and Romero-Trillo (2014), consumer capitalism was particularly associated with Anglo culture (in countries such as the UK and USA and also Australia and Canada). The economic development of Russia and Spain, by contrast, was associated with a significantly smaller impact on consumerism. In particular, Russia’s nascent consumerism in the second half of the nineteenth century and the beginning of the twentieth century was limited to a very narrow upper class (Stearns 2006). This was followed by communist rule in the twentieth century that cut short the development of consumerism. In the case of Spain, industrialization was slow relative to that of Britain and the USA in the nineteenth and twentieth centuries (Landes 1999; Bhattacharyya 2011, 2020). This fact led to a weaker development of consumerism and, possibly, to a more reduced attention to the “commercialization” of the five senses. Our hypothesis is that this difference in historical and economic development (context) across three cultures could be the reason for the enlargement of the meaning of beautiful to the five senses in English and the absence of a similar phenomenon in Russian and Spanish. Therefore, with the case of beautiful and its equivalents in Russian and Spanish we were able to demonstrate the influence of a broader social and historical context on language use. The NSM served as a reliable tool in the semantic analysis that allowed us to identify this influence. We will now discuss the technique of cultural scripts.

10.4.4 Cultural Scripts Goddard and Wierzbicka (2004: 153) define cultural scripts as a “powerful new technique for articulating cultural norms, values, and practices in terms which are clear, precise, and accessible to cultural insiders and to cultural outsiders alike.” In cultural scripts the NSM is used to articulate cultural norms, ideas, and understandings that are shared by speakers of a language and that are reflected in that language. Cultural scripts are grounded in linguistic data, such as conversational routines, common sayings and proverbs, frequent collocations, and varieties of formulaic and semi-formulaic speech (Wierzbicka 2003/1991, 2002). Discourse particles, interjections, and terms of address and reference also commonly embody cultural attitudes. They can reflect communicative practices and capture specificity of a communicative style. Evidence for cultural scripts is also acquired from data of bilingual use of language (e.g., Besemeres and Wierzbicka 2007). Cultural scripts as a technique of formulating cultural norms have a number of advantages. First, cultural scripts represent the cultural insiders’ perspective. Second, the use of simple and universal concepts allows for the scripts to be verified by cultural insiders. Third, the scripts are devoid of an ethnocentric bias. The use of the same inventory of semantic primes in constructing explications and cultural scripts allows one to show explicitly the relation between the

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

239

240

ANNA GLADKOVA

meaning of a word and cultural ideas and beliefs that it is related to. At the same time, cultural scripts differ from semantic explications. While semantic explications represent the meaning of a particular concept, cultural scripts reflect broader cultural norms. These norms concern ways of thinking, speaking, and behaving that are salient in a particular culture. As it has been mentioned, cultural scripts have a linguistic grounding in cultural key words, proverbs and common sayings, terms of address, phraseological patterns, among others (Goddard 2006), but they formulate ideas that are broader than the meaning of one word or concept. We will illustrate the application of cultural scripts technique analysis using the example of traditional Russian modes of interaction. In particular, we will analyze ways of talking depending on social categorization between svoi ‘close’ and č užie ‘distant’ people (as presented in Gladkova 2013a, 2017). Arguably, such ways of interaction undergo change in modern society; however, they are still recognized as salient discourse-forming concepts. Russian styles of interaction are often characterized as “polar.” In particular, anthropological studies of Russian culture note a considerable distinction in the ways in which Russian people communicate with people they know and those they don’t know (Boym, 1994; Pesmen, 2000). Consequently, “close” people receive a warm and engaged treatment, while “distant” people are treated with distinct reservation and even hostility. Arguably, the “polarity” of Russian styles of interaction is consistent with the “polarity” of the concepts svoi ‘one’s own’ and č užie ‘alien/strangers/foreigners.’ These indigenous terms encode interpersonal relationships associated with group inclusion and exclusion and represent a “folk” conceptualization of relationships. The following discussion outlines the meanings of these terms and demonstrates their relationship to Russian communicative styles. The social category svoj ‘one’s own’ (svoi in plural) refers to people of one’s immediate circle, which can include family members, members of close-knit groups, or people on the same side as oneself in a conflict. The following examples from the Russian National Corpus illustrate some of its uses: (5)

Poedu provedaju svoix – uznaju, č em oni tam zanimajutsja, pofotkaju. ‘I’ll go and check on svoi, will find out what they are doing there and take photos.’

(6)

Pomniš’, v tom godu s’’exalos’ mnogo naroda? Kak-to oni tam vse sgovorilis’. Roditeli, ix druz’ja. Tol’ko svoi. ‘Do you remember that many people came that year? They somehow arranged it among themselves. Parents and their friends. Only svoi.’

The relationship between svoi is characterized by complete trust. It involves sharing information about oneself that is not supposed to be shared with other people. One can open up with svoi without thinking what impression one’s speech and actions will produce, because svoi accept the person as he or she is. Treating other people as svoi presupposes openness and trust and extends to the domain of behavior, not just speech. In the following example from the

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

10 Natural Semantic Metalanguage and Context

corpus, a woman describes her life in a communal apartment, where several families shared kitchen and bathroom facilities while having individual rooms for living. She compares living with other people to living with svoi: (7)

Oč en’ xorošaja byla kvartira – kak svoi, my nikogda ne zakryvali ni kuxnju, ni dveri. Nikto ne xodil, nič ego drug u druga ne brali, a naoborot – pomogali. Kto zaboleet – i xvorostu napekut, i vrač a vyzovut, kak rodnaja mat’ pridet sosedka. ‘It was a very good apartment. We were like svoi and never locked the kitchen or the rooms. No one entered each other’s rooms or took anything from each other. On the contrary, we helped each other. If someone fell sick, others would bake biscuits and call a doctor. A neighbor would come like one’s own mother.’

This example also reinforces the idea that people are expected to be generous, hospitable, and helpful to svoi. To summarize, svoj has the following elements of meaning: group membership, close ties, openness and mutual knowledge, a caring attitude, a positive emotional attitude toward this person, a positive feeling from being in the company of this person, and reciprocity of the attitude. Čužoj (č užie in plural) refers to people one does not know or does not know well and whom one considers to be different from oneself. The attitude toward č užoj involves the feeling of uneasiness, alienation, and difference. Čužoj entails more than not knowing another person. A person can be labeled č užoj due to perceived difference and a lack of understanding, rather than a lack of mutual knowledge. Example (8) demonstrates that č užoj can be applied to people one knows and even shares one’s life with on an everyday basis: (8) Oni, koneč no, byli v zakonnom brake, no ostavalis’ nastol’ko č užimi ljud’mi, č to mne kažetsja, daže s trudom terpeli drug druga. ‘They were, of course, legally married, but remained č užie ljudi to each other, and I even think that they could hardly stand each other.’

A typical attitude toward someone perceived as č užoj implies a significant degree of distrust and hostility. The manner of interaction with č užoj is reserved and restricted. In particular, things that are perceived as very dear and intimate (called sokrovennyj ‘hidden/innermost’) are not to be communicated to č užoj (Gladkova, 2013b). To summarize, č užoj refers to a person one does not know or does not know well, and whom one considers to be different from oneself. Consequently, one does not share information about oneself with such a person and does not feel obliged to provide care and assistance. The attitude toward č užoj involves a negative feeling, and one can also feel uncomfortable being in the company of such a person. A contextual analysis of the terms svoi and č užie demonstrates that categorizing a person in either of these ways influences the choice of communicative strategies by the speaker. “Openness” and sharing of information of all kinds characterize interaction with svoi. A common phrase, Zdes’ vse svoi ‘everyone is svoj here,’ encourages the interlocutor to speak openly and truthfully.

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

241

242

ANNA GLADKOVA

Similarly, in interaction with svoi one does not have to follow codes of behavior that are appropriate with people one does not know or does not know well. Consequently, the rules of “courteous” behavior, which are suitable in formal settings, are not observed with svoi. Interaction with svoi is characterized by frankness and can involve even saying not very pleasant things. “Openness” in the manner of interaction with svoi extends not only to the domain of speech but also to the domain of behavior; it involves allowing oneself to express one’s desires and to speak and behave as one wants. There is an expectation that people like svoi will not be hurt by this kind of behavior and talk, and will not use the acquired information against the speaker. With č užie one employs the opposite communicative strategies. Čužie are people with whom one is deliberately reserved and even avoids interaction. Particularly, dear and intimate (sokrovennyj) thoughts are not to be communicated to č užie. This rule, however, applies to any kind of information that can make one vulnerable. Interaction with such people involves not saying much, not offering help, and not expecting help. Not sharing one’s feelings, particularly not displaying a positive attitude, is characteristic of interaction with č užoj. We now can summarize the results of analysis in NSM formulae. Explications [N] and [O] represent meanings of the concepts svoi and č užie. [P] and [Q] are cultural scripts formulating cultural norms relating to these concepts. Explication [N] svoi (These people are svoi.) (a) some people (b) I think about these people like this: (c) these people are like one something (d) I am one of these people (e) I do many things with these people (f) these people can know many things about me I don’t want many other people to know these things (g) I want good things to happen to these people (h) if one of these people wants me to do something good for them, I will do it because of this (i) I know that these people can think about me in the same way (j) I feel something good towards these people (k) I feel something good when I am with these people

Explication [O] č užoj (č elovek) (This person – č užoj (č elovek)) (a) (b) (c) (d) (e) (f) (g) (h)

someone I think about this someone like this: I don’t know this someone well this someone is not someone like me because of this, it is not good if this someone knows many things about me it is not good to think like this: it is good if I do something good for this someone when I am with this someone, I can feel something bad when I think about this someone, I don’t feel something good

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

10 Natural Semantic Metalanguage and Context

Explication [P] Cultural script spelling out modes of speaking and behaving with people like svoj (a) (b) (c) (d) (e) (f) (g) (h) (i)

[many people think like this at many times:] if I can think about someone like this: this someone is like svoj[M] when I am with this someone, it is like this: I can say many things to this someone I can’t say these things to many other people I can say many things as I want I can do many things as I want I know: this someone will not think something bad about me because I do something it is good if I want to do good things for this someone

Explication [Q] Cultural script spelling out modes of speaking and behaving with people like č užoj (a) (b) (c) (d) (e) (g) (h) (i)

[many people think like this at many times:] if I think about someone like this: this someone is like č užoj[M] it is not good if this someone knows many things about me because of this, when I am with this someone it is not good if I say many things to this someone when I say something to this someone, I don’t want this someone to think like this about me: this someone feels something good toward me it is not good to think like this: this someone can do something good for me it is not good to think like this: it is good if I do something good for this someone

10.5 Further Developments and Applications of NSM NSM has been a productive program which has undergone significant developments. Among most recent major NSM advances has been the development of Minimal Language. Wierzbicka (2014) introduces the concept of Minimal English in her book Imprisoned in English: The Hazards of English as a Default Language (chapter 14). She views it as “a common auxiliary inter-language for speakers of different languages.” As a base, this minilanguage uses the NSM but also includes more complex meanings, such as semantic molecules – universal, near-universal, and culture-specific. It is expected that language-specific versions of Minimal Languages can be identified and developed for each language (e.g., Minimal Korean, Minimal Spanish). While Minimal Language is a relatively new development, it has been successfully applied in the analysis of different types of contexts and discourses. In particular, there has been fruitful application in the areas of narrative medicine (Marini 2018), disability and mental health issues (Jordan 2017), diplomacy (Maley 2018; Farrelly and Wesley 2018), language teaching (Bullock 2012–2020; Sadow 2019), among others. The advances in the theory and application of Minimal Language are presented in Goddard (2018b) and (2021).

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

243

244

ANNA GLADKOVA

The power and potential of Minimal Language has been extensively demonstrated in Wierzbicka’s recent book What Christians Believe: The Story of God and People (2019). The book retells and re-thinks the basics of Christian faith in simple words and linguistic constructions. The text is an experiment in semantics and theology simultaneously, and it is also an experiment in crosscultural communication through a Minimal Language. Two other versions of this book have been already published in Polish and Russian (Wierzbicka 2017 and 2021b) and, potentially, can be studied further in order to identify the universality and translatability of Minimal Language.

References Apresjan, Ju. (1992). Lexical Semantics: User’s Guide to Contemporary Russian Vocabulary. [Translation of Ju. Apresjan 1974. Leksič eskaja semantika: sinonimič eskie sredstva jazyka. (In Russ.)] Ann Arbor: Karoma Publishers. Besemeres, M., and Wierzbicka, A., eds. (2007). Translating Lives: Living with Two Languages and Cultures. St. Lucia: University of Queensland Press. Bhattacharyya, S. (2011). Growth Miracles and Growth Debacles: Exploring Root Causes. Cheltenham: Edward Elgar. Bhattacharyya, S. (2020). A History of Global Capitalism: Feuding Elites and Imperial Expansion. Cham: Springer Nature. Boym, S. (1994). Common Places: Mythologies of Everyday Life in Russia. Cambridge/ London: Harvard University Press. Bromhead, H. (2018). Landscape and Culture: Cross-linguistic Perspectives. Amsterdam: John Benjamins. Bullock, D. (2012–2020). Learn These Words First: Multi-layer Dictionary for SecondLanguage Learners of English. https://learnthesewordsfirst.com. Farrelly, N., and Wesley, M. (2018). Internationalizing Minimal English: Perils and parallels. In C. Goddard (ed.), Minimal English for a Global World: Improved Communication Using Fewer Words, 95–112. Cham: Palgrave Macmillan. Geertz, C. (1973). The interpretation of cultures. Selected essays by Clifford Geertz. London: Hutchinson. Gladkova, A. (2013a). “Is he one of ours?” The cultural semantics and ethnopragmatics of social categories in Russian. Journal of Pragmatics, 55, 180–194. Gladkova, A. (2013b). “Intimate” talk in Russian: Human relationships and folk psychotherapy. Australian Journal of Linguistics, 33, 322–343. Gladkova, A. (2017). Communication modes, Russian. In Young Yun Kim, (ed.) The International Encyclopedia of Intercultural Communication. Hoboken, NJ: Wiley. Gladkova, A., and Larina, T. (2018). Anna Wierzbicka, Words and the World. Russian Journal of Linguistics, 22(3), 499–520. doi: 10.22363/2312-9182-2018-223-499-520. Gladkova, A. and Romero-Trillo, J. (2014). Ain’t it beautiful? The conceptualization of beauty from an ethnopragmatic perspective. Journal of Pragmatics, 60, 140–159. Goddard, Cliff. (2000). Polysemy: A problem of definition. In Y. Ravin and C. Leacock (eds.), Polysemy: Theoretical and Computational Approaches (pp. 129–151). Oxford: Oxford University Press.

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

10 Natural Semantic Metalanguage and Context

Goddard, C. (2006). Ethnopragmatics. A new paradigm. In C. Goddard (ed.), Ethnopragmatics: Understanding Discourse in Cultural Context (pp. 1–30). Berlin: Mouton de Gruyter. Goddard, C. (2011). Semantic Analysis: A Practical Introduction, 2nd ed. Oxford: Oxford University Press. Goddard, C. (2015). Verb classes and valency alternations (NSM approach), with special reference to English physical activity verbs. In A. Malchukov and B. Comrie (eds.), Valency Classes in the World’s Languages (pp. 1649–1680). Berlin: Mouton de Gruyter. Goddard, C. (2016). Semantic molecules and their role in NSM lexical definitions. Cahiers de Lexicologie, 2(109), 13–34. Goddard, C. (2018a). Ten Lectures on Natural Semantic Metalanguage: Exploring Language, Thought and Culture Using Simple, Translatable Words. Leiden: Brill. Goddard, C. (2018b). Minimal English for a Global World: Improved Communication Using Fewer Words. London: Palgrave Macmillan. Goddard, C. (2020). Natural Semantic Metalanguage. In W. Xu and J. R. Taylor (eds.), The Routledge Handbook of Cognitive Linguistics (pp. 93–110). New York: Taylor & Francis Group. Goddard, C. (2021). Minimal Languages in Action. Cham: Springer Nature. Goddard, C., and Wierzbicka, A., eds. (1994). Semantic and Lexical Universals: Theory and Empirical Findings. Amsterdam: John Benjamins. Goddard, C., and Wierzbicka, A., eds. (2002). Meaning and Universal Grammar: Theory and Empirical Findings, Vols. I, II. Amsterdam: John Benjamins. Goddard, C., and Wierzbicka, A. (2004). Cultural scripts: What are they and what are they good for? Intercultural Pragmatics, 1–2, 153–166. Goddard, C., and Wierzbicka, A. (2014). Words and Meanings: Lexical Semantics Across Domains, Languages, and Cultures. Oxford: Oxford University Press. Goddard, C., and Wierzbicka, A. (2021). Semantics in the time of coronavirus: “Virus,” “bacteria,” “germs,” “disease” and related concepts. Russian Journal of Linguistics 25 (1). 7–23. Howes, D. (2005). Hyperesthesia, or, the sensual logic of late capitalism. In D. Howes (ed.), Empire of the Senses: The Sensual Culture Reader (pp. 281–303). Oxford: Berg. Iordanskaja, L. (1979). The semantics of three Russian verbs of perception: Vosprinimat’ “(to) perceive,” ošč ušč at’ “(to) sense” and č usvstvovat’ “(to) feel.” Linguistics, 17, 825–842. Jordan, P. (2017). How to Start, Carry on and End Conversations: Scripts for Social Situations for People on the Autism Spectrum. London: Jessika Kingsley Publishers. Landes, D. (1999). The Wealth and Poverty of Nations: Why Some Are So Rich and Some So Poor? London: Abacus. Leibniz, G. W. (1956). Preface to an edition of Nizolius. In Philosophical Papers and Letters: A Selection, trans. and ed., with intro. by L. E. Loemker (Vol. I, pp. 186–202). Chicago: University of Chicago Press. Leibniz, G. W. [1678] (1987). The analysis of languages. In M. Dascal, Leibniz, Language, Signs and Thought: A Collection of Essays (pp. 161–165). Amsterdam: John Benjamins. Levisen, C. (2012). Cultural Semantics and Social Cognition: A Case Study on the Danish Universe of Meaning. Berlin: Mouton de Gruyter.

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

245

246

ANNA GLADKOVA

Levisen, C. (2017). Communication modes, Danish. In Young Yun Kim (gen. ed.), K. L. McKay-Semmler (associate ed.), The International Encyclopedia of Intercultural Communication. New York: John Wiley & Sons. Levisen, C., and Waters, S., eds. (2018). Cultural Keywords in Discourse. Amsterdam: John Benjamins. Maley, W. (2018). Minimal English and diplomacy. In C. Goddard (ed.), Minimal English for a Global World: Improved Communication Using Fewer Words (pp. 71–93). Cham: Palgrave Macmillan. Marini, M. G. (2019). Languages of Care in Narrative Medicine: Words, Space and Time in the Healthcare Ecosystem. Cham: Springer Nature. Mel’č uk, I. (2016). Language: From Meaning to Text. Moscow/Boston: Academic Studies Press. The Natural Semantic Metalanguage Homepage. https://intranet.secure.griffith.edu.au /schools-departments/natural-semantic-metalanguage. Peeters, B., ed. (2006). Semantic Primes and Universal Grammar: Empirical Evidence from the Romance Languages. Amsterdam: John Benjamins. Pesmen, D. (2000). Russia and Soul: An Exploration. Ithaca: Cornell University Press. Sadow, L. (2019). An NSM-based cultural dictionary of Australian English: From theory to practice. Ph.D. thesis, Australian National University. Stearns, P. (2006). Consumerism in the World History: The Global Transformation of Desire, 2nd ed. New York/London: Routledge. Wierzbicka, A. (1972). Semantic Primitives. (Trans. A. Wierzbicka and J. Besemeres.) Frankfurt: Athenäum Verlag. Wierzbicka, A. (1996). Semantics: Primes and Universals. Oxford: Oxford University Press. Wierzbicka, A. (1997). Understanding Cultures through Their Key Words: English, Russian, Polish, German, and Japanese. Oxford: Oxford University Press. Wierzbicka, A. (2002). Russian cultural scripts: The theory of cultural scripts and its applications. Ethos, 30(4): 401–432. Wierzbicka, A. (2003/1991). Cross-Cultural Pragmatics, 2nd ed. Berlin: Mouton de Gruyter. Wierzbicka, A. (2006). English: Meaning and Culture. Oxford: Oxford University Press. Wierzbicka, A. (2014). Imprisoned in English: The Hazards of English as a Default Language. New York: Oxford University Press. Wierzbicka, A. (2015). Natural Semantic Metalanguage. In Tracy, Karen (ed.), The International Encyclopedia of Language and Social Interaction. New York: John Wiley & Sons. Wierzbicka, A. (2017). W co wierzą chrześcijanie? Opowieść o Bogu i o ludziach (What Christians Believe: The Story of God and People). Cracow: Znak. Wierzbicka, A. (2019). What Christians Believe: The Story of God and People in Minimal English. Oxford: Oxford University Press. Wierzbicka, A. (2021a). “Semantic primitives,” fifty years later. Russian Journal of Linguistics, 25(2), 317–342. Wierzbicka, A. (2021b). Vo č to verjat xristiane: Istorija Boga i ljudej. [What Christians Believe: The Story of God and People.] Moscow: Languages of Slavonic Cultures. (Trans. from English into Russian by Anna Gladkova.)

https://doi.org/10.1017/9781108989275.011 Published online by Cambridge University Press

11 Relevance Theory and Context Deirdre Wilson

11.1 Introduction Here is a fairly standard definition of pragmatics: “Pragmatics studies the use of language in context, and the context-dependence of various aspects of linguistic interpretation” (Lycan 1995: 588). According to this definition, providing an adequate account of context is not just an optional extra for pragmatics but part of the job description. Yet there is no general consensus in pragmatics on the nature of context and its role in utterance production or interpretation. As Meibauer (2012: 9) puts it, “Although most scholars would accept that the notion of context is fundamental not only for pragmatics, but also for linguistics in general, robust theories of context are lacking.” The lack of an adequate account of context in pragmatics is due less to disagreement about the goals of pragmatic theory than to doubts about whether such a theory is achievable at all. Since the pioneering work of Grice (1957, 1967, 1969), much research in pragmatics has centered on a type of communication, arguably unique to humans, in which the communicator intends not only to inform the audience of something, but also to make clear to the audience that she has this intention. Utterances are prime examples of such overtly intentional (ostensive) communicative acts, and a main goal of pragmatics is to explain how hearers recognize the informative intentions behind them. It is generally agreed that there is no fail-safe procedure for doing this. The intended import of an utterance cannot be decoded but only non-demonstratively inferred, by constructing a “best hypothesis” about the speaker’s meaning based on evidence provided by the linguistic meaning of the sentence uttered, together with the context (Horn and Ward 2004; Allan and Jaszczolt 2012; Huang 2017). In this chapter I will consider the treatment of context in Relevance Theory, a cognitively oriented pragmatic theory which sees human communication and cognition as governed by the search for relevance (Sperber and Wilson I would like to thank Robyn Carston for valuable comments on an earlier version.

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

248

DEIRDRE WILSON

1986/1995; Carston 2002; Wilson and Sperber 2012; Clark 2013). Relevance Theory approaches the context in cognitive terms and argues that it is not fixed in advance of the utterance but constructed in the course of the comprehension process, using information derived from perception, memory, and/or inference. As Sperber and Wilson (1986/1995: 15–16) put it: A context is a psychological construct, a subset of the hearer’s assumptions about the world. It is these assumptions, of course, rather than the actual state of the world, that affect the interpretation of an utterance. A context in this sense is not limited to information about the immediate physical environment or the immediately preceding utterances: expectations about the future, scientific hypotheses or religious beliefs, anecdotal memories, general cultural assumptions, beliefs about the mental state of the speaker, may all play a role in interpretation.

The fact that there seems to be no principled limit to the type of contextual information used in utterance interpretation raises an immediate question for pragmatics. How do hearers more often than not find the right contextual information and thus succeed in recognizing the speaker’s informative intention?

11.1.1 Skepticism about Pragmatics It is sometimes argued, by both linguists and philosophers, that there is no satisfactory general answer to this question, and that an adequate pragmatic theory is therefore not achievable at all. Chomsky (1992: 120) maintains that utterance interpretation is too complex to be a suitable topic for theoretical investigation: There is . . . a . . . problem, which we can formulate in vague terms but which cannot be studied in practice: namely to construct an “interpreter” which includes the parser as a component, along with all other capacities of the mind . . . and accepts non-linguistic as well as linguistic input. The interpreter, presented with an utterance and a situation, assigns some interpretation to what is being said by a person in this situation. The study of communication in the actual world of experience is the study of the interpreter, but this is not a topic of empirical enquiry for the usual reasons: there is no such topic as the study of everything.

Similar views are expressed by the psycholinguist Ray Gibbs (2017: 324), who argues that whatever regularities exist in pragmatics cannot be isolated from the operation of the cognitive system as a whole: It may very well be true that people often act co-operatively in linguistic interactions, and appear to be interested in what others may know and are intending to do. Yet these regularities are emergent from the human system overall and not the output of some specialized “pragmatics” part of the mind.

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

11 Relevance Theory and Context

Or, as the philosophers Herman Cappelen and Ernie Lepore (2005: 190) conclude, “there can be no systematic theory of speech act content.” Another philosopher who was famously skeptical about the prospects for theorizing in pragmatics was Jerry Fodor. Surveying the state of the art in cognitive science, Fodor (1983) distinguished local (typically modular) cognitive processes, which have access only to information from some fixed cognitive domain, from global (typically central) processes, which have free access to information from any source. In his view, while some progress had been made in studying local processes, global processes remained a mystery, and he went on to propose Fodor’s First Law of the Nonexistence of Cognitive Science: “The more global . . . a cognitive process is, the less anybody understands it” (p. 107). Much of the skepticism in philosophy and linguistics about the prospects for constructing an adequate pragmatic theory is linked to the fact that pragmatic processes (e.g., disambiguation, reference resolution, figurative interpretation, identification of implicatures,1 and so on) are global rather than local processes, whose outcome may be affected by contextual information derived from anywhere in the cognitive system (on the issues raised in this section, see Carston 2002: 1–14; Allott 2019; Allott and Wilson 2021; Wilson 2022).

11.1.2 Minimizing the Role of Context One reason why more progress has been made in studying local rather than global processes is that local processes have so far proved more amenable to formal treatment. It is therefore worth considering how the pragmatic processes involved in recognizing a speaker’s informative intention might be formalized too. Here I will look at two attempts to move toward a more formal approach to utterance interpretation which seem to me to set off in the wrong direction, by trying to reduce appeals to context to a minimum. The first approach centers on Grice’s (1967/1989: 37) distinction between particularized and generalized implicatures. As Grice describes them, particularized implicatures (illustrated in footnote 1) are “entirely dependent on special features of the context,” whereas in generalized implicatures, “the use of a certain form of words in an utterance would normally (in the absence of special circumstances) carry such-and-such an implicature or type of implicature.” The association of generalized implicatures with particular “forms of words” seems to open the way to formal treatments not available for particularized cases, and there is now a considerable literature on generalized implicatures and the linguistic constructions which give rise to them (Horn 1984, 2004; Hirschberg 1991; Levinson 2000; Noveck and Sperber 2007; Noveck 2018; Zufferey et al. 2019). 1

An implicature is a proposition which is implicitly communicated without being linguistically encoded (e.g., saying “I’m driving” when offered a drink might implicate that you don’t want a drink) (Grice 1967/1989: 22–40; Horn 2004; Allott 2018).

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

249

250

DEIRDRE WILSON

Perhaps the best-known examples of generalized implicature arise in scalar utterances such as (1a), where use of the less informative scalar term some rather than the stronger alternative all would often implicate ‘not all,’ as in (1b): (1)

a. Some of the food was delicious. b. Not all of the food was delicious.

Stephen Levinson (2000) proposes a semi-formal account of generalized implicatures such as (1b) as “default,” or “preferred,” interpretations, which are automatically assigned to utterance types and accepted as part of the speaker’s meaning unless explicitly or contextually cancelled (e.g., by adding “In fact, all the food was top notch”). On this approach, contextual information plays no role in the derivation of generalized implicatures and contributes nothing to their content: Its only function is to cancel any such implicature that combines with it to yield a contradiction (Geurts 2010: chapter 1.5). There is thus a clear distinction between the roles of context in the derivation of generalized implicatures and particularized cases, which are still seen as heavily context dependent. For Levinson, the main advantage of his default account is that it improves the speed and efficiency of communication, sparing speakers the effort of putting generalized implicatures into words and hearers the effort of inferring them from contextual information. Noveck and Sperber (2007) compare this default account with an alternative relevance-theoretic account which treats the generalized–particularized distinction as one not of kind but only of degree, with context making important contributions to both. As they point out, the two accounts make different predictions about the processing of scalar utterances such as (1a). According to the default account, hearers should find it easier to interpret (1a) as carrying the automatically generated implicature in (1b) than go to the effort of cancelling it, whereas according to the relevance-theoretic account, hearers should find it easier to interpret (1a) as lacking the implicature in (1b) than go to the effort of inferring it from contextual information. Noveck and Sperber (2007: 198–209) report a series of experiments which confirm the predictions of the relevance-theoretic account. Moreover, as Noveck and Sperber point out, Levinson’s claim that the default account improves the speed and efficiency of communication is valid only on the assumption that generalized implicatures such as (1b) are routinely accepted as part of the speaker’s meaning and rarely have to be cancelled. Noveck (2018: 99–100) reports a series of experiments which substantially undermine this assumption, indicating, on the contrary, that the derivation of scalar implicatures is not a common occurrence, nor is it automatic . . . adults do not necessarily carry them out routinely and context is critical to making them occur. It is hard to argue – based on a wide range of data – that scalar enrichments arise automatically.

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

11 Relevance Theory and Context

Levinson’s attempt at the partial formalization of generalized implicatures is therefore questionable on both theoretical and empirical grounds. Moreover, it does not answer the central question for pragmatics, since it leaves the global nature of the pragmatic processes involved in contextually cancelling generalized implicatures or inferring particularized implicatures unaddressed (for further discussion, see Carston 1995, 1998; Sperber and Wilson 1995: 276–278; Noveck and Sperber 2007; Dupuy et al. 2016). A more radical move in a similar direction is made by Ernie Lepore and Matt Stone (2015: 6), who argue that “much more of interpretation than one might have expected is encoded in the rules of language.” Their aim is to show that a massively expanded set of linguistic and discourse conventions, generating a much wider range of semantic interpretations than are available on standard truth-conditional accounts, makes it possible to drop the notion of implicature entirely, and with it much of the machinery of inferential pragmatics. On this approach, the two possible interpretations of a scalar utterance such as (1a) are treated as distinct conventional linguistic meanings, cued by different types of discourse context; Lepore and Stone go on to claim that not only generalized implicatures but even many particularized implicatures can be reanalyzed using this richer set of linguistic conventions. Within this framework, the only function of contextual information is to choose between conventional linguistic meanings. As Lepore and Stone (2015: 83) put it, “Pragmatics can be, at most, a theory of disambiguation; pragmatic reasoning never contributes content to utterances.” However, this reduction in the scope of pragmatics comes at a price. Grice’s most original contribution was to show that as long as the addressee has some way of recognizing the communicator’s informative intention, communication can take place even in the absence of a code. This opens up the possibility of a unitary inferential pragmatic account of the role of context not only in nonverbal communication, but in both conventional and creative uses of language too. By arguing that any aspect of communicated content which cannot be treated as the output of linguistic conventions lies beyond the scope of pragmatics, Lepore and Stone rule out the possibility of such a unitary pragmatic account. Here is what they say about metaphor: The commonplace view about metaphorical interpretation is that it can be characterised in traditional semantic and pragmatic terms, thereby assimilating metaphor to other familiar uses of language. We will reject this view, and propose in its place the view that, though metaphors can issue in distinctive cognitive and discourse effects, they do so without issuing in metaphorical meaning and truth, and so without metaphorical communication. (Lepore and Stone 2010: 1)

They go on to claim that the effects not only of metaphor but of irony, humor, hinting, and other creative uses of language are governed by “distinct imaginative processes” which are too heterogeneous and open-ended to be dealt with in pragmatic theory as traditionally conceived (Lepore and Stone 2015: 155). As they see it, there is no set of “general interpretive principles” which explain

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

251

252

DEIRDRE WILSON

how these creative uses of language could be understood based only on “the meaning of what the speaker uttered, background knowledge, and overarching pragmatic principles” (p. 160). What they are offering, then, is a pragmatic theory that deals only with literal utterances – at a time when the interpretation of metaphor, irony, humor, hinting, and other creative uses of language is crying out for explanation and the literal–figurative distinction itself is increasingly under attack (Searle 1978; Atlas 2000; Ariel 2002; Recanati 2004; Sperber and Wilson 2008; Carston 2012). In fact, even on Lepore and Stone’s minimalist view of pragmatics, there is little of theoretical interest to say. While acknowledging that utterance interpretation involves some form of intention recognition, they believe that this relies on no special-purpose principles other than those used in interpreting contributions to joint activity in general. In particular, [it] doesn’t appeal to anything like Grice’s Cooperative Principle, and typically exploits shallow cues rather than deep inferences about the speaker’s mental state. Although this is pragmatics in the sense that it is concerned with explaining the behavior of rational agents, it is a very different enterprise from the pragmatics assumed in most previous theories. (Lepore and Stone 2015: 165)

Thus, Lepore and Stone join the company of those who dismiss the idea of a dedicated pragmatic theory as either unnecessary or unachievable, while leaving the role of context in a wide range of uses of language unexplained (for further discussion, see Chierchia 2004; Bezuidenhout 2016; Carston 2016; Harris 2017; Stojnić 2021).

11.1.3 Context and Common Ground However broadly or narrowly the role of context is construed, Levinson, Lepore, and Stone, like most people working on pragmatics, see the context for comprehension as drawn largely from the speaker and hearer’s common ground. This is described as consisting of “mutually recognized shared information” (Stalnaker 2002: 704): that is, information which is not only shared by speaker and hearer, but believed to be shared, and believed to be believed to be shared, and so on ad infinitum (Lewis 1969). According to Clark and Marshall (1981: 27), restricting the context to information drawn only from the common ground – while unachievable in practice because of the infinite series of checks that would be required – is “an ideal people strive for because they . . . want to avoid misunderstanding whenever possible.” The proposed link between context and common ground has stimulated a flood of research and continuing efforts to develop an empirically useful notion of common ground (Smith 1982; Sperber and Wilson 1990; Clark 1996: chapter 4; Eilan et al. 2005). However, there is little evidence that speakers and hearers strive to restrict the context for comprehension to common-ground information, or that it would improve their chances of successful communication if they did.

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

11 Relevance Theory and Context

Here is an example from Dan Sperber (2021) in which the hearer ignores a potential referent from the common ground and chooses a non-commonground referent instead: (2) Mary and Jane have a common friend, Mike Smith. Mary and Jane are also both friends with a Mike Jones, but they have never seen him together or talked to one another about him. Mary actually knows that Jane is also friends with Mike Jones, but Jane does not know that Mary knows. One day Mary sees Jane and Mike Jones kissing in the park. That evening, she says to Jane, “I didn’t know that Mike was your boyfriend!” Who is she talking about? Mike Jones, who was not in the common ground!

In the common-ground literature, such cases are described as involving accommodation, a process by which “the context is adjusted quietly and without fuss to accept the utterance of a sentence that imposes certain requirements on the context in which it is processed” (von Fintel 2008: 137). Some of these requirements are semantic, while others are properly pragmatic (Lewis 1979; Stalnaker 1999; Roberts 2015; Garcia-Carpintero 2016). In interpreting Mary’s utterance in (2), Jane chooses a referent from outside the common ground for pragmatic reasons. If she had stuck to common-ground information and refused to extend the context, she would have misunderstood. In fact, even restricting the context to common-ground information may not be enough to avoid misunderstanding. In (3), for instance, two flatmates may have “mutually recognized shared information” about several pens, the loss of any of which would make the utterance true, but only one of which might be valuable enough for its loss to be worth remarking on: (3)

I lost my pen.

In interpreting (3), the potential common-ground referents would be narrowed to a single one for pragmatic reasons, and the hearer would take the intended referent to be the valuable pen. More generally, two people who share a physical environment, language, and culture have access to a huge amount of “mutually recognized shared information,” and restricting the context to information drawn only from the common ground falls far short of explaining how hearers choose the right contextual information and thus succeed in recognizing the speaker’s informative intention. Although examples (2) and (3) both involve the use of non-common-ground information in reference resolution, the problem is quite general. Consider (4), a recent headline in a London newspaper: (4)

PM turns up for meeting. (Metro, August 11, 2022)

PM is a standard abbreviation for prime minister. For anyone who knows that the UK Prime Minister at the time was Boris Johnson, it is easy enough to recognize what is being asserted; however, to see the point of the utterance, a lot of further contextual information must be supplied. The fact that the

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

253

254

DEIRDRE WILSON

Prime Minister had turned up for a meeting could be headline news only in a context where carrying out his regular prime-ministerial duties is the last thing you would expect him to do. Whatever your opinion of Boris Johnson, on reading the headline in (4), you would automatically construct a context of this type for pragmatic reasons, whether or not it is drawn from the common ground, and an adequate pragmatic theory should help to explain why. To sum up the arguments of this section, the context for utterance interpretation is not fixed in advance but constructed in the course of the comprehension process, drawing on information from anywhere in the cognitive system. Explaining how hearers find the right contextual information to use in recognizing the speaker’s informative intention is a central goal of pragmatics. Although common ground between communicators vastly increases their possibilities of communication, the idea that the context for comprehension should be drawn only from the common ground is a leftover from the code model of communication. In an inferential framework, where communication takes place at a risk and hypotheses about the speaker’s meaning are accepted or rejected on the basis of an adequate set of pragmatic principles or mechanisms, there is no reason to restrict the context in this way. I will now consider what light Relevance Theory can shed on these issues.

11.2 Relevance in Communication and Cognition Relevance Theory is based on the assumption that human cognition and communication are both governed by the search for relevance. Relevance is seen as a potential property not only of utterances and other ostensive acts, but of any perceptual stimulus (say, the sight of clouds overhead) or internal representation (say, the memory of what happened last night) that provides an input to cognitive processes. Sights, sounds, smells, actions, thoughts, memories, or conclusions of inferences may all be relevant (to an individual, at a certain time), and context plays a crucial role in establishing their relevance, whether or not communication is taking place (for brief overviews of Relevance Theory, see Wilson, 2017, 2019).

11.2.1 Relevance and Cognition In discussing relevance, linguists tend to focus on discourse-specific notions such as relevance to a topic or relevance to a question under discussion. Sperber and Wilson (1986/1995: chapter 3) define two more general notions – relevance in a context and relevance to an individual – which relate to both communication and cognition. A context consists of mentally represented information which may be of any type – beliefs, doubts, hopes, wishes, plans, goals, intentions, questions – and is selected from a range of potential contexts accessible to the individual via perception, memory, and/or inference.

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

11 Relevance Theory and Context

An input to cognitive processes is relevant in a context when it interacts with that context to produce a worthwhile cognitive effect: say, by answering a question, confirming a suspicion, overturning an expectation, suggesting a hypothesis or a plan of action, or yielding a contextual implication (inferable from input and context together, but from neither alone).2 A single input may achieve several such cognitive effects: For instance, the sight of clouds overhead as I set off for work may make me revise my plans to have lunch outside, strengthen my suspicion that the weather forecast always gets things wrong, and contextually imply that I should go back for an umbrella. Other things being equal, the greater the cognitive effects achieved, and the smaller the mental (or processing) effort required to produce those effects, the greater the relevance (of the input, in that context) (Wilson and Sperber 2002, section 4). In processing a certain input, the individual has a range of potential contexts available, which vary in both the amount of effort they take to construct and the cognitive effects they would yield when combined with that input. An input is described as relevant to an individual when it is relevant in one of the contexts accessible to him. Maximizing the relevance of an input involves selecting the best possible context in which to process it: that is, the one that provides the best possible balance of effort and effect given the range of accessible contexts. A central claim of Relevance Theory is that as a result of constant selection pressures toward increasing cognitive efficiency, the human cognitive system has evolved a variety of mental mechanisms or heuristics that conspire to allocate attention and processing resources to inputs with the greatest expected relevance, and to process these inputs in a context that maximizes their relevance. Why do we automatically pay attention to bright lights, loud noises, sudden movements, human faces or voices? Because our perceptual systems are geared to picking out such stimuli, which are potentially highly relevant to us. Why, seeing clouds overhead as we set off for work, do we recall the weather forecast, our plans for the day, and the umbrella we left upstairs? Because our memory and inferential systems are geared to activating contextual information which is likely to maximize the relevance of the inputs we are processing. Much of what is known about the organization of human perceptual, memory, and inferential systems is consistent with the suggestion that they tend to be geared to maximizing relevance (for experimental evidence, see Van der Henst et al. 2002; Van der Henst and Sperber 2004; Sperber 2005). An utterance, like other inputs to cognitive processes, may produce a variety of cognitive effects in the addressee, not all of which will have been intended by the speaker. As we talk, you may be making inferences about my social status and origins, my prejudices and preferences, the social relations between us, and so on, which may help to make my utterance relevant to you, whether or not I intended you to draw them. It is therefore useful to distinguish between comprehension and interpretation, where comprehension is the 2

On what makes a cognitive effect worthwhile, see Sperber and Wilson (1995: 260–266).

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

255

256

DEIRDRE WILSON

process of recognizing the intended import of an utterance, and interpretation includes a broader process of drawing one’s own conclusions as part of the overall cognitive search for relevance (Wilson 2022). Ostensive communication, then, takes place against a background where the addressee is permanently engaged in a cognitive search for relevance, and identifying the intended import of an utterance is only one of his goals (albeit an important one, since much of what we know could only have been acquired through communication). What pragmatic theory has to explain is not how the addressee draws his own conclusions from an utterance (this falls within the domain of a theory of cognition), but how he constructs a hypothesis about its intended import, by disambiguating ambiguous expressions, assigning reference to referential expressions, adjusting lexical meanings, assembling an appropriate set of contextual assumptions, and deriving implicatures and other cognitive effects, in a way the speaker could manifestly have foreseen and intended. On this approach, communication and cognition are intimately intertwined: Both are governed by the search for relevance, and the same perceptual, memory, and inferential mechanisms play important roles in both. What prevents the resulting account amounting to a “theory of everything” (as skeptics about the prospects for theorizing in pragmatics maintain) is that ostensive communication presents certain challenges, and exhibits certain regularities, not found in other domains, which form the basis for a dedicated pragmatic theory.

11.2.2 Relevance and Communication Relevance Theory’s account of communication centers on the notion of an ostensive act: an overtly intentional communicative act designed to attract the addressee’s attention and focus it on the communicator’s intentions. Common cues to ostension include catching someone’s eye, touching them, pointing, showing them something, speaking, and writing. Since attention tends to go automatically to what is most relevant at the time, a communicator, by the very act of addressing someone, communicates that her utterance (or other ostensive act) is relevant enough to be worth the effort required to process it. Thus, ostensive acts create expectations of relevance not raised by other stimuli, and this regularity in the domain of ostensive communication forms the basis for an automatic heuristic that addressees can use (of course at a risk) to infer the intended import of ostensive acts addressed to them. How relevant an utterance has to be to satisfy the addressee’s expectation of relevance depends on what else is going on in his cognitive environment at the time. The more relevant the other inputs competing for his attention, the more relevant an utterance has to be to out-compete them. Thus, a comment on the weather which might be relevant enough in a casual chat is unlikely to be relevant enough when addressed to someone fighting a fire, whose attention is likely to be more productively allocated elsewhere. The addressee of an

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

11 Relevance Theory and Context

utterance is therefore entitled to expect that, on the intended interpretation, it will be at least relevant enough to be worth the processing effort required: otherwise, he will not attend to it or will transfer his attention elsewhere. But the addressee can also reasonably expect something more than this. The speaker wants to be understood. It is therefore in her interest to go beyond this minimal level of relevance by reducing the processing effort required and increasing the cognitive effects achieved – to the extent that this is compatible with her own abilities and preferences – thus improving her chances of holding the addressee’s attention and getting her point across. When an utterance is not only (a) relevant enough to be worth the processing effort required, but also (b) the most relevant one compatible with the speaker’s abilities and preferences, it is described as optimally relevant. Thus, the expectation raised in the addressee of an utterance (or other ostensive act) is one of optimal (rather than maximal) relevance (Sperber and Wilson 2002, 2005). On this approach, the best way for the addressee to recognize the speaker’s informative intention is to construct an overall interpretation that satisfies the expectation of optimal relevance raised by the utterance. This involves enriching the decoded sentence meaning at the explicit level (e.g., by resolving ambiguities and referential indeterminacies) and complementing it at the implicit level (e.g., by supplying contextual assumptions and deriving contextual implications) to a point where it yields enough cognitive effects, at a low enough processing cost, to be relevant in the expected way. There is, then, an automatic heuristic that hearers can use to achieve this goal: Relevance-guided comprehension heuristic Follow a path of least effort in deriving cognitive effects. Test interpretive hypotheses (e.g., disambiguations, reference resolutions, lexical adjustments, contextual assumptions and contextual implications) in order of salience. Stop when you have enough cognitive effects to satisfy the expectation of optimal relevance raised by the utterance.

What makes it reasonable for the hearer to follow a path of least effort is that the speaker is expected (within the limits of her abilities and preferences) to have made the intended import as easy as possible for him to recognize. What makes it reasonable for him to stop at the first interpretation which satisfies the expectation of optimal relevance is that a speaker who knowingly produced an utterance with two or more significantly different interpretations, each with the expected level of cognitive effect, would put the addressee to the gratuitous extra effort of choosing between them, and the resulting interpretation (if any) would not be relevant in the expected way. Thus, when an addressee following the path of least effort finds an interpretation that satisfies his expectation of optimal relevance, this is the best possible interpretive hypothesis in the absence of contrary evidence. Since communication takes place at a risk, this hypothesis, however plausible and well evidenced, may well turn out to be false; but it is the best a rational hearer can do.

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

257

258

DEIRDRE WILSON

The relevance-guided comprehension heuristic is seen as part of a dedicated comprehension mechanism or module which exploits a regularity that exists only in the domain of ostensive communication (Sperber and Wilson 2002, 2005). A lot of interesting work on communicative development in the last twenty years supports this approach to pragmatics. For instance, there is evidence that pre-verbal infants are sensitive to certain cues to ostension (Csibra 2010); respond differentially to ostensive and non-ostensive acts (Csibra and Gergely 2009; Schulze and Tomasello 2015); and form pragmatic expectations comparable to those of adults (Southgate et al. 2009). In the next section, I will consider how this approach might help to answer the central question for pragmatics: How do hearers more often than not use the right contextual information and thus succeed in recognizing the speaker’s meaning?

11.3 Constructing the Context for Comprehension I have tried to show that utterances raise expectations of optimal relevance not raised by other stimuli, and that hearers are equipped with a relevance-guided heuristic for inferring the speaker’s meaning. I will now look more closely at how the utterance interpretation process might go, and at how new information may be added to the context in order to satisfy expectations of relevance.

11.3.1 Explicatures and Implicatures Most approaches to pragmatics distinguish between explicit and implicit communication (or what is said and what is implicated). While Grice was mainly concerned with the implicit side of the distinction, relevance theorists see pragmatic factors as playing an equally rich and inferential role on the explicit side, in disambiguating ambiguous expressions, assigning reference to referential expressions, narrowing the interpretation of semantically vague expressions, fixing the domain of quantifiers, restoring ellipsed material, interpreting figurative expressions, and so on (Carston 2002: chapter 2.3; Carston and Hall 2012). Sperber and Wilson (1986/1995: 182) introduce the term explicature for communicated propositions which are identified by developing an encoded (incomplete) sentence meaning into a fully propositional form. Everything else communicated is an implicature. On this approach, the explicit–implicit distinction is exhaustive: a communicated proposition is either an explicature or an implicature (Sperber and Wilson 1986/1995: 182–202, 217–224). To illustrate how the relevance-guided comprehension heuristic might help with the identification of explicatures and implicatures, suppose you and I are academic colleagues; you ask if I’d like to see a film with you tonight and I reply, Sorry, I have to finish a paper. In interpreting my utterance, you have to resolve a number of indeterminacies in the encoded sentence

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

11 Relevance Theory and Context

259

meaning: deciding what type of paper I have in mind (newspaper, academic paper, parliamentary paper . . .), what I have to finish doing to it (reading it, writing it, shredding it . . .), when I must finish it (this year, next year, sometime, never . . .), and so on. Depending on how these indeterminacies are resolved, the utterance will have different cognitive effects, and therefore be relevant to you in different ways. Your goal (as described in the relevanceguided comprehension heuristic) is to resolve these indeterminacies in a way that satisfies your expectation of optimal relevance. You proceed by testing candidate interpretations in order of salience, and stop when you have enough cognitive effects to make my utterance relevant in the expected way. The interpretation process is seen as taking place in parallel rather than in sequence: it is not a matter of first identifying the explicature, then selecting an appropriate set of contextual assumptions, then deriving a set of implicatures. Instead, tentative hypotheses about explicatures, contextual assumptions and implicatures are accessed in any order and mutually adjusted, with each other and with the expectation of relevance, until a pragmatically satisfactory overall interpretation is reached. The process is heavily context-dependent, and the speaker’s goal in formulating her utterance is to provide enough clues to make the appropriate disambiguations, reference resolutions, lexical adjustments, contextual assumptions and implications salient enough to be selected by a hearer using the relevance-guided comprehension heuristic (Wilson and Sperber 2004, section 4). Table 11.1 shows, in outline, how the interpretation of my utterance I have to finish a paper might proceed, with your interpretive hypotheses on the left

Table 11.1 Interpretation of my utterance I have to finish a paper (5a) DW has said, ‘Xspeaker has to finish doing Yaction to Zpaper Result of embedding the decoded (incomplete) meaning of at ttime’ the sentence “I have to finish a paper” into a description of my ostensive act. (5b) DW’s utterance will be optimally relevant to me. Expectation raised by your recognition of my ostensive act and acceptance of the presumption of relevance it conveys. (5c) DW’s utterance will achieve relevance by explaining why Expectation raised by (5b), together with the fact that such she is not accepting my invitation to see a film with me an explanation would be most relevant to you at this point. tonight. (5d) Having to finish writing an academic paper very soon is First contextual assumption to occur to you which, together a reason for refusing an invitation to see a film tonight. with other appropriate premises, might satisfy expectation (5c). Accepted as an implicit premise of my utterance. (5e) DW has to finish writing an academic paper very soon. First enriched interpretation of (5a) to occur to you which might combine with (5d) to lead to the satisfaction of (5c). Accepted as an explicature of my utterance. (5f) DW is refusing my invitation to see a film with me Inferred from contextual assumption (5d) and explicature tonight because she has to finish writing an academic paper (5e), satisfying expectation (5c) and accepted as an very soon. implicature of my utterance. (5g) DW may accept an invitation to see a film with me From (5f) plus further contextual assumptions. One of another time. several possible further implicatures which, together with (5f), satisfy expectation (5b).

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

260

DEIRDRE WILSON

and a note on your basis for deriving them on the right.3 At (5a), you embed the decoded (incomplete) sentence meaning, with its semantic indeterminacies (informally indicated by subscripts) still unresolved, into a description of my ostensive act. At (5b), you assume that my utterance, decoded as in (5a), will be optimally relevant to you. Since what you want to know at this point is why I am refusing your invitation, you assume at (5c) that my utterance will achieve relevance by answering this question. Given our academic connections, the decoded sentence meaning provides easy access to the contextual assumption in (5d), that having to finish writing an academic paper very soon is a good reason for refusing an invitation. You could use this as an implicit premise in deriving the expected explanation of my refusal, provided that you resolve the indeterminacies in (5a) so as to yield the explicature in (5e), that I have to finish writing an academic paper very soon. By combining the implicit premise in (5d) with the explicature in (5e), you arrive at the implicated conclusion in (5f), that I am refusing your invitation because I have to finish writing an academic paper very soon; from this, further implicatures, including (5g) and others, can be derived. The resulting interpretation satisfies your expectation of relevance. On this account, contextual assumptions, explicatures, and implicatures are arrived at by a process of mutual parallel adjustment, with hypotheses about both being considered in order of salience until enough implications to satisfy the expectations of relevance raised by the utterance are derived (Wilson and Sperber 2002, section 5; Carston and Hall 2012).

11.3.2 Implicit Premises and “Accommodation” The mutual adjustment process offers some insight into how new information may be “accommodated” into the context in the course of comprehension (Section 11.1.3). Consider the exchange in (6) (Sperber and Wilson 1986/ 1995: 16): (6)

a. Max: Would you like some coffee? b. Jan: Coffee would keep me awake.

Interpreting Jan’s utterance in (6b), Max will expect it to achieve relevance by implying a response to his offer of coffee. In different circumstances, the utterance could be construed as either an acceptance or a refusal. However, in the absence of contrary evidence, a hearer using the relevance-guided comprehension heuristic is likely to interpret it on the lines in (7), as both refusing his offer of coffee and explaining the refusal: (7)

3

a. Contextual assumption: Jan doesn’t want to stay awake tonight. b. Explicature: Drinking coffee would keep Jan awake tonight. c. Implicature: Jan is refusing coffee because it would keep her awake tonight.

I have presented these hypotheses in English, but for the hearer, they would be in whatever is the medium of conceptual thought and need not correspond very closely to my paraphrases.

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

11 Relevance Theory and Context

This interpretation is favored by considerations of both effort and effect. In the first place, the contextual assumption in (7a) is easily derived from, and supported by, the generalization that most people don’t want to stay awake at night; given that the organization of memory is relevance-oriented, it may therefore be the first to occur to Max. Further support for (7a) comes from the well-evidenced and frequently encountered association between drinking coffee and having difficulty getting to sleep at night, as a result of which Max may have a ready-made schema for interpreting (6b) along the lines in (7). Whether or not he uses such a schema, Jan’s indirect answer to his question will require more processing effort than an explicit Yes or No, and a speaker aiming at optimal relevance must have expected this extra effort to be offset by extra effects. The interpretation in (7), on which Jan is implicitly both refusing his offer of coffee and explaining her refusal, satisfies this expectation; by contrast, acceptances need no explanation and are rarely left implicit. In the absence of contrary evidence, this interpretation is therefore likely to be the first to occur to a hearer using the relevanceguided comprehension heuristic. The exchange in (6) shows how an under-evidenced assumption which is tentatively added to the context in the course of the mutual adjustment process may be retroactively strengthened by the fact that it leads to a pragmatically satisfactory overall interpretation. Sometimes, in order to arrive at such an interpretation, the hearer may have to add to the context an assumption he does not believe at all. This is illustrated by the following (attested) exchange between a charity flag-seller and a passerby (Sperber and Wilson 1986/1995: 121–122): (8)

a. Flag-seller: Would you like to buy a flag for the Royal National Lifeboat Institution? b. Passerby: No thanks, I always spend my holidays with my sister in Birmingham.

In the first part of (8b), the passerby refuses the flag-seller’s request, creating the expectation that she will go on to explain her refusal. However, some hearers may have difficulty recognizing what that explanation might be. What possible link could there be between holidaying in Birmingham and refusing to support the Royal National Lifeboat Institution? Well, a salient fact about the Royal National Lifeboat Institution is that it is dedicated to saving lives at sea. On the other hand, anyone who has looked at a map of England is capable of realizing that Birmingham is about as far from the sea as it is possible to be. Given these assumptions, it follows that someone who spends her holidays in Birmingham is unlikely to need the services of the Royal National Lifeboat Institution. Then all that is needed to complete the link is the assumption in (9): (9) Someone who is unlikely to need the services of a charity cannot be expected to subscribe to that charity.

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

261

262

DEIRDRE WILSON

Interpreted in this context, (8b) yields the implicature in (10), which explains the speaker’s refusal to buy a flag, thus satisfying the hearer’s expectation of relevance: (10) The passer-by cannot be expected to subscribe to the Royal National Lifeboat Institution since she has no need of its services.

For a hearer who shares the contextual assumption in (9), or who has encountered it before, arriving at the intended interpretation may be fairly straightforward. For others, it may take more effort, and a thoughtful speaker, anticipating this, could have increased the relevance of her utterance by providing more clues. However, lapses of this type are not necessarily a bar to understanding. As Sperber and Wilson (1995: 266–271) point out, a speaker cannot reasonably be expected to go beyond her own abilities and preferences in choosing what to say or how to say it, and this is built into the expectation of optimal relevance as described in Section 11.2.1. The interpretation of utterance (8b) underlines the fact that in constructing a context for utterance comprehension, what matters is not whether the assumptions used are part of the speaker and hearer’s common ground, but whether they help to satisfy the expectation of optimal relevance raised by the utterance. As discussed in Section 11.1.3, Lepore and Stone (2015) propose to drop the notion of implicature entirely, and with it much of the machinery of inferential pragmatics, by reanalyzing implicatures as conventional meanings generated by a massively expanded set of sentence and discourse conventions. In their view, the only function of contextual information is to choose between conventional meanings (‘Pragmatics can be, at most, a theory of disambiguation; pragmatic reasoning never contributes content to utterances’). Utterance (8b), in which the contextual assumption in (9) makes a crucial contribution to the content of the implicature in (10), is a counterexample to this claim. Even a hearer equipped with a discourse convention to the effect that refusal of an offer is generally followed by an explanation may fall far short of recognizing what that explanation is. For this, he has to rely on the contextual information in (9) and derive the implicature in (10), and these are the very things that Lepore and Stone want to do without. Another case where it may be easy enough to see what is being asserted but hard to see the point of the utterance without making additional contextual assumptions is the headline in (4) (Section 11.1.3, repeated here for convenience): (4) PM turns up for meeting. (Metro, August 11, 2022)

Headlines raise high expectations of relevance. To satisfy those expectations, (4) should be more relevant than anything else published in the newspaper that day. Yet a prime minister is expected to turn up to meetings as part of his regular prime-ministerial duties, and in normal circumstances the fact that Boris Johnson had turned up to a meeting should not be relevant enough to qualify as headline news.

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

11 Relevance Theory and Context

At the time this headline was published, Boris Johnson had agreed in principle to resign but stayed on as prime minister for several months while the search for a successor went on. During this period, he held several farewell parties, arranged a series of photo opportunities, and went on honeymoon, but failed to make the sort of decisions a prime minister is expected to make at a time of national crisis. This provides a clue to the sort of context the headline writer was trying to evoke: one in which Boris Johnson’s neglect of his primeministerial duties was so egregious that even his turning up to a single meeting would be more relevant than anything else worth publishing that day. This headline might thus be seen as a case of “layering” in communication, where an ostensive act on one level becomes the object of a second-order communicative intention (Sperber and Wilson 1987: 751; Clark 1996: chapter 12). By behaving on one level as if it went without saying that the Prime Minister had been grossly negligent, the headline writer communicates, on another level, a degree of contempt that would have been hard to express overtly (for further discussion of the treatment of context in relevance theory, see Mazzone 2015; Assimakopoulos 2017).

11.4 Concluding Remarks In this chapter I have outlined a relevance-theoretic account of how hearers more often than not find the right contextual information to use in interpreting an utterance, and thus succeed in recognizing the speaker’s meaning. The particular problem for pragmatics is that there is no principled limit on the type of information that may be required, raising doubts about whether a robust theory of context is achievable at all. This problem becomes more tractable given two assumptions: first, that the human cognitive system tends to be geared to the search for relevance, and second, that speakers want their intentions to be recognized, are actively helping the hearer to recognize them and would acknowledge them if asked. It is then possible to envisage a relevanceguided comprehension heuristic by which tentative hypotheses about explicit content, context and intended implications are mutually adjusted, with each other and with the expectation of relevance, until a pragmatically satisfactory overall interpretation is reached. Relevance Theory is a result of many collaborations over the years and is still developing. On possible directions for future research, see Sperber and Wilson (2015); Sperber (2018); Scott et al. (2019); Wilson and Carston (2019); Heintz and Scott-Phillips (2022); Wharton et al. (2022).

References Allan, K., and Jaszczolt, K., eds. (2012). The Cambridge Handbook of Pragmatics. Cambridge: Cambridge University Press.

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

263

264

DEIRDRE WILSON

Allott, N. (2018). Conversational implicature. In M. Aronoff (ed.), Oxford Research Encyclopedia of Linguistics. Oxford: Oxford University Press. DOI:10.1093/acrefore/ 9780199384655.013.205. Allott, N, (2019). Scientific tractability and relevance theory. In K. Scott, B. Clark, and R. Carston (eds.), Relevance, Pragmatics and Interpretation (pp. 29–41). Cambridge: Cambridge University Press. Allott, N., and Wilson, D. (2021). Chomsky and pragmatics. In N. Allott, T. Lohndal, and G. Rey (eds.), A Companion to Chomsky (pp. 433–447). Hoboken, NJ: Wiley– Blackwell. Ariel, M. (2002). The demise of a unique concept of literal meaning. Journal of Pragmatics, 34, 361–402. Assimakopoulos, S. (2017). Context selection in relevance theory. In J. Blochowiak, C. Grisot, S. Durrleman, and C. Laenzlinger (eds.), Formal Models in the Study of Language: Applications in Interdisciplinary Contexts (pp. 221–242). Cham: Springer. Atlas, J. (2000). Logic, Meaning, and Conversation: Semantical Underdeterminacy, Implicature, and Their Interface. Oxford: Oxford University Press. Bezuidenhout, A. (2016). What properly belongs to grammar: A response to Lepore and Stone. Inquiry, 59, 175–194. Cappelen, H., and Lepore, E. (2005). Insensitive Semantics: A Defense of Semantic Minimalism and Speech Act Pluralism. Oxford: Blackwell. Carston, R. (1995). Quantity maxims and scalar implicature. Lingua, 96, 213–244. Carston, R. (1998). Informativeness, relevance, and scalar implicature. In R. Carston and S. Uchida (eds.), Relevance Theory: Applications and Implications (pp. 179–236). Amsterdam: John Benjamins. Carston, R. (2002). Thoughts and Utterances: The Pragmatics of Explicit Communication. Oxford: Blackwell. Carston, R. (2012). Metaphor and the literal–non-literal distinction. In K. Allan and K. Jaszczolt (eds.), The Cambridge Handbook of Pragmatics (pp. 469–492). Cambridge: Cambridge University Press. Carston, R. (2016). Linguistic conventions and the role of pragmatics. Mind and Language, 31, 612–624. Carston, R., and Hall, A. (2012). Implicature and explicature. In H. J. Schmid (ed.), Cognitive Pragmatics (pp. 47–84). Berlin: Mouton de Gruyter. Chierchia, G. (2004). Scalar implicatures, polarity phenomena and the syntax/pragmatics interface. In A. Belletti (ed.), Structures and Beyond (pp. 39–103). Oxford: Oxford University Press. Chomsky, N. (1992). Language and interpretation: Philosophical reflections and empirical enquiry. In J. Earman (ed.), Inference, Explanation and Other Frustrations: Essays in the Philosophy of Science (pp. 99–128). Berkeley, CA: University of California Press. Clark, B. (2013). Relevance Theory. Cambridge: Cambridge University Press. Clark, H, (1996). Using Language. Cambridge: Cambridge University Press Clark, H., and Marshall, A. (1981). Definite reference and mutual knowledge. In A. Joshi, B. Webber, and I. Sag (eds.), Elements of Discourse Understanding (pp. 10–63). Cambridge: Cambridge University Press. Csibra, G. (2010). Recognizing communicative intentions in infancy. Mind and Language, 25, 141–168. Csibra, G., and Gergely, G. (2009). Natural pedagogy. Trends in Cognitive Sciences, 13, 148–153.

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

11 Relevance Theory and Context

Dupuy, L., Van der Henst, J.-B., Cheylus, A., and Reboul, A. (2016). Context in generalized conversational implicatures: The case of some. Frontiers in Psychology, 7(381), March 22. https://doi.org/10.3389/fpsyg.2016.00381. Eilan, N., Hoerl, C., McCormack, T., and Roessler, J., eds. (2005). Joint Attention: Communication and Other Minds. Oxford: Oxford University Press. von Fintel, K. (2008). What is presupposition accommodation, again? Philosophical Perspectives, 22: 137–170. Fodor, J. (1983). Modularity of Mind. Cambridge, MA: MIT Press. Garcia-Carpintero, M. (2016). Accommodating presuppositions. Topoi, 35, 37–44. Gibbs, R. (2017). Experimental pragmatics. In Y. Huang (ed.), The Oxford Handbook of Pragmatics (pp. 310–325). Oxford: Oxford University Press. Geurts, B. (2010). Quantity Implicatures. Cambridge: Cambridge University Press. Grice, H. P. (1957). Meaning. Philosophical Review, 66: 377–388. Repr. in Grice (1989), pp. 213–223. Grice, H. P. (1967). Logic and conversation. William James Lectures, Harvard University. Repr. in Grice (1989), pp. 1–143. Grice, H. P. (1969). Utterer’s meaning and intentions, Philosophical Review, 78, 147–177. Repr. in Grice (1989), pp. 86–116. Grice, H. P. (1989). Studies in the Way of Words. Cambridge, MA: Harvard University Press. Harris, D. (2017). Review of Imagination and Convention: Distinguishing Grammar and Inference in Language by E. Lepore and M. Stone. Philosophical Review, 26, 554–558. Heintz, C., and Scott-Phillips, T. (2022). Expression unleashed: The evolutionary and cognitive foundations of human communication. Behavioral and Brain Sciences, Jan. 5. doi: 10.1017/S0140525X22000012. Hirschberg, J. (1991). A Theory of Scalar Implicature. New York: Garland. Horn, L. (1984). Toward a new taxonomy for pragmatic inference: Q-based and R-based implicature. In D. Schiffrin (ed.), Meaning, Form, and Use in Context: Linguistic Applications (pp. 11–42). Washington, DC: Georgetown University Press. Horn, L. (2004). Implicature. In L. Horn and G. Ward (eds.), The Handbook of Pragmatics (pp. 3–28). Oxford: Blackwell. Horn, L. and Ward, G., eds. (2004). The Handbook of Pragmatics. Oxford: Blackwell. Huang, Y. (2017). The Oxford Handbook of Pragmatics. Oxford: Oxford University Press. Lepore, E., and Stone, M. (2010). Against metaphorical meaning. Topoi, 29, 165–180. Lepore, E., and Stone, M. (2015). Imagination and Convention: Distinguishing Grammar and Inference in Language. Oxford: Oxford University Press. Levinson, S. (2000). Presumptive Meanings: The Theory of Generalized Conversational Implicatures. Cambridge, MA: MIT Press. Lewis, D. (1969). Convention. Cambridge, MA: Harvard University Press. Lewis, D. (1979). Scorekeeping in a language game. Journal of Philosophical Logic, 8: 339–359. Repr. in Lewis (1983), pp. 233–249. Lewis, D. (1983). Philosophical Papers. Oxford University Press. Lycan, W. (1995). Philosophy of language. In R. Audi (ed.), The Cambridge Dictionary of Philosophy (pp. 586–589). Cambridge: Cambridge University Press. Mazzone, M. (2015). Constructing the context through goals and schemata: Top-down processes in comprehension and beyond. Frontiers in Psychology, May 19. https://doi .org/10.3389/fpsyg.2015.00651.

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

265

266

DEIRDRE WILSON

Meibauer, J. (2012). What is a context? Theoretical and empirical evidence. In R. Finkbeiner, J. Meibauer, and P. Schumacher (eds.), What Is a Context? Linguistic Approaches and Challenges (pp. 9–32). Amsterdam: John Benjamins. Noveck, I. (2018). Experimental Pragmatics: The Making of a Cognitive Science. Cambridge: Cambridge University Press. Noveck, I., and Sperber, D. (2007). The why and how of experimental pragmatics: The case of “scalar inferences.” In N. Burton-Roberts (ed.), Pragmatics (pp. 184–212). Basingstoke: Palgrave Macmillan. Recanati, F. (2004). Literal Meaning. Cambridge: Cambridge University Press. Roberts, C. (2015). Accommodation in a language game. In B. Loewer and J. Schaffer (eds.), A Companion to David Lewis (pp. 345–366). Hoboken, NJ: Wiley-Blackwell. Schulze, C., and Tomasello, M. (2015). 18-month-olds comprehend indirect communicative acts. Cognition, 136, 91–98. Scott, K., Clark, B. and Carston, R., eds. (2019). Relevance, Pragmatics and Interpretation. Cambridge: Cambridge University Press. Searle, J. (1978). Literal meaning. Erkenntnis, 13, 207–224. Smith, N. V. (1982). Mutual Knowledge. London: Academic Press. Southgate, V., Chevallier, C., and Csibra, G. (2009). Sensitivity to communicative relevance tells young children what to imitate. Developmental Science, 12, 1013–1019. Sperber, D. (2005). Modularity and relevance: How can a massively modular mind be flexible and context-sensitive? In P. Carruthers, S. Lawrence, and S. Stich (eds.), The Innate Mind: Structure and Content (pp. 53–68). Oxford: Oxford University Press. Sperber, D. (2018). Rethinking ostension, (1) and (2). Dan Sperber’s blog, cognitionand culture.net. Sperber, D. (2021). Rethinking common ground. Conference presentation, MK40: Common knowledge, common ground and context in communication. University College London. Sperber, D., and Wilson, D. (1986/1995). Relevance: Communication and Cognition. Oxford: Blackwell. Sperber, D., and Wilson, D. (1987). Précis of Relevance: Communication and Cognition. Behavioral and Brain Sciences, 10, 697–751. Sperber, D., and Wilson, D. (1990). Spontaneous deduction and mutual knowledge. Behavioral and Brain Sciences, 13, 179–184. Sperber, D., and Wilson, D. (1995). Postface to D. Sperber and D. Wilson (1995), Relevance: Communication and Cognition, 2nd ed. Oxford: Blackwell. . Sperber, D., and Wilson, D. (2002). Pragmatics, modularity and mindreading. Mind and Language, 17, 3–26. Repr. in Wilson and Sperber (2012), pp. 261–278. Sperber, D., and Wilson, D. (2005). Pragmatics. In F. Jackson and M. Smith (eds.), Oxford Handbook of Contemporary Philosophy (pp. 468–501). Oxford: Oxford University Press. Repr. in Wilson and Sperber (2012), pp. 1–27. Sperber, D., and Wilson, D. (2008). A deflationary account of metaphors. In R. Gibbs (ed.), The Cambridge Handbook of Metaphor and Thought (pp. 84–105). Cambridge: Cambridge University Press. Sperber, D., and Wilson, D. (2015). Beyond speaker’s meaning. Croatian Journal of Philosophy, 15, 117–149. Stalnaker, R. (1999). Context and Content: Essays on Intentionality in Speech and Thought. Oxford: Oxford University Press. Stalnaker, R, (2002). Common ground. Linguistics and Philosophy, 25, 701–721.

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

11 Relevance Theory and Context

Stojnić, U. (2021). Context and Coherence: The Logic and Grammar of Prominence. Oxford: Oxford University Press. Van der Henst, J.-B., and Sperber, D. (2004). Testing the cognitive and communicative principles of relevance. In I. Noveck and D. Sperber (eds.), Experimental Pragmatics (pp. 141–171). Basingstoke: Palgrave. Repr. in Wilson and Sperber (2012), pp. 279–306. Van der Henst, J.-B., Sperber, D., and Politzer, G. (2002). When is a conclusion worth deriving? A relevance-based analysis of determinate relational problems. Thinking and Reasoning, 8, 1–20. Wharton, T., Jagoe, C., and Wilson, D., eds. (2022). Special issue, Relevance Theory: New Horizons. Journal of Pragmatics, 124. Wilson, D. (2017). Relevance theory. In Y. Huang (ed.), The Oxford Handbook of Pragmatics (pp. 79–100). Oxford: Oxford University Press. Wilson, D. (2019). Relevance theory. In M. Aronoff (ed.), Oxford Research Encyclopedia of Linguistics. Oxford: Oxford University Press. DOI: 10.1093/acrefore/ 9780199384655.013.201. Wilson, D. (2022). Communication, comprehension, and interpretation. In H. Colston, T. Matlock, and G. Steen (eds.), Dynamism in Metaphor and Beyond (pp. 143–155). Amsterdam: John Benjamins. Wilson, D., and Carston, R. (2019). Pragmatics and the challenge of “nonpropositional” effects. Journal of Pragmatics, 145, 31–38. Wilson, D., and Sperber, D. (2002). Truthfulness and relevance. Mind, 111, 583–632. Repr. in Wilson and Sperber (2012), pp. 47–83. Wilson, D., and Sperber, D. (2004). Relevance theory. In L. Horn and G. Ward (eds.), The Handbook of Pragmatics (pp. 607–632). Oxford: Blackwell. Zufferey, S., Moeschler, J., and Reboul, A. (2019). Implicatures. Cambridge: Cambridge University Press.

https://doi.org/10.1017/9781108989275.012 Published online by Cambridge University Press

267

12 The Interplay of Linguistic, Conceptual, and Encyclopedic Knowledge in Meaning Construction and Comprehension Istvan Kecskes 12.1 Introduction Linguistics research has always been characterized by separatist and synergistic endeavors. The debate is about the basic question of whether language is a relatively independent cognitive faculty or an essential part of overall cognitive development. Recent events point to the fact that synergetic endeavors are receiving more attention than ever before. Current debate has been not about the legitimacy of meaning and performance research (usage), but about the extent to which grammar-research and usage-research can be combined, and what role communication plays in shaping grammar. In other words, whether one can and/or needs to support the other, or they should not interact at all. Linguists have never denied that there is a link between structure and usage. Chomsky (1975: 56) wrote that “surely there are significant connections between structure and function”: So this has never been in doubt. Searle (1979) argued that ‘‘it is reasonable to suppose the needs of communication influenced [language] structure.” “I agree,” said Chomsky (1975: 56–58). Newmeyer (2003) claimed that mental grammar contributes to language use, but usage is not represented in the grammar itself. Knowledge of grammatical structure is only one of many systems that underlie usage. Recent linguistic research (e.g., Hauser et al. 2002; Pinker and Jackendoff 2005) differentiates between aspects of language that are special to language code (“Narrow Language Faculty,” NLF hereafter) and the faculty of language in its entirety, including parts that are shared with other psychological abilities found elsewhere in cognition (“Broad Language Faculty,” BLF hereafter). The

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

12 Linguistic, Conceptual and Encyclopedic Knowledge

BLF (All Mechanisms Involved in Language)

NLF (System of Signs)

Figure 12.1 Recent linguistic theory of language

system of signs

conceptual base sociocultural background Figure 12.2 Kecskes’ definition of language

lexicon can be considered as an interface that ties NLF to the other elements of the BLF. See Figure 12.1. Kecskes’ definition (2013) of language attempts to pull the two sides together as follows: “Language is a system of signs operated by a conceptual base that is the reflection of the socio-cultural background in which the system of signs is put to use.” This definition emphasizes the unique relationship of the sociocultural background (BLF) with the system of signs (NLF) through the conceptual base (see Figure 12.2). So, the question is how these three factors: system of sign, sociocultural background, and conceptual base interact with one another.

12.2 Sociocultural Background Knowledge Language can never be all-inclusive to convey every aspect of meaning. Gumperz argued that, as a consequence, some combination of “practical reasoning” and “unstated, taken-for-granted background knowledge” is needed to

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

269

270

ISTVAN KECSKES

fill in what is left unsaid (Gumperz 2001: 216). So what is this “unstated, takenfor-granted background knowledge”? It is sociocultural background knowledge that comprises sociocultural practices and experiences that a member of a speech community (temporary or relatively stable) is and has been exposed to and uses to make sense of the world around him/her. All this is reflected in core common ground, collective salience, and beliefs for individuals in a given speech community, which has a profound effect on language behavior of speech community members. The more homogeneous and stable (relatively) a speech community is, the more significant the effect of the core common ground and collective salience is. Conversely, the more diverse a speech community is, the more limited the reliance on core common ground and collective salience is, the more is there a need for co-construction of common ground. In other words, smooth communication is hardly possible without some, at least, minimal common ground. Speech community members gain sociocultural knowledge from having been and being acculturated in their speech community. This is the basis for core common ground and collective salience that help speech community members to behave as insiders in the community. If this is lacking, or limited, interlocutors may feel like outsiders, as in the case of intercultural communication where interlocutors do not feel and, most of the time, do not want to feel like insiders of an English L1 speaking community. The following conversation between a Korean student and a Chinese student demonstrates this very clearly.1 (1) K:

CH: K: CH: K: CH: K:

CH: K:

And then language problem. Sometimes I obviously look like a foleign . . . foreign person . . . foreigner here . . . so they assume I don’t speak English so they sometimes . . . I don’t know . . . they sometimes don’t understand what I’m saying . . . even though I’m speaking English. It hurts me a lot . . . I don’t know. Could you follow them? Of course. But they find it hard to follow you? Mhmm I don’t know why. I think it’s because of my . . . how I look like you know. I don’t know it hurts me a lot. I don’t think it matters very much because just for your physical appearance. Did you try slowing down your space? Yes eventually they understand I can speak English but still in their mind they have strong strategy . . . I mean . . . I’m sorry . . . stereotypes prejudice like . . . you look foreign. Foreigner. And you probably don’t speak English so they don’t even bother themselves to speak to me.

Both the Korean student and the Chinese student express their frustration about not feeling like members of the Albany speech community in which they have been living. They talk about their physical appearance, their language use 1

Example from Kecskes (2019a).

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

12 Linguistic, Conceptual and Encyclopedic Knowledge

and the prejudice of some members of the speech community. What frustrates them most is being misunderstood. As the Korean student says: “they sometimes don’t understand what I’m saying . . . even though I’m speaking English.” Although he uses the same code system as native speakers do, something is still missing. This “something” is the other side of the language use: The sociocultural background knowledge that would help him use the right intonation, select the preferred words in a given context, use the appropriate situation-bound utterances, understand implications and contextualization cues, including both verbal and nonverbal, and so on. Broadly speaking, everything that goes with a language minus the code system can be considered sociocultural background knowledge. Nevertheless, the nature and functioning of this knowledge – what exactly it includes and how people access and activate the appropriate knowledge – in a given conversational moment is still not clear and is under investigation.

12.3 The Relationship of Sociocultural Background Knowledge to Linguistic Knowledge and Conceptual Knowledge 12.3.1 Understanding the Three Knowledges One of the most complicated issues of present-day linguistics is the relationship of the three knowledges: linguistic knowledge, conceptual knowledge, and encyclopedic knowledge. Language philosophy, theoretical linguistics, cognitive linguistics, computational linguistics, and pragmatics have been addressing this issue from different perspectives. Kecskes’ definition of language (see above) is based on the relationship between these three knowledges. There are three important factors in that definition: system of signs, conceptual base, and sociocultural background. It would be logical to assign linguistic knowledge to the system of signs, conceptual knowledge to the conceptual base, and encyclopedic knowledge to the sociocultural background. This, however, is not that simple. Goddard is probably right when he said: “knowledge of all kinds is integrated in the mind to such an extent that it does not make any sense to partition it into two distinct realms” (1998: 15). But we still need to partition knowledge if we want to understand how language works. Also, this may help us understand better what happens in intercultural interactions when not all these knowledges are equally available. Some works in theoretical linguistics (e.g., Gruber 1985; Kiefer 1990; Bierwisch and Schreuder 1992) talked about the need to distinguish between three types of knowledge: linguistic knowledge, which, roughly speaking, concerns the core meaning (mostly literal meaning) of lexical items, conceptual knowledge, which has to do with the possible modifications, conceptualizations of the core meaning in various actual situational contexts, and encyclopedic knowledge, which comprises the rest, that is, world knowledge that is associated with a word, but which is not immediately relevant to linguistic structure. Take, for instance, the lexical item plate whose core meaning is something like ‘a flat

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

271

272

ISTVAN KECSKES

dish.’ The fact that this lexical item can be used also as a non-physical entity, as in the sentences: “I have too much on my plate” or “dinner is twenty dollars a plate,” belongs to our conceptual knowledge. Actually, conceptual knowledge comprises the different possible senses of the lexical item. Everything beyond that, like the knowledge that plate is used to eat food from, that it is usually made of porcelain, metal, or some artificial material, and it is usually oval or round belong to our encyclopedic knowledge. Another example can be the verb shoot. Its core meaning is ‘kill or wound (a person or animal) with a bullet or arrow.’ Our conceptual knowledge comprises different other possible senses of the label “shoot,” such as “to shoot questions at someone,” or fling (“The volcano shot lava into the air”) or to direct suddenly or swiftly as in “He shot a smile at his girlfriend.” At the same time, there are some things that we know about the word shoot which, however, do not play any role (at least most of the time) in its semantic interpretation. Beyond what was said above, we know that we need an instrument or tool to execute the action shoot, like a weapon or a volcano or metaphorically a laugh. We also know that shoot presupposes a swift and rapid move. This approach to three different knowledges makes sense, but how does this relate to what we call “sociocultural background knowledge”? Can sociocultural background knowledge be equated with encyclopedic knowledge, as is usually done in some of the existing literature (e.g., Fillmore 1982; Croft and Cruse 2004; Evans 2006)? One can argue that sociocultural background knowledge comprises everything we know about the world. So, should we distinguish conceptual knowledge from encyclopedic knowledge, or should we handle them as one entity? The answer to this question is not straightforward, but we can look at intercultural rather than L1 communication for an answer. We know that linguistic knowledge is relatively easy to define: that is coded, and (relatively) standardized. Languages do not differ that much with respect to core meanings of lexical items denoting basic objects and actions in the world, but they may exhibit essential differences as to conceptual knowledge and, to some extent, encyclopedic knowledge. What the core meaning of the verb cut denotes ‘make an opening, incision, or wound in something with a sharp-edged tool or object’ exists in many languages. However, the various conceptualizations of that action are at odds in different languages. Take, for instance, cut to the chase, which indeed is a very American way to mean ‘make your point.’ Likewise, there is no English translation for Hungarian megvágtam egy kis pénzzel, which means ‘I squeezed some money out of him’ in which the literal equivalent of cut (vág) is used.

12.3.2 Approaches to Separating Linguistic Knowledge from Conceptual Knowledge Linguistic theories (e.g., generative linguistics and lexical semantics) usually focus on and discuss linguistic knowledge and encyclopedic knowledge. Linguistic knowledge (dictionary knowledge) is supposed to cover the

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

12 Linguistic, Conceptual and Encyclopedic Knowledge

idiosyncrasies of particular words, whereas encyclopedic knowledge covers everything regarding the underlying concepts. In cognitive linguistics, however, semantic representations which constitute the semantic pole of a linguistic sign are equated with “conventionalized conceptualizations” (Langacker 1988: 94). This view integrates encyclopedic knowledge into the lexicon. So, there is no separate linguistic knowledge and encyclopedic knowledge. In cognitive linguistics, meaning emerging from language use is a function of the activation of conceptual knowledge structures as guided by context. Consequently, there is no principled distinction between semantics and pragmatics (e.g., Fauconnier 1997; Evans 2006; Paolucci 2021). Thus, language is seen as a repository of world knowledge and a structured collection of meaningful categories which help us deal with new experiences and store information about old ones (Geeraerts 1997). It can be deduced that our linguistic knowledge is rooted in our experience with the world; this is what language in use expresses. So, if language is a system for categorization of the world, one can argue that there does not seem to be any need to postulate a systemic and structural level of meaning that differs from the level where world knowledge is associated with linguistic forms. However, each language categorizes the same world out there differently (see Van Olmen and Tantucci 2022). Americans make money, Hungarians search for money (pénzt keres), and Russians work for money (зарабатывать). What is more, Slobin (1997) argued that language is a transmitter of real-world experiences and that these experiences are filtered through language into verbalized events. If that were not the case, second language speakers/learners would not face the constant problem of linguistic and conceptual differences in Chinese English, Bulgarian English, Indian English, German English, etc. The differences between languages are most significant at the conceptual level. There is no one-toone relationship between a word and a concept in languages, and especially not across languages. This is where the different world views of different speech communities are conventionalized and developed into a relatively standardized code system (cf. He 2021; Hopkinson 2021). That is the main reason why intercultural communicators consider linguistic knowledge as relatively similar in each variety of English, no matter what their L1 is. This is why they treat the code system as their core common ground. In a way everybody who speaks English uses the same code system. For instance, English L2 users all know at a particular level of proficiency the semantics of the word patronize – to act as a patron of support – such as in the sentence “they patronized the local stores.” However, what not all L2 speakers may know is the conceptual load that is attached to this word depending on its direct object. If the direct object is inanimate – take, for instance, the sentence above – there is no problem: the meaning is neutral or positive. But if the direct object is animate the meaning becomes negative. It refers to a condescending treatment, some kind of superiority as in the following encounter: (2) Kelly: Bill:

I am sure you did your best. The accident was not your fault. Please do not patronize me. I will take the responsibility.

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

273

274

ISTVAN KECSKES

Bill does not like Kelly’s patronizing tone. He is fully aware that he was wrong and is ready to take the responsibility. In most cases, linguistic knowledge in English is available for the L2 speakers (depending on their proficiency), while conceptual knowledge may or may not be accessible.2 This is where the major difference lies between L1 use and L2 use. While speakers in L1 seek and establish common ground on the conceptual level, L2 speakers usually do that on the linguistic level. The separation of linguistic knowledge and conceptual knowledge is quite clear in L2 (e.g., Kecskes 2003; Baker 2015). Of course, this does not mean that common ground cannot be established at the conceptual level in L2 communication. In fact, what happens is that the more time a temporary speech community spends together the more common ground is created at the conceptual level, just as in L1. Cognitive linguists have attempted to keep linguistic knowledge separate from conceptual knowledge. Bierwisch (1981) argued that there is a clear distinction between a word’s purely linguistic meaning and the interpretation that a word may have relative to conceptual knowledge (cf. Bierwisch 1981; Bierwisch and Schreuder 1992; Taylor 2000). This two-level model distinguishes between a linguistic-semantic level of meaning and an essentially nonlinguistic, conceptual level of interpretation. Encyclopedic knowledge is provided only in the act of interpreting the word in its context, while the word meaning itself is underdetermined with respect to its possible interpretations in different kinds of context. This position seriously constrains polysemy in the mental lexicon and relegates it to the conceptual level. So, words have a number of meanings because of their various possible conceptual interpretations. It could be argued that the goal of dictionary knowledge (linguistic knowledge) is only to collect all possible interpretations under a unitary semantic entry. This corroborates Taylor’s point that “on the two-level model, interpretation of an expression emerges through the interaction of the unitary semantic representation with conceptual information, relative to a context” (2000: 130). Consequently, the dictionary knowledge (linguistic knowledge) – encyclopedic knowledge controversy opposes two extreme conceptions of word definitions. According to the dictionary approach, a word is described in terms of linguistic elements only, without recourse to world knowledge, whereas an encyclopedic definition includes “an indication of the different species or different stages of the object or process denoted by the word, the main types of behavior of this object or process” (Mel’cuk and Zholkovsky 1984: 88). Let me highlight that the debate on the separation of the conceptual and linguistic-lexical level does not necessarily go on between generativists and cognitivists, although it concerns the basic difference between the two perspectives. Rather, it involves linguists from both sides. For instance, Bierwisch is a generativist, and Wierzbicka is a cognitivist, still they are on the same side 2

This may depend on factors such as frequency of occurrence or noticing.

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

12 Linguistic, Conceptual and Encyclopedic Knowledge

although their reasons are unalike. Also, several researchers, who are considered as cognitive linguists, argued for keeping lexical semantics separate from conceptual semantics. Haiman, for example, claimed that while theoretically untenable, the distinction between dictionaries and encyclopedias has “the happy property of working very well in practice” (1980: 355). Similarly, Gibbs (1996) claimed that cognitive linguists would not argue against the idea that there is a mental lexicon that might be independently accessed during sentence processing. In his view, part of the confusion about the role of cognitive structure in language use and processing results from the failure to distinguish between different levels at which cognition and language interact. Moreover, Wierzbicka (1996) strongly believes that word meaning can and has to be defined at the linguistic (lexical) level. She argued that the acquisition of another language demonstrates very well how important lexical semantics is. She said that “for anyone seriously trying to learn another language and understand another culture, the proposition that words cannot be defined can hardly be anything but bad news (p. 256). Also, she said that the “belief that a dictionary definition represents nothing other than a selection from a (real or imaginary) encyclopedic entry, with the choice being determined by practical considerations and having no theoretical justification, leads to stagnation in lexical semantics” (p. 336). Wierzbicka clearly makes the distinction between linguistic knowledge and encyclopedic knowledge for she thinks that linguistic knowledge is essentially shared between all speakers of a particular language, while encyclopedic (real-world) knowledge is not. She pointed out that there is linguistic evidence that the human mind itself draws a distinction between a mental dictionary and a mental encyclopedia (p. 344). On that account, the nature of available linguistic evidence is summarized by Taylor (2000: 16) who argued that acquisition is not a process of building up a concept from its constituent parts, but it consists in the gradual elaboration of a knowledge network. Another important observation comes from Harnad (1990) who said that “once one has the grounded set of elementary symbols provided by a taxonomy of names (and the iconic and categorical representations that give content to the names and allow them to pick out the objects they identify), the rest of the symbol strings of a natural language can be generated by symbol composition alone, and they will all inherit the intrinsic grounding of the elementary set” (pp. 343–344). This implies that higher-order symbols can be interpreted without direct acquaintance with reality. We should not think, however, that these higher-order symbols are not grounded in experience at all. They, together with their underlying symbolic representations, are derived from the sensory representations and thus indirectly grounded in experience. Actually, the real issue here is not whether the content of lexical units is grounded in experience or not, because in most cases they are grounded in experience either directly or indirectly. What is important for us is that linguistic symbols can entirely be cut off from their original source and live the life of their own, which is independent from the conceptual system to a particular extent and from a particular perspective. Just think about

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

275

276

ISTVAN KECSKES

this issue from a second language perspective. Although L2 users have access to the symbols (words) in English, they have limited direct experience with the target language reality in which those symbols are grounded for the native speakers. Consequently, English as a Lingua Franca (ELF) speakers may have linguistic knowledge but lack – or have only limited access to –conceptual knowledge. Based on the instances of mono- and multilingual development, it would be far-reaching to claim that linguistic units (words) reflect specific, autonomous linguistic knowledge that is separate from various types of conceptual knowledge. However, it might very well be that some linguistic knowledge that is attached to the word form is partly autonomous from the rest of our conceptual system (see Gibbs 1996; Kecskes 2003, 2019b, 2021). We can take the expression chicken out as an example. The basic meaning of the expression is ‘to decide not to do something because you are too frightened.’ No actual situational context can cancel the negative conceptual load attached to the expression. For example, in the utterance “at the last minute Sally chickened out of the race,” no contextual support is needed for the expression to call upon the negative conceptual load. Consequently, we can say that the conceptual load is already tied to the semantics of the lexical unit. In other words, the conceptual load becomes its word-specific semantic property. In that regard, Croft and Wood argued that “it is not the case that any time we think we must conceptualize our experience the way that our language requires us to. But it is the case that any time we express our thoughts in language, we must conceptualize our experience in the way that our language requires us to. Cognition may be linguistically neutral, but language is not semantically neutral” (2000: 55). Slobin (1991) also made a similar point when he described “thinking for speaking” as the appropriate domain for the influence of language on thought. The issue here is not necessarily whether language influences thought or not, but to what extent. The world is always given to us under some construal for our purposes. Thus, the remaining question is: How much of this construal is provided by the language we speak?

12.3.3 The Multilingual Perspective To answer this question, we must go back to Wierzbicka’s thought which reflects a second language perspective. She argued that basically the process of second language learning is the proof that the proposition that words cannot be defined is wrong. She rejects the belief that a dictionary definition represents nothing other than a selection from a (real or imaginary) encyclopedic entry, with the choice being determined by practical considerations and having no theoretical justification. Wierzbicka makes the distinction between linguistic knowledge and encyclopedic knowledge because she is convinced that linguistic knowledge is essentially shared between all speakers of a particular language, while encyclopedic (real-world) knowledge is not. So far, she is right, but she does not clarify the nature of what she calls

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

12 Linguistic, Conceptual and Encyclopedic Knowledge

“encyclopedic knowledge” because she equates it with what I have called sociocultural background knowledge. So, the problem is not solved yet. But we are on the right track because we are using a multilingual view, the perspective of L2 or Lx speakers. When bringing a multilingual view into the debate, we should say that the relationship between the three knowledge types is way more complex than just a simple trichotomy. The main problem is that world knowledge is available for human beings in two ways: (1) A part of world knowledge is encapsulated in the lexical items that we use in a language. That is why cognitive linguists argue that it does not make sense to separate linguistic knowledge from encyclopedic knowledge. In a way they are right because when we use a word, that triggers what they call linguistic knowledge (grammatical information about the word, its relations to other words, synonymy-antonymy, etc.) as well as encyclopedic knowledge, which comprises what Melcuk and Zholkovsky (1984) called an indication of the different species or different stages of the object or process denoted by the word, the main types of behavior of this object or process. This kind of encyclopedic knowledge sounds like something encoded in the word which we called “conceptual knowledge.” But, then, where is the other part of world knowledge that is not encoded in the lexical items? (2) It looks like the “second” type of world knowledge is not directly connected with our language system; rather, it comprises behavior practices, models, and social frames that are called upon when we use a language. We will take a closer look at these two sides of sociocultural background knowledge – conceptual knowledge and encyclopedic knowledge – in the following section.

12.4 A Possible Model of Knowledge Distribution 12.4.1 The Relationship of Three Knowledges in the Model I propose a model in which we have the linguistic knowledge on one side, and the sociocultural background knowledge (world knowledge) on the other side. There is constant interaction between the two sides in any language use. For analytic reasons, within the sociocultural background knowledge we should distinguish between conceptual knowledge and encyclopedic knowledge, as Figure 12.3 shows. According to this model, meaning is constructed in the dynamic interplay of actual situational context and lexical items, with the context representing the actual, present, situational, ever-changing side of sociocultural background and the lexical item(s) embodying previous experiences and relations in the

Linguistic knowledge

Sociocultural background knowledge conceptual knowledge ---- encyclopedic knowledge

Figure 12.3 Conceptual knowledge and encyclopedic knowledge

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

277

278

ISTVAN KECSKES

sociocultural background (see Kecskes 2003, 2013a). This is how the three knowledges are intertwined and function together. The lexical items with their semantic properties (linguistic knowledge) represent prior reoccurring experience (conceptual knowledge), and the actual situational context triggers the other part of world knowledge that we previously called encyclopedic knowledge. The difference between the two types of sociocultural background knowledge is that the conceptual knowledge part is immediately tied to linguistic knowledge, while the other type of sociocultural background knowledge (encyclopedic knowledge) is called upon as needed in the actual language use. Second language users have the least problem with linguistic knowledge that is mainly reflected in their language proficiency. Linguistic knowledge is learnable both systematically (schools, institutions) or asystematically (natural environment). It represents core knowledge that L2 interlocutors share and rely on, no matter which variety of English (Chinese, German, French, Russian, etc.) they speak. This linguistic core knowledge gives some core common ground for the interlocutors to rely on. However, conceptual knowledge is different. The problem is that conceptual knowledge that is tied to the semantics of a lexical item is quite conservative and occasionally seems to be old-fashioned or obsolete because it may be rooted in the past experience of the given speech community (cf. Kecskes 2019a; Werkmann Horvat et al. 2021). Language only partly reflects the actual social world around us because of this delay in conceptual coding. It will take a while for some neologisms such as sharrow (road marking), procott (intentionally buy things), or placemaking to be fully part of common conceptual knowledge. Words with their semantics are needed not only to symbolize concepts but also to stabilize them, keep them tidy, and make them definable. This, however, can only be done if a word or formula has some elements in its meaning that are relatively constant, that is, part of the language system and not totally dependent on actual use and context. So I need to quote Leibniz’s argument: “Si nihil per se concipitur, nihil omnino concipietur,” that is, ‘if nothing can be understood by itself nothing at all can ever be understood (1903: 430).” Human communication would hardly be possible if words did not have any relatively permanent elements in their meaning and meaning formation would only be the matter of online, in-process production. If, however, these relatively constant elements of meaning really exist, cognitive linguists’ claim that word meanings may not apply at all without the supporting cognitive structures and mechanisms becomes questionable (at least to some extent).

12.4.2 Words and Concepts (the Interplay of Linguistic Knowledge and Conceptual Knowledge) The central element of lexical pragmatics is the way we understand the relationship of word and concept, that is, linguistic knowledge and conceptual knowledge. This issue needs a deeper analysis because this is where languages

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

12 Linguistic, Conceptual and Encyclopedic Knowledge

differ from each other, and it is the key to understanding difficulties in L2 and Lx use. The conceptual system and linguistic system are related through the interdependence of concepts and words. Most of the controversy centers around the question of how concepts and words are related to one another. A natural language possesses an inventory of lexical forms, and these are mapped onto the concept network (Cruse 1992: 290). The question is: how does this mapping take place? It is true that concepts are represented by words, but this representation is rather contradictory. The contradiction is caused by two facts. First, there is no one-to-one relationship between the concept and the word: The overall concept usually extends beyond the sections labeled with a word. Second, the semantic domain of a word also contains specific properties which are not present in the concept. As a result, sometimes there are several different lexical routes to one and the same concept. While discussing the relationship between concept and word, a distinction needs to be made between two approaches. Kurt Baldinger described this distinction in the following way: “Semasiology . . . considers the isolated word and the way its meanings are manifested, while onomasiology looks at the designations of a particular concept, that is, at a multiplicity of expressions which form a whole” (1980: 278). The semasiological perspective focuses on the word and explores how concepts (conceptual meanings) are associated with that word. So, semasiology targets polysemy and the multiple applicability of a lexical item. In contrast, the onomasiological approach takes the concept as its starting point and investigates what words can express a particular concept. The subject of onomasiology is synonymy, name-giving, and the selection of an expression from among a number of possible alternatives (cf. Geeraerts 1997). It is clear from this distinction that semasiology refers to processes of meaning, while onomasiology involves naming. As a result of the thought–word interaction in language production, thought usually undergoes several changes as it turns into speech. During this process conceptual categories (concepts) are mapped on linguistic categories (words). Here is an example of how this happens when English as an L2 is used. (3)

A male native speaker of English is having a conversation with a Thai female student. He wants to know whether she likes sports or not. The Thai student initially says she likes “football” and then immediately replaces that with “soccer.” Non-Native Speaker (Thai woman) – Native Speaker (American man) NS: . . . . So do you like sports? NNS: I love sports. NS: What sports do you love? NNS: I love football . . . Ops- in here we call soccer. NS: Oh okay.

When replacing football with soccer, the NNS probably wanted to provide the NS with an accurate identification of the sport she liked. According to collective salience of her culture, the Thai woman used the right label “soccer.”

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

279

280

ISTVAN KECSKES

However, she realized that that label for the concept might not work in the actual situation context for the American man. In addition, she likely anticipated that if NS understood her to be identifying American football, he could ask follow-up questions that NNS would be unable to answer. In the NNS’s mind “football” is the label that has priority based on her L1 experience. However, she realized that for the American the word has another most salient meaning. To avoid misunderstanding, the Thai student changed the word for “soccer” that matched the actual situational context.

12.4.3 Encyclopedic Knowledge Now we need to examine how the uncoded part of sociocultural background knowledge affects language production. In our understanding, encyclopedic knowledge refers not only to extralinguistic information about the world but also to a structured system of knowledge, organized as a network. Rosch’s (1977) schema theory is a theoretical framework which has been developed in the cognitive sciences over the last several decades. The central concept, “schema,” has gone by a number of other names, including “cultural model” (Holland and Quinn 1987; D’Andrade 1992), “frame” (Minsky 1975; Fillmore 1982), “mental model” (Johnson-Laird 1983), “idealized cognitive model” (Lakoff 1987), “folk model” (D’Andrade 1987), and “script” (Schank and Abelson 1977). According to most of these theories, encyclopedic knowledge is mostly represented in cultural models and schemas that provide scenarios or action plans for individuals on how to interpret speech situations and behave in a particular situation or how to interpret the behavior of others in one or another situation. Cultural schemas for social interaction are cognitive structures that contain knowledge for face-to-face interactions in a person’s sociocultural environment. The following encounter between a Hungarian man and three Chinese women illustrates how these cultural models work. (4)

The four people are going into a restaurant in Albany, NY. The Hungarian stops at the desk of the receptionist, while the three Chinese women are walking into the restaurant looking for an empty table. H: Where are you going? CH1: There is a vacant table over there. H: Please wait till we are seated.

According to US customs, when someone walks into a restaurant they have to wait till a member of waiting staff walks them/their party to a table. In China there is a different model: usually (though not always) it is acceptable to occupy whichever table is available. Not knowing the US cultural model, the three Chinese women acted according to their own. I argued before that cultural models provide a kind of reference library for possible plans of action for oneself or possible interpretation of the actions of others (see Kecskes 2013). These models are not learned directly as models but are inferred by each of us from what we see and experience with those other people in our speech

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

12 Linguistic, Conceptual and Encyclopedic Knowledge

community. But what we experience are never the models themselves. What we infer from experience is pieces of information, images, features that keep a model together. What we infer depends directly on what parts of the given scenario are saliently and repetitively present in the messages we experience for us to pull out the regularities on which we will base our construction of the model behind them. Thus, systematic and repeated changes in speech or cultural behavior in one generation will be learned by the next generation as part of the givens of language or culture. The core of cultural models that people in the same speech community share changes diachronically through systematic and repeated shifts that can come from sociopolitical changes, technological changes, environmental changes, and the like. For instance, currently gender-specific expressions such as policeman, chairman, and freshman are replaced by gender-neutral expressions such as police officer, chairperson, and first-year student. Of course, usually there is trouble if people try to force the linguistic system to accept something that is conceptually unfounded. An example can be Canadian prime minister Justin Trudeau’s attempt to change mankind for peoplekind. These attempts are the result of a sociopolitical practice that prefers the use of gender-neutral terms in language. Those sociopolitical practices may prevail if the given speech community will find them important to keep and use them repeatedly.

12.5 Sociocultural Background Knowledge in L2 To justify the proposed model, we need to see how it works for L2 and ELF users. Baker (2009) argued that linguistic and cultural forms expressed through ELF are likely to be hybrid, dynamic, and continuously adapting to local needs, global influences, and the demands of communicating across cultures. He lamented that apart from Meierkord’s (2002) study, there is little empirical evidence from ELF settings, and none from expanding circle environments,3 that is specifically concerned with understanding how cultural frames of reference and communicative practices operate in such a liminal manner in intercultural communication through English. The situation has not changed much as we can see in Baker’s new book (2015). It is worth looking at House’s argument (2014) who said that when English is used as a lingua franca for communication, it is in principle neutral with regard to the different sociocultural backgrounds of its users. Kecskes (2019a) argued that House may not be right. Risager (2006) took House’s argument even further. She spoke not only about neutrality of the sociocultural background but also about separation of language and culture. Building on the understanding of dynamism in critical theory, or as Risager views it, the complex and global flow of language and culture, she claimed that from the perspective of users of English as a second or 3

Kachru refers to the “extended circle of English” where English is learned/used as a foreign language such as Russia, Germany, and China.

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

281

282

ISTVAN KECSKES

foreign language, languages and cultures can be separated. In order to explain what she means by this, she makes a distinction between cultures and languages in the generic sense and in the differential sense (Risager 2006: 4–5). In the generic or universal sense, language and culture are intertwined in the way that some theories such as linguistic relativity proposed (e.g., Whorf 1939; Gumperz and Levinson 1996). Language is always an enactment and embodiment of culture, so the two cannot be separated. However, Risager believes that when examining specific languages and cultures such as English, in the differential sense, language and culture can be separated. Her main argument is that all languages, and especially international languages such as English or Spanish, can in practice (i.e., during actual instances of use) take on new cultural meanings. She refers to those as languacultures (Risager 2006: 110), which depend on users and context. In this sense “the link between language and culture is created in every new communicative event” (Risager 2006: 185). As a result, a language such as English will have as many languacultures as there are speakers of the language, and in this sense, there is no identifiable culture to which a language is inseparably tied. She goes even further than that. She believes that at the individual psychological level, that is, at the level of an individual’s linguistic competence or resources, language and culture are again inseparable, and develop in tandem based on the individual’s life experiences. She believes that this individual perspective may have led to confusion over the relationship between specific languages and cultures, and consequent claims that languages such as English are in some way directly tied to British or US culture as in strong forms of linguistic relativity (Whorf 1939). What is important for us here is that while maintaining that languages are never culturally neutral for their users, Risager’s approach to the relationship between languages and cultures allows English use in ELF to be separated from the cultures of the inner circle countries. However, this is not quite so if we accept what I suggested earlier: a significant part of inner circle culture (whether it be Australian, American, or British English) is encapsulated in the lexical items in English. So Chinese, German, Russian, Korean, etc. learners/users of English have this conceptual knowledge available to them to a certain extent depending on their proficiency level. So, it is a mistake to suggest that target language culture is separated from the use of English as a lingua franca. The problem is that the ELF interlocutors cannot rely on that conceptual knowledge as the inner circle interlocutors can because it is unknown how much of this conceptual knowledge of inner circle English the outer and expanding circle interlocutors share. This is especially true for formulaic language, as the example below clearly demonstrates. (5)

Jianwei, Mary, Andy, and Liya are sitting at a table and eating. Jianwei: Can I eat that last piece of sandwich? Liya: Be my guest. Jianwei: But it’s not yours.

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

12 Linguistic, Conceptual and Encyclopedic Knowledge

This example shows quite well why we cannot separate inner circle culture from L2 and ELF. There are three non-native speakers and one native speaker at the table. The Chinese student asks for permission to eat the last piece of sandwich. The Russian student uses a formulaic expression to respond, which is a situation-bound utterance, encouraging or allowing someone else to take action like in the following encounter: A: Do you mind if I order another glass of beer? B: Not at all – be my guest. The possessive pronoun my in the set phrase “be my guest” does not necessarily refer to possession. However, the Chinese student in example (5) processes the utterance based on its literal meaning because he may not know the conceptual load of the situation-bound utterance in this figurative use “be my guest.” So, a slight misunderstanding occurs. This issue was discussed by Kramsch from a second language acquisition perspective. Kramsch (1993: 233) viewed second language communication as operating in a “third place” between the users’ first language and culture (L1/C1) and the target language and culture (L2/C2) but being part of neither. She claimed that second languages operate along a ‘‘cultural faultline’’ (p. 205) in which communicative practices are freed from the norms of both L1/C1 and L2/C2, which opens up new perspectives on languages and cultures. Baker (2015) suggested that the notion of liminality, as proposed by Rampton (1995), shares many features with Kramsch’s third places. Rampton studied communication between different ethnic groups within the United Kingdom. He identified ‘‘liminal moments’’ or ‘‘crossings’’ (Rampton 1995: 167) when language users who are not part of a given language community adopt that language for their own purposes or needs at least temporarily. This results in a code-alteration (p. 280) of the L2 by minority or outside users. Rampton argued that such crossings are quite common in the L2 classroom and are a part of L2 teaching and learning practice. Relying on Rampton’s approach, Brumfit (2006) proposed that L2 learning and use can be considered a liminal process that leads users into new areas, in which languages and their cultural codes are unique to each individual and communicative encounter. Baker (2015) emphasized the importance of these approaches and noted that although none of these authors had talked specifically about ELF communication, the notion of third places and liminality have much to offer in highlighting the fluid, dynamic, and novel communicative practices and language–culture connections that might be expected in such contexts. Definitely, both the third space approach and liminality are relevant to our discussion about the interplay of linguistic, conceptual, and encyclopedic knowledge in meaning production and comprehension in ELF. However, the question is whether ELF users do something different from what L1 users do. Seemingly, the answer is “yes.” But this is just the tip of the iceberg. What is beneath is more complex. L1 users rely on the relative, synchronically

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

283

284

ISTVAN KECSKES

available “end-product” of language development4: a relatively coherent system of signs and a relatively coherent sociocultural background knowledge that serves as a support system. This is what LaPolla (2010: 13) claims about language development: Language is the cumulative result of the actions of many individual humans, but their actions are not with the intention to create language; language is the unintended byproduct of their attempts to communicate effectively (constrain the addressee’s inferential process effectively) on an individual level. It forms as if guided by some invisible hand, much the way economies and paths in fields develop.

But, how is that related to ELF? We should think about the following: ELF users enter into a communicative encounter with their full knowledge of L1, different levels of linguistic, conceptual, and encyclopedic knowledge of L2 and an individual understanding of the actual situational context. From there, they start the cumulative process, which is usually very short: to put together what they have so as to create core common ground that they can rely on while communicating with each other. Although this is a short process (ELF speech communities are only temporary), people generally try to do similar things to what they do in L1, but some of those things are more articulated than in L1. So, ELF users also try to primarily rely on prefabricated language (the idiom principle works), and if they do not have enough of it, they produce new ones. They seek common ground, and if they do not have enough of it, they create some. They try to make sense of the actual situation context, and if that does not fit in their existing repertoire, they co-construct one, and so on and so forth. So one could argue that they create their own variety of ELF for the time they spend together. And in this endeavor, it is linguistic knowledge that helps them most, since that is what they all share to some extent. Conceptual knowledge is mostly co-constructed on the basis of this shared linguistic knowledge.

12.6 Summary In this chapter we discussed the relationship and interplay of three knowledges: linguistic knowledge, conceptual knowledge, and encyclopedic knowledge. A model was proposed in which there is linguistic knowledge on one side, and sociocultural background knowledge (world knowledge) on the other side with constant interaction between the two sides. For analytic reasons, within the sociocultural background knowledge there was a distinction made between conceptual knowledge and encyclopedic knowledge. According to this model, meaning is constructed in the dynamic interplay of actual situational context and prior context encapsulated in lexical items. The context represents the 4

There is no real end-product because language keeps changing.

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

12 Linguistic, Conceptual and Encyclopedic Knowledge

actual, present, situational, ever-changing side of sociocultural background, and the lexical item(s) used in the interaction encapsulate previous experiences and relations in the sociocultural background. It was claimed that differences between the three types of knowledges are most visible in L2 or ELF use. Access to the three knowledges represented by the target language is a graded phenomenon for L2 and ELF users, with the most accessible knowledge being the linguistics code, followed by encyclopedic knowledge. L2 interlocutors have the least access to conceptual knowledge tied to the target language. The problem is not just that they do not have enough experience in the target language to develop enough of that type of knowledge, but also that when interlocutors use L2 and ELF they cannot be sure how much of that conceptual knowledge they share with their partners. The interplay of three knowledges requires more research both in L1 and L2 communication not only to further clarify their interrelations but also to analyze the outcome of their interplay in different types of speech communities.

References Baker, W. (2009). The cultures of English as a Lingua Franca. TESOL Quarterly, 43(4), 567–592. Baker, W. (2015). Culture and Identity through English as a Lingua Franca. Boston/ Berlin: De Gruyter Mouton. Baldinger, K. (1980). Semantic Theory: Towards a Modern Semantics. Oxford: Blackwell. Bierwisch, M. (1981). Basic issues in the development of word meaning. In W. Deutch (ed.), The Child’s Construction of Language (pp. 341–387). London: Academic Press. Bierwisch, M., and Schreuder, R. (1992). From concepts to lexical items. Cognition, 42, 23–60. Brumfit, C. (2006). A European perspective on language as liminality. In C. MarMolinero and P. Stevenson (eds.), Language Ideologies, Policies and Practices. Language and Globalization (pp. 28–43). Palgrave Macmillan, London. https://doi .org/10.1057/9780230523883_3. Chomsky, N. (1975). The Logical Structure of Linguistic Theory. New York: Plenum Press. Croft, W., and Cruse, D. A. (2004). Cognitive Linguistics. Cambridge University Press. Croft, W., and Wood, E. J. (2000). Construal operations in linguistics and artificial intelligence. In L. Albertazzi (ed.), Meaning and Cognition: A Multidisciplinary Approach (pp. 51–78). Amsterdam: John Benjamins. Cruse, D. A. (1992). Antonymy revisited: Some thoughts on the relationship between words and concepts. In Frames, Fields, and Contrasts (pp. 289–306). Hillsdale, NJ: Lawrence Erlbaum. Cruse, D. A. (2002). Language, meaning and sense: semantics. In N. Collinge (ed.), An Encyclopedia of Language (pp. 87–104). London: Routledge. D’Andrade, R. (1987). A folk model of the mind. In D. Holland and N. Quinn (eds.), Cultural Models in Language and Thought (pp. 113–147). Cambridge: Cambridge University Press.

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

285

286

ISTVAN KECSKES

D’Andrade, R. (1992). Schemas and motivation. In R. D’Andrade and C. Strauss (eds.), Human Motives and Cultural Models (pp. 23–44). New York: Cambridge University Press. Evans, V. (2006). Lexical concepts, cognitive models and meaning construction. Cognitive Linguistics, 17(4), 491–534. Fauconnier, G. (1997). Mappings in Thought and Language. Cambridge: Cambridge University Press. Fillmore, C. J. (1982). Frame semantics. In Linguistic Society of Korea (ed.), Linguistics in the Morning Calm (pp. 111–137). Seoul: Hanshin Publishing Co. Geeraerts, D. (1997). Diachronic Prototype Semantics. Oxford: Clarendon Press. Gibbs, R. (1996). Why many concepts are metaphorical. Cognition, 61, 309–319. Goddard, C. (1998). Semantic Analysis: A Practical Introduction. Oxford: Oxford University Press. Gruber, J. S. (1985). Lexical, conceptual, and encyclopedic meaning. Quaderni de Semanlica, 2, 254–267. Gumperz, J. J. (2001). Contextualization and ideology in intercultural communication. In A. D. Luzio, S. Gunthner, and F. Orletti (eds.), Culture in Communication: Analysis of Intercultural Situations (pp. 35–54). John Benjamins. Gumperz, J. J., and Levinson, S., eds. (1996). Rethinking Linguistic Relativity. Cambridge: Cambridge University Press. Haiman, J. (1980). Dictionaries and encyclopedias. Lingua, 50, 329–357 Harnad, S. (1990). The Symbol Grounding Problem. Physica D, 42, 335–346. Hauser, M. D., Chomsky, N., and Fitch, W. T. (2002). The faculty of language: What is it, who has it, and how did it evolve? Science, 298, 1569–1579. He, Sui. 2021. Cognitive metaphor theories in translation studies: Toward a dual-model parametric approach. Intercultural Pragmatics, 18(1), 25–52. Holland, D., and Quinn, N., eds. (1987). Cultural Models in Language and Thought. Cambridge: Cambridge University Press. House, J. (2014). Managing Academic Institutional Discourse in English as a Lingua Franca. Amsterdam/Philadelphia: John Benjamins. Hopkinson, Christopher. (2021). Realizations of oppositional speech acts in English: A contrastive analysis of discourse in L1 and L2 settings. Intercultural Pragmatics, 18(2): 163–202. Johnson-Laird, P. N. (1983). Mental Models. Cambridge, MA: Harvard University Press. Kecskes, I. (2003). Situation-Bound Utterances in L1 and L2. Berlin/New York: Mouton de Gruyter. Kecskes, I. (2013). Intercultural Pragmatics. Oxford: Oxford University Press. Kecskes, I. (2019a). English as a Lingua Franca: The Pragmatic Perspective. Cambridge: Cambridge University Press. Kecskes, I. (2019b). Impoverished pragmatics? The semantics–pragmatics interface from an intercultural perspective. Intercultural Pragmatics, 16(5): 489–517. Kecskes, I. 2021. Processing implicatures in English as a Lingua Franca communication. Lingua, 256. https://doi.org/10.1016/j.lingua.2021.103067. Kiefer, F. (1990). Linguistic, conceptual and encyclopedic knowledge: Some implications for lexicography. In T. Magay and J. Zigány (eds.), BudaLEX ’88 Proceedings: Papers from the 3rd International EURALEX Congress. Budapest: Akademiai Kiado.

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

12 Linguistic, Conceptual and Encyclopedic Knowledge

Kramsch, C. (1986). From language proficiency to interactional competence. The Modern Language Journal, 70(4), 366–372. Kramsch, C. (1993). Context and Culture in Language Teaching. Oxford: Oxford University Press. Lakoff, G. (1987). Women, Fire and Dangerous Things: What Categories Reveal about the Mind. Chicago: University of Chicago Press. Langacker, R. W. (1988). An overview of cognitive grammar. Topics in Cognitive Linguistics, Vol. III. Amsterdam: John Benjamins. LaPolla, R. J. (2010). Arguments against “Subject” and “Direct Object” as Viable Concepts in Chinese. Frankfurt: Universitätsbibliothek Johann Christian Senckenberg. Leibniz, G. W. (1903). Opuscules et fragments inédits de Leibniz: Extraits des manuscrits de la Bibliothèque royale de Hanovre, ed. Louis Couturat, Paris: Félix Alcan. Meierkord, C. (2002). “Language stripped bare” or “linguistic masala”? Culture in lingua franca communication. In K. Knapp and C. Meierkord (eds.), Lingua Franca Communication (pp. 109–134). Frankfurt am Main: Peter Lang. Mel’cuk, I., and Zholkovsky, A. (1984). Explanatory Combinatorial Dictionary of Modern Russian. Vienna: Wiener Slawistischer Almanach. Minsky, M. (1975). A framework for representing knowledge. In P. H. Winston (ed.), The Psychology of Computer Vision (pp. 211–277). New York: McGraw-Hill. Newmeyer, F. J. (2003). Grammar is grammar and usage is usage. Language, 79, 682–707. Paolucci, Claudio. 2021. The distinction between semantics and pragmatics: The point of view of semiotics. Intercultural Pragmatics, 18(3): 293–307. Pinker, S., and Jackendoff, R. (2005). The faculty of language: What’s so special about it? Cognition, 95, 201–236. Rampton, B. (1995). Crossing: Language and Ethnicity among Adolescents. London: Longman. Risager, K. (2006). Language and Culture: Global Flows and Local Complexity. Clevedon: Multilingual Matters. Rosch, E. (1977). Human categorization. In N. Warren (ed.), Advances in Cross-cultural Psychology (Vol. I, pp. 1–49). London: Academic Press. Schank, R. C., and Abelson, R. P. (1977). Scripts, Plans, Goals, and Understanding: An Enquiry into Human Knowledge Structures. Hillsdale, NJ: Lawrence Erlbaum. Searle, J. R. (1979). Expression and Meaning: Studies in the Theory of Speech Acts. Cambridge: Cambridge University Press. Slobin, D. (1991). Learning to think for speaking: Native language, cognition, and rhetorical style. Pragmatics, 1(1), 7–25. Slobin, D. I. (1997). Mind, code, and text. In J. Bybee, J. Haiman, and S. A. Thompson (eds.), Essays on Language Function and Language Type: Dedicated to T. Givón (pp. 437–467). Amsterdam/Philadelphia: John Benjamins. Taylor, J. (2000). The network model and the two-level model in comparison. In B. Peeters (ed.), The Lexicon–Encyclopedia Interface (pp. 115–143). Amsterdam: Elsevier. Van Olmen, D., and Tantucci, V. (2022). Getting attention in different languages: A usage-based approach to parenthetical look in Chinese, Dutch, English, and Italian. Intercultural Pragmatics, 19(2): 141–181. Werkmann, H., Marianna Bolognesi, A., and Kohl, K. (2021). The status of conventional metaphorical meaning in the L2 lexicon. Intercultural Pragmatics, 18(4), 447–467.

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

287

288

ISTVAN KECSKES

Whorf, B. L. (1939). The relation of habitual thought and behavior to language. In J. Carroll (ed.), Language, Thought and Reality: Selected Writings of Benjamin Lee Whorf. Cambridge, MA: MIT Press. Wierzbicka, A. (1996). Semantics: Primes and Universals. Oxford: Oxford University Press.

https://doi.org/10.1017/9781108989275.013 Published online by Cambridge University Press

13 Corpus Pragmatics Karin Aijmer

13.1 Introduction Corpus pragmatics is a hybrid discipline representing an alliance between corpus linguistics and pragmatics where the corpus-linguistic study provides the subject matter and pragmatics is used to explain the corpus data. Corpus pragmatics is still a new discipline characterized by varying degrees of dependence on corpus evidence. However, great progress is being made by studies extending the boundaries of pragmatic issues which can be studied by means of corpora and corpus-linguistic techniques. The interaction between the fields implies both challenges and possibilities. This chapter will give a state-of- the-art overview of the themes and issues in corpus pragmatics and describe new directions in the field represented by empirical corpus studies where synchronic pragmatic variation and change are analyzed in a broader social and cultural perspective. The rest of the chapter is structured in the following way. Section 13.2 introduces the objectives, methods, and scope of pragmatics and how these have changed over time. Section 13.3 brings up the advantages provided by recent attempts to annotate corpora with pragmatic information and gives an overview of corpora suited for pragmatic research. The major topics which have been dealt with in corpus pragmatics are illustrated in Section 13.4. Section 13.5 discusses problematic cases where corpora are used to analyze a pragmatic function. Section 13.6 deals with the extension of corpus pragmatics to social and interactive contexts. The chapter ends with some challenges for the future confronting corpus pragmatics.

13.2 Objectives, Methods, and Scope of Pragmatics The emphasis on “meaning as use” is essential for pragmatic approaches to language (Traugott and Dasher 2002: 16). However, the scope of pragmatics is difficult to circumscribe and the boundary between semantics and pragmatics is fuzzy.

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

290

KARIN AIJMER

We can distinguish several pragmatic orientations with different theoretical origins and research agendas. In the Anglo-Saxon tradition, pragmatics has its roots in ordinary language philosophy and in speech act theory (Austin 1962; Searle 1969). Within this tradition the aim of pragmatics is to study how speakers and hearers engage in the production and interpretation of utterances. The speaker’s intention is to convey a certain meaning and the hearer arrives at the speaker’s intended message by making pragmatic inferences on the basis of communicative principles, such as Grice’s (1975) maxims or the Relevance principle (Sperber and Wilson 1986), of the content of the message, and of information from the context. From early on, pragmatics has been closely associated with linguistics. Mey regards pragmatics as an interdisciplinary enterprise with a close connection to linguistics: “In particular, linguistics needs pragmatics to obtain a proper perspective for its functions . . . but also, pragmatics needs linguistics to provide the data to prove the pragmatic points” (2017: 4). Issues in “inferential pragmatics” have not usually been explored by corpus-linguistic methods. Instead, experimental methods have been proposed as a methodology to study how people actually produce and understand speech acts (Gibbs and Colston 2020). An influential contribution to the discussion of the scope and objectives of pragmatics comes from the European or continental tradition characterized by its focus on language in its social and cultural context (Verschueren 1999). In this perspective, pragmatics opens the field wide “and takes in the total context of the situation, understood historically as not just what is present here and now, but including that which came before, as well as that which may be expected or predicted” (Mey 2017: 16). A large number of phenomena in language are sociopragmatically grounded and require access to the context for their interpretation. This situation raises the problem of how the context should be defined and how the influence of contextual features on the interpretation of linguistic expressions can be integrated with a corpus-linguistic approach. We can conclude that an issue in corpus pragmatics is to decide which aspects of the context are relevant in the communication situation. Aspects of the context which may be relevant to utterance interpretation are collocating linguistic elements, the topic discussed, the preceding discourse, text type (genre), sociopragmatic information about the speakers, and general background knowledge.

13.3 Corpus Pragmatics The advantages of using corpora and corpus techniques to study language are well known. Corpora and corpus-linguistic methods offer opportunities to investigate the frequencies of lexical expressions and reveal their patterns of use on the basis of large amounts of authentic data. Leech et al. (2009: 32) describe corpus linguistics as a combination of quantitative and qualitative analysis:

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

13 Corpus Pragmatics

Whereas in other disciplines these terms are often regarded as incompatible opposites, in corpus linguistics quantitative and qualitative methods are extensively used in combination. It is also characteristic of corpus linguistics to begin with quantitative findings, and work toward qualitative ones.

Another useful distinction discussed by Leech et al. (2009) is that between formal identification and functional interpretation. On a formal level, we are interested in distinguishing distributional patterns of occurrence of structures. On the functional level, the concerns are with interpreting these functions in terms of factors external to corpus data, for instance cognitive, social or historical. Corpus linguistics can be combined with other fields of linguistics. However, in pragmatics corpora are a relative newcomer and it is not self-evident to what extent the two areas can benefit each other (Rühlemann and Aijmer 2015:1). Pragmatics is primarily concerned with the use of language in interaction where a specific speaker (writer) and addressee (reader) can be identified. Most research has been carried out on the basis of spoken language. Given the emergent and usage-based character of spoken language, we can expect the relation between the forms of lexical expressions and their functions to be complex. Ways of combining the two approaches and exploiting the strength of each are now being explored. The procedure involves going back and forth between the quantitative and the qualitative analysis of the data. The analysis begins with calculating the frequency of a certain form (such as well or you know). In the next stage the extracted occurrences are classified according to a specific theory or classification scheme and the frequencies of the different functions are calculated. Finally, the results of the quantitative analyses can be explained in a theoretical framework such as Conversation Analysis or Relevance Theory. It should be clear that corpora lend themselves to searching for a lexical word or construction in the corpus and a quantitative analysis of the form in different contexts. We therefore have many corpus analyses of pragmatic markers (e.g., you know or well), while it has been difficult to analyze politeness, “hedges,” or speech acts using corpus methods. However, there is now a great deal of interest in finding strategies to get around the problem of searching for a function (O’Keeffe 2018). One way forward is indicated by the possibility to annotate corpora with pragmatic information. However, the pragmatic annotation of corpora meets many challenges.

13.3.1 Pragmatic Annotation Morphosyntactic annotation of words in general corpora has been available for a long time. A similar “revolution” has now also reached pragmatics (Weisser 2017: 44). However, the pragmatic annotation of corpus data is a thorny issue. 1

There are by now several introductions to corpus pragmatics (e.g., Romero-Trillo 2008; Jucker and Taavitsainen 2014; Aijmer and Rühlemann 2015; Jucker 2018) with illustrations of the topics covered.

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

291

292

KARIN AIJMER

The complexity of the task is due to the fact that it is not clear what pragmatic features should be annotated, many features are difficult to identify lexically, and it is difficult to do the annotation automatically or semi-automatically (Weisser 2015).2 There have been several attempts in computer pragmatics to develop annotation schemes based on the automatic detection of dialogue/speech acts (see Weisser 2015 for details). One of the better-known schemes is the DAMSL model (Dialogue Act Markup in Several Layers) which goes beyond the classification suggested in philosophical speech act theory and annotates utterances on different levels, for example, depending on whether they have a forwards-looking or backwards-looking function. However, the existing annotation schemes in computer pragmatics have been accused of being unnecessarily complex and more work is needed to make them accessible to corpus pragmaticians (Weisser 2015). (For more discussion of pragmatic annotation schemes and speech/dialogue acts in computer pragmatics, see Archer et al. 2008.) Recently, there has also been a growing interest in annotating existing corpora for pragmatic information. The SPICE-Ireland Corpus has been annotated manually with features such as pragmatic markers, speech acts, quotatives, tags, and prosodic elements (Kallen and Kirk 2012). The pragmatic annotation of the function of speech acts was made on the basis of existing taxonomies which means that some functions may not appear (see below). Pragmatic markers cause problems because of their multifunctionality. What looks like a pragmatic marker can be ambiguous between different interpretations. In I think Bill is fat, it is not clear without considering the broader context whether “I think” is a comment clause with pragmatic function or is used in its grammatical function (cf. Brinton 2008 on “comment clauses”). The pragmatic annotation depends on the research interests of the linguist and on the theoretical framework within which the pragmatic expressions are analyzed. Rühlemann and O’Donnell (2012) excerpted the narratives from the demographic part of the British National Corpus (BNC) and annotated the texts with features such as report mode, quotatives, and participation framework. The Narrative Corpus is a fairly small corpus which was compiled for the specific goal to study narratives and not for general purposes (see Rühlemann and O’Donnell 2012). Projects of this kind involving the co-operation between computational linguistics and pragmatics will probably become more frequent in the future. Pragmatic annotation strengthens the potential of corpus pragmatics by making it possible to code pragmatic features, which makes it easier for the linguist to collect data. However, pragmatic annotation meets several challenges. The problem is how to find the “right criteria” or “shared guidelines” for identifying the expressions one wants to annotate. Annotation therefore 2

Archer and Culpeper (2018: 510) call for more approaches that focus on sociopragmatic features. They demonstrate a scheme implemented in the Sociopragmatic Corpus which annotates speaker/hearer relationships, social roles, etc.

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

13 Corpus Pragmatics

needs to be carried out manually. The researcher reads the corpus line by line in order to identify the items which need to be annotated and can then be searched for automatically. This process is very time-consuming. A balance therefore has to be struck between the degree of detail of the annotation and what it adds to the qualitative analysis carried out on the basis of the corpus. If the classification is too fine-grained, it may be difficult to find enough examples of a certain category for meaningful statistical comparisons.

13.3.2 Corpora Suited for Pragmatic Research Pragmatics is primarily concerned with the use of language in interaction where a specific speaker (writer) and addressee (reader) can be identified. A large number of pragmatic phenomena are specific to spoken language or occur more frequently there. For discourse-pragmatic studies spoken corpora are therefore essential, which has led to an increasing interest in building spoken corpora. However, outside English large spoken corpora are still unusual, since collecting a spoken corpus is time-consuming and the project requires a big budget. For English several large corpora of spoken language are available. The BNC (1994) with a spoken component of 10 million words is a gold mine for studying the frequency and distribution of linguistic expressions in spoken English. The corpus is also useful for pragmatic studies, since it includes demographic information about the age, gender, social class, and regional provenance of the speakers. The BNC (1994) has now been supplemented with the Spoken British National Corpus 2014 (Spoken BNC2014) (Love et al. 2017), which is modeled as far as possible on its predecessor. Other corpora providing “big data” are the Corpus of Spoken American English (COCA; Davies 2009) and the Corpus of Historical American English (COHA; Davies 2010). Another exciting new addition to the collection of spoken corpora is the London–Lund Corpus of Spoken British English 2 (Pöldvere et al. 2017), a halfa-million-word corpus of spoken language with data recorded in 2014–2019. The corpus was designed in the same way as the London–Lund Corpus of Spoken English (LLC), one of the earliest spoken corpora with recordings from the 1960s and 1970s, constituting a fairly small corpus of half a million words (Svartvik 1990). For many research questions in pragmatics, large corpora are not necessary (Clancy 2011). A new tendency in corpus pragmatics is to use smaller specialized corpora, for example, corpora of political discourse, family discourse, and academic speech to study pragmatic phenomena. Smaller corpora make it easier to analyze pragmatic meanings because it is more likely that the analyst has a good knowledge of the speech situation and the speakers (Garcia McAllister 2015: 46). Another advantage is that the corpus can be coded with features specific to the type of research carried out. Pragmatic items (elements having a form–function analysis) are mainly associated with spoken language and are therefore best studied in spoken

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

293

294

KARIN AIJMER

corpora. However, they can also serve interactional and interpersonal functions in written discourse. Several studies have used written corpora to study pragmatic phenomena. For example, Gray and Biber (2015) wanted to study the adjectives and nouns indicating the writer’s attitudes or evaluative stance in a corpus of academic writing. In order to identify the structures which were new in academic writing, they documented the patterns of use in three subcorpora from the Longman Spoken and Written English corpus. Building on politeness theory, Diani (2015) investigated the use of mitigated critical comments in English and Italian book review articles.

13.4 Topics in Corpus Pragmatics 13.4.1 Pragmatic Markers Corpora and corpus-linguistic techniques are increasingly used to describe phenomena which do not have a natural place in grammar but have an important role in the communication situation. Ariel (2012: 30) uses the term form/ function pragmatics to describe a line of pragmatics concerned with pragmatic and discourse meanings (functions) associated with specific lexical items (or constructions). An important landmark in this tradition is Schiffrin’s monograph on discourse markers (1987), where she described the markers well, now, so, but, oh, because, or, I mean, y’know, and then in a small corpus of interviews with American Jews. Following Schiffrin’s pioneering research, an increasing number of studies have examined these and similar items in spoken corpora. Several different terms have been used besides “discourse marker,” which is now often used only with reference to expressions signaling a sequential relationship between the basic message and the preceding context (Fraser 1999). In this chapter I will use the term “pragmatic marker” as a broad term for lexical expressions which are not integrated in the syntactic structure and do not have propositional meaning. (For a more detailed description, see, e.g., Beeching 2016; Brinton 2017.) Pragmatic markers have functions in the evolving discourse much like Gumperz’ contextualization cues leading the hearer to resort to their background knowledge and by an inferential process retrieve the intended message (Gumperz 1996: 379). Corpus methods can be combined with many different theories and approaches to describe pragmatic markers. Corpora make it possible to search for pragmatic markers automatically and chart their frequencies in different contexts. The advantages of the method are illustrated by the rich output of corpus-based studies of pragmatic markers. Aijmer (2002) investigated the functions of selected pragmatic markers (now, oh, ah, just, sort of, and that sort of thing, actually) in the LLC (see Section 13.3.2). The pragmatic markers were analyzed with regard to the context (text type), collocational patterns, prosody, and function. Pragmatic markers are now also investigated on the basis of a combination or mixture of different methods and corpora. This is illustrated by Beeching’s

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

13 Corpus Pragmatics

(2016) study of the meanings of the pragmatic markers well, just, you know, like, sort of, I mean from a sociolinguistic and diachronic perspective. The full BNC (1994, 100 million words) was selected for the purpose of investigating the variation in the frequency and use of the pragmatic markers in spoken interaction. In addition, a small specialized “young corpus” consisting of data from role plays was used to provide information on the most recent developments of the selected pragmatic markers. In addition, there are diachronic corpora which can be used as a corpus-pragmatic resource. Beeching used the Proceedings of the Old Bailey, 1674–1913, which contains spoken language to trace the changes of pragmatic markers over time. 13.4.1.1 Pragmatic Markers in an Extended Sense More and more grammatical features are now regarded as pragmatic markers in a broad sense and discussed from a corpus-pragmatic perspective. Examples include vocatives (Clancy 2015; Palacios-Martínez 2018), interjections (Norrick 2009, 2015; Stange 2019), general extenders (Cheshire 2007; Wagner et al. 2016), “emojis” (Li and Yang 2018), and dislocations (“tails”) (Timmis 2015). It can be shown that these elements can occupy a slot outside the sentence structure, they have functions rather than meanings and they have different functions depending on the context. As such, they qualify for pragmatic marker status. I will discuss a few case studies demonstrating this on the basis of corpora. Palacios Martínez (2018) argued that vocatives should not be regarded simply as forms of address but that they should be analyzed as pragmatic markers because of their position in the clause and function. A selection of common “familiarizers” (man, mate, brother, and boy) were extracted from recent corpora from two successive periods of time (the London English Corpus, 2004–2007 and the Multicultural London English Corpus, 2007– 2011).3 The corpus analysis showed that the familiarizers have increased in frequency over time and that their variation and changes were correlated with sociolinguistic factors such as the age of the speakers. Another finding was that the familiarizers occurred in both the left periphery and the right periphery of the clause with different functions. Interjections, too, can be regarded as pragmatic markers (in occurrences where they are “discursive” and not used as exclamations) (Stange 2019). They are potentially more challenging to interpret than other pragmatic markers since they lack corresponding forms having propositional meaning. Stange (2019) explored the discursive uses of interjections in the Spoken BNC2014 in order to demonstrate this. She showed that they had recurrent pragmatic functions, they were characteristic of spoken language and were typically found as “inserts” (Biber et al. 1999) or in free-standing positions. However, there were differences between various types of interjections. Phatic interjections like mhm require an addressee and are used with a back-channelling 3

The corpora were compiled within the Linguistic Innovators project (Cheshire et al. 2011).

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

295

296

KARIN AIJMER

function in response to what is said. Furthermore, they occurred in combinations with other interjections. Emotive interjections such as ow, ouch, ugh, yuck, whoops allow both for uses where the interjection expresses the speaker’s spontaneous emotions and discursive uses as a pragmatic marker. The findings of the study showed that their frequency in discursive uses varied significantly with regard to the emotion expressed. Interjections of disgust (ugh! Yuck!) and “spill-cries” (oops!) were, for instance, more frequently used as pragmatic markers than interjections expressing pain (ow! ouch!). Stange also found that young female speakers differed from other speaker groups in that they used emotive interjections such as ow, ouch, ugh, yuck, whoops more frequently than males. Ow and whoops were not used at all by male speakers. To sum up, the number of elements which are now treated as pragmatic markers and analyzed with quantitative methods is increasing as a result of the corpus-pragmatic turn. The “new pragmatic markers” also include graphic pragmatic symbols such as “emojis,” which are used in communication on the Internet and in social media. Li and Yang (2018) found that emojis were used in a corpus on the Internet with the function of enhancing emotion but also with interactive functions such as turn-taking or turn-giving. Filled and unfilled pauses have been studied by means of experimental methods and in different spoken corpora (Stenström 2011). There are several reasons for studying them systematically using a corpus-pragmatic approach. They occur frequently in certain genres in particular. They are found in positions where they are not integrated in the utterance (e.g., initially) with discourse management or planning-indicating functions. Another reason is that filled pauses are used differently depending on the speakers’ age, gender, and social class. Tottie (2011) argued that filled pauses (er, erm) in the BNC (1994) could be regarded as sociolinguistic (pragmatic) markers that differentiated between registers of English along with the gender, age, and socioeconomic class of the speakers. 13.4.1.2 Position, Function, and Corpus Pragmatics The study of pragmatic markers cannot ignore the relation between a marker’s placement and function. In a pioneering study, Clift (2001) showed that the pragmatic marker actually performed different conversational tasks in clauseinitial and clause-final position in the utterance or the turn. From a diachronic point of view, it has been suggested that elements have moved from the position inside the clause to the left periphery and then to the right periphery (Traugott and Dasher 2002). Basing themselves on this hypothesis, Degand and Fagard (2011) investigated, on the basis of diachronic corpora, how the meanings of the French pragmatic marker alors in the peripheries have been driven by syntactic and positional changes gradually evolving over time. Beeching and Detges (2014) desired to test the universality of the hypothesis that pragmatic markers receive different functions depending on whether they are placed in the leftmost and rightmost periphery of the utterance (or turn). According to this hypothesis, the leftmost position is best suited for expressing

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

13 Corpus Pragmatics

subjective meanings while pragmatic elements in the rightmost position have an intersubjective (or hearer-oriented) meaning. The contributions to the volume test the Beeching–Detges hypothesis on the basis of corpus data from Asian languages (Chinese, Japanese, Korean) alongside English, French, and Italian data. The research questions for corpus pragmatics are, first, whether there is a relationship between position and function and second, whether the leftmost and rightmost positions in the utterance are linked to different functions. On the basis of the chapter findings, Beeching and Detges conclude that pragmatic markers are not constrained to a particular function in the left or the right periphery and that therefore the hypothesis cannot be upheld in a strong form but that the quantitative data showed that pragmatic markers tend to be used differently depending on their position in the utterance (turn). The contributions in Van Olmen and Šinkūnienė (2021) continue the debate about the relationship between position and function from a synchronic and diachronic perspective. The case studies include more pragmatic markers from both well-researched and less-researched languages.

13.5 Corpus Pragmatics and Function-to-Form Analyses 13.5.1 Politeness Topics such as politeness, evaluation, subjectivity, or hedging, which are at the heart of pragmatics, can be studied using a combination of pragmatic and corpus-linguistic methods. The number of such studies using corpus-linguistic methods trying to integrate a quantitative and qualitative analysis is now rapidly increasing.4 The challenges for a corpus-pragmatic approach are nicely illustrated by Culpeper and Gillings’s (2018) study of politeness. Culpeper and Gillings set out to examine whether there is a north–south politeness variation in England linked to the type of politeness used by speakers. Their method was to search for formulaic expressions (e.g., thanks, please, can you) in the Spoken BNC2014 which could be associated with negative or positive politeness (Brown and Levinson 1978). The problem with this approach was finding the right criteria for diagnosing an expression as polite. Moreover, the number of polite expressions which could be analyzed in this way was necessarily limited. As may be expected, the method also necessitated time-consuming manual reading and weeding out the occurrences that had other non-polite meanings in context. Culpeper and Gillings’s study points to new ways to study politeness by focusing on individual politeness expressions which can be analyzed by means of corpora with regard to functional, formal, and regional properties. However, similar studies are still rare because of the problems of analyzing politeness and identifying the items which can be searched for in the corpus. 4

On evaluation, see Goź dź -Roszkowski and Hunston (2016); on subjectivity see Butler (2008); and on hedging see Hulleberg-Johansen (2020).

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

297

298

KARIN AIJMER

13.5.2 Speech Acts Speech acts are a central notion in pragmatics. They are difficult to identify on the basis of form, and there is an ongoing debate about how they should be searched for in corpora. There are many challenges for the corpus-pragmatic analysis of speech acts. Speech acts do not have a fixed form and a single function. The same form (sorry) may realize speech acts other than an apology (e.g., to ask a person to repeat something). An apology can be interpreted sincerely or ironically depending on the context. Syntagmatically, the speech act may have a complex form consisting of a head and modifier, and it needs to be analyzed with regard to both the strategies it performs and its function. The patterning of a speech act also includes the response in the next speaker turn. These properties make them difficult to describe with the help of corpora. As a result, the focus in corpus-based speech act studies has been on those forms that are easy to identify. “Social acts” such as thanks, requests, and apologies have a routinized, searchable form and their functions can be described on the basis of corpora (Aijmer 1996). A particular speech act (e.g., thanking) is identified on the basis of (I)thank (you) (an illocutionary force indicating device “IFID”), which can be varied in different ways (e.g., thank you very much). The usage conditions can be analyzed along several contextual dimensions including speaker attitude, expected response, degree of conventionalization, and aspects of the social situation (e.g., what one is thanking for) (cf. also Jucker and Taavitsainen 2000). Aijmer (1996) used the notion (shared) analytical frame to account for what the speaker and hearer know about the function and context in which a particular speech act form is used. The data for Aijmer’s study comes from the LLC collected during the 1960s and 1970s and containing about half a million words. Recent studies have used a similar methodology on a larger set of materials. Deutschmann (2003) investigated the speech act of apologizing in a sub-corpus of the BNC (1994), using as a starting point a list of conventionalized forms extracted from existing taxonomies of apologies. The corpus (c. 5 million words) was used to describe the frequency and distribution of apologies in different speaker groups and in various conversational settings. It was also employed to study the correlation between a particular variant and situational factors (e.g., the type and imposition of the offense one is apologizing for). Aijmer and Deutschmann both used a pre-selected list of lexical items to study the functions of a speech act. However, the full range of realizations of a speech act may not be discovered using this methodology. Lutzky and Kehoe (2017) addressed this methodological problem in a study they carried out of apologies in the Birmingham Blog Corpus. The authors used a novel form of analysis to reveal “hidden manifestations” of apologies in the corpus by examining the collocates of the illocutionary force identifier sorry and its synonyms. In this way they discovered that the form oops (and its phonological variants) could be used as an apologizing IFID in online media.

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

13 Corpus Pragmatics

However, the methodology used cannot detect any forms that do not co-occur with sorry. The interplay between language and context is an important factor determining the hearer’s interpretation of the function of a pragmatic element in corpus-pragmatic studies. There is now a tendency to focus more on how speech acts emerge in the discourse and how they are identified in the ongoing negotiation between the speaker and the hearer. As a result, several new ways of using corpora have been explored to discover the variations of a speech act in different discourse contexts. Garcia Mc is Allister (2015) used a corpus-driven bottom-up approach to identify the ways in which the directive (general request) category was expressed which implied that she read through the transcripts of the corpus (the spoken component of the TOEFL 2000 Spoken and Written Academic Language Corpus) taking into account the prosodic and contextual cues used to identify a speech act in the general directive category. Corpus tools were then used to assign further linguistic information as indicators of the speaker’s intentions (e.g., the presence of a modal) to the examples identified as directives in the corpus. This approach made it possible to answer further research questions, for example what directives (requests, advice, warnings, etc.) people used in different situation types (study groups, office hours) and to make a detailed pragmatic analysis of the speech acts. The study also opened up for the empirical study of many specific, so far unexplored, speech acts with a directive function (e.g., giving directions or repeat a request).

13.5.3 Corpus Pragmatics and Turn-Taking Turn-taking is important to explain the structures emerging in the spoken interaction. Speakers take turns in the conversation “evoking the context,” which itself is the object of negotiation in the following speaker turn (Sacks et al. 1974). Exactly how a turn ends or continues will not be possible to predict on the basis of the ongoing discourse. However, in retrospect it can be shown that certain patterns of co-constructed turns recur indicating that certain contexts provide the opportunity for the second speaker to complete the turn along with other variations. We can get a more detailed and precise picture of what happens at the end of the turn using corpora. The potential of the corpus-based approach to search for patterns in the conversation where the second speaker completes or extends an utterance by the first speaker is illustrated by Clancy and McCarthy’s (2015) study of types of co-constructed turn-taking in a pragmatic variational perspective. The data is taken from the CANCODE Corpus, a collection of largely informal British English (compiled during the 1990s) and the Limerick Corpus of Irish English paralleling the design of CANCODE (Clancy and McCarthy 2015: 437). However, the search method is not self-evident. The authors focused on the formal side and searched for if, when, and which in turn-initial position, manually removing “false” examples where the subordinate clause was not

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

299

300

KARIN AIJMER

followed by a completion of the turn by a second speaker. Their findings confirmed certain patterns which had been pointed out in studies not using corpora and showed that there were many variations of the co-constructional patterns which had not been observed earlier.

13.6 Corpus Pragmatics and Context The meaning and use of language are increasingly studied in corpora in the social and cultural context where it is produced. Context has been defined both in linguistics and in other disciplines. In linguistics the context of a lexical item refers to the words it collocates with. It has also been analyzed in discoursepragmatic terms as including the preceding and following discourse. In corpus pragmatic studies, “context” is now also used to refer to the social context. The social context has been difficult to define. However, anthropologists and sociolinguists have made important contributions to the description of the social or situational features which play a role for the production and processing of meaning in the communication situation. In a dynamic theory of pragmatics, indexicality plays an important role in these disciplines to explain how talk is linked to social and situational factors (Ochs 1996). Both (conventionalized) speech acts and pragmatic markers are indexically associated with sociocultural features which can be activated in the speech situation, such as “social identity,” understood as social persona or relationships (familiars versus non-familiars), group identity (gender, social class, generation, ethnic membership), and rank. Another important situational dimension is stance. To give an example, the speech act of thanking (thank you) is indexically associated with affective stance (“feeling gratitude”). The formulaic can you indexes a request and more indirectly interpersonal attitudes and politeness. Certain pragmatic markers may also index “youth talk” or “femaleness.” The indexical or social meanings of pragmatic elements can be described in a corpus-pragmatic perspective, since many new spoken corpora provide information about the speakers’ age, gender, and regional provenance. We can, for instance, study how speakers belonging to a particular age group use pragmatic markers to express a social identity.

13.6.1 Corpus Pragmatics and Sociolinguistic Variability Corpus pragmatics has recently been influenced both by sociolinguistics and by variational pragmatics because of the shared interest in the influence of the context on language. The terminology varies between the disciplines. Variationist sociolinguists use the term discourse-pragmatic feature (or variable) (rather than pragmatic marker or discourse marker) to refer to ‘a heterogeneous category of items or constructions’ which have discourse or pragmatic functions (Pichler 2016: 3).

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

13 Corpus Pragmatics

In sociolinguistics the interest in discourse-pragmatic variation is a recent phenomenon. Variationist methods taking into account sociolinguistic factors have been employed mainly to analyze phonological and morphological variation where the variants have the same meaning. Interestingly, the same methods have now also been applied to the study of pragmatic phenomena. However, the multifunctionality of (some) discourse-pragmatic features makes it problematic to conceptualize them as variables and to investigate their variation in a systematic way (Pichler 2016: 5; Waters 2016: 421). A number of solutions have therefore been proposed with the purpose to define the pragmatic variable in such a way that it can be used to study (discourse-)pragmatic variation. According to D’Arcy (2017: 165), the feature that is varied need not even be defined in a semantic or functional way. She argues that in order to carry out a variationist analysis of like the starting point should be structural (like occurs in structurally delimited contexts of variation) (cf. also Denis and Tagliamonte 2016: 90; Barron 2017b). In addition to the variationist studies, there is now a growing body of research using corpus-linguistic methods to identify discourse-pragmatic features and study their sociolinguistic variation. Both corpus linguistics and (variationist) sociolinguistics are concerned with describing language in its social contexts. However, the two methods differ and do not give comparable results. The corpus-linguistic methodology is characterized by a form-based approach where the frequency of tokens of a pragmatic marker such as like represents normalized rather than raw frequencies, which makes it possible to compare its frequencies across a range of speakers belonging to different age bands and either gender. The sociolinguistic approach can also be based on form (and corpora), but the frequency of a variable feature is measured in relation to other variants which could have been selected as variants and not in absolute terms (D’Arcy 2017: 163). We can now see that the interest in combining the two paradigms is growing, sparked by the possibility of using corpora which provide sociolinguistic data about the speakers. Special search tools make it possible to extract automatically the occurrences of a pragmatic marker in its linguistic context together with sociolinguistic information about the speakers’ age, gender, social class, and region. The quantitative analysis can then provide detailed information about the influence of both linguistic and nonlinguistic factors on the frequency and distribution of the form. Pragmatic markers, speech acts, and many other elements which vary depending on the social context have been studied in corpora. In a pioneering study, Andersen (2001) investigated the pragmatic markers innit and like in the Bergen Corpus of London Teenage Language (COLT) which was compiled in the 1990s. The corpus was used to study their form, function, and social variation. The analysis showed that the frequencies of like and innit were correlated with the speakers’ age and the locality in London where they came from. Another finding was that the ethnic minority group of the speakers

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

301

302

KARIN AIJMER

played an important role in the use and spread of innit/is it but a lesser role in the use of like (Andersen 2001: 293).

13.6.2 Corpus Pragmatics Meets Variational Pragmatics Variational pragmatics aims to “systematically describe synchronic variation in the patterns of human interaction within one language due to such factors as region, gender, socio-economic status, ethnic identity and age” (Barron 2017b: 91). The methodology is contrastive and an effort is made to make the contrasted varieties as comparable as possible with regard to the speakers’ age, gender, social class, etc. The focus in variational pragmatics has been on speech acts, which have mostly been studied under controlled conditions on the basis of role plays or discourse completion tests (DCTs) (Félix-Brasdefer and Hasler-Barker 2017). Corpora and corpus-linguistic analysis have not generally been seen as a suitable method, the reason being that speech acts cannot be searched for without first identifying the forms. Moreover, speech acts are context-bound, which makes it difficult for the researcher to control the social and situational factors and norms influencing the choice of a particular variant (see Section 13.5.2). However, the situation is now changing thanks to the availability of comparable corpora which can be used to study similarities and differences in language use between varieties of the same language. The International Corpus of English (ICE-Corpus), compiled in the 1990s and consisting of sub-corpora representing national varieties of English, has been important for the development of variational corpus pragmatics.5 The spoken data in the corpora represent different text types such as informal face-to-face conversation, telephone conversation, and more formal speech. The ICE-Corpus has so far not been fully exploited as a resource to study regional variation. Kallen (2015) has shown one way in which the ICE-Corpus can be used to study variation. He compared the frequency and positions of actually in the Irish English component of the ICE-Corpus with the results of a similar investigation of actually in the other varieties. The analysis showed that actually occurred less frequently in the clause-marginal positions than many other varieties. Aijmer (2022) carried out a similar study of the pragmatic marker anyway. She observed that anyway occurred more frequently in the ICE-Ireland Corpus than in the other varieties of English in the corpus. Anyway was most common in the right periphery of the utterance in Irish English, British English, New Zealand English, Australian English, and Singapore English, whereas it dominated in the left periphery of the utterance in American, Canadian, and Philippine English. Anyway was significantly more frequent in the right periphery in Irish English, which she suggested may be due to its association with Irish friendliness. 5

The International Corpus of English (ICE), http://ice-corpora.net/ice/index.html.

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

13 Corpus Pragmatics

Comparable corpora are also suitable for variational pragmatic research on speech acts. The use of corpora and a quantitative analysis of different patterns transfers the interest from the function of the speech act to the ways it is realized, how the speech act is expressed, who uses it, and when. Barron (2017a) compared the speech act of “offers” in the British and the Irish components of the ICE-Corpus. The corpus analysis allows the researcher to investigate similarities and differences between the varieties at the level of form and strategies chosen to perform offers. Barron identified several methods to search for the speech act of offer. The search strings were based on previous research on offers and included a performative verb (offer), conventionalized patterns (shall I . . .), modifiers (if you want), routine responses (thanks), and descriptive comments or topic-related words (coffee, tea). However, there were problems such as devising criteria for the differentiation between offers and related speech acts (cf. Section 13.5.2). The quantitative analysis was followed by a qualitative in-depth analysis showing that the speakers of Irish English used more forceful offers, thus providing empirical evidence for the generally held opinion about the hospitality of the Irish. Another contribution to the corpus-pragmatic study of speech acts and regional variation is Jautz’s (2013) ambitious project of comparing thanking formulae in (a selection of spoken texts from) the BNC and the Wellington Corpus of Spoken New Zealand English (1 million words collected between 1988 and 1994).6 The initial quantitative analysis showed that British speakers used more (and different) gratitude expressions than New Zealand speakers, but the latter used thanking more for phatic talk. The qualitative analysis concerned how thanking formulae should be described in different politeness theories. There are also interesting initiatives to combine a variational pragmatic analysis with a study of changes over time. Ronan (2019) employed data from COCA and the Web-Based Global English (GloWbE)7 in order to study the use of the “x-much” construction (silly much). In addition, COHA (Davies 2010) was used to examine the time depth of the construction and its change over time. Ronan found that the “x-much” construction was extending its use to new grammatical and semantic areas and also that it varied in frequency across the varieties of English with the highest frequency in Singapore English.

13.6.3 Diachronic Corpus-Pragmatic Analyses Studies in historical corpus pragmatics at the intersection between historical linguistics, pragmatics, and corpus linguistics have emerged as an area of growing interest in the last twenty years (Jucker and Taavitsainen 2014). 6

7

The Wellington Corpus of Spoken New Zealand English (WSC), www.wgtn.ac.nz/lals/resources/corpora-default/cor pora-wsc. Corpus of Global Web-based English (GloWbE), www.english-corpora.org/glowbe/.

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

303

304

KARIN AIJMER

Brinton (2017) has described the developments of pragmatic markers such as I mean or you know within the grammaticalization framework. They arise from lexical elements and acquire grammatical qualities as the result of a number of syntactic and semantic changes. Lutzky (2012) is carried out a synchronic corpus-pragmatic analysis of three discourse markers (marry, well, and why) from a qualitative and a quantitative perspective but restricted to Early Modern English. She used a combination of Early Modern English corpora containing speech-based or speech-related data to analyze the functions of the discourse. She also used a specialized sociopragmatic corpus to investigate the social status and gender of the users of the interjections. We can now also use spoken corpora to study short-time changes in real time in conversational interaction. Many different words can be studied with regard to change in a sociopragmatic perspective. Aijmer (2021) studied the variation and short-time changes of well as an intensifying adverb before adjectives by comparing its frequency and uses in comparable contexts in the BNC (1994) and in the Spoken BNC2014. The intensifying use of well was first distinguished from other uses of well (e.g., as a pragmatic marker) and searched for in the context of a following adjective. The results showed that well as an intensifier was used differently by males and females, by that it increased in the speech of females over the twenty-year gap in time.

13.6.4 Corpus Pragmatics, Activity Type, and “Communicative Practice” Language is used differently depending on the activity we are engaged in. This is true about the lexis employed, but the activity also affects how and when we apologize to someone and what types of requests we make. Several terms (genre, text type) have been used to refer to specific communicative activities. I will use Levinson’s term “activity type” with reference to “a category whose focal members are goal-defined, socially constituted, bounded events with constraints on participants, setting, and so on” (1979: 69).8 Activity types are less often discussed than region or other macro-social variables in variational pragmatics, but they have an important role to explain how pragmatic markers and speech acts are used with different frequencies and for different purposes depending on the activity. The interaction between activity type and pragmatics can be studied on the basis of corpora representing different text types. Aijmer (2013) studied the frequencies and distribution of the pragmatic markers (well, actually, in fact) across a wide range of text types in the ICECorpus (e.g., conversation, telephone conversation, classroom lesson, broadcast discussions, legal cross-examinations, parliamentary debates). In this way she showed that in fact was more frequent in cross-examinations than in other

8

A more generally used term is “genre” (see Aijmer and Lewis 2017). The term “text type” is used in corpus linguistics “to refer to a group of related texts in a corpus.”

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

13 Corpus Pragmatics

text types where it was used in questions by the counsel when the answer to the question was already known (Aijmer 2013: 92). In addition to activity type, one can identify more fine-grained “communicative practices” (Linell 1998) influencing the use of language. Both the topical domain and the tasks performed in the speech situation may affect the meaning and use of a lexical expression (including pauses). Staley and Jucker (2021) investigated the frequency, position and function of the pause-filler uhm using a corpus-pragmatic approach. They used a specialized corpus consisting only of restaurant server talk, the Los Angeles Restaurant Corpus compiled in 2011. The restrictions to a small range of speech situations allowed the authors to relate the frequency and function of uhm to a specific task performed by the server (highlighting the food item when serving food to guests).

13.6.5 Corpus-Based Interlanguage Pragmatic Studies Another area of pragmatic studies where corpora are now useful is interlanguage pragmatics, defined as the area at the intersection between pragmatics and learner language. Traditionally, this discipline focused on experimental methods and written DCTs. In recent years, learner corpora have been used as an additional methodology especially for the purpose of studying learners’ use of pragmatic markers in actual language use (see Romero-Trillo 2018 for more discussion). A considerable number of studies have by now investigated pragmatic markers from a corpus-based interlanguage pragmatic perspective. Müller (2005) compared the use of the pragmatic markers so, well, now, and like by American and German students based on a corpus compiled for that purpose. Interlanguage corpus pragmatic studies can now also be carried out on the basis of computerized learner corpora. The Louvain International Database of Spoken English Interlanguage (LINDSEI Corpus) contains corpora representing the language produced by advanced learners of English from different mother tongue backgrounds. Alongside the learner corpora, a similar corpus has been compiled with the difference that the speakers are native speakers of English. We can therefore compare the spoken language produced by learners with that of native speakers. The individual learner corpora are carefully annotated with information about the learner’s native language, number of years that they have studied English, contact with native speakers, etc. Aijmer (2011) used the Swedish component of the LINDSEI Corpus to compare the Swedish learners’ use of well with the native speakers’ use of the marker. By comparing learners’ use of the pragmatic marker with native speaker’s use, she could establish that well was overused by learners in relation to the native speaker norm. The qualitative analysis showed that well was more frequently used by learners than by native speakers as a fluency device to cope with speech management problems. She also found that there were large differences between the individual learners with respect to their use of well, suggesting that pragmatic markers are acquired in different ways and to different extents. In another study, Buysse

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

305

306

KARIN AIJMER

(2017) used learner corpus data from the LINDSEI Corpus to test the hypothesis that you know was employed less frequently by learners than by native speakers and that learners used it for different purposes than native speakers. The findings demonstrated that you know was used less frequently in the Dutch, French, German, Spanish component than in the reference corpus but that the learners used you know in all the nine functions that were attested in the corpus. The quantitative deviations between the learners could also be explained from a qualitative perspective. It was seen that learners used you know more in the editing function (e.g., marking a repetition or reformulation) than native speakers, as may be expected. Another factor explaining the learners’ use of you know was their native language. The findings showed that German learners used you know less frequently than learners with other mother tongues.

13.6.6 Multimodal Corpus Pragmatics An issue in corpus pragmatics is which features of spoken communication are relevant to pragmatics because they are used by the hearer as cues to the interpretation of the utterance. Spoken communication does not only involve the transfer of a verbal message, but the speaker’s intentions are additionally signaled nonverbally by gestures and face expressions, Recently, we have witnessed a rapprochement between corpus pragmatics and multimodal approaches to the study of spoken language production. As a consequence, there are now several attempts to build multimodal corpora which can be used to facilitate and enrich pragmatic studies. Multimodal corpora provide rich information about the context in which the communication occurs with respect to gestures and face expressions. There are now several ongoing projects extending the study of corpus pragmatics in this direction (see, e.g., Huang 2021). Some pioneering research has been carried out on the basis of the Nottingham Multi-Modal Corpus combining a pragmatic study of back-channelling in the conversation with the speakers’ use of head-nods (Knight and Adolphs 2008). Back-channels were defined as short response items which did not appear to take over a speaker turn commonly expressed by yeah. The authors found that length of the nods co-occurred with whether the back-channel items were used to mark information receipt, or whether they were used to express speaker–hearer convergence.

13.7 Challenges for the Future A growing number of studies have shown how the quantitative corpus-based analysis of pragmatic phenomena can be the entry-point for qualitative, indepth studies. However, a number of challenges for corpus pragmatics remain and need to be considered.

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

13 Corpus Pragmatics

A problematic issue in corpus pragmatic studies concerns the context. A major challenge for future research is to ensure that the corpora make available to the researcher as much as possible of the contextual information that is relevant to the hearer’s interpretation of the utterance. At the present time, many corpora represent different text types or activities (such as service encounter or office talks) offering the opportunity to study how they influence the interpretation. However, a large number of other linguistic and extralinguistic features may be behind the selection of a certain variant. These include sociopragmatic features such as the speaker’s age, gender, and social class. Another challenge concerns the extent to which we can use corpora to study pragmatic phenomena such as politeness, evaluation, hedging, turntaking, and indirect speech acts. Researchers are now taking steps toward using corpora for bottom-up searches to identify elements which can be associated with a particular pragmatic function. The advantage of this method is that it avoids the intuition-based selection of items as candidates for analysis, thereby providing a more reliable picture of what the variants are (Andersen 2018: 473). As has been shown above, we have many corpora of spoken language which serve as a resource for pragmatic studies. However, we need more corpora which are suited to study the function of pragmatic phenomena in different genres. New genre-based corpora should also include blogs, discussion forums, and Twitter. Finally, there is a demand for spoken corpora for languages other than English. Corpus pragmatics interacts with other disciplines to a greater or lesser extent. Just as we now have historical corpus pragmatics (Jucker and Taavitsainen 2014), there are also alliances between corpus linguistics and variational pragmatics and between corpus linguistics and sociopragmatics. Other combinations are multimodal corpus pragmatics and interlanguage corpus pragmatics. Corpora and corpus-linguistic methods also hold the potential to contribute to crosslinguistic pragmatic studies (Aijmer and SimonVandenbergen 2006; Aijmer 2009; Aijmer and Lewis 2017). The comparison between the languages is made possible by the availability of parallel corpora consisting of translations between languages. These new hybrid disciplines have in common that they combine a qualitative analysis with a quantitative corpus analysis. The research questions depend on the particular discipline. The empirical corpus-based analysis leads to more reliable results than alternative approaches because they are based on large amounts of real data and can give rise to new research questions which can be tested. The corpus-based study of pragmatic phenomena is no longer neglected but much research remains to be done. We are at the beginning of an era where more and more pragmatic elements are studied from a sociopragmatic corpus perspective (in different situations, regions, and activity types). The pragmatic elements which have been investigated in corpora are mainly pragmatic markers and speech acts. An increasing number of elements categorized as adverbs, vocatives, interjections, and pauses have been shown to be better analyzed as

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

307

308

KARIN AIJMER

pragmatic markers with interactive and interpersonal functions. On the agenda for corpus pragmatic studies is to study more phenomena from a pragmatic variational perspective. There are no doubt many lexical expressions, constructions, and whole utterances in language which are indexically associated with social and other contextual features and which vary with regard to social and discourse-pragmatic variables in the communication situation. A great deal of cross-pragmatic research also remains to be done on the basis of corpora which are comparable with regard to genre and sociolinguistic features. New possibilities for variational corpus-based pragmatic studies are opened up by projects having the aim to compile big spoken corpora which are comparable with existing corpora such as the new Spoken BNC2014.

References Aijmer, K. (1996). Conversational Routines in English: Convention and Creativity. London: Longman. Aijmer, K. (2002). English Discourse Particles: Evidence from a Corpus. Amsterdam/ Philadelphia: John Benjamins. Aijmer, K., ed. (2009). Special issue on contrastive pragmatics. Languages in Contrast, 9(1), 1–181. Aijmer, K. (2011). Well, I’m not sure I think . . . The use of well by non-native speakers. International Journal of Corpus Linguistics, 16(2), 231–254. Aijmer, K. (2013). Understanding Pragmatic Markers: A Variational Pragmatic Approach. Edinburgh: Edinburgh University Press. Aijmer, K. (2021). “That’s well good”: A re-emergent intensifier in current British English. Journal of English Linguistics, 49(1), 18–38. Aijmer, K. (2022). “Well he’s sick anyway like”: Anyway in Irish English. Corpus Pragmatics: International Journal of Corpus Linguistics and Pragmatics, 6(2), 101–125. Aijmer, K., and D. Lewis, eds. (2017). Contrastive Analysis of Discourse-Pragmatic Aspects of Linguistic Genres. Cham: Springer. Aijmer, K., and Simon-Vandenbergen, A.-M. (2006). Pragmatic Markers in Contrast. Amsterdam: Elsevier. Aijmer, K., and Rühlemann, C. (2015). Corpus Pragmatics: A Handbook. Cambridge: Cambridge University Press. Andersen, G. (2001). Pragmatic Markers and Sociolinguistic Variation. Amsterdam/ Philadelphia: John Benjamins. Andersen, G. (2018). Corpus construction. In A. H. Jucker, K. Schneider, and W. Bublitz (eds.), Methods in Pragmatics (pp. 467–494). Berlin/Boston: De Gruyter Mouton. Archer, D., Culpeper, J., and Davies, M. (2008). Pragmatic annotation. In M. Kytö and A. Lüdeling (eds.), Corpus Linguistics: An International Handbook (pp. 613–642). Berlin: Mouton de Gruyter. Archer, D. and Culpeper, J. (2018). Corpus annotation. In A. H. Jucker, K. Schneider, and W. Bublitz (eds.), Methods in Pragmatics (pp. 495–525). Berlin/Boston: Mouton de Gruyter.

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

13 Corpus Pragmatics

Ariel, M. (2012). Research paradigms in pragmatics. In K. Allan and J. M. Jaszczolt (eds.), The Cambridge Handbook of Pragmatics (pp. 23–45). Cambridge: Cambridge University Press. Austin, J. L. (1962). How to Do Things with Words: The William James Lectures Delivered at Harvard University in 1955. Oxford: Oxford University Press. Barron, A. (2017a). The speech act of “offers” in Irish English. World Englishes, 36(2), 224–238. Barron, A. (2017b). Variational pragmatics. In A. Barron, G. Yueuo, and G. Steen (eds.), The Routledge Handbook of Pragmatics (pp. 91–104). London/New York: Routledge. Beeching, K. (2016). Pragmatic Markers in British English: Meaning in Social Interaction. Cambridge: Cambridge University Press. Beeching, K., and Detges, U. (2014). Introduction. In K. Beeching and U. Detges (eds.), Discourse Functions at the Left and Right Periphery: Crosslinguistic Investigations of Language Use and Language Change (pp. 1–23). Brill: Leiden. Biber, D., Johansson, S., Leech, G., Conrad, S., and Finegan E. (1999). The Longman Grammar of Spoken and Written English. London: Longman. Brinton L. J. (2008). The Comment Clause in English: Syntactic Origin and Pragmatic Development. Cambridge: Cambridge University Press. Brinton, L. J. (2017). The Evolution of Pragmatic Markers in English: Pathways of Change. Cambridge: Cambridge University Press. Brown, P., and Levinson, S. C. (1978). Universals in language usage: Politeness phenomena. In E. N. Goody (ed.), Questions and Politeness: Strategies in Social Interaction (pp. 56–311). Cambridge: Cambridge University Press. Butler, C. S. (2008). The subjectivity of basically in British English: A corpus-based study. In Romero-Trillo, J. (ed.), Pragmatics and Corpus Linguistics: A Mutualistic Entente (pp. 37–63). Berlin/New York: Mouton de Gruyter. Buysse, L. (2017). The pragmatic marker “you know” in learner Englishes. Journal of Pragmatics, 121, 40–57. Cheshire, J. (2007). Discourse variation, grammaticalization and stuff like that. Journal of Sociolinguistics, 11, 155–193. Cheshire, J., Kerswill, P., Fox, S., and Torgersen, E. (2011). Contact, the feature pool and the speech community: The emergence of Multicultural London English. Journal of Sociolinguistics, 15(2), 151–196. Clancy, B. (2011). Complementary perspectives on hedging behaviour in family discourse: The analytical synergy of variational pragmatics and corpus linguistics. International Journal of Corpus Linguistics, 16(3), 371–390. Clancy, B. (2015). “Hurry up baby son all the boys is finished their breakfast”: Examining the use of vocatives as pragmatic markers in Irish Traveller and settled family discourse. In C. Amador-Moreno, K. McCafferty, and E. Vaughan (eds.), Pragmatic Markers in Irish English (pp. 229–247). Amsterdam/Philadelphia: John Benjamins. Clancy, B., and McCarthy, M. (2015). Co-constructed turn-taking. In K. Aijmer and C. Rühlemann (eds.), Corpus Pragmatics: A Handbook (pp. 430–453). Cambridge: Cambridge University Press. Clift, R. (2001). Meaning in interaction. Language, 77, 246–291. Culpeper, J., and Gillings, M. (2018). Politeness variation in England: A north–south divide? In V. Brezina, R. Love, and K. Aijmer (eds.), Corpus Approaches to Contemporary British Speech: Sociolinguistic Studies of the Spoken BNC2014 (pp. 33–59). New York/London: Routledge.

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

309

310

KARIN AIJMER

D’Arcy, A. (2017). Discourse-Pragmatic Variation in Context: Eight Hundred Years of LIKE. Amsterdam: John Benjamins. Davies, M. 2009. The 385+ million word Corpus of Contemporary American English (1990–present). International Journal of Corpus Linguistics, 14(2): 159–190. Davies, M. (2010). The Corpus of Historical American English: 400 million words, 1810– 2009. http://corpus.byu.edu/coha/. Degand, L., and Fagard, B. (2011). Alors between discourse and grammar: The role of syntactic position. Functions of Language, 18(1), 29–56. Denis, D., and Tagliamonte, S. A. (2016). Innovation, right? Change, you know? Utterance-final tags in Canadian English. In H. Pichler (ed.), Discourse-Pragmatic Variation and Change in English: New Methods and Insights (pp. 86–112). Cambridge: Cambridge University Press. Deutschmann, M. (2003). Apologising in British English. Institutionen för moderna språk. Umeå: Umeå universitet. Diani, G. (2015). Politeness. In K. Aijmer and C. Rühlemann (eds.), Corpus Pragmatics: A Handbook (pp. 169–191). Cambridge: Cambridge University Press. Félix-Brasdefer, J. C., and Hasler-Barker, M. (2017). Elicited data. In A. Barron, G. Yueuo, and G. Steen (eds.), The Routledge Handbook of Pragmatics (pp. 27–40). London/ New York: Routledge. Fraser, B. (1999). What are discourse markers? Journal of Pragmatics, 14, 219–236. Garcia McAllister, P. (2015). Speech acts: A synchronic perspective. In K. Aijmer and C. Rühlemann (eds.), Corpus Pragmatics: A Handbook (pp. 29–51). Cambridge: Cambridge University Press. Gibbs, R. W., and Colston, H. L. (2020). Pragmatics always matters: An expanded vision of experimental pragmatics. Front. Psychol., July 24. https://doi.org/10.3389 /fpsyg.2020.01619. Goźdź-Roszkowski, S., and Hunston, S. (2016). Corpora and beyond: Investigating evaluation in discourse. Introduction to the special issue on corpus approaches to evaluation Corpora, 11(2), 131–141. Gray, D., and Biber, D. (2015). Stance markers. In K. Aijmer and C. Rühlemann (eds.), Corpus Pragmatics: A Handbook (pp. 219–246). Cambridge: Cambridge University Press. Grice, H. P. (1975). Logic and conversation. In P. Cole and J. M. Morton (eds.), Syntax and emantics 3: Speech acts (pp. 41–58). New York: Academic Press. Gumperz, J. J. (1996). The linguistic and cultural relativity of inference. In J. J. Gumperz and S. C. Levinson (eds.), Rethinking Linguistic Relativity (pp. 374–406). Cambridge: Cambridge University Press. Huang, L. (2021). Toward multi-modal corpus pragmatics: Rationale, case, and agenda. Digital Scholarship in the Humanities, 36(1). DOI:10.1093/llc/ fqz080. Hulleberg Johansen, S. (2020). “I just need to like see what I can do”: A contrastive study of hedging strategies in English and Norwegian informal spoken conversations. Ph.D. dissertation, Department of Literature, Area Studies and European Languages, University of Oslo. Jautz, S. (2013). Thanking Formulae in English: Explorations across Varieties and Genres. Amsterdam/Philadelphia: John Benjamins. Jucker, A. H., and Taavitsainen, I. (2000). Diachronic speech act analysis: Insults from flyting to flaming. Journal of Historical Pragmatics, 1(1), 67–95.

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

13 Corpus Pragmatics

Jucker, A. H., and Taavitsainen, I. (2014). Diachronic corpus pragmatics: Intersections and interactions. In A. H. Jucker, I. Taavitsainen, and J. Tuominen (eds.), Diachronic Corpus Pragmatics (pp. 3–26). Amsterdam: John Benjamins. Jucker, A. H. (2018). Introduction to part 5: Corpus pragmatics. In A. H. Jucker, K. Schneider, and W. Bublitz (eds.), Methods in Pragmatics (pp. 455–466). Berlin/ Boston: Mouton de Gruyter. Kallen, J. L. (2015). “Actually, it’s unfair to say that I was throwing stones”: Comparative perspectives on uses of actually in ICE-Ireland. In C. AmadorMoreno, K. McCafferty, and E. Vaughan (eds.), Pragmatic Markers in Irish English (pp. 135–155). Amsterdam/Philadelphia: John Benjamins. Kallen, J. L., and Kirk, J. (2012). SPICE-Ireland: A User’s Guide. Belfast: Cló Ollscoil na Banríona. Knight, D., and Adolphs, S. (2008). Multi-modal corpus pragmatics: The case of active listenership. In J. Romero-Trillo (ed.), Pragmatics and Corpus Linguistics: A Mutualistic Entente (pp. 175–190). Berlin/New York: Mouton de Gruyter. Leech, G., Hundt, M., Mair, C., and N. Smith. (2009). Change in Contemporary English: A Grammatical Study. Cambridge: Cambridge University Press. Levinson, S. C. (1979). Activity types and language. Linguistics, 17, 365–379. Li, L., and Yang, Y. (2018). Pragmatic functions of emoji in internet-based communication: A corpus-based study. Asian Journal of Second and Foreign Language Education, 16(3). https://doi.org/10.1186/s40862-018-0057-z Linell, P. (1998). Approaching Dialogue Talk, Interaction and Contexts in Dialogic Perspectives. Amsterdam/Philadelphia: John Benjamins. Love, R., Dembry. C., Hardie. A., Brezina,V., and McEnery, T. (2017). The Spoken BNC2014: Designing and building a spoken corpus of everyday conversations. International Journal of Corpus Linguistics, 22(3), 319–344. Lutzky, U. (2012). Discourse Markers in Early Modern English. Amsterdam/ Philadelphia: John Benjamins. Lutzky, U., and Kehoe, A. (2017). “Oops, I didn’t mean to be so flippant.” A corpus pragmatic analysis of apologies in blog data. Journal of Pragmatics, 116: 27–36. Mey, J. L. (2017). Interdisciplinarity in pragmatics and linguistics. In A. Barron, Y. Gu, and G. Steen (eds.), The Routledge Handbook of Pragmatics (pp. 4–18). London/ New York: Routledge. Müller, S. (2005). Discourse Markers in Native and Non-native English Discourse. Amsterdam/Philadelphia: John Benjamins. Norrick, N. N. (2009). Interjections as pragmatic markers. Journal of Pragmatics, 41, 866–891. Norrick, N. N. (2015). Interjections. In K. Aijmer and C. Rühlemann (eds.), Corpus Pragmatics: A Handbook (pp. 249–275). Cambridge: Cambridge University Press. Ochs, E. (1996). Linguistic resources for socializing humanity. In J. J. Gumperz and S. C. Levinson (eds.), Rethinking Linguistic Relativity (pp. 407–437). Cambridge: Cambridge University Press. O’Keeffe, A. (2018). Corpus-based function-to-form approaches. In A. H. Jucker, K. Schneider, and W. Bublitz (eds.), Methods in Pragmatics (pp. 587–618). Berlin/ Boston: Mouton de Gruyter. Palacios-Martínez, I. (2018). ‘Help me move that, blood’ A corpus-based study of the syntax and pragmatics of vocatives in the language of British teenagers. Journal of Pragmatics, 130, 33–50.

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

311

312

KARIN AIJMER

Pichler, H. (2016). Introduction: Discourse-pragmatic variation and change. In H. Pichler (ed.), Discourse-Pragmatic Variation and Change in English: New Methods and Insights (pp. 1–18). Cambridge: Cambridge University Press. Pöldvere, N., Paradis, C., and Johansson, V. (2017). The London-Lund Corpus 2: A New Corpus of Spoken British English in the Making (LLC 2). Lund University. https:// portal.research.lu.se/en/publications/the-london-lund-corpus-2-a-new-corpus-ofspoken-british-english-i. Romero-Trillo, J., ed. (2008). Pragmatics and Corpus Linguistics: A Mutualistic Entente. Berlin/New York: Mouton de Gruyter. Romero-Trillo, J. (2018). Corpus pragmatics and second language pragmatics: A mutualistic entente in theory and practice. Corpus Pragmatics, 2, 113–127. Ronan, P. (2019). Silly much? Tracing the spread of a new expressive marker in recent corpora. Corpus Approaches into World Englishes and Language Contrasts (Studies in Variation, Contacts, and Change in English, Vol. 20), ed. H. Parviainen, M. Kaunisto, and P. Pahta. Helsinki: VARIENG. Rühlemann. C., and Aijmer, K. (2015). Introduction. Corpus pragmatics: Laying the foundations. In K. Aijmer and C. Rühlemann (eds.), Corpus Pragmatics: A Handbook (pp. 1–26). Cambridge: Cambridge University Press. Rühlemann, C., and M. B. O’Donnell (2012). Introducing a corpus of conversational narratives: Construction and annotation of the Narrative Corpus. Corpus Linguistics and Linguistic Theory, 8(2), 313–350. Sacks, H., Schegloff, E., and Jefferson, G. (1974). A simplest systematics for the organization of turn-taking for conversation. Language, 50(4), 696–735. Schiffrin, D. (1987). Discourse Markers. Cambridge: Cambridge University Press. Searle, J. R. (1969). Speech Acts: An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. Sperber, D., and Wilson, D. (1986). Relevance: Communication and Cognition. Oxford: Blackwell. Staley, L., and Jucker, A. H. (2021). The uh-deconstructed pumpkin pie: The use of uh and um in Los Angeles restaurant server talk. Journal of Pragmatics, 172, 21–34. Stange, U. (2019). The social life of emotive interjections in spoken British English. Scandinavian Studies in Linguistics, 10(1), 174–193. Stenström, A.-B. (2011). Pauses and hesitation. In G. Andersen and K. Aijmer (eds.), Pragmatics of Society (pp. 537–567). Berlin: Mouton de Gruyter. Svartvik, J. (1990). The London–Lund Corpus of Spoken English: Description and Research. Lund: Lund University Press/Bromley: Chartwell-Brat. Timmis, I. (2015). Tails. In K. Aijmer and C. Rühlemann (eds.), Corpus Pragmatics: A Handbook (pp. 304–327). Cambridge: Cambridge University Press. Tottie, G. (2011). Uh and um as sociolinguistic markers in British English. The International Journal of Corpus Linguistics, 16(2), 173–196. Traugott, E. C., and Dasher, R. B. (2002). Regularity in Semantic Change. Cambridge: Cambridge University Press. Van Olmen, D., and Šinkūnienė, J., eds. (2021). Pragmatic Markers and Peripheries. Amsterdam/Philadelphia: John Benjamins. Verschueren, J. (1999). Understanding Pragmatics. London: Arnold. Wagner, S. E., Hesson, A., and Little, H. M. (2016). The use of referential general extenders across registers. In H. Pichler (ed.), Discourse Pragmatic Variation and

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

13 Corpus Pragmatics

Change in English: New Methods and Insights (pp. 211–231). Cambridge: Cambridge University Press. Waters, C. (2016). Practical strategies for elucidating discourse-pragmatic variation. In H. Pichler (ed.), Discourse-Pragmatic Variation and Change in English: New Methods and Insights (pp. 41–55). Cambridge: Cambridge University Press. Weisser, M. (2015). Speech act annotation. In K. Aijmer and C. Rühlemann (eds.), Corpus Pragmatics: A Handbook (pp. 84–110). Cambridge: Cambridge University Press. Weisser, M. (2017). Corpora. In A. Barron, G. Yueuo, and G. Steen (eds.), The Routledge Handbook of Pragmatics (pp. 41–52). London/New York: Routledge.

https://doi.org/10.1017/9781108989275.014 Published online by Cambridge University Press

313

14 Prosodic Pragmatics in Context Jesús Romero-Trillo and Yalda Sadeghi

14.1 Introduction The relationship between context and prosody is undoubtedly one of the most intuitive ones in language. At the same time, it is one of the most difficult to describe because it is based on acoustic cues that need only milliseconds to create an image in our brain. However, speakers of a language can generally understand their interlocutors’ emotional and cognitive status through their prosodic realization. Paraphrasing the article published by Bolinger (1972) and entitled “accent is predictable (if you’re a mind reader),” we can say: prosody is predictable (if you’re a contextual reader). Pragmatics mainly focuses on spoken language and investigates how speakers communicate and understand information with the aid of the contextual features that are present in any interaction. Nevertheless, the description of the synergic relationship between prosody and pragmatics is an ongoing task that has not been fully approached by many linguists. In fact, Levinson’s (1983: 36) famous dictum is still valid: “Various syntactic rules seem to be properly constrained only if one refers to pragmatic conditions; and, similarly, for matters of stress and intonation.” Prosodic pragmatics (Romero-Trillo 2012, 2019) is the branch of pragmatics that attempts to identify the intentionality of the speaker’s meaning in a real context based on the analysis of the suprasegmental aspects of speech production. If prosody is the study of the way an utterance is pronounced in unison with the perceptual features of pitch, length, and loudness, prosodic pragmatics is the study of the acoustic and cognitive contextual parameters in conversation. A significant phenomenon that needs to be highlighted in the description of the prosodic system of any language is the interplay between the intonational and the lexical-grammatical forms (Bolinger 1978; Cruz-Ferreira 1987; Ladd 1996), as prosody interacts with the grammatical, pragmatic, and emotional This work was supported by the Ministerio de Economía y Competitividad (Spain) under Grant PID2019-105678RB-C22.

https://doi.org/10.1017/9781108989275.015 Published online by Cambridge University Press

14 Prosodic Pragmatics in Context

levels of language production in real contexts (Snow and Balog 2002). In this sense, slight variations in prosodic performance, such as the prominence, the duration, or the speech rate of an utterance, can influence the linguistic and emotional interpretation of a message, as well as condition the relationship between the speaker and hearer. In return, the interpretation of prosody can also alter the contextual variables in which a message is embedded.

14.2 Pragmatics and Prosody We must differentiate between the general patterns of prosody that function at the phonological level in a language, and the specific acoustic analysis of prosody in the interpretation of meaning at the micro-level of conversation, bearing in mind all idiolectal features of the speakers in terms of sex, age, emotional state, dialectal origin, etc. There are different models of prosodic analysis that are generally classified under what can be described as continuous or categorial. The continuous models are mainly interested in the description of the variation of the F0, i.e., the fundamental frequency, while the categorial models try to identify the F0 movements according to an established paradigm that assigns a communicative meaning to the prosodic variation. Within the categorial approach we can mention two competing systems: the Tone and Break Indices system (ToBI) (Pierrehumbert 1980; Pierrehumbert and Hirschberg 1990), which is based on what is called the autosegmental theory of intonation; and the Nuclear Tone Theory (Cruttenden 1997), which focuses on the nuclear accent of the utterance and develops the original descriptions by Halliday (1967) and O’Connor and Arnold (1973), and is the preferred one in prosodic pragmatics research (Romero-Trillo 2012, 2019). In the analysis of context, one of the prosodic elements that is more significant in any communicative event is prominence, which is used by the speaker to signal the focus of the utterance both at the word level and at the utterance level. Lexical stress (Bolinger 1958; Lehiste 1970) refers to the inherent prosodic prominence of a syllable, generally louder, stronger, and longer than the rest of the syllables within the same word. For this purpose, numerous studies have analyzed “stress” or “accent” at a segmental level by considering some of the acoustic correlates of stress (duration, frequency, amplitude, and intensity). For instance, Fry (1954) studied the influence of the intensity and duration cues on the perception of linguistic stress patterns; Liberman (1960) concentrated on how duration influenced the judgment of what is being stressed; and Pollock et al. (1993) observed how fundamental frequency, duration, and intensity indicate word stress. However, although every word has a stressed syllable, not every word within an utterance receives the same emphasis. The pseudo-equivalent terms “focus,” “prominence” (Brazil 1975), “pitch accent” (Bolinger 1958), or “tonic” – the one we shall predominantly use in the chapter – (Halliday

https://doi.org/10.1017/9781108989275.015 Published online by Cambridge University Press

315

316

JESÚS ROMERO-TRILLO AND YALDA SADEGHI

1967; Halliday and Matthiessen 2014) refer to the additional emphasis assigned to a specific part of the message by means of some acoustic properties to one of the stressed syllables, the focus, which becomes the most meaningful element in the tone unit. Therefore, the focus has been acoustically analyzed at a suprasegmental level by, among others, Cooper et al. (1985) and Nooteboom and Kruyt (1987), who investigated how accentuation of words in context was related to fundamental frequency and duration, both separately and in interaction. Also, Terken (1991) discussed the importance of fundamental frequency in the perception of prominence within accented words, and Eefting (1991) underlined the importance of accentuation in the duration of words in context. The study of the tonic centers around the relationship between the form of the message (duration, fundamental frequency, and intensity) and the discourse meaning conveyed to establish coherence within the specific context. In fact, although the traditional unit of analysis of the tonic used to be the phoneme or syllable within individual words (Pollock et al. 1993; De Jong and Zawaydeh 2002), or within syntactic units: clauses or sentences (Cooper et al. 1985; Eady and Cooper 1986; Heldner and Strangert 2001), other approaches have selected the tone unit as natural environment to study the tonic, due to its pragmatic role in the structure of the message (Romero-Trillo 1994). One of the most significant approaches in the study of the tonic inside the tone unit has been the description of its contrastive function in communication (Halliday 1970; Bolinger 1972; Cooper et al. 1985; Cruttenden 1997; Halliday and Matthiessen 2014). The other approach has been the identification of tonicity to the status of information whereby the tonic highlights the new information, and the rest of the unit the given information, which remains nonemphasized (Halliday 1967, 1970; Bolinger 1972; Chafe 1974; Ladd 1980; Gussenhoven 1984; Nooteboom and Kruyt 1987; Ladd 1996; Heldner and Strangert 2001; Ward and Birner 2001; Ward and Birner 2003; Halliday and Matthiessen 2014). It is interesting to notice in this sense that contextually predictable words are read faster than the information that is presented for the first time (Lieberman 1963). As mentioned above, for the analysis of interaction from a prosodic pragmatic perspective, Romero-Trillo (1994) proposed the Nuclear Tone theory due to the transformation of the quantitative acoustic measurements to a distinct qualitative classification relating the basic contours of English intonation to the pragmatic functions realized in a specific context. The model follows the premise that each intonation group has a nuclear accent – the primary stressed syllable (Vanderslice and Ladefoged 1972; Crystal 1986) – which is (part of) the word that marks the most salient information in the group. This element functions as the “punting” cognitive element in the specific contextual moment of the conversation. The acoustic relevance of the Nuclear Tone theory is that the nuclear tone coincides with the place of the utterance that realizes the most salient grammatical information, as well as the pragmatic and contextual information.

https://doi.org/10.1017/9781108989275.015 Published online by Cambridge University Press

14 Prosodic Pragmatics in Context

In acoustic terms, the nuclear tone is characterized by cued phrase-final lengthening (Snow and Balog 2002), and the contour direction of the nuclear tone is mainly used to convey intentionality and attitude. In many cases, this acoustic peak is generally produced toward the right edge of the tone group following the Hallidayan principle of “end focus” (Halliday 1967). However, this general principle – called “unmarked tonicity” – is frequently altered in casual conversation due to the communicative contextual needs related to emphasis and contrast. In these cases, we would speak about “marked tonicity.” For Halliday, intonation is the result of the systemic realization of three different autonomous but interrelated systems that are managed by the speaker according to his/her communicative intentions and the understanding of the context that is shared with the recipient. The three systems are the following: • Tonality; the division of the talk into chunks of information selected by the speaker to organize the message (tone units). • Tonicity: the identification of the nuclear accent. • Tone: the contour direction of the tonic. The representation of the contours of the English essential tones, called “primary tones,” are as follows: • • • • •

Tone 1: falling \ Tone 2: rising / Tone 3: level-(rise) =(/) Tone 4: (rise)-fall-rise (/)\/ Tone 5: (fall)-rise-fall (\)/\

To these primary tones, Halliday identified some standard combinations, called “compound tones,” that are composed of the sequence of the primary tones 13 and 53. The system was completed with a selection of the so-called “secondary tones,” which are slight modifications of the primary ones. Romero-Trillo (2001) enlarged Halliday’s model with the addition of Tone 0 for the cases in which an element has no tonicity or tone, which often occurs in the case of pragmatic markers in final position after the tonic element (Romero-Trillo 2018b).

14.3 Information and Context One of the essential features of the relationship between prosody and pragmatics is the capacity of the speakers to move forward in a conversation in a parallel and coordinated way. This phenomenon has been labeled “Pragmatic punting,” which can be defined as “the cognitive process that certifies that the communicative transfer from the speaker to the listener in a conversation has achieved its goal” (Romero-Trillo 2014: 209). The signals that realize this function can be gestural, e.g., nodding, or vocal/verbal. In the case of vocal-verbal signals, the

https://doi.org/10.1017/9781108989275.015 Published online by Cambridge University Press

317

318

JESÚS ROMERO-TRILLO AND YALDA SADEGHI

role of prosody is tantamount to the lexical value of the markers that realize any pragmatic function in conversation (Romero-Trillo and Newell 2012). The notion of “pragmatic punting” comes from the image of a punt sailing along the water with the only aid of a punting pole whose function is both steering and propulsion. Likewise, a conversation flows dynamically in the context via some pragmatic and prosodic cues. These cues – the pragmatic markers – are inextricably linked to their intonation contour, which typically have a falling or level contour when they signal neutral feedback but use a rising tone when they request repetition or a halt in the interaction for clarification or disagreement purposes. Another essential concept in the relationship between prosody and context is the distinction between given and new information, i.e., what is known and not known by the speakers in a conversation, realized through tonicity. Although in the past, and in some language teaching methods, this concept was conceived as a yes/no feature, i.e., tonic vs. non-tonic, most linguists nowadays have opted for a dynamic understanding of this notion, which we will classify as the cognitive, and the discourse-pragmatic approaches. Cognitivists believe that the status of information, i.e., givenness or newness, is dependent on the context represented in the addressee’s consciousness and is related to the speaker’s guess about the shared knowledge between speaker and listener (Chafe 1974; Haviland and Clark 1974; Clark and Haviland 1977; Chafe 1987). Therefore, the information status of some concepts may vary throughout a conversation. The discourse-pragmatic approach assigns the information status on the grounds of recoverability (Geluykens 1991; Romero-Trillo 1993; Halliday and Matthiessen 2014) or evokedness (Prince 1981). In other words, this approach relies on the explicit information found in the immediate context. Therefore, given is “something that is present or derivable from the situation; or simply because the speaker takes it for granted and assumes that the hearer will do likewise” (Halliday 1970: 42), whereas new relates to the “information which the speaker has decided to present as not being already available to the hearer; that is, as something the hearer could not have supplied for himself. It may be something that has not been mentioned before or is unexpected or contrary to what has been implied and so on” (p. 40). In sum, we can say that the focus is what makes the message cognitively salient for the listener, who orients his/her interpretation of the surrounding context vis-à-vis the succession of foci in the speaker’s message. In ordinary face-to-face conversation the listener can also identify the nuclei of the message by observing the speaker’s accompanying gestures in what has been defined as “prosodic pointing” (Madella and Romero-Trillo 2019). However, sometimes paralinguistic phenomena may be absent, as in the case of phone conversations, where only acoustic cues are available to the interactants. On other occasions the paralinguistic cues may be misinterpreted due to different cultural norms in different languages (e.g., in India a quick head wobble typically means “yes, OK”).

https://doi.org/10.1017/9781108989275.015 Published online by Cambridge University Press

14 Prosodic Pragmatics in Context

14.4 Adaptive Context and Prosody The role of context as a key element in the interpretation of language has been studied since the early 1920s from different linguistic traditions that try to account for the sometimes elusive, but always real, relationship between language and context. This relationship, which has been labeled “contextualism,” can be defined as “the position that the truth-conditions of knowledgeascribing and knowledge-denying sentences (sentences of the form ‘S knows that P’ and ‘S doesn’t know that P’ and related variants of such sentences) vary in certain ways according to the context in which they are uttered” DeRose (1999: 187). The main approaches of contextualism throughout history can be summarized as follows: the ethno-naturalistic approach (Malinowski 1923; Goodwin and Duranti 1992); the conversational approach (Heritage 1984; Pomerantz 1984); the functional-systemic approach (Halliday and Hasan 1989; LeckieTarry 1995); the discourse-semantic approach (Glandzberg 2002; Fetzer and Oishi 2011); and the pragmatic approach (Searle 1969; Levinson 1983; Sperber and Wilson 1986/1995). As it can be observed, the principle of “contextualism” has become increasingly influential, and it is important to highlight that many of the theories share ontological affinities in which the pivoting element is the ascription of different degrees of “radicality,” in the etymological sense of “root.” In other words, theories differ in how radical are utterances in relation to context to understand the conditions in which a given utterance can be more or less open to ambiguity or creativity. The “mild contextualists” will consider that “any piece of contextual information may turn out to be relevant to establishing the correct interpretation for the speech act” (Recanati 2002: 107). More radical approaches to contextualism, like the Graded Silence Hypothesis, advocate for the independent essence of lexical meanings: “coded meanings will be accessed upon encounter, regardless of contextual information or authorial intent” (Giora 2003, 10). In other words, salient meanings are only contextualized by speakers when they collide with context in the first phase of comprehension and, as a result, they require a reinterpretation by the addressee. One of the most relevant approaches that also studies the interaction between L1 and L2 speakers is the socio-cognitive theory, which considers that any act of communication is the realization of a cognitive process that follows a trial-and-error dynamic mental process in a conversation (Kecskes and Fenghui 2009). This theory, developed within the Dynamic Model of Meaning (DMM), avers that the participants in an interaction must share and mutually understand a: (1) “Private context,” i.e., “the world knowledge that is represented in the head of the interlocutors” (Kecskes 2008, 403). (2) “Linguistic context,” the actual linguistic output: “Our mind exists simultaneously both in the head and in the world. So linguistic context is what

https://doi.org/10.1017/9781108989275.015 Published online by Cambridge University Press

319

320

JESÚS ROMERO-TRILLO AND YALDA SADEGHI

is uttered (or written) ‘out there’ in the world by a speaker . . . . So the linguistic context is created online” (Kecskes 2006: 234). (3) “Situational context,” defined as “the world knowledge that is outside in the world as interpreted by the interlocutors” (Kecskes 2008, 404). It is important to highlight that for the DMM, the three contexts are equally important and need to co-occur – although not simultaneously – to achieve successful communication. In other words, they can occur at different stages in the communication process: “Both private context and actual situational context have a decisive role in the communication process, but at different stages” (Kecskes 2008: 403). The interest of the DMM and its trial-and-error process is its impossible autonomous verification by the speaker, i.e., it demands the aid of other speakers, or of other objects, that test the veracity of one’s cognitive construct. In other words, I can only know if my utterance has been understood as a promise if the addressee understands it as such and confirms that this is the case. This empirical conception is analogous to the perception of an element: I am certain that an object can roll if my senses (my touch and possibly my sight in this case) verify that this is the case. Therefore, from a contextualist perspective it can be said the objective of any act of communication is to reach mutual comprehension by interlocutors and, as a result, the avoidance of mis/ non-understanding (Bazzanella and Damiano 1999). To understand the role of prosody in contextualism, Romero-Trillo and Maguire (2011) developed the notion of “Adaptive Context” as a complement to the DMM (see Figure 14.1). Adaptive Context can be defined as the process that allows the participants in a conversation to adapt the private, lexical, and situational contextual elements in order to avoid misunderstanding. In other words, Adaptive Context contributes to the scaffolding of information by feeding the prosodic and discursive information through an online positive feedback process. The ultimate function of Adaptive Context is to enable participants in a conversation to reach the most likely optimal meaning from a pragmatic repertoire.

Lexical Context

Adaptive Context Private Context

Situational Context

Figure 14.1 The function of Adaptive Context

https://doi.org/10.1017/9781108989275.015 Published online by Cambridge University Press

14 Prosodic Pragmatics in Context

This online positive feedback process is called “contextual sifting,” defined as “the process of cognitive filtering that leaves out the incorrect assumptions in a given communicative situation, and sieves through the correct elements to guarantee successful communication” (Romero-Trillo and Maguire 2011: 234). This theoretical update adds the phenomenon of misunderstanding as an intrinsic and productive mechanism in interaction. The important thing is that this process does not only consider prosodic or pragmatic factors but also includes cultural elements that can interfere in the process (Wierzbicka 2010).

14.5 Pragmatic Markers and Adaptive Context Among the many strategies that speakers have at their disposal for successful communication, acoustic cues are possibly the most efficient to emphasize and make certain elements in the discourse semantically prominent. However, the notion of prominence is a complex issue, especially when communication occurs between native and nonnative speakers of a language. In fact, the mechanics of the suprasegmental features of a language may not be immediately transparent to the learner, and their incorrect use may result in communication failures or, in the long run, in pragmatic fossilization when the grammatical and lexical competence of the language learner evolves at a higher speed than the pragmatic competence, which remains anchored in a prior learning stage (Romero Trillo 2002). Pragmatic markers can be considered the prototypical elements that realize Adaptive Context, as they are used by the addressor and addressee to confirm that a message is processed correctly. In this sense, when misunderstanding arises, pragmatic markers promote the clarification of meaning through supportive strategies – within the realm of Adaptive Context – that do not interfere with the autonomy of the linguistic context of the message, nor with the private and situational contexts. On the contrary, the process guarantees that repair can be triggered at any moment of the interaction without loss of face for the parties. It can be said that Adaptive Context is a positive politeness strategy (Brown and Levinson 1978), as it avoids showing lack of understanding on the part of the addressee and prevents face-threatening comprehension confirmations by the addressor. For this reason, one essential aspect in the study of pragmatic markers is the description of their intonation contour vis-à-vis their contextual realization. From a prosodic perspective, pragmatic markers can appear with multiple intonation patterns and have a distinct prosodic value if they are the only elements in a tone unit, or if they appear inside a tone unit accompanied by other (usually lexical) elements (Romero-Trillo 2012). For example, in the case of English, the use of the pragmatic marker well in the London–Lund Corpus appears with the five primary tones described by Halliday (1967), realizing the function of neutral disagreement in 62.2 percent of the cases

https://doi.org/10.1017/9781108989275.015 Published online by Cambridge University Press

321

322

JESÚS ROMERO-TRILLO AND YALDA SADEGHI

(Romero-Trillo 2001). This multiplicity of prosodic features, together with their variation in pragmatic meaning, are the reasons why pragmatic markers are difficult to learn by nonnative speakers of a language (Romero-Trillo 1997; Romero-Trillo and Newell 2012). The following examples from the London–Lund Corpus of Spoken English (Svartik 1990) show the role of pragmatic markers as Adaptive Management tools: (1)

B B A A A

^are you in t\ouch with the St {B/eès} cr/owd# / or / ^w\ell# / you ^kn/ow#. / to a ^certain ext/ent#

In Example (1), the markers “well” and “you know,” each in a separate unit and with diverse tones (falling + rising), indicate the indeterminacy of the response to the question “are you in touch with the St. Bee’s crowd?, or . . .” Here, Adaptive Context would be responsible for the modulation of the response using the pragmatic parkers to show the stance of the speaker and thus help the addressee to adjust his/her cognitive expectations. In fact, the contextual effect would have a completely different orientation if the response had offered no additional elements apart from the direct information, as shown in Example (2): (2)

B B A

^are you in t\ouch with the St {B/eès} cr/owd# / or / to a ^certain ext/ent#

In fact, we can say that Example (2) is probably counterintuitive as the use of a rising intonation in the response – typical of questions – would be anomalous, if it were not preceded by a marker that serves as a pragmatic trampoline to allow for a response with an interrogative intonation. Therefore, we cannot separate the essential function of Adaptive Context from the role of prosody (Romero-Trillo 2001). In fact, it is very likely that the response would have realized a falling intonation, with a meaning of full assertion, if Adaptive Context had not been present, as indicated by Example (3): (3) B B A

^are you in t\ouch with the St {B/eès} cr/owd# / or / to a ^certain ext\ent#

Another interesting fact is that pragmatic markers can complement each other when they appear together. In Example (4) we can see the appearance of one of the two pragmatic markers would give a different meaning to the response: (4) B B A A

^are you in t\ouch with the St {B/eès} cr/owd# / or / ^w\ell# / to a ^certain ext/ent#

https://doi.org/10.1017/9781108989275.015 Published online by Cambridge University Press

14 Prosodic Pragmatics in Context

Or alternatively, (5) B B A A

^are you in t\ouch with the St {B/eès} cr/owd# / or / you ^kn/ow# . / to a ^certain ext/ent#

One special case of the use of pragmatic markers in Adaptive Context is when they realize the function of feedback. Feedback can be defined as the use of “some linguistic elements to show that the listener is following the ideas expressed in the message by the speaker” (Romero-Trillo 2001: 536). The use of feedback in a conversation confirms that the message is being processed and accepted by the recipient to maintain the cognitive rhythm of a conversation without disruption. Feedback signals that the participants in a conversation are active in the “here-and-now” coordinates that feature context. In fact, feedback is the most frequently realized function by pragmatic markers in the London–Lund Corpus, totaling the 39.57 percent of the cases (Romero-Trillo 2001).

14.6 Conclusion and Further Research Context is a personal construal of reality that has a dynamic nature and is difficult to share in its full shape with other speakers. In this sense, conversation is the paradigmatic language activity for the cognitive construction of context, as it evolves with each turn, although the shared contextual information between speaker and hearer may differ to the extent that misunderstandings may arise, among other phenomena, with prosodic misinterpretations. We can identify some challenges in prosodic pragmatics, such as the intercultural understanding of acoustic cues and intentionality in relation to emotion research in the narratives of refugees and migrants (RomeroTrillo, 2023). Also, the role of prosodic pragmatics in second language teaching remains a pending issue as a way to transfer the pragmatic competence in the L1 to the pragmatic functions in an L2 through prosody (Romero-Trillo et al. 2022). To sum up, the present chapter has shown the intimate relationship between prosody, information, and context in communication. Starting from the essential acoustic parameters of speech, it has revised the most influential theories of intonation and has illustrated the use of the Nuclear Tone Theory in prosodic pragmatics as an essential tool to understand the communicative process and to avoid misunderstanding through the cognitive adaptation of a message in a specific context. Also, the increasing multimodal approach to the field will surely render very interesting results about the mutual dependency of prosodic pragmatics and context.

https://doi.org/10.1017/9781108989275.015 Published online by Cambridge University Press

323

324

JESÚS ROMERO-TRILLO AND YALDA SADEGHI

References Bazzanella, C., and Damiano, R. (1999). The interactional handling of misunderstanding in everyday conversations. Journal of Pragmatics, 31, 817–836. Bolinger, D. (1958). A theory of pitch accent in English. Word, 14, 109–149. Bolinger, D. (1972). Accent is predictable (if you’re a mind reader). Language, 37, 83–96. Bolinger, D. (1978). Intonation across languages. In J. H. Greenberg, C. A. Ferguson, and E. A. Moravcsik (eds.), Universals of Human Language, Vol. II: Phonology (pp. 471–524). Stanford: Stanford University Press. Brazil, D. (1975). Discourse Intonation 1/2. Birmingham: Birmingham University Press. Brown, P., and Levinson, S. C. (1978). Politeness: Some Universals in Language Usage. Cambridge: Cambridge University Press. Chafe, W. L. (1974). Language and consciousness. Language, 50, 111–133. Chafe, W. L. (1987). Cognitive constraints on information flow. In R. S. Tomlin (ed.), Coherence and Grounding in Discourse (pp. 21–51). Amsterdam: John Benjamins. Clark, H. H., and Haviland, S. E. (1977). Comprehension and the given-new contract. In R. D. Freedle (ed.), Discourse Production and Comprehension (pp. 1–40). Norwood: Ablex Publishing Corporation. Cooper, W. E., Eady, S. J., and Mueller, P. R. (1985). Acoustical aspects of contrastive stress in question–answer contexts. Journal of the Acoustic Society of America, 77, 2142–2156. Cruttenden, A. (1997). Intonation, 2nd ed. Cambridge: Cambridge University Press. Cruz-Ferreira, M. (1987). Non-native interpretative strategies for intonational meaning. In J. A. Leather and J. Leather (eds.), Sound Patterns in Second Language Acquisition (pp. 103–120). Dordrecht: Foris Publications. Crystal, D. (1986). Prosodic development. In P. J. Fletcher and M. Garman (eds.), Studies in First Language Development (pp. 174–197). New York: Cambridge University Press. De Jong, K., and Zawaydeh, B. (2002). Comparing stress, lexical focus and segmental focus: patterns of variation in Arabic vowel duration. Journal of Phonetics, 30, 53–75. DeRose, K. (1999). Contextualism: An explanation and defense. In J. Greco and E. Sosa (eds.), The Blackwell Guide to Epistemology (pp. 187–206). Oxford: Blackwell. Eady, S. J., and Cooper, W. E. (1986). Speech intonation and focus location in matched statements and questions. Journal of the Acoustical Society of America, 80, 402–415. Eefting, W. 1991. The effect of ‘information value’ and ‘accentuation’ on the duration of Dutch words, syllables and segments. Journal of the Acoustic Society of America 89: 412–423. Fetzer, A., and Oishi. E., eds. (2011). Context and Contexts: Parts Meet Whole? Amsterdam: John Benjamins. Fry, D. B. (1954). Duration and intensity as physical correlates of linguistic stress. Journal of the Acoustic Society of America, 27, 765–768. Geluykens, R. (1991). Information flow in English conversation: A new approach to the Given-New distinction. In E. Ventola (ed.), Functional and Systemic Linguistics. Approaches and Uses (pp. 141–169). Berlin: Mouton de Gruyter. Giora, R. (2003). On Our Mind: Salience, Context and Figurative Language. Oxford: Oxford University Press. Glandzberg, M. (2002). Context and discourse. Mind and Language, 17, 333–375.

https://doi.org/10.1017/9781108989275.015 Published online by Cambridge University Press

14 Prosodic Pragmatics in Context

Goodwin, C., and Duranti, A. (1992). Rethinking context: An introduction. In A. Duranti and C. Goodwin (eds.), Rethinking Context: Language as an Interactive Phenomenon (pp. 1–42). Cambridge: Cambridge University Press. Gussenhoven, C. (1984). On the Grammar and Semantics of Sentence Accents. Dordrecht: Foris. Halliday, M. A. K. (1967). Intonation and Grammar in British English. The Hague: Mouton Halliday, M. A. K. (1970). A Course in Spoken English: Intonation. London: Oxford University Press. Halliday, M. A. K., and Hasan, R. (1989). Language, Context and Text: Aspects of Language in a Social-Semiotic Perspective, 2nd ed. Oxford: Oxford University Press. Halliday, M. A. K., and Matthiessen, M. I. M. (2014). Halliday’s Introduction to Functional Grammar, 4th ed. Oxford: Routledge. Haviland, S. E., and Clark, H. (1974). What’s new? Acquiring new information as a process in comprehension. Journal of Verbal Learning and Verbal Behaviour, 13, 512–521. Heldner, M., and Strangert, E. (2001). Temporal effects of focus in Swedish. Journal of Phonetics, 29, 329–361. Heritage, J. (1984). Garfinkel and ethnomethodology: Developments in Conversation Analysis. Sociolinguistics Newsletter, 1, 1–16. Kecskes, I. (2006). On my mind: Thoughts about salience, context and figurative language from a second language perspective. Second Language Research, 22, 219–237. Kecskes, I. (2008). Dueling contexts: A dynamic model of meaning. Journal of Pragmatics, 40, 385–406. Kecskes, I., and Fenghui, Z. (2009). Activating, seeking, and creating common ground: A socio-cognitive approach. Pragmatics & Cognition, 17, 331–355. Ladd, R. (1980). The Structure of Intonational Meaning: Evidence from English. Bloomington: Indiana University Press. Ladd, R. (1996). Intonational Phonology. Cambridge: Cambridge University Press. Leckie-Tarry, H. (1995). Language and Context: A Functional Linguistic Theory of Register. London/New York: Pinter Publishers. Lehiste, I. (1970). Suprasegmentals. Cambridge, MA: MIT Press. Levinson, S. C. (1983). Pragmatics. Cambridge: Cambridge University Press. Liberman, P. (1960). Some acoustic correlates of word stress in American English. Journal of the Acoustic Society of America, 32, 451–454. Lieberman, P. (1963). Some effects of semantic and grammatical context on the productions and perception of speech. Language and Speech, 6, 172–187. Madella, P., and Romero-Trillo, J. (2019). Prosodic Pointing in inferential comprehension: The application of Relevance Theory to L2 listening instruction. Letronica 4. http://dx.doi.org/10.15448/1984-4301.2019.4.34157. Malinowski, B. [1923] (1989). The problem of meaning in primitive languages. In C. Ogden and I. Richards (eds.), The Meaning of Meaning (pp. 451–510). New York: Harcourt, Brace and World. Nooteboom, S. G., and Kruyt, J. G. (1987). Accents, focus distribution and the perceived distribution of given and new information: an experiment. Journal of the Acoustic Society of America, 82, 1512–1524. O’Connor, J. D., and Gordon, F. A. (1973). Intonation of Colloquial English. London: Longman.

https://doi.org/10.1017/9781108989275.015 Published online by Cambridge University Press

325

326

JESÚS ROMERO-TRILLO AND YALDA SADEGHI

Pierrehumbert, J. (1980). The phonology and phonetics of English intonation. Ph.D. dissertation, Massachusetts Institute of Technology. Pierrehumbert, J., and Hirschberg, J. (1990). The meaning of intonation contours in the interpretation of discourse. In Philip R. Cohen, J. Morgan, and M. E. Polack (eds.), Intention in Communication (pp. 371–411). Cambridge, MA: MIT Press. Pollock, K. E., Brammer, D. M., and Hageman, C. F. (1993). An acoustic analysis of young children’s productions of word stress. Journal of Phonetics, 21, 183–203. Pomerantz, A. (1984). Agreeing and disagreeing with assessments: Some features of preferred/dispreferred turn shapes. In J. Atkinson and J. Heritage (eds.), Structures of Social Action (pp. 57–101). Cambridge: Cambridge University Press. Prince, E. F. (1981). Toward a taxonomy of given-new information. In P. Cole (ed.), Radical Pragmatics (pp. 223–255). London: Academic Press. Ramírez-Verdugo, M. D. (2021). Intonation in L2 Discourse: Research Insights. London: Routledge. Recanati, F. (2002). Does communication rest on inference? Mind & Language, 17, 105–126. Romero-Trillo, J. (1993). Reading aloud and the structure of information. Estudios Ingleses de la Universidad Complutense, 1, 133–142. Romero-Trillo, J. (1994). Ahm, ehm . . . You call it theme? A thematic approach to spoken English. Journal of Pragmatics, 22, 495–509. Romero-Trillo, J. (1997). Your attention please: Pragmatic mechanisms to obtain the addressee’s attention in English and Spanish conversations. Journal of Pragmatics, 28, 205–221. Romero Trillo, J. (2001). A mathematical model for the analysis of variation in discourse. Journal of Linguistics, 37, 527–550. Romero-Trillo, J. (2002). The pragmatic fossilization of discourse markers in non-native speakers of English. Journal of Pragmatics, 34, 769–784. Romero-Trillo,J., ed. (2012). Pragmatics and Prosody in English Language Teaching. Dordrecht: Springer. Romero-Trillo, J. (2014). “Pragmatic punting” and prosody: Evidence from corpora. In M. A. Gómez González, F. J. Ruíz de Mendoza Ibáñez, F. Gonzálvez García, and A. Downing (eds.), The Functional Perspective on Language and Discourse: Applications and Implications (pp. 209–221). Amsterdam: John Benjamins. Romero-Trillo, J. (2018a). Corpus pragmatics and second language pragmatics: A mutualistic entente in theory and practice. Corpus Pragmatics, 2, 113–127. Romero-Trillo, J. (2018b). Prosodic modeling and position analysis of pragmatic markers in English conversation. Corpus Linguistics and Linguistic Theory, 14, 169–195. Romero-Trillo, J. (2019). Prosodic pragmatics and feedback in intercultural communication. Journal of Pragmatics, 151, 91–102. Romero-Trillo, J. (2023). “Life is a journey” (literally): Conflict and emotion in the narratives of migrants and refugees. Talk delivered at King’s College London, January 18, 2023. Romero-Trillo, J., and Maguire, L. (2011). Adaptive context: The fourth element of meaning. International Review of Pragmatics, 3, 228–241. Romero-Trillo, J. and Newell, J. (2012). Prosody and feedback in native and non-native speakers of English. In J. Romero-Trillo (ed.), Pragmatics, Prosody and English Language Teaching (pp. 117–132). Dordrecht: Springer.

https://doi.org/10.1017/9781108989275.015 Published online by Cambridge University Press

14 Prosodic Pragmatics in Context

Romero-Trillo, J., Ramírez-Verdugo, D., Jiménez-Vilches, R., Mestre-Mestre, E., Avila, Ledesma, Sadeghi, N. Y., and Madella, P. (2022). Teaching English Prosodic Pragmatics APP (TEPP). Available in Google Play and Apple Store. Searle, J. (1969). Speech Acts. Cambridge: Cambridge University Press. Snow, D., and Balog, H. L. (2002). Do children produce the melody before the words? A review of developmental intonation research. Lingua, 112, 1025–1058. Sperber, D., and Wilson, D. (1986/1995). Relevance: Communication and Cognition. Oxford: Blackwell. Svartvik, J. (1990). The London–Lund Corpus of Spoken English: Description and Research. Lund: Lund University Press/Bromley: Chartwell-Brat. Terken, J. (1991). Fundamental frequency and perceived prominence of accented syllables. Journal of the Acoustical Society of America, 89, 1768–1776. Vanderslice, R., and Ladefoged, P. (1972). Binary suprasegmental features and transformational word-accentuation rule. Language, 48, 819–838. Ward, G., and Birner, B. J. (2001). Discourse and information structure. In D. Schiffrin, D. Tannen, and H. Hamilton (eds.), Handbook of Discourse Analysis (pp. 119–137). Oxford: Basil Blackwell. Ward, G., and Birner, B. J. (2003). Information structure. In L. R. Horn and G. Ward (eds.), Handbook of Pragmatics (pp. 153–174). Oxford: Basil Blackwell. Wierzbicka, A. (2010). Cross-cultural communication and miscommunication: The role of cultural keywords. Intercultural Pragmatic, 7, 1–23.

https://doi.org/10.1017/9781108989275.015 Published online by Cambridge University Press

327

https://doi.org/10.1017/9781108989275.015 Published online by Cambridge University Press

Part IV

Applications of Context Studies

Published online by Cambridge University Press

Published online by Cambridge University Press

15 Language Learning and Assessment in Context Dale Koike and Elisa Gironzetti

15.1 Introduction All activity, including language learning, occurs in some kind of context, whether it is acknowledged or not in research. Language is situated behavior: It happens between and among people, at a given time and space, and with certain goals and effects. Language learning, teaching, and assessment, which are subfields of applied linguistics, are particularly sensitive to contextual factors that affect society in a broad sense. The very labels and terminology that are employed in applied linguistics to refer to first/native (L1), second (L2), and heritage (HL) languages reflect the important role of context. The learning of an L11 does not happen without a rich context that provides input and feedback to the baby. Children learn their L1 from their primary caregivers in a familial, informal setting and, only later, expand their knowledge in more formal and public settings through instruction in an institutional context. An L2 learner typically learns the target language in a formal setting after the L1 and can also be exposed to it in informal settings at a later stage, such as during a study abroad experience. The learning of an L2 in institutional settings is also shaped by contextual factors, such as historical events and social trends, theories, policies, ideas that are current at a given time, and the characteristics of the individuals involved. Last, an HL speaker learns the HL in a familial setting, as in the case of an L1, but without the opportunity to expand their knowledge in more formal and public settings until they are older, and only in some cases (see also Valdés 2005). The learning of an HL, like the L1 and L2, is also shaped by educational and societal contexts, and how people use the heritage and dominant languages in their lives. However, the HL represents special contextual circumstances, for which we have included it here. This

1

A discussion about what it means to learn or acquire a language is outside of the scope of this chapter. The interested reader can explore this debate further in Krashen (1981) and Ellis (1989), among others.

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

332

DALE KOIKE AND ELISA GIRONZETTI

chapter will discuss all three types of language learning, teaching, and assessment in relation to context in the following sections.

15.2 The Contexts of L1 Learning, Teaching, and Assessment The learning, teaching, and assessment of the L1, regardless of whether it is acquired at home or in school, is a crucial factor in human development, in all capacities (i.e., emotional, cognitive, social, linguistic, etc.). In this section, we begin by discussing internal factors that affect L1 development and how language is learned naturally at home, to then address L1 teaching and assessment in formal educational contexts. In learning an L1, before birth, the developing fetus can hear language through the walls of the mother’s womb, a context devoid of much external linguistic stimuli except for sound. However, the affordance of hearing the mother’s voice and perhaps those around her as she interacts with them starts the child on a journey to learning acoustic features of the mother’s language: phonemes, intonation, pitch, etc. (Partanen et al. 2013). When the baby is born, they begin to receive rich input from the context around them through all their senses (hearing, smell, touch, taste, and even sight), which, coupled with their innate disposition to learn languages, accelerate the language learning process. Absent this context, the baby will fail to learn the language, as documented cases have indicated (e.g., the case of Genie, in Curtiss et al. 1974). One need only look at examples of language-deprived children, which represent severe cases of child abuse, in which the child is usually deprived of linguistic input by strict confinement to a small space, with little to no social interaction. These examples show that context and language learning are inseparable, since due to this type of deprivation, the child cannot fully acquire the language. The L1 learning process in context continues throughout childhood.2 When the child reaches puberty, an age at which the brain loses its plasticity and the child typically undergoes changes that affect their social and personal life, L1 learning does not stop. However, linguistic information is then learned and stored in the brain differently.3 From these facts, we know that language learning is not completely dependent on the world external to the individual, but that the internal processes of the brain, the development of the vocal apparatus, and emotional and social growth of the individual, all play a part in interaction with external context. In addition, when considering L1 learning, 2

3

In this chapter, we discuss the language learning and acquisition process referring to a hypothetical child with no impediment that would affect their development. According to the Critical Period Hypothesis (Lenneberg 1967), after puberty one cannot reach native-like competence in an L2 due to changes in the brain plasticity. Since it was initially proposed in the 1960s, this hypothesis has been largely debated, as there is no consensus on a definition of critical period (some researchers, such as DeKeyser and Larson-Hall 2005, support some form of a critical period, while others reject it; e.g., Bialystok 1997). More recently, age effects initially attributed to the existence of a critical period have been revisited and interpreted as due to personal and societal factors, and not cognition-driven developmental factors (van der Silk et al. 2021).

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

15 Language Learning and Assessment in Context

we must distinguish between formal and informal contexts. On the one hand, in an informal context, the language is learned at home, through interaction with caregivers and family members, where certain functions are regularly expressed in daily living, and language use is limited to intimate and familiar registers. On the other hand, in a formal context, the language is learned at school, where it is used in an institutional setting through writing and the use of additional registers. The latter type of language extends the child’s linguistic domains and registers beyond the intimate and familiar registers, the here and now, to enable them to deal with complexities and abstractions and with the vocabulary and grammar of the institutional and formal domains. This expansion is important to allow broader modes of expression and understanding, to lead into adulthood. However, as we discuss in the next section, for students of a higher socioeconomic status, the difference between home and school language is not as great as it is for students from a lower socioeconomic status or those who belong to a minority group, who may use a distinct and often stigmatized variety of the L1 at home.

15.2.1 Societal Factors in L1 Teaching and Assessment Here, we discuss L1 teaching and learning in formal settings, while considering societal factors related to how language is taught and assessed. Although language instructors may be somewhat free to construct their own curricula, at least in part, they are still largely bound to follow the policies and guidelines set by their supervisors and institutions. Moreover, they also must adhere to the policies, guidelines, and trends set by a commission or department at local, regional, and national levels, which will affect the presentation and evaluation of the material for the students, for both L1 and L2 contexts. Societal trends influence the formation of policies that determine the way languages are taught. As one example for the teaching of L1 in public schools, each state in the United States used to have its own standards for proficiency that set goals for the curricula. But by 2012, a statement of Common Core State Standards4 for English language arts/literacy and mathematics was established for the entire country to follow, built on existing state standards with input from teachers, education experts and their research, and feedback from parents and the public in general. The standards set expectations for two general categories of language education: (a) kindergarten to the end of high school, and (b) college and career readiness. Thus, the belief in a collective effort instead of individual standards established a set of targeted common goals and expectations for language teaching and assessment for the country. On the other hand, the goals themselves have been questioned lately regarding whose language standards should be used, especially in the context of the current heightened tensions in the United States regarding race. While the examples we discuss are from the US context, they are comparable with other 4

See www.corestandards.org/about-the-standards/development-process/.

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

333

334

DALE KOIKE AND ELISA GIRONZETTI

countries where racialized practices and discourses are still pervasive in education (Rosa and Flores 2017). To be sure, racial issues in education have been raised before, for example, in the efforts to legitimize Ebonics or the belief that American Black English should be regarded as a language of its own instead of a dialect of English (e.g., Baugh 2000). Even today, the ways in which a language is taught within the school system present challenges, as language policies set by school districts and institutions have been shown to promote the learning and use of certain prestigious varieties of the dominant language over others, thus perpetuating social inequalities and the racialization of minority groups (Rosa 2019). Language policies are a key contextual factor that affects how L1s are learned in school and include activities that influence the structure, function, use, or acquisition of language, are intentional and unintentional, overt and covert, de jure and de facto; and are engaged in by agents across multiple levels of language policy creation, interpretation, and appropriation, from the macro-levels of national planning and policy to the micro-level of language use. (Johnson 2011: 268)

These activities encompass not only actions, but also the underlying beliefs of a community that lead to those actions (Spolsky 2004). In the context of the United States, like other countries with a multilingual and multicultural population, the implementation of policies that conceptualize L1 learning as the mastery of an ill-defined “academic” language modeled on monolingual English (McSwan 2018) affects minority and bilingual learners who are also learning English in school. As we discuss in Section 15.4, an example of how this view of L1 learning negatively affects bilingual learners who enter the school system can be observed in the case of HL speakers. These learners grew up learning their HL at home while being exposed to the dominant language (in the United States, English) in the broader society. When they start school in the majority language, they are often categorized as “English Language Learners” (ELL) for their inability to engage in monolingual language practices (e.g., the use of “academic monolingual English”) or simply because they speak a language other than English (regardless of their English proficiency) and thus are assigned to take remedial courses. The emphasis on academic language in the United State was underscored with the promulgation of the No Child Left Behind law in 2001 that further strengthened the monolingual paradigm in language policy and contributed to widening the literacy gap (Wiley and Rolstad 2014), rather than narrowing it. These language policies for education affect not only what is taught, but also what is assessed and how. As an example, ELLs face numerous difficulties when trying to place out of remedial courses (Lillie et al. 2012; Slama 2012) because the language tests to assess their progress are also based on a monolingual academic language. Language policies, including goals and standardized curricula and assessment, and the informal context of use of a given language, influence which L1 is learned and, in the case of formal learning contexts, how it is taught and

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

15 Language Learning and Assessment in Context

assessed, creating issues yet to be resolved. In the next section, we discuss the multiple ways in which context affects L2 learning, teaching, and assessment, highlighting the societal factors that have shaped the field.

15.3 The Contexts of L2 Learning, Teaching, and Assessment A child can learn two or more languages simultaneously or sequentially, one after the other. Many societies promote bilingualism, and sometimes multilingualism, to enhance intercultural communication and preserve other languages, such as indigenous languages, to some extent. Here we examine the historical and societal events that, throughout the years, informed L2 teaching and learning and L2 assessment in the context of formal education; that is, in institutional settings, such as the classroom, with defined, predetermined goals.

15.3.1 Historical and Societal Contexts of L2 Teaching and Learning The history of L2 teaching and learning approaches and methods is often linked to societal events and scholarly works of broad impact: The popularity of different teaching methods reflects contextual trends of the times while also responding to the limitations of preceding ones. It should be noted, however, that although each new teaching method or approach reacted against some of the principles of prior methods, they still incorporated some elements of previous ones. Here, we review some of the major trends in L2 teaching that characterize the last 100 years as we establish connections among them and with societal and scholarly trends. We also consider the ways in which the view of L2 and L2 learning has changed, specifically regarding the way in which context has figured into the conceptualization of what a language is, as reflected in teaching and assessment practices. Since medieval times in the Western world, the formal teaching of Latin and ancient Greek motivated the application of the Grammar Translation Method (GTM) to the teaching of other L2s. The GTM was employed to teach these classical languages and their grammars so that clerics could learn to read religious and classical texts; thus, when other languages began to be taught in school, the same method was applied. Classes were conducted in the students’ L1 and learning activities included the memorization of declensions and conjugations, grammatical rules that were presented deductively (i.e., explained by the teacher, memorized, and then applied to complete other activities), vocabulary lists, and translations of sentences from one language to the other. No attention was devoted to speaking and listening skills, as the main goal (that is, what learning an L2 meant at the time) was to be able to read literature in the target language and translate a text from the source language into the target language, or vice versa.

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

335

336

DALE KOIKE AND ELISA GIRONZETTI

The rise of natural methods in L2 teaching, including the Direct Method5 (DM) during the early 1900s, was linked to studies of L1 development in children, which in turn led scholars to claim the superiority of the spoken language over its written form and to argue for an orally delivered teaching method for L2s. This method prioritized speaking but also promoted a focus on everyday vocabulary and an exclusive use of the target language in class. Typical class activities included highly scripted dialogues that moved beyond the focus on single sentences, seen in the GTM. This method was very popular in Europe until the 1920s, but not as much in the United States, where a focus on reading was preferred (Coleman 1929). The increased mobility and ease of communication among people in the mid-nineteenth century led to more travel to other countries and an interest in speaking other languages, thus paving the way to further developments in L2 teaching. World War II also had a large impact that favored the Audio-Lingual Method (ALM), in answer to a need for military personnel to acquire a nativelike pronunciation in the target L2. The ALM was strongly influenced by structuralism, a movement that conceived of language learning as the learning of the structural properties of a given language, and by the behaviorist theory of learning that emphasized the formation of habits through repetition and practice, thus distancing itself from the previous focus on grammar and translation in the GTM. The rise of the Cognitive Code (CC) approach in the 1960s was a response to Chomsky’s Transformational Generative Grammar model and a growing attention to the brain and psycholinguistic processes in language learning, while also reacting against behaviorist trends (e.g., habit formation and repetition in the ALM) that were popular at the time. Chomsky (1959) emphasized that language derives from an innate capacity of the human mind and is rulegoverned, rather than based on the formation of habits. Under the CC approach, language learners were encouraged to work out grammar rules for themselves, following an inductive learning process (as opposed to a deductive approach followed under the GTM). While during most of the past century L2 pedagogy was dominated by cognitive views of language, understood as a structured, mental system, the last four decades beginning in the 1980s saw increased attention to social views of language as a semiotic and social communication system. This change of orientation first became evident with the rise of the Communicative Approach (CA) to language teaching, possibly the most influential one of the last century. This approach was based on Krashen’s and Terrell’s work on the Natural Approach, which prioritized vocabulary learning in a naturalistic language acquisition process (Krashen and Terrell 1983), derived from Krashen’s 5

The Direct Method, also known as the Natural Method and the Natural Approach, are often confused due to the similarity of their names. Krashen and Terrell (1983) explain that while these share some similarities, the Natural Approach is more in line with other communicative approaches that were popular at that time and focuses on exposure to input and a “silent period” (delayed production of output in the L2). The Natural or Direct Method, originating in the early 1900s in England, denotes the exclusive use of the target language.

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

15 Language Learning and Assessment in Context

Monitor Model. This approach emerged as a criticism of the ALM in the United States, and in response to Chomsky’s “competence” and “performance” notions. That is, while Chomsky viewed knowing a language as competence, thus abstracted from its context of use, the CA focused on language learning that emphasized linguistic performance, or the actual use of language in a given communicative context. For the first time, language teaching put principles and concepts of sociolinguistics and pragmatics at the forefront, while syntax and phonology (the teaching of grammar and pronunciation) became relegated to a secondary role. In contrast to previous methods and approaches, the CA is an umbrella term that encompasses different subapproaches and methods that have evolved through the years. Common to all is the idea that the goal of language teaching should be language use, so classroom activities try to expose learners to authentic materials and input as a means to prepare them for using the L2 in real life. The notional-functional and pragmatic approach that took hold in the 1970s and 1980s (Trim et al. 1973) focused on language functions and the realization of speech acts to accomplish concrete actions (e.g., in order to take a trip, one must be able to ask for information, express their needs, complain, etc.). This early CA approach was influenced by research in the linguistic field of pragmatics and the works of Austin (1962) and Searle (1969) on speech acts. It was promoted in Europe as scholars and language experts recognized the need for a common frame of reference for L2 teaching that could promote cooperation and mobility among Europeans. This project is the basis for the development of the standards found in the Threshold Level (Van Ek and Alexander 1980) and its later development into the Common European Framework of Reference (CEFR; Council of Europe 2001, 2020). The 1980s gave way to the Intercultural Approach (IA, Byram 1997, 2021), influenced by research in sociolinguistics as well as social anthropology and ethnography, and promoted within the context of the European Union project to foster communication among people from different countries. The IA is based on a view of language as a way to communicate with people who live in a different country, speak a different language, and belong to a different culture. This approach saw an increased interest toward the communicative needs of a group within a given sociocultural context, which translated in the L2 classroom as a focus on the cognitive, affective, and interactional dimensions of communication (see Deardorff 2006, 2009). The 1990s were dominated by Task-Based Instruction (TBI) or Task-Based Language Teaching (TBLT) (Liddicoat and Scarino 2013; Long 2016), still popular today (see Norris 2021), which emphasized the use of language to accomplish tasks in real life, and the integration of linguistic, cultural, and strategic content in the L2 classroom through a sequence of pedagogical tasks. TBLT originated in the work of Prabhu (1987) with the Bangalore Project, which challenged the idea that L2 learning should be based on direct instruction as in the ALM and aimed at meeting students’ expectations of an authentic, real-life learning experience. The international success of this approach

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

337

338

DALE KOIKE AND ELISA GIRONZETTI

was also supported by research in discourse analysis, which considered communication as the interpretation of meaning based on text but also the interlocutor’s prior knowledge of the world and the context in which communication takes place. The turn of the century saw two more variations of the CA that emerged in part from an increased focus on the psychology of language learning: the Competence-Based Language Teaching (CBLT) approach and the Strategic Approach (SA). CBLT was influenced by behaviorism and was initially employed to teach survival language skills to immigrants. Following this approach, L2 learners were taught the necessary skills to be able to use the language in real-life situations (see Pérez-Cañado 2013, for CBLT in higher education). SA was based on the work of Oxford and colleagues (Oxford 1991; Chamot et al. 1999) and the concept of strategic competence as a component of communicative competence (Canale and Swain 1980). It was influenced by studies on individual variation in L2 acquisition. This approach emphasized teaching not only how to use the L2 in context, but also the resources that learners need to learn the language and communicate in a variety of situations. The latest iteration of the CA is the Sociocultural Approach (SCA), informed by the work of Vygotsky (1978), who believed that all learning (not just L2 learning) depends on interaction with other people. Frawley and Lantolf (1985) were among the first to apply Vygotsky’s work to L2 teaching, promoting a shift toward a more social and interactional view of L2 learning (but also note Hatch’s [1978] earlier work proposing a discourse approach to L2 learning). One example of how the SCA translated into the classroom is the Presentation, Attention, Co-construction, and Extension (PACE) model (Donato and Adair-Hauck 2016), which stressed the role of conceptual thinking, interaction, and dynamic assessment for L2 teaching. Finally, in the last few years there has been a rise in popularity and scholarly interest in a new proposal (not limited to, nor initially developed for L2 teaching) that stems from the work of the New London Group (NLG 1996). The NLG’s pedagogy for multiliteracies considers teaching and educating as political acts (i.e., in the case of L2 learning, the variety of a language that instructors or institutions decide to teach, the texts and authors they use, the cultural perspectives they offer, etc.), and is grounded in critical pedagogy to foster social change through principles of representation, access, and equity (Freire 1970). In this view, texts are the key elements through which teaching and learning take place in a process of textual thinking that considers genres, agents, audience, as well as the sociohistorical context in which those texts were produced, the perspectives they offer, and the ideologies they reflect, among other elements. This model was brought about, in part, by the ubiquity of technology in the modern world and the increasing technological gap between generations and groups of people from different socioeconomic backgrounds. This focus on technology-mediated communication and language use also caused a more conscious effort to work with multimodal texts in L2 teaching, since the focus is gradually shifting from a linguocentric

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

15 Language Learning and Assessment in Context

approach (Erickson 2004) that prioritized verbal language, like nearly all previous L2 teaching approaches and methods, to multimodal communication; thus, including gestures, images, and other semiotic resources that people can mobilize to communicate. As shown through this brief historical overview, the role of context in relation to L2 teaching developments can be seen as twofold, attending to external or internal context with respect to the language classroom and approaches to learning. On the one hand, each language teaching approach or method was influenced by larger societal events, research trends, and prior approaches and methods, in a dynamic and nonlinear progression.6 On the other hand, the ways in which language was conceptualized within each approach also evolved in a more linear fashion, beginning with more decontextualized views that prioritized work with single sentences or even words, and did not consider where, by whom, or why those sentences would be used. Not until multimodal and socially situated approaches to L2 learning arose, in which language and context are treated as inseparable and context is understood broadly, did we see approaches that encompassed political, social, and historical uses of the language. The field of L2 teaching and learning gradually moved from a linguocentric approach aimed at studying forms and did not consider the context beyond the sentence level (e.g., in the GTM), to approaches that place the sociocultural context of language use at the forefront (e.g., in the SCA) and consider the multiple modalities and multiliteracies people employ to communicate (see Young 2009, for a discussion on contexts of learning). This shift affected assessment as well, as the field gradually moved from emphasizing accuracy (as in the GTM) to acknowledging and assessing students’ engagement with the community (community-based service courses), and the ability to reflect critically and draw connections and comparisons across cultures (sociocultural approach, critical pedagogy).

15.3.2 L2 Assessment and Context As described in the previous section, L2 assessment methods and instruments developed concurrently with the emergence of various L2 teaching approaches and the role of context in the conceptualization of what constitutes language in the L2 classroom. Regarding assessment in the GTM, for example, greater emphasis was placed on accuracy while other important contextual aspects, such as who the speaker/writer and the intended audience are, when the sentence was produced and for what purpose, etc., were not considered. Methods and approaches that emphasized communication, such as the CA discussed earlier, introduced the use of question-answer-elaboration oral interactions in the target language and, with them, a broader vision of language 6

Some scholars have argued in favor of a “postmethod” era or condition in language teaching (Kumaravadivelu 2006a) in which teachers should not try to find the perfect method but, rather, make informed choices and choose the most appropriate materials, goals, and classroom techniques and strategies free from the constraints of a given method and based on the consideration of contextual factors.

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

339

340

DALE KOIKE AND ELISA GIRONZETTI

that emphasized its use in real-life situations. Thus, assessment could incorporate elements of the real world, such as objects and artifacts that learners could refer to and manipulate, and opportunities were created for learners to express their own thoughts through the target language. With the Natural Method, the emphasis shifted from grammar and writing to speaking, but still with relatively little consideration of the communicative context in which language is used. With the ALM, students’ mastery of the L2 was assessed on the basis of their ability to reproduce sentences following a model provided by the instructor or to respond in the target language based on the input received. As these sentences occurred in artificial dialogues, little consideration was paid to contextual factors such as audience or media, since the emphasis was still placed on form and not meaning. The rise of the Cognitive Code approach shifted the focus of instruction once again toward the mastery of grammar rules and metalinguistic knowledge, leaving no room for learners to focus on the communicative context in which language is used, since Chomsky’s linguistic theory on which it was based was not concerned with linguistic performance but rather linguistic competence. Assessment once again focused on learners’ ability to transform sentences or short texts – for example, from one tense to another – in order to demonstrate their mastery of a given grammatical element. While sentences were often presented within short texts and thus their use was indeed contextualized in a given communicative situation, the communicative context was typically not relevant for the tasks learners were required to perform. With the CA model, assessments changed drastically as the communicative context in which language was used became key and, at times, even more important than accuracy. In fact, the CA was later criticized for not fostering the development of real-life communicative skills (Kumaravadivelu 2006b), not accounting for and so unable to fit different cultural contexts (e.g., Kumaravadivelu 2006b; Littlewood 2007), as well as for teaching skills and concepts that are actually unnecessary because they could be transferred from the L1 (e.g., Swan 1985a, 1985b). Students were no longer required simply to produce grammatically accurate sentences in the target language, but rather intelligible utterances that were appropriate in a given situation (i.e., considering the audience and their communicative purpose). Assessments also featured a prominent oral communication component in the form of role-plays and interviews, and L2 teachers employed more holistic rubrics to assess a wider set of features not limited to (and sometimes not including) grammatical accuracy. With the CA, for example, L2 learners could be required to produce extensive portfolios documenting their communicative abilities in the target language across skills, texts, audiences, and media. The last major change surfaced only recently with a pedagogy for multiliteracies, as teachers and learners were asked to pay attention not only to the immediate communicative context in which the language is used (which, within the CA, was sometimes socially de-contextualized and limited to a set

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

15 Language Learning and Assessment in Context

341

of practical communicative situations), but also to sociocultural and political factors that affect language use, such as the coexistence of Spanish and English in the United States and the ways in which language is used in the public space to define the identity of a neighborhood or community. Working with texts as the main units of meaning (as opposed to sentences in the GTM or questionanswer dialogues in the DM) also reflects the way learners are assessed. A sample instrument is seen in Integrated Performance Assessments (IPAs), which are holistic instruments that integrate principles of dynamic assessment (Adair-Hauck et al. 2006) and emphasize the teaching and formative value of assessment. IPAs are employed not only to determine whether students have met the requirements of a course or unit, but also to help them move forward in their learning in a dialogic and reflective process together with the instructor and other students. IPAs include cultural comparisons and reflections on the inferential meaning of utterances, among other elements. Another example is the development of digital storytelling projects, in which learners can rely on a variety of modes and media to create meaning in the target language. As a review of the main points discussed in Section 15.3, Table 15.1 summarizes the main societal events or trends (column 1) that led to the development of a given method or approach for L2 teaching (column 2), together with sample assessment instruments and foci (column 3), and a brief summary of the role of context in language teaching, learning and assessment (column 4), such as the use of decontextualized written sentences, scripted dialogues, etc. Table 15.1, thus, can be seen horizontally, from the societal context and events to the development of a certain method or approach and their characteristics, or vertically, to show how each category (societal context, SLA methods and approaches, etc.) changed over time.

Table 15.1 Historical overview of the relationship between context and L2 learning

Societal context Formal teaching of Latin and Greek to access religious and classical texts Studies of L1 development in children Mobility, tourism, World War II, behaviorism, structuralism Chomsky’s Transformational Generative Grammar Sociolinguistics and pragmatics

Critical pedagogy, Multimodality

SLA methods and approaches

Assessment

Language in context

Grammar-Translation Method

Translation and conjugation exercises

Decontextualized written sentences

Natural methods, Direct Method Audiolingual Method

Comprehension of simple utterances Question-answer

Repetitions, drills, form and accuracy Highly scripted dialogues Cognitive Code Approach Tests, fill in the blanks, memorization Sentences and short of linguistic rules texts (competence) Communicative Approach Oral interviews, role-plays, Socially de(including IA, TBLT, portfolios, tasks contextualized CBLT, SA, SCA) communicative situations Multiliteracies Integrated Performance Sociocultural and Assessments, digital storytelling political contexts of projects, dynamic assessment language use

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

342

DALE KOIKE AND ELISA GIRONZETTI

Up to this point in this chapter, we have reviewed L1 and L2 learning, teaching, and assessment in contexts of informal and formal instruction from a historical perspective, showing how approaches and practices have changed over the years along with societal and historical developments. We now present in Section 15.4 a unique example of how context has shaped language learning, teaching, and assessment in a very different way, which is the case of heritage language learners.

15.4 The Contexts of Heritage Language Maintenance and Assessment To illustrate another example of how context shapes language teaching, learning, and assessment, we introduce the case of HL learners, which was briefly mentioned in Section 15.2 as an example of how linguistic policies shape the teaching and learning of English in the school system. In broad terms, an HL is a minority language that is learned and spoken mostly in the familial context and coexists with a majority language used for schooling and in other public contexts. It is important to consider that the learning of an HL generally occurs before puberty, when the child has not yet reached what many consider to be the “critical period” of language learning, discussed earlier. In the case of HL speakers of Spanish in the United States, for example, Spanish is the minority language learned at home and usually it is not employed in as many contexts as the majority language, English, which HL speakers learn at school and use when speaking and writing in public spaces in the larger society. Owing to this division of functions of the two languages, and what the HL speakers themselves may perceive as linguistic inadequacies, they may suffer from anxieties and feelings of insecurity when using their HL with Spanish native speakers and even L2 learners of the same language, because their language practices are criticized by native speakers or because they are intimidated by the metalinguistic knowledge of L2 learners (Lynch 2018). The different definitions of an HL and HL speakers that are used today (e.g., Silva-Corvalán 1996; Fishman 2001; Peyton et al. 2001; Valdés 2001; Extra and Yagmur 2002; Van Deusen-Scholl 2003; Duff 2008) attempt to capture the complex dynamics and the social and political contexts that characterize HLs and their speakers. While some of these definitions focus on the linguistic abilities of HL speakers and their bilingual status (the so-called “narrow” definitions), others emphasize the personal connection of the individual with the heritage culture or ethnic group and thus include those HL speakers with limited or no practical knowledge of the language (receptive learners or speakers, emergent bilinguals). In the United States, which is the context with which we the authors are most familiar, Spanish is one of the most important HLs

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

15 Language Learning and Assessment in Context

due to the large numbers of Latinos/as7 in the country and the growing populations of HL learners in the public school system. Spanish heritage learners (SHL) form an important, expanding constituency and the issue of context is a critical factor in their language maintenance. The contexts that form the backdrop of this learner population are discussed in the next section.

15.4.1 Historical, Political, and Societal Contexts of Spanish as a Heritage Language in the United States The Latino/a population in the United States is diverse, and the historical, political, and societal contexts of its people are equally as diverse. Some of the Latino/a communities spread across the country are rooted as indigenous societies (e.g., in the Southwest, discussed below), while most arrived and continue to come as immigrants from various Latin American countries. Most of the immigrant population in the United States over the last thirty years originate in Spanish-speaking countries (e.g., Mexico, Venezuela, Cuba, El Salvador, and the Dominican Republic), but this can vary greatly depending on the particular US region and the specific year. For example, in the Greater Washington area (District of Columbia, Maryland, and Virginia), a majority of Caribbean and Central Americans families can be found who emigrated from Central America in the 1980s and from the Caribbean in the 1950–1960s. On the other hand, in the southern US border area there has been marked immigration from Central American countries in 2021, mainly due to the factors of political, criminal, and economic strife, conflict, and violence that make life unbearable in their home countries. To understand the relevance and the goals of HL teaching, we now look at the background of one specific example: the Southwestern US region. Much of this area was first claimed by Mexico as part of its territory, independent from Spain, in 1810. However, in 1848, Mexico ceded much of the land via the Treaty of Guadalupe Hidalgo after the Mexican-American war to the United States. What are now Texas and California were annexed around that time as well, and the people who lived in this vast part of the country were largely Spanish-speaking Latinos/as (along with some indigenous peoples as well) (Silva-Corvalán 2001). White settlers soon came to occupy these lands, and colonization began and continued at a rapid pace, often resulting in discrimination and injustices against the Latinos/as. This treatment continued in many areas into modern times until, during the 1960s and 1970s, existing issues of racial discrimination against people of color (i.e., all non-White people) finally surfaced in an expression of anger and rebellion against this treatment and the decades of inequalities of all forms; that is, economic, social, and educational 7

In this chapter, we use the term Latino/a to refer to people of Latin American origins living in the United States, rather than Hispanic, because the latter emphasizes the connection with Spain instead of Latin America, and the Spanish language, which many Latinos/as do not consider a defining factor of their identity.

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

343

344

DALE KOIKE AND ELISA GIRONZETTI

disparities and blatant segregation. Many people of color organized themselves into groups and movements. Among the Latino/a communities, we find, for example, La Raza Unida party, the Chicano Power movement, and the Mexican American Legal Defense and Education Fund (MALDEF), all of which demanded sweeping societal changes (e.g., Martínez and Train 2020). It is against this backdrop of social and political upheaval that the push to rectify discrimination and segregation against Latinos/as in the Southwest, such as in the public school system, was born. García and Castro (2011; as reported in Martínez and Train 2020: 121–123) describe past efforts in this region to change the general high school curricula so that Latino/a students were no longer directed into mainly vocational tracks, to become blue collar workers, but instead were also prepared for higher education possibilities. With the hiring of more Latino/a teachers as well, the curriculum gradually changed to reflect more closely their needs and interests, consider their attitudes and experiences, and become aware of their lived experiences (Martínez and Train 2020). These contexts can shed light on why many SHL in the Southwest may suffer from anxiety and insecurity, especially when confronting situations in which native Spanish speakers from outside their community may use a more prestigious, monolingual variety of Spanish, even though they may not explicitly criticize their language. And these demands for change by the Latinos/as may clarify a crucial difference between the SHL in the American Southwest as opposed to those in other parts of the country, or in other places such as Europe, where the historical context of heritage speakers has been different, usually driven by immigration alone. Efforts made to establish educational programs addressing issues facing the SHL population are described in the next section.

15.4.2 Spanish as a Heritage Language in Instructional Contexts in the United States Here we describe the instructional contexts that have been created to foster the maintenance and expansion of SHL in the United States. The background factors and developments described in Section 15.4.1 form the backdrop for schools to implement language programs that could be successful in teaching SHL learners, who have different characteristics from L2 learners. Their familial ties to the language, cultures, and communities, their historical background, the decades of prejudice, segregation, and insistence on assimilation into the larger American society and learning the English language (see Section 15.2 on the monolingual paradigm), not to mention the criticism of the Spanish variety they speak, have all had a great effect on whether or not SHL speakers maintain their HL, especially as there is greater assimilation into the general English-speaking society as time goes on (e.g., Zyzik 2016). The Pew Research Center reported that in 2015, 13.5 percent of US Latinos/ as were born in the United States, a number with important ramifications for the maintenance of Spanish (Potowski and Carreira 2010). That is, often as the

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

15 Language Learning and Assessment in Context

generations move from the first to the second and so on over time, fewer Latinos/as speak Spanish at home and encourage their children to speak it, so by the third generation, the number of Latinos/as who speak Spanish is much smaller, pointing to Spanish language loss. Between 2006 and 2015, the use of Spanish was reported to have declined (Krogstad and López 2017), partly due to the societal challenges that US Latinos/as regularly face, and the fact that most of the US Latino/a population does not believe that speaking Spanish is a key part of their identity; that is, it is not necessary to speak Spanish to consider oneself Latino/a (Pew Research Center 2015). At the same time, the census numbers also indicate that the population of Latino/a students in the school system is growing, and separate linguistic needs related to Spanish must be addressed, such as placement tests, communicating with parents to foster maintenance of the home language, school counseling, types of language courses offered, teacher preparation, and measures to ensure HL learners are not placed in L2 courses by default. Nevertheless, improvements have been made regarding Spanish language education for SHL, which are described next. Considering the contexts that shape the life of SHL speakers and their educational experiences in the United States, several recommendations for teaching SHL have been proposed, initially as a set of five core goals for HL teaching (Valdés 1995) and later expanded to reflect advances in the field (Martínez 2016; Valdés 2000). These recommendations, and the contextual motivations behind each one, are summarized as follows: (a) Because the home language tends to be used orally and for the domain of the home and intimate relationships, if at all, language use for other domains (e.g., academics, work) and language skills (writing, reading) develop in English. This usage indicates a need for work in contexts of more formal genres and in the reading and writing modalities in Spanish (Zyzik 2016; Shively 2018). (b) Owing to the linguistic insecurities and issues of identity and ideologies that many SHL face when learning their heritage language, these should be addressed in the curriculum. At the same time, the wide range of ideologies, communities, and cultural connections with Hispanic roots that are present in the SHL population need to be recognized and fostered (e.g., Fairclough 2016; Showstack 2017). (c) Because SHL learners’ proficiency in Spanish as a group varies greatly, their differences (strengths and weaknesses) need to be acknowledged and considered when designing a language course curriculum. Some SHL learners can speak and write in Spanish, while others display various degrees of bilingualism, with some able to understand only a few words and phrases of given speech registers as receptive bilinguals (e.g., Belpoliti and Gironzetti 2018). In terms of approaches to teaching and assessment, the field of HL teaching has undergone a movement parallel to that described for L2 teaching

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

345

346

DALE KOIKE AND ELISA GIRONZETTI

(Gironzetti and Belpoliti 2018). In its beginnings, it was characterized by a strong focus on grammatical accuracy as well as reading and writing, due to the influence of established practices in L2 pedagogy. The goal was to focus on linguistic information with which HL learners are usually unfamiliar, such as conventions of Spanish orthography, given their experiences with the oral and aural modalities. More recently, scholarly work and pedagogical efforts began to consider and give a more prominent role to affective, identity, cultural, and community-related factors and their relevance for language maintenance and expansion. HLs are valued and recognized in the classroom not only as communicative tools, but also for their ties to the family, community, and culture. Considering this recent shift, researchers argue that an effective HL program requires several important aspects that impact teaching practices as well as assessment goals and instruments. For example, Belpoliti and Gironzetti (2018) propose fostering a connection with the Spanish heritage community, addressing issues of identity, understanding language ideologies (e.g., Leeman 2012), expanding knowledge of Spanish registers and dialects (e.g., Fairclough 2016), and promoting literacy and knowledge of various discourse genres (e.g., argumentative texts). However, HL programs and instructors often struggle to find a balance between a narrower, linguocentric view of SHL, and a broader, multimodal, and humanistic one. A good example that strives to integrate both views in a meaningful way is the Sabine Ulibarrí SHL program at the University of New Mexico, which is one of the largest and most enduring in the country. The introductory courses in this program prioritize learners’ identity and their affective connections with Spanish, focusing on building their confidence and recognizing the value of the language and cultures associated with it, rather than centering exclusively on language accuracy or linguistic skills. Subsequent courses introduce more linguistic- and accuracyoriented goals toward the expansion of learners’ repertoire as well as continue to foster the development of their critical skills and an appreciation of their identity.

15.5 Future Trends and Implications In this chapter, we have shown relationships that exist between contextual factors and events, and language learning and assessment in L1, L2, and HL learning, teaching, and assessment. We have shown that, at least in the United States, L1 (language arts) teaching and assessment policies are largely determined at local, regional, and national levels, while in the case of L2 and HL language learning, teaching, and assessment, there has been a plethora of language theories, policies, approaches, and methods that aim at teaching the language according to each institution’s program and goals. Numerous changes have occurred over time regarding which and how languages are taught and assessed.

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

15 Language Learning and Assessment in Context

The profile of L1 learners is changing due to globalization, increased mobility, and the ubiquity of technology. Instructors of L1 at the elementary and secondary level are now teaching a more diverse student body that comprises not only monolingual L1 speakers, but also multilingual speakers who may not share the same L1 but are, nonetheless, learning the dominant language in school. The case of HL speakers in the United States is a clear example, but this situation is also a pressing reality in other countries (e.g., on Spanish as a minority or heritage language outside of the United States, see chapters 30 to 36 in Potowski 2018). We expect this trend will not stop and will eventually shape not only the teaching practices at the classroom level, but also, in a bottom-up fashion, institutional guidelines and language teaching policies at the state level. If L1 teaching is to serve the needs of our students, then L1 policies need not be modeled after a monolingual and ill-defined academic register but rather consider the richness of the multilingual practices students bring to the classroom. As societal trends and events and scholarly research continue to have an impact on the field, we also witness changes in how language is conceptualized. In the last decades, we have seen a gradual move toward a more humanistic, interdisciplinary, and less cognitive-focused approach to language teaching and assessment that places more emphasis on social, political, and affective factors of language use. At the same time, there is a growing interest in the different modalities and media that humans rely upon for communication, in part due to globalization and the impact of technology. From this perspective, then, language is seen as a complex dynamic system of conventions that transcend verbal language to include other semiotic resources that people mobilize to create meaning, such as facial expressions or gestures. This broader conceptualization of language is having an impact on language teaching and assessment, and current proposals such as multimodal and multiliteracies approaches stress that language learning should be fostered by all these elements. Just as history has shaped our language teaching and assessment practices over time, we are witnessing at this moment how the use of technology has affected our teaching and assessment practices, especially with regard to languages. Owing to the COVID-19 pandemic, the need to use technological innovations to connect with learners at a distance, and to adapt our practices, materials, and syllabi to this mode of delivery, have greatly affected the field. While the pedagogical approaches to teach L2s may not necessarily have changed notably due to the use of online tools, many instructors and students faced new challenges with regard to equity and access to education, assessment, and learners’ agency. Online teaching has the potential of reaching a wider student population, including those who live in rural areas or far away from schools. At the same time, it has also brought attention to the existing technology gap, as many students still do not have reliable internet capabilities at home or access to a computer to attend online classes. Similarly, assessment instruments have had to transform to allow for online

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

347

348

DALE KOIKE AND ELISA GIRONZETTI

communication (e.g., storytelling projects, blogs), often assuming students and teachers already possess technology literacy instead of recognizing the need for a multiliteracies-oriented pedagogy. Finally, students’ motivation, affect, and self-regulation assume even greater importance in the success of an online language program (Mercer 2011), as individual learners’ agency is key in a context in which instructors and students do not share a physical space or students may have to work independently more often. The fields of language teaching and assessment await new contexts in the future.

References Adair‐Hauck, B., Glisan, E. W., Koda, K., Swender, E. B., and Sandrock, P. (2006). The Integrated Performance Assessment (IPA): Connecting assessment to instruction and learning. Foreign Language Annals, 39(3), 359–382. Austin, J. (1962). How to Do Things with Words. Cambridge, MA: Harvard University Press. Baugh, J. (2000). Beyond Ebonics: Linguistic Pride and Racial Prejudice. Oxford: Oxford University Press. Belpoliti, F., and Gironzetti, E. (2018). Hablantes de herencia. In J. Muñoz-Basols, E. Gironzetti, and M. Lacorte (eds.), The Routledge Handbook of Spanish Language Teaching: Metodologías, Contextos y Recursos para la Enseñanza del Español L2 (pp. 447–462). London: Routledge. Bialystok, E. (1997). The structure of age: In search of barriers to second language acquisition. Second Language Research, 13(2), 116–137. Byram, M. (1997). Teaching and Assessing Intercultural Communicative Competence. Bristol: Multilingual Matters. Byram, M. (2021). Teaching and Assessing Intercultural Communicative Competence. Revisited. Bristol: Multilingual Matters. Canale, M., and Swain, M. (1980). Theoretical bases of communicative approaches to second language teaching and testing. Applied Linguistics, 1, 1–47. Chamot, A., Barnhardt, S., El-Dinary, P., and Robbins, J. (1999). The Learning Strategies Handbook. Boston, MA: Addison Wesley Longman. Chomsky, N. (1959). A review of B. F. Skinner’s Verbal Behaviour. Language, 35(1), 26–58. Coleman, A. (1929). The Teaching of Modern Foreign Languages in the United States: A Report Prepared for the Modern Foreign Language Study (Vol. 12). New York: Macmillan. Council of Europe. (2001). Common European Framework of Reference for Languages: Learning, Teaching, Assessment. Strasbourg: Council of Europe. Council of Europe. (2020). Common European Framework of Reference for Languages: Learning, Teaching, Assessment. Companion volume. Strasbourg: Council of Europe. Curtiss, S., Fromkin, V., Krashen, S., Rigler, D., and Rigler, M. (1974). The linguistic development of Genie. Language, 50(3), 528–554. Deardorff, D. (2006). Identification and assessment of intercultural competence as a student outcome of internationalization. Journal of Studies in International Education, 10(3), 241–266.

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

15 Language Learning and Assessment in Context

Deardorff, D. (2009). The SAGE Handbook of Intercultural Competence. Thousand Oaks, CA: SAGE. DeKeyser, R., and Larson-Hall, J. (2005). What does the critical period really mean? In J. F. Kroll and A. M. B. de Groot (eds.), Handbook of Bilingualism: Psycholinguistic Approaches (pp. 109–127). Oxford: Oxford University Press. Donato, R., and Adair-Hauck, B. (2016). PACE: A story-based approach for dialogic inquiry about form and meaning. In J. Shrum and E. Glisan (eds.), Teacher’s Handbook: Contextualized Foreign Language Instruction, 5th ed. (pp. 206–230). Boston, MA: Cengage Learning. Duff, P. (2008). Heritage language education in Canada. In D. Brinton, O. Kagan, and S. Bauckus (eds.), Heritage Language Education (pp. 71–90). Mahwah, NJ: Lawrence Erlbaum. Ellis, R. (1989). Understanding Second Language Acquisition. Oxford: Oxford University Press. Erickson, F. (2004). Origins: A brief intellectual and technological history of the emergence of multimodal discourse analysis. In P. LeVine and R. Scollon (eds.), Discourse and Technology. Multimodal Discourse Analysis (pp. 196–207). Washington, DC: Georgetown University Press. Extra, G., and Yagmur, K. (2002). Language Diversity in Multicultural Europe: Comparative Perspectives on Immigrant Minority Languages at Home and at School. Paris: UNESCO/The Most Programme. Fairclough, M. (2016). Incorporating additional varieties to the linguistic repertoires of heritage language learners. In S. Beaudrie, M. Fairclough, and A. Roca (eds.), Innovative Strategies for Heritage Language Teaching (pp. 143–165). Washington, DC: Georgetown University Press. Fishman, J. (2001). 300-plus years of heritage language education in the United States. In J. K. Peyton, D. A. Ranard, and S. McGinnis (eds.), Heritage Languages in America: Preserving a National Resource (pp. 81–98). Washington, DC/McHenry, IL: Center for Applied Linguistics & Delta Systems. Frawley, J., and Lantolf, J. P. (1985). Second language discourse: A Vygotskyan perspective. Applied Linguistics, 6, 19–44. Freire, P. (1970). Pedagogy of the Oppressed. New York: Seabury Press. García, M., and Castro, S. (2011). Blowout! Sal Castro and the Chicano Struggle for Educational Justice. Chapel Hill, NC: University of North Carolina Press. Gironzetti, E., and Belpoliti, F. (2018). Investigación y pedagogía en la enseñanza del español como lengua de herencia (ELH): Una metasíntesis cualitativa. Journal of Spanish Language Teaching, 5, 16–34. Hatch, E. (1978). Discourse analysis and second language acquisition. In E. Hatch (ed.), Second Language Acquisition (pp. 401–435). Rowley, MA: Newbury House. Johnson, D. C. (2011). Critical discourse analysis and the ethnography of language policy. Critical Discourse Studies, 8(4), 267–279. Krashen, S. (1981). Second Language Acquisition and Second Language Learning. Oxford: Pergamon Press. Krashen, S., and Terrell, T. (1983). The Natural Approach: Language Acquisition in the Classroom. New York: Macmillan. Krogstad, J. M., and López, M. (2017). Use of Spanish Declines among Latinos in Major U.S. Metros. www.pewresearch.org/fact-tank/2017/10/31/use-of-spanish-declinesamong-latinos-in-major-u-s-metros/.

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

349

350

DALE KOIKE AND ELISA GIRONZETTI

Kumaravadivelu, B. (2006a). Understanding Language Teaching: From Method to Postmethod. Mahwah, NJ: Lawrence Erlbaum. Kumaravadivelu, B. (2006b). TESOL methods: Changing tracks, challenging trends. TESOL Quarterly, 40(1), 59–81. Leeman, J. (2012). Investigating language ideologies in Spanish as a heritage language. In S. Beaudrie and M. Fairclough (eds.), Spanish as a Heritage Language in the United States: The State of the Field (pp. 43–59). Washington, DC: Georgetown University Press. Lenneberg, E. (1967). Biological Foundations of Language. New York: Wiley. Liddicoat, A., and Scarino, A. (2013). Intercultural Language Teaching and Learning. Hoboken, NJ: Blackwell. Lillie, K., Markos, A., Arias, M. B., and Wiley, T. (2012). Separate and not equal: The implementation of structured English immersion in Arizona’s classrooms. Teachers College Record, 114(9), 1–33. Littlewood, W. (2007). Communicative and task-based language teaching in East Asian classrooms. Language Teaching, 40(3), 243–249. Long, M. (2016). In defense of tasks and TBLT: Nonissues and real issues. Annual Review of Applied Linguistics, 36, 5–33. Lynch, A. (2018). A historical view of US Latinidad and Spanish as a heritage language. In K. Potowski (ed.), The Routledge Handbook of Spanish as a Heritage Language (pp. 17–35). London: Routledge. MacSwan, J. (2018). Academic English as standard language ideology: A renewed research agenda for asset-based language education. Language Teaching Research, 24, 1–9. Martínez, G. (2016). Goals and beyond in heritage language education: From competencies to capabilities. In M. Fairclough and S. Beaudrie (eds.), Innovative Strategies for Heritage Language Teaching (pp. 39–55). Washington, DC: Georgetown University Press. Martínez, G., and Train, R. (2020). Tension and Contention in Language Education for Latinxs in the United States: Experience and Ethnics in Teaching and Learning. London: Routledge. Mercer, S. (2011). Understanding learner agency as a complex dynamic system. System, 39(4), 427–436. New London Group. (1996). A pedagogy of multiliteracies: Designing social futures. Harvard Review, 66(1), 60–92. Norris, J. M. (2021). The long road to TBLT: 40 years of innovation, investigation, and inspiration. TASK, 1(2), 162–183. Oxford, R. (1991). Language Learning Strategies: What Every Teacher Should Know. New York: Newbury House. Partanen, E., Kujala, T., Näätänen, R., Liitola, A., Sambeth, A., and Huotilainen, M. (2013). Learning-induced neural plasticity of speech processing before birth. PNAS, 110(37), 15145–15150. Pascual y Cabo, D. (2018). Spanish as a heritage language in the US: Core issues and future directions. In K. Geeslin (ed.), The Cambridge Handbook of Hispanic Linguistics (pp. 478–495). Cambridge: Cambridge University Press. Pérez-Cañado, M. L., ed. (2013). Competency-Based Language Teaching in Higher Education. New York: Springer.

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

15 Language Learning and Assessment in Context

Pew Research Center. (2015). National Survey of Latinos. www.pewhispanic.org/2016/ 06/08/latinos-increasingly-confident-in-personalfinances-see-better-economictimes-ahead/ph_2016-06-08_nsl-economy-12/. Peyton, J. K., Ranard, D. A., and McGinnis, S. (2001). Heritage Languages in America: Preserving a National Resource. Language in Education: Theory and Practice. Washington, DC/McHenry, IL: Center for Applied Linguistics & Delta Systems. Pienemann, M., and Johnston, M. (1987). Factors influencing the development of language proficiency. In D. Nunan (ed.), Applying Second Language Acquisition Research (pp. 45–141). Adelaide: National Curriculum Resource Center. Potowski, K., ed. (2018). The Routledge Handbook of Spanish as a Heritage Language. London: Routledge. Potowski, K., and Carreira, M. (2010). Spanish in the USA. In K. Potowski (ed.), Language Diversity in the United States (pp. 66–80). Cambridge: Cambridge University Press. Prabhu, N. S. (1987). Second Language Pedagogy. Oxford: Oxford University Press. Rosa, J., and Flores, N. (2017). Unsettling race and language: Toward a raciolinguistic perspective. Language in Society, 46(5), 621–647. Rosa, J. (2019). Looking Like a Language, Sounding Like a Race: Raciolinguistic Ideologies and the Learning of Latinidad. Oxford: Oxford University Press. Searle, J. (1969). Speech Acts. Cambridge: Cambridge University Press. Shively, R. (2018). Spanish heritage speakers studying abroad. In K. Potowski (ed.), The Routledge Handbook of Spanish as a Heritage Language (pp. 403–419). London: Routledge. Showstack, R. (2017). Spanish identity among Latinos in the US. In Kim Potowski (ed.), The Routledge Handbook of Spanish as a Heritage Language (pp. 92–106). London: Routledge. Silva-Corvalán, C. (1996). Language Contact and Change. Oxford: Oxford University Press. Silva-Corvalán, C. (2001). Sociolingüística y Pragmática del Español. Washington, DC: Georgetown University Press. Slama, R. B. (2012). A longitudinal analysis of academic English proficiency outcomes for adolescent English language learners in the United States. Journal of Educational Psychology, 104(2), 265. Spolsky, B. (2004). Language Policy. Cambridge: Cambridge University Press. Swan, M. (1985a). A critical look at the communicative approach (1). ELT Journal, 39(1), 2–12. Swan, M. (1985b). A critical look at the communicative approach (2). ELT Journal, 39(2), 76–87. Trim, J. L. M., Richterich, R., van Ek, J. A., and Wilkins, D. A. (1973). Systems Development in Adult Language Learning: A Unit-Credit System for Modern Language Learning by Adults. Strasbourg: Council of Europe. Valdés, G. (1995). The teaching of minority languages as academic subjects: Pedagogical and theoretical challenges. The Modern Language Journal, 79(3), 299–328. Valdés, G. (2000). Introduction. In L. Sandstedt (ed.), Spanish for Native Speakers: AATSP Professional Development Series Handbook for Teachers K-16 (Vol. 1, pp. 1–20). New York: Harcourt College.

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

351

352

DALE KOIKE AND ELISA GIRONZETTI

Valdés, G. (2001). Heritage language students: Profiles and possibilities. In J. Peyton, J. Ranard, and S. McGinnis (eds.), Heritage Languages in America: Preserving a National Resource (pp. 37–80). Washington, DC/McHenry, IL: Center for Applied Linguistics and Delta Systems. Valdés, G. (2005). Bilingualism, heritage language learners, and SLA research: Opportunities lost or seized? The Modern Language Journal, 89, 410–426. van der Slik, F., Schepens, J., Bongaerts, T., and van hout, R. (2021). Critical Period Claim revisited: Reanalysis of Hartshorne, Tenenbaum, and Pinker (2018) suggests steady decline and learner-type differences. Language Learning, 72(1), 9–112. Van Deusen-Scholl, N. (2003). Toward a definition of heritage language: Sociopolitical and pedagogical considerations. Journal of Language, Identity, and Education, 2(3), 211–230. van Ek, J., and Alexander, L. G. (1980). Threshold Level English in a European Unit/ Credit System for Modern Language Learning by Adults. Oxford: Pergamon Press. Vygotsky, L. (1978). Mind in Society: The Development of Higher Psychological Processes. Cambridge, MA: Harvard University Press. Wiley, T. G., and Rolstad, K. (2014). The Common Core state standards and the great divide. International Multilingual Research Journal, 8(1), 38–55. Young, R. F. (2009). Discursive Practice in Language Learning and Teaching. Malden, MA: Wiley-Blackwell. Zyzik, E. (2016). Toward a prototype model of the heritage language learner. In M. Fairclough, S. Beaudrie, A. Roca, and G. Valdés (eds.), Innovative Strategies for Heritage Language Teaching: A Practical Guide for the Classroom (pp. 19–38). Washington, DC: Georgetown University Press.

https://doi.org/10.1017/9781108989275.016 Published online by Cambridge University Press

16 Linguistic Creativity and Humor in Context Delia Chiaro

16.1 Introduction Humor depends entirely on context; without it, humor could not exist. In order to get a joke, a wisecrack or an internet meme, we need to understand its frame of reference because both in its creation and in its reception, humor is inextricably linked to its setting. Incongruity, an essential feature of humor, consists in the inappropriateness of what we expect to be said, see, or hear in a certain situation. From the person who slips on the unseen banana skin to the clown who receives a custard pie in the face, a defeat in our expectations may spark a humorous response, so it is essential that we know what those given expectations are. This chapter will focus on spoken and written banana skins and custard pies, namely verbally expressed humor, that is to say, instances of humor that pivot mainly, although not necessarily exclusively, on humor that occurs either in speech or in writing. Although in a sense, the fact that verbally expressed humor deals with words may appear manifest, occurrences of humor in speech and writing rarely stand alone. A humorous trope, such as a joke, that is delivered orally is likely to be accompanied by significant facial expressions, body movement, gesture, and a judicious use of timing. Instances of humor in written form may occur in prose or online within an internet meme or in a text message accompanied by an emoji or a gif. The specificity of the term verbally expressed humor (VEH) allows us to select and center on the spoken and written aspects of humor even if it takes place within a wider setting. Having established this, however, as with all manifestations of language, VEH (Chiaro 2005) does not occur in a void and both its making and its response will rely on a wider setting, in other words, on context. Humor is a fuzzy concept. Although we all know what it is, it is an extremely complex notion that is everything but easy to define. Numerous philosophers and intellectuals from Ancient Greece to the present day have endeavored to define humor, so I shall certainly not attempt to do so. For the purposes of this chapter, my use of the term will rest on my favorite and somewhat jolly

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

354

DELIA CHIARO

definition provided by linguist Wallace Chafe, namely “that feeling of nonseriousness” (2007). While I am uneasy about defining something by what it is not, rather than what it is, I would rather employ this simple definition than get involved in designations that are entangled in humor’s cognitive, psychological, and sociological underpinnings, not to mention its close connection with such disparate emotions as happiness and disgust. Furthermore, it would be easy to become ensnared by the notion of sense of humor which, for those who possess it, is considered a positive personality trait. If humor is indeed “that feeling of non-seriousness”; having a sense of humor (SOH) may well be a virtue. SOH plays an important role in the felicitous outcome of interactions involving VEH but, to complicate matters further, we can never be quite sure whether an unresponsive recipient of VEH does not have a particularly strong SOH or whether they are simply unamused. As SOH depends on a combination of both state and trait, you might certainly have a great SOH under normal circumstances but not if you have just been through a traumatic experience (for an extensive discussion of SOH, see Martin 1998; Ruch 1998). Finally, and possibly most importantly, humor involves moral proximity and/ or distance between the humorous content conveyed by the speaker and received by the listener. Humor and its sister, laughter, have to do with comity, since both act as a social glue but can also be divisive depending, for example, on our closeness to a mocked target or our sensitivity to an object of ridicule. What I am getting at is that it is all very well to take a joke, a funny story, or an internet meme and parse it, link it to a number of well-known theories that range from incongruity and superiority to purely linguistic models, by depriving humor of its context; we are only up to speed with half the story. Without knowing who made the humorous remark, where, when, and to whom, we cannot become familiar with its whys and its wherefores, and we may well find ourselves, as has often been said of humor research, dissecting a frog. According to Pinker (2011), one of the reasons humans understand language better than computers lies in the difficulty of programming machines with all the knowledge we have. He argues that comedies often use robots’ lack of such extensive knowledge as a source of humor. A running joke in the sixties’ sitcom Get Smart, featuring Maxwell Smart and a robot called Hymie, consists of Maxwell asking the android to “Give me a hand.” Whenever Hymie hears this utterance, he will physically break off his hand and pass it over to Maxwell. Of course, Hymie does not understand that in context “give me a hand” is a request for help and not a literal call for him to detach his hand and deliver it to the speaker. It is the lack of pragmatic knowledge that is responsible for the misunderstanding and that in turn triggers the joke. VEH is very much about this kind of (mis)understanding of contextual knowledge. Although Attardo and Raskin’s (1991) linguistic theory of humor is one of abstract competence unaffected by context, providing us with a hypothesis that pivots upon linguistic ambiguity, incongruity, and the notion of opposing and overlapping scripts, their perspective, and several more, can be appreciated in terms of misplaced context.

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

16 Linguistic Creativity and Humor in Context

Context is of course an enormous topic, and I propose to limit my examination to linguistic play that relies for its ambiguity on context, and a selection of humorous phenomena that depend on context in interesting ways. In fact, humor and context concern a vast area of scholarship that touches several disciplines – if we consider linguistics alone, it involves at the very least semantics, pragmatics, sociolinguistics, psycholinguistics, and translation studies. This chapter can therefore provide only a short overview of where the two meet (see Attardo 2020 for an exhaustive discussion). Following an outline of the variety of linguistic resourcefulness required in its creation – brief owing to the breadth of such a huge topic – I will explore the intersection of linguistic and sociocultural competences necessary in order to understand VEH and the impact of context in both its construction and its reception. Backed up by examples, I will argue that the creation of VEH mirrors unintentional slips of the tongue, as both display an identical breach of the indistinctness that is present in and across all formal aspects of language. Following an overview of some of the linguistic technicalities involved in creating VEH, I will examine how the lack of conversational cooperation can also create humor, briefly touching on tropes such as irony and sarcasm. I will conclude this overview with a discussion of the role that context and moral distance between the object of humor and its recipient play in humorous discourse and survey whether it is acceptable to make fun of delicate issues, and if so, where, when, and with whom it might be appropriate to be funny about taboo subjects.

16.2 Humor and Linguistic Creativity All languages when used for specific purposes display a series of distinct characteristics. Legalese will differ from the language of science and technology, academic discourse will be quite diverse from journalistic genres and the language of knitting is likely to be different from that of dressmaking. These genres, whether scientific, academic, or within the realm of everyday reality, exemplify a sort of baseline of serious discourse in the sense that none is designed to amuse the recipient but to inform, explain, instruct, and so forth. On the contrary, humorous discourse, which aims to amuse, is strikingly different from such “serious” varieties owing to its extreme intricacy on numerous linguistic levels. This is not to say that serious discourse cannot be intricate, but I emphasize the word extreme. While the intricacy of legalese or a knitting pattern are due to the presence of highly context-specific lexis and syntactic structures, unlike humorous discourse they are devoid of occurrences of linguistic manipulation or distortion. Whether we are looking at a joke, a single comic utterance or a lengthy funny story, when it comes to VEH, we are dealing with exemplars of a special kind of linguistic inventiveness in which the speaker stretches and twists language rules, and at the same time, if the humorous utterance occurs in conversation, may intentionally and

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

355

356

DELIA CHIARO

legitimately flout maxims. The creation of VEH is not simply a matter of adopting specific lexis, register, or convention, as in more serious genres. In order to create VEH, linguistic norms require distortion and the recipient, at least for part of the content of the incoming message, needs to be deceived through that very same manipulation. In fact, the effort involved in the construction of VEH is such that it led Charles Hockett (1977) to define the joke form in terms of “layman’s poetry,” arguing that, like traditional verse, it is founded on the extreme exploitation of language to create the desired effect, which, in the case of the joke, is to amuse. Where traditional poetry exploits features of a language, such as rhyme, assonance, and rhythm, jokes – and by extension all VEH in general – capitalize on the ambiguity present at all levels of language from its sounds and its syntactic and grammatical structures, all the way up to its semantic properties and pragmatic functions, to create a humorous outcome. In addition, Hockett argues that the joke, like poetry, is arguably untranslatable, in the sense that, owing to the fact that no two languages possess identical formal features, upon which poetry heavily depends, much will be lost in translation. Luckily, for those who do not speak several foreign languages, poetry is indeed translated, but by default, the translated versions will differ to a certain extent from the target texts. Certainly, more is lost in the translation of poetry (and by extension literature tout court) than, say, a paper on mathematics. Jokes, for the same reason, are equally untranslatable. For an extensive discussion of the complexity involved in translating humor and the strategies employed to overcome them, see Chiaro (2008, 2017). In his reflections on puns, the eighteenth-century essayist Joseph Addison went as far as to claim that if you cannot translate a “Piece of Wit,” it consequently must be a pun: a conceit arising from the use of two words that agree in the Sound, but differ in the Sense. The only way therefore to try a Piece of Wit, is to translate it into a different language: If it bears the Test, you may pronounce it true; but if it vanishes in the Experiment, you may conclude it to have been a Punn. (Addison and Steele [1711] 1982: 343)

The punster thus engages in creativity at all levels of language by manipulating sounds, words, and syntax to generate something that is likely to lose its humorous function if translated into another language. Let us consider the following example that pivots, among other things, on the exploitation of the complexity of English phonology and its writing system: Q. How do you make a cat drink? A. Easy. Put it in a liquidizer and you get a cat drink.

This rather cruel and politically incorrect riddle shows that in the phrase “cat drink,” by placing stress on the term drink (i.e., cat DRINK) and giving it phonological prominence, the speaker is leading their recipient up a garden path by having them believe that they are being questioned on ways of

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

16 Linguistic Creativity and Humor in Context

quenching a feline’s thirst. The desired response, instead, is determined by a shift in prominence in what would otherwise seem to be an identical question. However, the question is different once contextualized by the answer. To justify the answer, prominence that had previously fallen on drink now falls on the term cat (i.e. CAT drink). In other words, noun + verb becomes and is to be understood as a compound noun. If someone reads the riddle, rather than hearing it, a similar reaction occurs when they see the answer that will make them double back on their first reading in search of a hidden script. In fact, the riddle is an instance of what Attardo and Raskin (1991), in their General Theory of Verbal Humor (GTVH), explain as a single script that contains two opposing yet perfectly overlapping scripts – one script involving cat nutrition and another script about a horrible cocktail. Both orally and graphically, the deceit is equally palpable because the two versions are flagrantly manipulating a verbal phrase by transforming it into a noun phrase. English does not make use of many diacritics, so the joke is likewise deceptive in its written version. According to Joel Sherzer, what has occurred in such a riddle is an “axial” clash. In order to be interpreted correctly “cat drink” needs to be understood as a single item, a compound noun, therefore according to the version of utterance Q that we wish to accept, the two items, in becoming a single item, will occupy different slots along the syntagmatic axis. If, intentionally or otherwise, as in the “cat drink” example, the wrong filler occupies the wrong slot along the language axes, the paradigmatic axis invades the syntagmatic leading Sherzer to define such wordplay as “a projection of the paradigmatic onto the syntagmatic . . . precisely the Jakobsonian definition of poetry” (1978: 341). We have thus come full circle and we return to Hockett’s notion of layperson’s poetry. Your Wit Is My Command is the title of a book by Tony Veale. From the title, knowing that the correct expression is “your wish is my command,” one is likely to deduce that the title contains a typo. One might even think that the book might be about a witty genie because the “correct” or more natural phrase is commonly attributed to Aladdin’s genie who utters it each time he is summoned from a magic lamp to grant wishes. The book in question is actually about humor and artificial intelligence, and its title makes use of allusive or distant homophony known as alliteratio a trope that exploits the beginning or the end of two words that sound the same – wit /wɪt/ and wish / wɪʃ/ begin with the same /wɪ / sound. The book’s subtitle, Building Ais with a Sense of Humor, makes contextual sense of the replacement of a /ʃ/ with a /t/ to turn the word wish into the term wit and place it within the setting of the term command, in the sense of a directive to a computer program that will make it perform a specific task. The subtitle leads us away from the genie script and toward the intended computer script. However, there is no escaping the feeling that the title contains an intentional “mistake.” In fact, much humor is created unintentionally, and when it does, just like in cases of deliberate VEH, context is key because the mistake will clash with the situation at hand and amuse those who come across it.

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

357

358

DELIA CHIARO

16.3 Unintentional Verbally Expressed Humor Freudian slips (or parapraxis) cause laughter because they accidentally occur where and when they should not, namely in the wrong context. One of the sources Freud used to talk about the language of the psyche was jokes. The other, of course, was dreams. Freud sees these two sources of data as tapping unconscious desires. From his patients’ verbal blunders, Freud developed his well-known personality theory that he related to their feelings of shame and primal desires, the point being that these slips of the tongue revealed new contexts that made sense. In other words, although we may laugh at the person who has made the verbal blunder for a number of reasons, the verbal faux pas usually creates perfect cohesion, albeit in the wrong context, which is precisely why it is amusing. What results from the mistake is cohesive and at the same time, ironic. Therefore, yes, the perpetrator has misplaced a sound, a word, or a phrase, but it is the accurate coherence in the wrong context that causes the humor. We laugh at people’s accidental mistakes all the time, so much so that many people pretend to make verbal blunders to show that they are clever and funny. Following an article on politicians’ slips of the tongue that appeared in the Guardian newspaper (Parkinson 2021), a reader added an example of a similar mistake: At my wedding a very long time ago, a friend made us a generous present of a kitchen appliance that he nervously described to the assembled guests as a perky copulator.

A wedding guest who presents a “perky copulator” (instead of a coffee percolator), the word copulator is technically a homeoteleuton, a word/words in which the substitution of one or two syllables occurs, tells an especially good and funny story precisely because of the context in which it was narrated. A wedding is traditionally followed by a honeymoon during which the newly-weds are expected to engage in plentiful sexual activity. A guest who makes a wedding speech in which they talk about having given the couple a percolator that accidentally (or otherwise, the story might be an urban legend) becomes a “copulator” is, under the circumstances, semantically appropriate yet, at the same time, funny because sex is one of those taboos people laugh about. The addition of the adjective perky, which has the meaning ‘frisky’ and ‘saucy’ among others, adds to the comic effect. Such accidental, or accidentally on purpose instances of wordplay are precisely those upon which “original” jokes, quips, funny remarks and the like are shaped. A Freudian slip is saying one thing when you mean your mother.

This one-liner that plays on the likeness between “your mother” and “another” – it is formally a combination of alliteratio and homeoteleuton – also exemplifies how the deliberate manipulation of language can closely

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

16 Linguistic Creativity and Humor in Context

resemble a fortuitous slip of the tongue. We can imagine the person uttering the quip in a deadpan manner and seemingly above comic suspicion, as though they truly did not purposefully deliver the line with humorous intent. Nevertheless, we suspect that the remark was intentional because contextually the notion of mothers primes so well with Freud’s work and his well-known Oedipus complex metaphor. Yet the one-liner resembles a slip of the tongue. Once more context is key, misplacing “another” with “your mother,” a word that sounds the same, has the same number of syllables, and also happens to rhyme, lodges well with Freud’s theory. Verbal blunders such as spoonerisms (or distant metathesis), malapropisms, and misplaced words are all accidental instances of VEH that are a source of enjoyment and that are often reported to others in order to amuse. We laugh at the person who makes the verbal faux pas and with the person who tells us about it. Homophones, homonyms, and homographs can also lead to misunderstanding and amusement when their ambiguity leads them away from the context at hand. Sherzer (1978) reports a poet ending her lecture by saying “There are some things that only happen to a woman. Period.” The American poet used the term period in the sense of “end of story” in order to emphasize what she had said in her lecture, namely that there was nothing more to add on the matter. Nevertheless, if we consider other possible senses of period, among which we find “menstruation,” we see how the poet’s choice of word was unfortunate because she adopts a homonym that also happens to denote some thing that only happens to women, that is, menstruation. So in order to determine what speakers mean (as opposed to what their words mean), listeners and readers need to assign sense to the words they hear and see. As we have seen, a very straightforward process such as assigning sense to words can cause problems if homonyms such as period, which can mean both ‘full stop’/’menstruation,’ are involved. An American package of a make of contraceptive pill containing two months’ supply is labeled “Twin Pack” (The Telegraph n.d.). Whoever was responsible for the wording on the packaging in question was linguistically careless. It is unlikely that they purposely went for the expression “Twin Pack” to cause hilarity, yet a joke has occurred precisely because of the use of the term twin to signal a package containing two cartons of the contraceptive pill whose aim it is to avoid pregnancy. Twin, which suggests a multiple birth, is an inappropriate choice of lexis within the context of contraception, which is precisely why it is a source of amusement. In addition, this mistake, which results in cohesive irony, was reported in a newspaper column and is a retelling in the writer’s own words of the incongruity they had spotted. The writer’s intention is that readers will laugh with them, and at the pharmaceutical company that made the verbal blunder.

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

359

360

DELIA CHIARO

16.4 The Treachery of Language Intentional VEH often mirrors slips of the tongue. In Blame It on the Bellboy, a filmic comedy set in Venice directed by Mark Herman (1992), Maurice Horton, an elderly, overweight, and unattractive Englishman, has been misled to believe that he is on a date with the young and beautiful Caroline Wright. In effect, Wright is not on a date with Horton; she is an estate agent who thinks that she is about to sell him a villa overlooking the Lagoon. Even without considering the dialogues, the plot contains all the necessary elements for a farce; however, linguistic ambiguity adds to the comic confusion. As Wright shows her client around the house saying, “Would you prefer to start inside?” “Shall we go upstairs,” and “Do you like what you see?” her utterances take on two different meanings. Wright thinks her client is there for a viewing of the villa that he intends to purchase, but Horton, who thinks he is in the company of an escort, makes very different contextual sense of what she is saying: Wright: Don’t you think we ought to get down to the nitty gritty? Horton: I’m sorry! We’ve only just met, I don’t want you to think I’m an old fuddy-duddy or anything, it’s just that . . . don’t you think it’s a bit early in the day for that sort of talk? Wright: Well, I suppose so . . . Horton; Well, I suppose you young people do things differently nowadays. I didn’t mean to rush you, but that’s what we’re here for isn’t it? Wright: Yes, yes, absolutely! The two characters are speaking within two distinct and unconnected contexts. For Wright, who is trying to sell a house, getting down to the “nitty gritty” means making decisions such as bargaining on the price, finalizing the sale, and nothing more. On the other hand, Horton is looking for a sexual encounter, and so for him, that same phrase is an invitation to just that. The ambiguity of “nitty gritty” generates a dialogue in which the two speakers interact at cross-purposes because each interlocutor is speaking within opposing contexts. Interpreting Wright’s utterance as a solicitation to engage in sex, at first Horton acts shocked and embarrassed about “That sort of talk” but soon doubles back with “but that’s what we’re here for isn’t it?” For Wright, the deictic “that” refers to negotiating about the sale of the villa, but Horton understands that same “that” to mean sexual intercourse. The fact that the two are thinking in two very diverse contexts triggers amusement, and the inclusion of sex gives the humor extra vivacity. In a later scene, over a meal in which Wright is convinced that she has sold the villa, the farcical, cross-purpose conversation continues: Wright: Some people do it over the phone. Horton: Sounds as if you’re not new at this game. Wright: No, I’ve been doing it for years.

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

16 Linguistic Creativity and Humor in Context

Wright is still uniquely talking about the sale of the house, stating that negotiations can take place over the phone. However, Horton, who is still fixated on the idea of engaging in sex with Wright, has no idea that she is trying to sell him a house and misunderstands the idiom “doing it.” In British English, “it” is a euphemism, in which do acts as a pro-form standing in for an activity that combined with the pronoun it takes on the meaning of ‘engaging in sexual intercourse.’ Fresh from the earlier conversation about the “nitty gritty,” Horton takes Wright’s utterance out of her intended context and into his own. In addition, Horton is especially content to see that she is not a novice at this “game,” a word which according to the context in which Wright is operating means working in real estate, but Horton understands that she has been on the game “for years,” in other words that she is someone experienced in having sex for money. These instances of incorrect assignment of sense are down to homonymy, words and expressions that have different meanings, but the same problem can also occur because of homophony, two words or expressions that sound the same. Thomas (1995: 25) provides an example from the minutes of a management meeting at the BBC: The solicitor reported that the BBC was being sued in Ireland by a man who thought he had been described as having herpes. The BBC’s defence was that it had accused him of having a hairpiece.

The issue arose because the newsreader in question spoke in a Northern Irish accent in which the term herpes is pronounced in the same way as the word hairpiece, i.e., /ˈhɜː(r)piːz/. The exact same ambiguity is exploited, this time deliberately, in the script of the film An Everlasting Peace (directed by Barry Levinson, 2000) about two men, a Catholic and a Protestant, selling wigs (hairpieces) during the Troubles in Northern Ireland. The title of the film itself intentionally plays on the homophony of the word peace that, during the 1970s and 1980s when the film was set, was a much sought after condition in Northern Ireland, and, as it is also a homophone of piece we therefore have an appropriate title for a film revolving around the IRA and wigs. The term is contextualized twice over in terms of its “war/wig” scripts. The likeness of the terms hairpiece/herpes, uttered in a Northern Irish accent, generated a dialogue in the film in which a customer allocates the wrong sense to the term and takes offense at the implication that he might have herpes: Mr. Black: [angry] Look, I told you: I don’t know who you’ve been talking to but he’s a fucking liar! You’ll find no herpes here! Colm: [surprised] Herpes? Mr. Black: That’s what you said. George: No sir. [points to his head] Hair pieces. Mr. Black: Oh. I thought you said “herpes.”

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

361

362

DELIA CHIARO

Ambiguity does not stop short at lexis, structural uncertainty can also be the source of humor: Oozing across the floor, Marvin watched the salad dressing.

The hearer has to decide whether it is Marvin or the salad dressing that is flowing, and although common sense points to the latter, the speaker has omitted the subject of the opening clause, creating doubt. Basic world knowledge leads the hearer to understand that it is unlikely that Marvin is “oozing” therefore leading to a humorous effect. Similarly, the traditional church hall notice “Ladies, please rinse out your teapots and stand upside down in sink” plays on the same ambiguity of the dangling participle caused by a missing subject and object in the second imperative (see Chiaro 1992: 42). Like these slips of the tongue, deliberate VEH often has an accidental-but-on-purpose feel about it. This is because the same linguistic mechanisms involved in unintended verbal gaffes are adopted in the intentional creation of VEH. Uncooperative conversational behavior represents another significant base for VEH. A panhandler who asks a passerby for a cigarette with, “Have you got a cigarette?” does not receive the cigarette they had asked for, but instead, the person they had asked just answers “Yes” and walks away. Likewise, a passerby could ask another “Have you got the time?” to which the respondent does not inform them of the time of day but just says “Yes” and walks off. In both situations the passerby’s uncooperative behavior flouts Grice’s maxim of relevance, and is indulging in what Mizzau (1991) has labeled sciopero bianco delle parole or ‘words working to rule’ (my translation). Words go on strike and work to rule when a respondent reacts to the literal sense of an utterance directed at them and not to the intention or the force of the speech act. Strictly speaking, the speaker simply makes the minimum effort to understand the speech act behind the utterance and takes it at face value. By doing this, the speaker contravenes the basic conversational principle of cooperation. In these examples, the requests for an object, a cigarette, or for the time of day are deliberately interpreted respectively as appeals to find out whether the respondent is carrying a cigarette and knows the time, and not whether they are willing to offer the ciggie or tell them the time. In terms of humor, the respondent is punching down at the person making the request. The kind of humor conveyed to audiences by Sheldon, a character in the sitcom The Big Bang Theory, works on this type of uncooperative conversational style. Sheldon is a very intelligent nuclear physicist who repeatedly makes words work to rule and accordingly obtains a comic effect. In a scene that takes place at a diner, a friend, directing his gaze at a bottle of ketchup close to Sheldon, politely asks, “Sheldon is that ketchup on that table?” Instead of responding to his friend’s request by fetching the ketchup, Sheldon answers, “Yes it is,” and launches into a lengthy and complex tirade about the ingredients of the sauce. When Sheldon has finished speaking, his friend gets up and says “I’ll get it,” causing further comedy. This kind of non-cooperation does not rely solely on the recipient of an utterance but can also be triggered by the

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

16 Linguistic Creativity and Humor in Context

initiator of an exchange. Catch riddles such as the notorious “Why did the chicken cross the road?” trick the recipient into thinking they are being asked to solve a conundrum and that a far-fetched and ingenious answer is expected of them (see seminal work on riddles by folklorists Opie and Opie 1959: 73–86). Instead, the answer is not a complex play on words but a straightforward truth, in the case of the peripatetic chicken, the traditional response is “to get to the other side.” Constantinople is a very long word. How do you spell it? Where did King John sign the Magna Carta? At the bottom.

These examples of deliberate playground humor work in exactly the same way as the riddle about the rambling chicken. The riddler, who provides both question and answer, is being uncooperative by providing solutions that are both evident and uninformative. The notorious “Constantinople” riddle is a verbal ambush that has been delighting children for decades, the answer is, of course “it.” However, regarding the riddle about the Magna Carta, if a teacher were to ask a class that same question, a child who answers “at the bottom” would be judged as having given an inappropriate answer that is likely to be taken as sarcasm, especially if uttered by the class clown. This is because the teacher is not operating within a play frame and is not riddling but testing the class. A timely example of this kind of behavior is a schoolchild’s answer in a gap-filling exercise to the question “Who killed Goliath?” The answer sheet provides a blank space preceding the letters VID (i.e. “. . . VID”), to which one member of the class inserted the letters “CO” to create the word covid. Whether the child was deliberately trying to be funny we do not know, but the teacher took the response as sarcasm, as is clear from her response “See me after class” at the bottom of the page.

16.5 Encyclopedic Knowledge Whether creating or receiving an instance of VEH, not only is it essential that interactants share the language in which it is delivered, but awareness of a number of circumstantial features is also indispensable. An important contextual wall that VEH hits is that of encyclopedic knowledge, a concept that in itself is extremely complex because it does not stop short at the sentience of a standard set of facts that vaguely go by the name of culture. The kind of knowledge required to get humor clearly stretches way beyond knowing the answers to a standard pub quiz, as it often includes information shared only by a restricted group of people, and here we see the concept of proximity/ distance raising its head. So far, in most of the examples we have examined the ambiguity that triggered the VEH hinged on basic world knowledge, with the exception of the one-liner about the Freudian slip that a person unfamiliar with the psychoanalyst’s work would not get.

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

363

364

DELIA CHIARO

I shall begin by stating the obvious, namely that in order for VEH to work, the language in which a witty quip, remark, or even full-blown joke is couched has to be shared by the interactants. VEH is created through the exploitation of linguistic features present in a given language in order to create the opposition and overlap first proposed in Raskin’s Semantic Script Theory (1984) and later expanded by Attardo and Raskin in the GTVH (1991). Attardo and Raskin argue that a humorous script must essentially consist of two overlapping scripts, one of which is obvious but that simultaneously shields another, less noticeable, script that is not immediately apparent. These two scripts must be in opposition to each other in order for humor to result. Why did the window box? Because it saw the garden fence.

To native speakers of English, this classic children’s riddle is evident. The opposing “gardening” script (window box and garden fence) versus the “sport” script (boxing and fencing) overlap to perfection rendering the riddle both ambiguous in its contrasts but at the same time unique in its words’ ability to conceal the two meanings. In translation the riddle just does not work – unless, of course, we find a language in which the words for both box and fence contain the same homonymic sense as English. However, let us move on to instances of VEH where all interactants speak the same language. According to the Roman orator Cicero, who among other things was notorious for his verbal wit, “a witty saying has its point sometimes in facts, sometimes in words, though people are most particularly amused whenever laughter is excited by the union of the two” (II, LXI, 248) (1965: 383). Specifically, according to Cicero, the best witty remarks, jokes, etc. combine language and world knowledge. Over and above proficiency in English, only those familiar with the Plastic Ono Band will appreciate that a vegetarian is someone who gives peas a chance and a minimum knowledge of English and French history is required to realize that Richard Coeur de Lion was the first heart transplant. For such VEH to travel successfully from initiator to recipient, two types of expertise are indispensable, linguistic expertise and encyclopedic or sociocultural knowledge, because, as emerges from the examples, these two competences may intersect. Expressly, you cannot do much if you just have awareness of one and not the other. These remarks exploit the options of English and general knowledge, too, and so far so good. However, as mentioned earlier, sociocultural knowledge is not as straightforward as one might think. Things get more complicated when knowledge becomes more specialized. Like everything else in life, humor changes in time. Rather, humor based upon situations that were once timely are unlikely to be funny today, simply because the relevant context is lost. In the 1950s, folklorists Opie and Opie (1959: 106) report the following playground joke: Do you know a teddy boy’s just been drowned – in his drainpipe trousers?

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

16 Linguistic Creativity and Humor in Context

It is questionable that schoolchildren today will be familiar with teddy boys, a British youth subculture of the 1950s, and the type of high-waisted “drainpipe” trousers they wore. We need not go back so far in time to find that jokes and witty remarks soon go stale owing to their invisible sell-by date that anchors them to a particular moment in time. The scandal involving Bill Clinton and Monica Lewinsky in the 1990s spawned countless jokes, and it is implausible that the average millennial will see two opposing and overlapping scripts in the claim that what Clinton found attractive in Lewinsky was “the prettiest smile he had ever come across” – but a Boomer is liable to see both. Whether anyone finds the quip funny or not is another matter and involves sense of humor and issues connected to good and bad taste that are beyond the scope of this chapter. Apart from temporal context, most individuals’ encyclopedias are bound to be limited to their world and to their interests – we do not live in a world of polymaths. I had no idea that engineers confuse Halloween and Christmas because Oct. 31 = Dec. 25. According to two numerical systems known as Octal (which can be abbreviated to Oct.) and Decimal (or Dec.), Octal 31 is equal to Decimal 25. At this point linguistic knowledge enters the equation (pun intended) as both look like month abbreviations and therefore we have a joke whose understanding is limited only to those familiar with the two numerical bases. I could go on providing data of VEH that is restricted to people with extremely specialized knowledge of specific topics, which could require knowing about practically anything from medicine to football, but I think that I have made my point: overlap and oppositeness involve language and world knowledge. Owing to its inherent linguacultural ambiguity, VEH is by default cryptic. As we have seen, it is about playful deceit, and sometimes, as in the case of sarcasm, somewhat cruel deception. Our immediate response to the query “What comes steaming out of cows backwards?” would be dung, although the right answer would be the Isle of Wight ferry – Cowes, a homophone of cows, is the island’s main seaport. Contexts shift. Both dung and ferries may steam, but in order to solicit laughter or a smile the contexts must seamlessly overlap. For the creator of VEH, linguistic competence and encyclopedic competence are, however, not enough. The creator also needs comic competence to be able to spot linguacultural duplicity and partake in Sontag’s notion of “partial knowing” (2004). According to Sontag “the theory of knowing that is relevant to the comic is essentially a theory of non-knowing, or pretending not to know, or partial knowing” –the creator of VEH must pretend not to know, they must pretend to be slipping up.

16.6 In and Out of the Play Frame Having examined the significance of linguistic and sociocultural context in VEH, I shall now look at situational context. Where and when it is acceptable to joke and engage in banter varies across cultures. Although I have no hard

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

365

366

DELIA CHIARO

data to back this up, VEH is extremely pervasive among native speakers of English, or to quote anthropologist Kate Fox “The important ‘rule’ about English humour is its dominance and pervasiveness. Humour rules. Humour governs. Humour is omnipresent and omnipotent” (2017: 61). While in many cultures there is a special time and place for humor, for the English there appears to be a permanent and ubiquitous undercurrent in all spoken interactions. Even phatic communion about the weather might include the ironic “Nice weather for ducks” if it happens to be a rainy day, while a response to “How are you?” could well be “How long have you got?” if the speaker does not want to provide the standard cooperative reply “Fine thanks.” The English are also renowned for understatement, as illustrated in this brief exchange between P.G. Wodehouse’s Bertie Wooster and his valet Jeeves: “Have you seen Mr Fink-Nottle Jeeves?” “No Sir.” “I am going to murder him.” “Very good, Sir.”

Tropes like irony, hyperbole, and understatement, all of which violate the conversational maxim of quality and that are highly context dependent play an important part in the creation of satire (for an exhaustive discussion of satire, see Simpson 2003). Satirical magazines and TV shows rely, among other things, on highlighting the incongruity of newsworthy stories. In October 2021, the cover of the satirical news magazine Private Eye (Issue 1557) shows former UK Prime Minister Boris Johnson in the company of Jeff Bezos, founder and chairman of Amazon. The speech bubble coming out of Johnson’s mouth reads, “That’s amazing, you actually deliver goods!” Once more, we are looking at two contexts, Bezos’ successful online marketplace and Johnson’s frequent inability to “come up with the goods” or to furnish the nation with what is expected of him, politically and socially, and therein lies the irony. Another cover of Private Eye in the same month (1159), in tune with United Nation’s Climate Change Conference held in Glasgow, shows Russian Prime Minister Vladimir Putin shaking hands with Chinese President Xi Jinping. Putin is saying, “We’ve saved a lot of energy” and Xi Jinping ironically replies, “By not going to COP26.” Have I Got News for You (HIGNFY) is a BBC satirical television panel show in which panelists answer questions on various news stories of the week prior to each episode. In an episode of October 2021, the compere reports how “Jack the Chipper,” a fish and chip shop in London, had been criticized over its name. Owing to the play on words recalling the terrible murderer Jack the Ripper, the owner was accused of both supporting violence and hating women. In response, and presumably unaware of the irony he was about to create, the owner decided to hold a “ladies’ night” when women could buy fish and chips at half price.1 In another episode, audiences learn that the Kunsten Museum of 1

Have I Got News for You, Series 62, Episode 4, 10/29/21.

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

16 Linguistic Creativity and Humor in Context

Modern Art in Aalborg, Denmark had given the artist Jens Haaning £65,000 with which to create a framed display that would represent the average wage of people in Denmark and Austria. Instead of framing banknotes, as requested by the museum, Haaning produced a blank canvas entitled “Take the Money and Run.” As though the tongue-in-cheek title of the opus was not ironic enough, in his defense the artist said, “The work is what I have taken the money for.” To add to the causticness, the quiz commentator jested that soon to appear at the gallery was a painting entitled “Artist with Broken Legs” by Frankie the Knuckle.2 Finally, several topics that are often the subject of humor are considered delicate and therefore not always appropriate to joke about. These subjects include taboo areas such as death, race, religion, sex, and bodily functions. Such humor that is in bad taste is likely to offend, yet the fact that it exists shows that it must appeal to some people. Traditionally disasters such as 9/11 and the death of Diana Spencer would trigger cycles of oral jokes (see, for example, the work of Dundes 1987; Davies 1999; Kuipers 2015), at the time of writing similar tragedies are more likely to prompt cycles of internet memes disseminated via smartphones. Whether passed round orally or via phones, these cycles display people’s humorous creativity. Furthermore, whether it is right or wrong to joke about such matters is a Sisyphean debate which, at the end of the day, boils down to context. Presumably, a person who sends a distasteful meme will do so only to someone who shares a similar sense of (bad) taste. However, social media platforms are dangerous ground for sharing icky jokes and memes because of a huge and variegated audience whose tastes in humor the person posting is unlikely to know. Moreover, the closer a person is to the target of a potentially offensive joke, the less likely they are to appreciate it. Collecting data on joke evaluation, I once asked respondents “Why is the Tory party known as the cream of society?” Before hearing the answer, “Because they’re rich and thick and full of clots,” one person remarked, “I don’t think I’m going to like this,” thus revealing her sympathies with the Conservatives. She knew the joke would punch up at the Tories with whom she shared her political views. The morally closer we are to the target of a joke, the more likely we are not to respond positively to ridicule about them. An example of the fine line between serious and non-serious discourse, and above all, moral proximity to a target of humor, can be exemplified by an incident that occurred in 2012 during a live performance in which stand-up comedian Daniel Tosh told a rape joke. Whether or not the joke was in good or bad taste is not at issue, but what is at issue is the comedian’s response to a woman in the audience who heckled him and shouted out that rape is never funny. To this, Tosh replied, “Wouldn’t it be funny if that girl got raped by, like, five guys right now? Like right now?” Clearly, Tosh tried to humiliate the woman by asserting his power, as stand-up comedians will do when heckled. However, his comment caused much controversy on social media and soon led 2

Have I Got a Bit More News for You, Series 62 Episode 2, 10/20/21.

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

367

368

DELIA CHIARO

to his having to apologize. Tosh’s initial joke may have been in bad taste, yet people do joke about everything and anything. The problem with Tosh’s response to his heckler was that it sounded like an invitation to rape and a celebration of a violent crime. In addition, and this is likely to have been the issue, he had stepped outside the play frame for his riposte. Although he was standing on stage, and in a comedy club, and the rape joke disputed by the heckler had taken place within a play frame, context matters. Tosh was knowingly or unknowingly approving and giving authenticity to a felony by stepping out of the play frame.

16.7 Closing Remarks Both creating and transmitting VEH successfully is heavily dependent on context – without context humor could not exist. Whoever creates VEH must possess linguistic competence, sociocultural competence, and, of course, comic competence in their ability to use the abstruseness present in language and culture to make it work as a comic device. The joker must be able to exploit the wide range of linguistic options available in the language to create a script, whether that script is in the form of a joke, a one-liner, or a meme as long as it contains a trope that displays VEH. The humorous script should contain two scripts, although the receiver will only see one – unless they have heard the joke before – resulting in a double-faced trap enticing the recipient to fall and then find comic relief. In addition, the joker must also pretend not to know that they are playing with two scripts in one. Comic competence ideally also comprises an awareness of the receiver’s sensitivities. In the case of sensitive subjects, the closer the receiver is to the target of the witty remark, and the more physical and moral elements are being shared with the target, the more likely it is that the receiver will be offended by a remark that punches down. Rather than take offense, on the contrary, the receiver can respond with equally creative banter, as did Obama when asked by comedian Zach Galifianakis on a spoof celebrity interview what it was like to be the last black President. Obama, noticing the ambiguity in the term last, creatively twists it away from Galifianakis’ pseudo-racist remark and answers, “Seriously? What’s it like for this to be the last time you ever talk to a President?” illustrating his skill in lingua-comic competence (ABC News 2014). Or, for example, when the Rolling Stones’ front man Mick Jagger uses irony in response to a journalist criticizing Bob Dylan’s voice. Dodging the journalist’s denigration of Dylan, Jagger ironically replies “Dylan has never been one of the great tenors of our time” impeccably displaying low-key comic competence.3 Context in humor is all-encompassing and this chapter has simply touched the tip of the iceberg.

3

The Beat J. Mick Jagger about Bob Dylan’s voice. www.youtube.com/watch?v=JqHG1YYV5yI.

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

16 Linguistic Creativity and Humor in Context

References ABC News (2014, March 12) President Obama and Zach Galifianakis: “Between Two Ferns” [Video]. https://abcnews.go.com/WNN/video/president-obama-zachgalifianakis-between-two-ferns-22872421. Addison, J., and Steele, R. [1709–1712] (1982). In A. Ross (ed.), Selections from the Tatler and the Spectator of Steele and Addison. Harmondsworth: Penguin. Attardo, S., and Raskin, V. (1991). “Script Theory revis(it)ed: Joke similarity and joke representational model.” Humor: International Journal of Humor Research, 4(3), 293–347. Attardo, S. (2020). The Linguistics of Humour: An Introduction. Oxford: Oxford University Press. Chafe, W. (2007). The Importance of Not Being Earnest: The Feeling behind Laughter and Humor. Amsterdam: John Benjamins. Chiaro, D. (1992). The Language of Jokes: Analysing Verbal Play. London: Routledge. Chiaro, D. (2005). Verbally expressed humor and translation: An overview of a neglected field. Humor: International Journal of Humor Research, 18(2), 135–145. Chiaro, D. (2008). Verbally expressed humor and translation. In V. Raskin (ed.) The Primer of Humor Research (pp. 596–608). Berlin: De Gruyter. Chiaro, D. (2017). Humor and translation. In A. Salvatore (ed.), The Routledge Handbook of the Linguistics of Humor (pp. 414–429). New York: Routledge. Cicero, M. Tullis. (1965). De Oratore, Libri tres. Heidesheim: Olm. Davies, C. (1999). Jokes about the death of Diana, Princess of Wales. In T. Walter (ed.), The Mourning for Diana (pp. 253–268). Oxford: Berg. Dundes, A. (1987). Cracking Jokes: Studies of Sick Humor Cycles and stereotypes. Berkeley: Ten Speed Press. Fox, K. (2017). Watching the English: The Hidden Rules of English Behaviour. London: Hodder and Stoughton. Hockett, C. F. (1977). Jokes. In The View from Language. Selected Essays 1948–1964 (pp. 257–289). Athens: University of Georgia Press. Kuipers, G. (2015). Good Humor, Bad Taste: A Sociology of the Joke. Berlin: De Gruyter. Martin, R. (1998). Approaches to the sense of humour: A historical review. In W. Ruch (ed.), The Sense of Humor: Explorations of a Personality Characteristic (pp. 15–62). Berlin: Mouton de Gruyter. Mizzau, M. (1991). Strategie nel confitto nei dialoghi di Ivy Compton Burnett. In S. Stati, E. Weigand, and F. Hundsnurscher (eds.), Dialoganalyse III. Berlin: De Gruyter. Opie, I. and Opie, P. (1959). The Language and Lore of School Children. Oxford: Oxford University Press. Parkinson, H. J. (2021, August 20). I relish a good Freudian slip: That revealing giveaway of the tongue. The Guardian. www.theguardian.com/lifeandstyle/2021/ aug/20/i-relish-freudian-slip-revealing-giveaway-of-the-tongue#comment -151355524. Pinker, S. (2011). Steven Pinker on Language Pragmatics. www.youtube.com/watch? v=VKbp4hEHV-s. Raskin, V. (1984). Semantic Mechanisms of Humor. Dortrecht: Reidel.

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

369

370

DELIA CHIARO

Ruch, W. (1998). Sense of humour: A new look at an old concept. In W. Ruch (ed.), The Sense of Humor: Explorations of a Personality Characteristic (pp. 3–13). Berlin: Mouton de Gruyter. Sherzer, J. (1978). Oh! That’s a pun and I didn’t mean it. Semiotica, 22(3/4), 335–350. Simpson, P. (2003). On the Discourse of Satire. Amsterdam: John Benjamins. Sontag, S. (2004). Performance Art. In PEN America Issue 5: Silences (pp. 92–96). New York: PEN American Centre. Thomas, J. (1995). Meaning in Interaction: An Introduction to Pragmatics. London: Routledge. Veale, T. (2021). Your Wit Is My Command: Building AIs with a Sense of Humor. Cambridge, MA: MIT University Press.

https://doi.org/10.1017/9781108989275.017 Published online by Cambridge University Press

17 Context in Translation and Interpreting Studies Luis Pérez-González

17.1 Introduction Popular perceptions of translation and interpreting assume that these forms of interlingual and intercultural mediation are based on an essential determinacy of meaning. This lay understanding of the written and oral modes of interlingual transposition that translation and interpreting entail is largely informed by a conception of original texts as “entit[ies] with a stable, definable meaning” that translated texts should aim to deliver in the target language (Mason 2006: 359). This view prevailed also in early translation scholarship published between the late 1940s, when translation emerged as an academic discipline (Tymoczko 2013), and the late 1980s. During this period translation was regarded as a subfield of linguistics, with theorization on translation initially revolving around a rigid understanding of textual equivalence as a form of identity rather than similarity (Tymoczko 2013). These early views were gradually superseded by other perspectives that placed less emphasis on referential or denotative equivalence, acknowledging instead the role that a range of linguistic units and ranks, text-normative constraints, and textual effects played in the production of equivalent translated texts (Kenny 1998: 96). The premise that translators and interpreters operate in a complex and often sensitive cross-linguistic and cross-cultural environment that calls for the exercise of professional judgment in context is now firmly entrenched in disciplinary discourses. However, the construct of context has not been the subject of systematic theorization on translation and interpreting until relatively recently. This chapter sets out to explore how scholarly thinking on context has informed research in translation and interpreting studies since the early 1990s. It begins with a historical overview of the contribution that linguistics has made to the emergence, development, and consolidation of translation and interpreting studies as a self-standing discipline. It then proceeds to critique a range of perspectives on context in translation scholarship. These range from

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

372

LUIS PÉREZ-GONZÁLEZ

cognitive approaches exploring how participants in each communicative encounter come to share and make use of a given set of contextual assumptions, to conceptions of context as a field of power play where identities are dynamically negotiated. This exploration is illustrated with examples from different domains of translation and interpreting research to foreground the breadth of theoretical and methodological orientations that converge within the discipline.

17.2 Linguistic Approaches to Translation Recognition of translation and interpreting as social phenomena, and of translators and interpreters as agents who contribute to the stability or subversion of social structures, constitutes a recent development in the evolution of translation studies as a disciplinary domain. During the late 1940s, early scholars of translation chose to align themselves with linguistics – at a time when the latter was emerging as a “scientific” discipline “with a highly formalized and rigorous machinery for describing languages” – in a bid to secure academic recognition (Baker 2005: 287). Between the 1950s and the late 1970s, translation scholars thus set out to formalize processes of lexical and syntactic transfer across languages as a basis for the elaboration of taxonomies of translation strategies, often for pedagogical purposes, informed by transformational grammar and the concept of deep structure (Saldanha 2009). Early conceptualizations of translation were considerably influenced by the notion of equivalence, understood as the replacement of textual material in the source language with written or spoken utterances in the target language that may or may not fully map onto the same aspect(s) of the extralinguistic world. Initially, equivalence was understood as a semantic category. From this perspective, translating involved finding utterances or sentences in the target language that referred to the same “slice of reality” (Rabin 1958: 123) as its source language counterpart, irrespective of the specific constraints at play in the communicative situation. Over the next few decades, the treatment of equivalence by translation scholars incorporated additional dimensions beyond semantics. In this revised approach, translations were regarded as equivalent if they managed to reproduce the “same effect” that their respective source texts had on their readers (Nida 1964; Nida and Taber 1969), or to match the main “textual function” or communicative purpose of the original text (Reiss 1971; House 1977/1981). As the disciplined developed, the emphasis placed by these early theories on formal linguistic analysis – rather than on translation as a social, professional, or political phenomenon – and on the sentence as the uppermost unit of analysis came to be perceived as unnecessarily mechanistic. By the mid1980s, a growing consensus emerged, particularly among German scholars, about the difficulty of measuring the effect of translated texts on their readers (Vermeer 1989; Nord 1991). The focus was therefore shifted to the fact that

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

17 Context in Translation and Interpreting Studies

translated texts often need to serve a different function from the one that motivated the production of the original text in the first place. For proponents of functionalist perspectives, including skopos theory, “adequacy” is achieved when translations successfully comply with the commissioner’s requirements (Vermeer 1989), thereby acquiring a function of their own within the target sociocultural context (Nord 1991). With the influence of linguistics gradually eroding, the contribution of other disciplines in the humanities to translation studies, most notably cultural studies and literary theory, gained growing recognition. This alternative orientation, often referred to as the “cultural turn,” represented a shift from the positivism fostered by linguistic approaches to a more relativist stance on the study of translation (Marinetti 2011), ultimately informed by the premise that the meaning of a text is unstable and culturally constructed. By highlighting the impact of sociocultural constraints on the translational activity (Hermans 1985, 1996; Venuti 1995), the new approach challenged “long-held tenets of translation such as the very notion of S[ource] T[ext]-T[arget] T[ext] equivalence” (Munday 2009: 11) and paved the way for the politicization of translation studies. The epistemological challenge that cultural approaches to translation studies ushered in, combined with developments in various strands of linguistic research, widened the range of theoretical frameworks available to translation scholars with an interest in textual analysis. House’s translation quality assessment model (1997), for example, elaborated on her earlier application of Hallidayan linguistics (1977/1981) to examine how translation is influenced by register, which encapsulates the connection between texts and their situational micro-contexts in each communicative encounter, and genre, namely the ritualized conventions of the culture in which texts are used to serve specific purposes. Also drawing on the Hallidayan tradition, Hatim and Mason (1990) explored how discourse and genre – conceptualized in their work as social planes that capture the ritualized linguistic expressions of sociocultural attitudes and the culturally recognized verbal means through which speakers pursue social goals, respectively – find expression in texts. Hatim and Mason’s (1997) understanding of translation as a form of mediation across these semiotic dimensions and their emphasis on the mutually shaping relationship between translation and the social situations where it is embedded brought into sharp relief the influence of ideology and power on translators’ and interpreters’ behavior. The contribution of translators’ decisions to reinforcing or countering the use of language as an instrument of ideological control in original texts emerged as a distinct object of inquiry in studies informed by critical discourse analyses of texts in political, institutional, and journalistic settings (Schäffner 2003, 2012; Calzada Pérez 2007; Kang 2007). Baker’s (2006a) application of socio-narrative theory in translation studies provided scholars in the field with a framework to account for translators’ and interpreters’ behavior in terms of the alignment or clash between the narratives or views of the world that they subscribe to and the narratives

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

373

374

LUIS PÉREZ-GONZÁLEZ

that circulate in their environment, including those inscribed in the texts they translate. Baker’s original conceptualization of narratives and the revised typologies proposed by subsequent scholars (Boéri 2008; Baker 2010; Harding 2012; Bassi 2015) have revealed how the intersection between personal, public, conceptual, and meta-narratives can influence translators’ and interpreters’ interventions in the target language and “re-narrate” the texts they work with. For translators, re-narrating a text – for example, as a form of resistance against narratives that create the conditions for the emergence or perpetuation of moral or armed conflict – involves exercising agency and discharging their moral duty to decide whether “to reproduce existing ideologies as encoded in the narrative elaborated in the text or utterance, or to dissociate themselves from those ideologies, if necessary by refusing to translate the text or interpret in a particular context at all” (Baker 2006a: 105). As the theoretical foundations of the discipline have broadened, the range of models and methods used in translation and interpreting research are no longer limited to those originally borrowed from literary criticism and contrastive linguistics. While translation studies continues to rely mainly on qualitative methods of textual analysis (Chesterman 2000), there is a growing trend to make use of interviews (Diriker 2004; Torikai 2009; Sapiro 2014; Marín Lacarta 2020; Zwishchenberger 2020) and ethnographic methods (Koskinen 2008; Tesseur 2014; Yu 2017). Process-oriented research, on the other hand, has drawn on and adapted methods from the field of cognitive psychology to empirically test the processes of cognitive management that interpreters activate under stress conditions, access translators’ and interpreters’ thought processes through think-aloud protocols, and monitor other process-related decisions through eye-tracking (Fonseca 2015; Hvelplund 2017; Lacruz 2017) or keystroke logging devices (Lörscher 1991; Jääskeläinen 1999; Jakobsen 2006, Halverson 2017; Rodríguez-Inés 2017). Corpus linguistics has also made an important contribution to translation and interpreting studies since the 1990s (Laviosa 2002). Parallel corpora consist of computer-held sets of texts in the source language and their target language versions in one or more languages, whereas comparable corpora are collections of texts written in the same language that share certain features or selection criteria (e.g., period of publication, topic, genre) (Kenny 2009). The use of corpora in translation studies research is driven by the premise that translational activities are bound to leave traces in the target text (Baker 2003). Using concordancing software, corpus-based translation studies have focused on the distinctive patterns of language use observed in translated language, also referred to as “typical features of translation” (Olohan 2004), including the tendency of translated texts to be more explicit and use more conventional grammar and lexis than texts originally written in either the source or target languages. The compilation and study of corpora continue to serve as a robust methodology in advancing applied (Ferraresi et al. 2010; Kübler and Volanschi 2012; Tagnin and Teixeira 2012; López-Rodríguez 2016; Rodríguez-Inés 2017) and theoretical research agendas, including but not limited to the study of

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

17 Context in Translation and Interpreting Studies

creative language (Dirdal 2014; Saldanha 2014), multimodal texts (Bernardini et al. 2018; Soffritti 2019), or translation processes as part of mixed-methods approaches (Halverson 2017 and Szymor 2017). As this historical overview shows, early linguistic approaches to translation focused on systemic differences between languages and developed semantic notions of equivalence to formalize internal mechanisms of textual transfer at sentence level. As the positivist epistemological orientation underpinning this body of scholarship began to lose its explanatory power, various disciplines in the humanities shifted the analytical lens onto more relativistic dimensions of translation – including the role of communicative situation, culture, and power. Since the 1990s, these aspects located at the interface between language and context have become more salient also in research work undertaken by linguists capitalizing on the affordances of new qualitative and quantitative research methods.

17.3 Translation and Context Baker’s insight that the notion of context “is hardly ever subjected to scrutiny in its own right,” despite “being routinely invoked in much of the literature on translation and interpreting” (2006b: 322), drives her critique of three conceptualizations of context developed by scholars in the fields of pragmatics and linguistic anthropology that have the potential to facilitate new developments in translation studies. This section examines how each of these approaches to the study of context has been applied in translation research, drawing on examples from various areas within the discipline.

17.3.1 Social versus Cognitive Approaches The first perspective is articulated around the distinction between social and cognitive approaches to context. In social approaches to context, the translation activity is examined against the sets of preexisting situational parameters and entities that mold real-world communicative encounters. Social perspectives on context in the field of translation studies, Baker (2006b) argues, are ultimately fashioned after Hymes’ (1964) S P E A K I N G model and its constitutive dimensions: S E T T I N G , pertaining to specific physical circumstances and the wider culture that encounters are embedded in; P A R T I C I P A N T S , including members acting as overhearers or bystanders; E N D S , understood as sets of outcomes and goals that may be individually or collectively pursued; A C T S E Q U E N C E , pertaining to the way in which a text unfolds; K E Y , referring to the mood or tone of the communicative encounter; I N S T R U M E N T A L I T I E S , as in channels and forms of speech; N O R M S , or socially agreed forms of behavior or performance expected in any given setting; and G E N R E , which designates conventionalized types of communicative encounters. Context, as realized through this set of parameters, assists text producers and receivers in assigning

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

375

376

LUIS PÉREZ-GONZÁLEZ

meaning to linguistic form, particularly when “the linguistic form on its own supports a whole range of meanings,” some of which can then be “eliminated or downplayed” (Baker 2006b: 323). As noted in Section 17.2, explorations of the impact of situational and cultural parameters on translational decisions have been informed primarily by applications of key Hallidayan concepts, notably register and genre, in translation studies (House 1977/1981; Hatim and Mason 1990, 1997; Bell 1991). But as sociological approaches to translation have gained recognition in recent years, a growing body of researchers have turned their attention to new aspects of the relationships between translators and other actors (Chesterman 2006). Gambier (2006), who advocates a differentiation between the sociology of translations and of their translators, makes a strong case to explore translators’ life stories, professional status, career paths, or working conditions under the widening remit of translation studies. This call has been heeded, among others, by scholars working on translation in digital culture. Technological advances have enabled the creation of networked communities of participatory translation, often consisting of individuals without formal training in interlingual and intercultural mediation (PérezGonzález and Susam-Saraeva 2012). With translation, whether crowdsourced or unsolicited, emerging as a form of immaterial labor that enables the flow of translated content outside traditional publishing and media industry circles, it has become ever more important to understand the motivations of these individuals (P A R T I C I P A N T S ) to become involved in the translation of digital commodities as members of wider assemblages of affective production (S E T T I N G ). Drawing on surveys as her core research method, Mcdonough Dolmaya (2012) investigates whether and why Wikipedia volunteer English translators donate their labor for a range of other crowdsourced translation projects. The information these individuals provide on their previous translation experience and current occupation allows Mcdonough Dolmaya to identify a number of “intrinsic” and “extrinsic” motivations to translate. These reveal their personal perceptions of translation (N O R M S ) and signal their willingness to translate for other crowdsourced translation projects (e.g., “cause-driven” vis-à-vis “product-driven” or “outsourcing-driven” initiatives) (G E N R E S ). Other studies (Mcdonough Dolmaya 2011) focusing on blogs run by translation scholars and practitioners identify personal circumstances and the content of posts (K E Y ) as explanatory parameters of bloggers’ attitudes toward translation, as well as the technological and ethical challenges associated with certain practices – thus opening new perspectives on the sociological dimensions of translation. But social perspectives on context also drive research on more traditional areas of inquiry in translation studies. Placing the concept of retranslation at the center of his study, Jones (2019) harnesses the power of corpora to scrutinize two language versions of Thucydides’ History of the Peloponnesian War, separated by more than one and a half centuries (Bloomfield 1829; Lattimore 1998), and unveil how the ordinary population of Thucydides’

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

17 Context in Translation and Interpreting Studies

Greece (referred to in the Greek original by a range of expressions, including but not limited to “the common people”) is characterized in two different periods of modern English history. Bloomfield’s version published in the first half of the nineteenth century, on the one hand, shows a preference for “the multitude” – a derogatory expression signaling a certain degree of contempt for the lower social classes in early modern England. By contrast, Lattimore tends to favor phrases like “common people” or “the majority” with more positive connotations. While Bloomfield’s characterization of the ordinary population in Thucydides’ time foregrounds the heterogeneity and the competing desires of the group, Lattimore’s emphasizes solidarity and a sense of unity of will. Jones’ analysis explores a number of situational perspectives – including the translators’ political views at different points in history – to account for these choices: The extent to which these choices are the result of deliberate, conscious strategies on the part of Lattimore or Bloomfield is of course extremely difficult if not impossible to assess. . . . [M]any forces are at play in the retranslation process and the differences observed must consequently be explained with reference not only to each translator’s personal politics, but also to broader trends of linguistic and ideological change within society at large. . . . Thucydides himself was a member of the Athenian aristocracy and it is now widely accepted that he was rather ambivalent in his opinion of democracy, remaining undecided in his assessment of the suitability of the ordinary Athenians to hold political power . . . [I]t is in my view likely that the patterns discussed above can be attributed at least in part to the radically diverging sets of stereotypical knowledge that have formed the schemata of interpretation for each translator’s representation of the non-elite members of the ancient Greek societies described. (Jones 2019: 6–7)

The context (conceptualized as S E T T I N G in Hymes’ model) in which the two translators worked is bound to have provided Bloomfield and Lattimore with diverging experiences and social role knowledge of the ordinary class in their respective communities, thus filtering their personal interpretations of the way in which “the common people” behaved and acted in Thucydides’ History of the Peloponnesian War. In this case, studying translation from a situational perspective helps us to understand how different translators retrieved different meanings from the same linguistic expressions, based on the impact that the social structures they were embedded in had on their interpretation of Thucydides’ discourse. Cognitive perspectives, on the other hand, regard context as a psychological construct. Under this approach, translational decisions are to be accounted in terms of the translator’s assumptions about who the target readers or viewers of a translated text are. Cognitive perspectives of context – as elaborated in the field of discourse studies by Van Dijk (2001a, 2001b) – have informed the work of researchers interested in pragmatic aspects of the translation activity. Leppihalme (1997), for example, examines whether Finnish readers are able

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

377

378

LUIS PÉREZ-GONZÁLEZ

to access the implicated premises that allusive passages in the original English texts were able to evoke in their original readership. As the English and Finnish versions are bound by their own genre conventions and influenced by their respective cultural traditions and intertextual networks, readers’ comprehension of implicatures in the target language may be facilitated or hampered by translators – specifically, by their assumptions about Finnish readers’ capacity to bring relevant contextual information to the text. The extent to which meaning can be inferred from the interaction between a text and its context has also been examined in the context of Bible translation. Hill (2007) draws on the relevance-theoretic premise that readers expect texts to be as easily comprehensible as possible, thus following the “path of least interpretive effort” in processing said texts (Wilson and Sperber 2004: 613–614). More pointedly, Hill’s study focuses on the potential mismatch between the implicatures – namely the contextual assumptions or implications (Sperber and Wilson 1995: 194–195) – that the original text evoked among its readers and those that can be realistically retrieved by a contemporary target audience. Hill argues that only by striking an optional “effort–benefit” balance can translators draw safe assumptions as to whether the audience of the translated text may be able to infer the intended meaning. Following Gutt (2000), Hill (2007: 56–46) suggests that one way of tackling problems arising from the differences between the cognitive environments of source- and target-language readers is to “explicitate” the relevant implicated premises. This “contextual adjustment” or “contextual bridge” can be achieved both within the text, i.e., with the insertion of an explanatory word or a lengthier linguistic expression, or outside the text, e.g., by means of a footnote (pp. 45, 82–85). The relevance of specific sets of assumptions to understand how translated allusive texts are comprehended would have been unverifiable, were it not for the methodological innovations that translation studies has witnessed in recent decades. The reception studies undertaken by Leppihalme and Hills drew on interviews to establish whether readers had managed to work out intended implicated premises. Experimental reception studies have also been carried out by other researchers exploring how translators and their readers negotiate the interface between their respective cognitive environments (Desilla 2012, 2014) – i.e., the set of assumptions that they share, as opposed to those that translators assume are shared (Hill 2007: 14). More recently, cognitive perspectives on context have informed scholarly research on and professional practices in audio description. In this form of intersemiotic mediation, visual aspects of an audiovisual text that play an important role in driving the plot forward are “described” in spoken form for the benefit of visually impaired viewers during stretches of the film that feature no dialogue. Early scholarship on audio description therefore assumed that describers fashion their narration by drawing on their professional judgment and deciding how much and what type of information viewers need at any given point (Orero 2008). Translational decisions made by practitioners were thus held to be motivated by their impressions as to which elements of the

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

17 Context in Translation and Interpreting Studies

visual context had to be encoded in spoken form to assist viewers in their comprehension of the film. The arbitrariness stemming from the fact that the describers’ assumptions were unverifiable went unchallenged during the formative years of audio description studies. Relying on practitioners’ assumptions as a product of their own cognitive processes was acceptable because “[e]ven after watching the same film different people have different recollections and interpretations, and in some cases some details are observed by some while going unnoticed by others” (Orero and Vilaró 2012: 297–298). But methodological advances enabled by technological developments have sought to address the arbitrariness that lies at the heart of cognitive perspectives on context. Eye-tracking analyses have been undertaken to gain a more systematic understanding of the reasons why viewers’ attention tends to be drawn to certain visual elements and the reasons why that is so. The premise underpinning this research strand is that, by establishing which visual components are more prominent for sighted viewers, researchers will be able to understand the role they play in facilitating the viewers’ comprehension of multimodal texts. By transferring these evidence-based insights to audio description practice through a set of empirically derived guidelines, audio describers should be able to “re-narrativize film more effectively for blind viewers” (Kruger 2012: 69). Research produced to date indicates that, based on their assumptions about the needs of sight impaired viewers, describers’ narrations have tended to prioritize visually salient elements. This was the case even when visually prominent elements did not have high “narrative saliency” – a distinction that audio description scholars have only recently become aware of, following the deployment of eye-tracking studies to complement and verify cognitively grounded assumptions (Fryer 2019). Social and cognitive approaches to the study on context in translation studies are not mutually excluding. Indeed, in trying to elucidate the type and scope of contextual information that their readers would be able to bring to the text, Bible translators like the ones placed at the center of Hills’ study will have drawn on their cognitive insights. Their understanding and perception of the situational context in which their contemporary readership is embedded will have acted as a lens, refracting their interlingual and intercultural mediation of the actors and events narrated in the source text.

17.3.2 Static versus Dynamic Approaches For decades, translation scholars have tended to approach contextual variables as static parameters. From this standpoint, communicative encounters and the settings in which they are embedded tend to be theorized as stable environments “which the analyst can simply document and use to generate an analysis of events and behaviour” (Baker 2006b: 325). By focusing on ritualized conventions of culture and situational aspects of register as determinants of translational behavior, linguistic approaches to translation have prioritized

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

379

380

LUIS PÉREZ-GONZÁLEZ

the analysis of translation products, to the detriment of the process that led to their production. In the case of interpreter-mediated communication, for example, it was widely assumed by scholars endorsing static approaches that the role, status, or identities of the interactants could not shift in unforeseen ways during the encounters they were party to. From their perspective, interpreters could only aim to put foreign defendants “on an equal footing” with native speakers of the language spoken by the court, in line with professional conduct guidelines. Under these ethical codes of practice, interpreters should not attempt to explicate or elucidate elements which barristers or attorneys may have chosen to formulate in ambiguous, implicit, or unclear terms (Schweda-Nicholson and Martinsen 1997; Morris 1999). Under the influence of early conceptualizations of context in the field of sociopragmatics, this early body of research on courtroom interpreting sought to understand “how activities make up patterns of activity and how, through their interrelation, they produce and reproduce the activities they compose” (Sharrock and Anderson 1986: 293). However, in doing so, they took for granted the interactants’ tacit knowledge of and orientation to contextually appropriate behavior. Ultimately, this meant that early courtroom interpreting studies expected interpreters to show awareness of institutional conventions and be a “pane of glass, through which light passes without alteration or distortion” (Schweda-Nicholson 1994: 82). Notions like “frame” and “footing” (Goffman 1974, 1981) and Gumperz’ notion of “contextualization cues” (1992) served to “advance and develop a more complex and dynamic analysis of the ‘context’ of interaction” (Drew and Heritage 1992: 9). This dynamic perspective acknowledged the participants’ capacity to steer the unfolding of a speech encounter, based on the ongoing realignment of the speakers’ talk to ratify, contest, or ignore the trajectory of the interaction. This shift from context to contextualization in sociopragmatic scholarship drew a significant amount of scholarly attention to the study of “interaction-cum-interpretation.” This label encompasses “what is variously referred to in English as Community, Public Service, Liaison, Ad Hoc or Bilateral Interpreting” (Mason 1999: 147) and typically takes the form of “triadic exchanges” (Mason 2001) or “three-way interactions.” Mason and Stewart’s (2001) work on this interpreting mode is informed by applications of Goffman’s (1981) “participation framework” in interpreting studies (Wadensjö 1998; Roy 2000). In their capacity as gatekeepers between institutional representatives and ordinary people, interpreters emerge “as speaking agents who are critically engaged in the process of making meaningful utterances that elicit the intended response from, or have the intended effect upon, the hearer” (Davidson 2002: 1275). Central to the participatory role that interpreters claim for themselves is their perceived ownership of the meanings attaching to an interpreter’s output. In sites of linguistic and cultural difference – but even more so in sites of linguistic, cultural or situational inequality – the ability of each participant to

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

17 Context in Translation and Interpreting Studies

control/preserve his/her own identity will be affected in a number of complex and interesting ways. (Mason 2004)

Acknowledging the potential for ongoing recontextualization – often in the form of interpreter-initiated turns, to ensure that the trajectory of the interaction is not unduly influenced by the bilingual dimension of the encounter – brings into sharp relief the involvement of interpreters in the co-construction of interpersonal meaning. Indeed, interpreters’ (inadvertent) decisions have been shown to alter the interactional expectations that lawyers attempt to set by choosing a specific type of question. Pérez-González (2006) focuses on lawyers’ use of nonrestrictive relative clauses as a noncoercive examination strategy to invite the defendant to elaborate freely on a previous response. As illustrated in Example (1), the nonrestrictive relative clause deployed in the lawyer’s question (Turn 5) is rendered by the interpreter as a restrictive polar interrogative in the defendant’s language (Turn 6): (1) (Pérez-González 2006: 403) [Participants: (S1) lawyer; (S2) defendant; (I) interpreter] 1 ( S 1 ) La vez anterior ya le había dicho dónde estaba la llave de la puerta del> e:::h del garaje de Las Matas, ¿no? [The previous time [she] had already told you where the key of the door of the garage in Las Matas was, right?] 2 (I) But she had already told you in your last visit where the key of the garage in Las Matas was, didn’t she? 3 ( S 2 ) She had told the three of us. [She had told us three.] 4 (I) Nos lo había dicho a las tres. 5 ( S 1 ) La cual seguramente sería demasiado pesada para abrirla con una mano en caso de necesidad. [Which would probably be too heavy to open with one had if it were necessary for you to do so.] 6 (I) And do you think you would manage to open it with only one hand, if you had to? 7 ( S 2 ) I shouldn’t think so. [I don’t think so.] 8 (I) No creo. 9 ( S 1 ) Ah> Así que ya la había abierto antes, ¿no? [Oh> so you had opened it before, right?]

As Pérez-González’s analysis shows, the interpreter’s version does not represent a major interpersonal shift – in terms of coerciveness, for example. However, it “modifies the pragmatic force of the original elicitation by substantially restricting the responder’s degree of freedom to design their own contribution” (2006: 404). By eliciting a much more compact reply, the interpreters’ rendition disrupts the trajectory that the lawyer had envisaged for the parties’ negotiation of their interactional common ground; in

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

381

382

LUIS PÉREZ-GONZÁLEZ

particular, it alters the interactional frame against which the laywer (and the jury) assess the defendant’s eventual response. Amid the ongoing shift from context to contextualization, the current groundswell of interest in nonverbal behavior is yielding new insights into the range of multimodal practices used to co-construct interactional meaning in interpreter-mediated encounters. The analysis of doctor–patient interviews, for example, has revealed how interpreters often deploy mediation strategies involving a “co-ordinated manipulation of several semiotic resources” (Pasquandrea 2011: 477), whose relevance has been attested to also in other settings, including but not limited to business meetings (Bao-Rozée 2016) and group-work sessions involving deaf and hearing students (Slettebakk Berge 2018). In a similar development, the study of interactional recontextualization in immigration hearings (Mason 2012) and pedagogical settings (Davitti 2012) has revealed that all participants in triadic interaction make use of gaze shifts to manage turn-taking and signal (dis)engagement from other interactants. The combined deployment of gaze, body position, proxemics, object manipulation, and other gestures (Davitti 2015) are therefore being construed as instrumental to set up new participation frameworks at different points in the conversation (Davitti and Pasquandrea 2017: 105). While not engaging explicitly with context, a growing body of studies informed by the notion of paratext is heightening awareness of the extent to which translators’ interventions can bring about a recontextualization of written texts. Genette (1987, 1997) elaborated the concept of paratext outside the field of translation studies to designate any element that delivers authorial commentary on a text or influences its reception. His typology of paratextual elements ranges from material elements, including typesetting and binding, to more discursive elements, such as prefaces, reviews, or interviews. When developing this typology of paratexts, Genette had in mind the reception of texts within the same community where they had been published. However, translation scholars have extended its domain of application to explore how it can inform studies on the reception of texts traveling across languages and cultures. As far as their recontextualizing role is concerned, paratexts have been shown to influence the perception and image of individual authors or entire countries by target cultures. Watts (2005) and Kung (2010), for example, have shown how Western publishers produce promotional blurbs and choose covers for translated novels with a view to enhance the “otherness” of the source culture or accentuate stereotypical representations of foreignness that target readers subscribe to. Although translators are not directly involved in the choice of paratextual elements at play in these examples, other studies show that translators can be directly involved in making paratextual choices. Alkroud (2018), for example, addresses the political, linguistic, and cultural conflict between Arab hegemony and Berber counter-hegemony in Morocco and Algeria. Gaining independence from France allowed both countries to assert themselves as Arab nation-states, deploying a homogenization strategy to promote Arabization and eradicate ethnic and cultural diversity within their

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

17 Context in Translation and Interpreting Studies

territories. Marginalized within the newly created states, the Berbers disagreed on whether and how to negotiate the relationship between their culture and their own language (Tamazight), on the one hand, and the majority Arab identity, on the other hand. The range of perspectives participating in these debates is paratextually reflected in the publication of various translations of the Holy Quran into Tamazight between 1999 and 2017. Translators chose to signal their political stance through their choice of script for the Tamazight versions of the Quran. The Arabic script was chosen by the translator seeking to present the Berber identity as constitutive of the Arab culture and to foreground the religious significance of the text. Tifinagh, the Berber script, was favored by the translator wishing to celebrate the ancient indigenous Berber civilization and lay a claim to ethnic idiosyncrasy. By contrast, the Latin script was used by the translator aiming to anchor the new Berber identity in a context of modernity and technological innovation. As Batchelor (2018) suggests, the notion of paratext as a recontextualizing device has attracted particular attention from literary translation scholars – in particular, those working with texts that have been subjected to censorship, inscribe contentious historical narratives, or focus on gender struggles, as they often make the mediation of the translator more visible to the reader’s eye. Recent publications (Batchelor 2022, Qi 2021) are widening the scope of paratext-based research to the fields of audiovisual and news translation, including the participatory networks of text production and translation at the heart of digital media culture.

17.3.3 Neutral versus Power-Sensitive Approaches The third and final perspective contrasts neutral with power-sensitive approaches to the study of context. This final strand of the interface between translation and context is inextricably interlinked with my earlier critique of static and dynamic approaches to context. Most traditional conceptualizations of context in translation studies – whether they treat it as a static set of constraints on how communication should unfold or contemplate the possibility that translators and interpreters may activate dynamic processes of contextualization at any given point – “often grant [it] an inert neutrality” (Lindstrom 1992: 103, quoted in Baker 2006b: 329). Under both approaches, “context is a neutral field for the play of speech events, or is the cumulation of cognitive schemata that are cued to foreground past understanding” (Lindstrom, quoted in Baker 2006b: 329). But even when translators and interpreters deploy contextualizing strategies to address communication challenges following their professional judgment, rather than their audience’s or readership’s expectations, the participation framework that they set up may not be necessarily ratified by other actors in the spoken or written communicative encounter. One of the most important challenges to the centrality of neutral perspectives on context is Baker’s (2006a) application of socio-narrative theories in

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

383

384

LUIS PÉREZ-GONZÁLEZ

translation studies. Narrative theory challenges entrenched assumptions in the discipline that translators are bridges enabling dialogue between cultures. In Baker’s view, this stance overlooks the extent to which translation, particularly when it is carried out in sites of conflict, is enmeshed in the negotiation of political and moral issues. The widely held view among translation scholars that translators and interpreters are neutral enablers overlooks the fact that, in their everyday work, all translators are either upholding or resisting the narratives that circulate in their environment, including those “that create the intellectual and moral environment for violent conflict” (2006a: 2). Should they choose to exercise their agency by dissociating themselves from the narratives that the text subscribes to, a “re-narration” of the original text will obtain. Re-narrations are often motivated by activist or ethical considerations (Baker 2019) and involve challenging powerful political or institutional actors. Translators’ decisions to intervene and reconfigure the narratives inscribed in the original text, however, may also aim to pander to commercial interests for a variety of reasons. The latter form of re-narration can be illustrated by the canonization of “life writing” works by Central American women located in the “periphery of world literature” (Casanova 2004). Through their translation into English, these revolutionary writings featuring ethnic female protagonists have managed to access the curriculum in Anglo-American universities, the US literary market and, more widely, the world literature canon. De Inés Antón’s (2017) narrative analysis shows how the consecration of these revolutionary writings, translated from a “semi-peripheral” language (Spanish) to a “hyper-central” one (English), involves a significant reframing of the narratives underpinning the original Latin American works located at the periphery of world literature. The narratives of political radicalism and women’s agency inscribed in the source texts are re-narrated so that they can better resonate with mainstream narratives circulating in US universities and the publishing industry shaping world literature. For example, Gioconda Belli’s memoir El país bajo mi piel. Memorias de amor y guerra (2001) is a love story narrated against the backdrop of an exotic and romanticized revolution – i.e., the Sandinista movement that Belli was involved in. Among other distinctive features, Belli places female subjectivity at the center of her memoir, writing openly about aspects of women’s intimate lives, including menstruation, birth, eroticism, and sex. De Inés Antón’s analysis shows how the translation published for the US market (The Country under My Skin, 2002) re-narrates Belli’s narratives of motherhood and female subjectivity, aligning them with conventional public narratives of Western feminism. Of particular note is the fact that, while political and sexual radicalism were inextricably interwoven in the original text, the English translation disentangles these narratives and presents them as two separate radicalisms. On the whole, this instance of re-narration shows how recent Central American “literary discourse has been disempowered politically, while paradoxically, being empowered as a commodity by globalising trends”

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

17 Context in Translation and Interpreting Studies

(Arias 2007: 25). The fact that Belli herself was involved in the production of the English translation is of particular interest here. Within translation studies, neutral and power-sensitive perspectives on context have tended to foreground the extent to which institutional, political, and corporate commissioners have curtailed translators’ exercise of their professional judgment and focused on the impact that such restrictive or homogenizing codes of professional practice have had on the textual fabric of translated works (Baker and Pérez-González 2011). However, the advent of digitization and the emergence of virtual communities of participatory translation have mounted an important challenge to the centrality of equivalence in translation studies and, more specifically, to the premise that translations should ultimately aim to convey approximate linguistic representations of the verbally encoded meanings or communicative intentions inscribed in the source text. This shift is particularly evident in subtitling, one of the most salient forms of translation in digital culture. In this site of media content production and consumption, subtitling is increasingly carried out by activist, fandom, or crowdsourcing communities of like-minded individuals (PérezGonzález 2014), who become involved in subtitling to advance a cause, support idols, or assist in the development of commercial products. Significantly, in these networks of collaborative translation affectivity emerges as a powerful nonrepresentational variable. Rather than aiming to deliver an equivalent (understood as “accurate” or “faithful”) version of the source dialogue or narration, subtitles seek to intervene performatively in the reception of audiovisual content, “forging mutual recognition and social bonds between subtitlers and receivers” (Lee 2021: 6). In order to attend to their audiences’ expectations, these communities of participatory translation challenge the profit-oriented logic of the market economy and the legal apparatus that props up sanctions for copyright infringement – an offense that lies at the very heart of unsolicited translation. Although the media industry has by and large opted to take legal action against these networks, communities of participatory subtitling continue to challenge the subservience of translators to the dictates of economic and political power and corporate regulation during the era of patronage-driven translation.

17.4 Conclusion In keeping with the overarching theme of this volume, this chapter has surveyed key developments on the role that context has performed and will continue to play in driving translation studies forward, against the shift in scholarly discourses on context in the language sciences – away from static perspectives toward the socially constituted processes of negotiation captured by the notion of contextualization. Structured around three perspectives on the study of context, each of which has been conceptualized as a cline between a pair of polarities, this chapter has

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

385

386

LUIS PÉREZ-GONZÁLEZ

shown how the discipline of translation studies has been considerably enriched through its more intense and sustained engagement with dynamic and powersensitive understandings of context. As the transition from the twentiethcentury electronic culture to the twenty-first-century digital culture continues to unfold and translation studies becomes fully emancipated from linguistics, scholars in the discipline are increasingly aligning themselves with dynamic conceptualizations of context. While power asymmetries and their concomitant sociocultural constraints on the production and reception of translated texts cannot be overlooked, translations are increasingly construed and investigated as sites of negotiation between actors furthering their own agendas and capitalizing on more democratic access to technologies of text production and distribution. Against this backdrop, the emergence of ever more complex forms of translation authorship and agency is exposing the limitations of traditional disciplinary discourses on translator agency and translation ethics. More importantly, this development is also raising new questions about how translation studies can engage with fast-evolving forms of digital literacy in the predominantly multimodal textual genres that underpin processes of political deliberation and affective sociality in the information and knowledge society.

References Alkroud, E. (2018). Renarrating the Berbers in three Amazigh translations of the Holy Quran: Paratextual and framing strategies. Unpublished doctoral thesis, University of Manchester. Arias, A. (2007). Taking Their Word: Literature and the Signs of Central America, Minneapolis: University of Minnesota Press. Baker, M. (2003). Corpus linguistics and translation studies: Implications and applications. In M. Baker, G. Francis, and E. Tognini–Bonelli (eds.), Text and Technology (pp. 233–250). Amsterdam: John Benjamins. Baker, M. (2005). Linguistic models and methods in the study of translation. In H. Kittel, A. P. Frank, N. Greiner, T. Hermans, W. Koller, J. Lambert, and F. Paul (eds.), Übersetzung, Translation, Traduction (pp. 285–294). Berlin/New York: Walter de Gruyter. Baker, M. (2006a). Translation and Conflict: A Narrative Account. London/New York: Routledge. Baker, M. (2006b). Contextualization in translator- and interpreter-mediated events. Journal of Pragmatics, 38, 321–337. Baker, M. (2010). Narratives of terrorism and security: “Accurate” translations, suspicious frames. Critical Studies on Terrorism, 3(3), 347–364. Baker, M. (2019). Audiovisual translation and activism. In L. Pérez-González (ed.), Routledge Handbook of Audiovisual Translation (pp. 453–467). London/New York: Routledge. Baker, M., and Pérez-González, L. (2011). Translation and interpreting. In J. Simpson (ed.), Routledge Handbook of Applied Linguistics 1st ed. (pp. 39–52). London/ New York: Routledge.

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

17 Context in Translation and Interpreting Studies

Bao-Rozée, J. (2016). Using multimodal analysis to investigate the role of the interpreter. Unpublished doctoral thesis, University of Stirling. Bassi, S. (2015). Italy’s Salman Rushdie: The renarration of “Roberto Saviano” in English for the post-9/11 cultural market. Translation Studies, 8, 48–62. Batchelor, K. (2018). Translation and Paratexts. London/New York: Routledge. Batchelor, K. (2022). Translation, media and paratexts. In E. Bielsa (ed.), Routledge Handbook of Translation and Media (pp. 122–135). London/New York: Routledge. Bell, R. (1991). Translation and Translating: Theory and Practice. London/New York: Longman. Belli, G. (2001). El país bajo mi piel: Memorias de amor y guerra. New York: Random House. Belli, G. (2002). The Country under My Skin: A Memoir of Love and War, trans. K. Cordero and G. Belli. London: Bloomsbury. Bernardini, S., Ferraresi, A., Russo, M., Collard, C., and Defrancq, B. (2018). Building Interpreting and Intermodal Corpora. In M. Russo, C. Bendazzoli, and B. Defrancq (eds.), Making Way in Corpus-Based Interpreting Studies (pp. 21–42). Singapore: Springer. Bloomfield, S., trans. (1829). The History of Thucydides, Newly Translated into English. London: Longman. Boéri, J. (2008). A narrative account of the Babels vs. Naumann controversy: Competing perspectives on activism in conference interpreting. The Translator, 14 (1), 21–50. Calzada-Pérez, M. (2007). Transitivity in Translating: The Interdependence of Texture and Context. Bern: Peter Lang. Casanova, P. (2004). The World Republic of Letters. Cambridge, MA: Harvard University Press. Chesterman, A. (2000). A causal model for translation studies. In M. Olohan (ed.), Intercultural Faultlines: Research Models in Translation Studies (pp. 12–27). Manchester: St. Jerome. Chesterman, A. (2006). Questions in the sociology of translation. In J. Ferreira Duarte, A. Assis Rosa, and T. Seruya (eds.), Translation Studies at the Interface of Disciplines (pp. 9–27). Amsterdam/Philadelphia: John Benjamins. Davidson, B. (2002). A model for the construction of conversational common ground in interpreted discourse. Journal of Pragmatics, 34, 1273–1300. Davitti, E. (2012). Dialogue interpreting as inter-cultural mediation: Integrating talk and gaze in the analysis of mediated parent–teacher meetings. Unpublished doctoral thesis, University of Manchester. Davitti, E. (2015). Gaze. In F. Pöchhacker (ed.), Routledge Encyclopedia of Interpreting Studies (pp. 168–169). London/New York: Routledge. Davitti, E., and Pasquandrea, S. (2017). Embodied participation: What multimodal analysis can tell us about interpreter-mediated encounters in pedagogical settings, Journal of Pragmatics, 107, 105–128. De Inés Antón, T. (2017). Translating Central American life writing for the anglophone market: A socio-narrative study of women’s agency and political radicalism in the original and translated works of Claribel Alegría, Gioconda Belli and Rigoberta Menchú, Unpublished doctoral thesis, University of Manchester. Desilla, L. (2012). Implicatures in film: Construal and functions in Bridget Jones romantic comedies, Journal of Pragmatics, 44(1), 30–35.

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

387

388

LUIS PÉREZ-GONZÁLEZ

Desilla, L. (2014). Reading between the lines, seeing beyond the images: An empirical study on the comprehension of implicit film dialogue meaning across cultures. The Translator, 20(2), 194–214. Dirdal, H. (2014). Individual variation between translators in the use of clause building and clause reduction. Corpus-Based Studies in Contrastive Linguistics, 6(1), 119–142. Diriker, E. (2004). De-/Re-contextualising Simultaneous Interpreting: Interpreters in the Ivory Tower? Amsterdam/Philadelphia: John Benjamins. Drew, P. & Heritage, J. (1992). Analysing Talk at Work: An Introduction. In P. Drew and J. Heritage, eds., Talk at Work, Cambridge: Cambridge University Press, pp. 3–65. Ferraresi, A., Bernardini, S., Picci, G., and Baroni, M. (2010). Web corpora for bilingual lexicography: A pilot study of English–French collocation extraction and translation. In R. Xiao (ed.), Using Corpora in Contrastive and Translation Studies (pp. 337–359). Newcastle: Cambridge Scholars Publishing. Fonseca, N. B. (2015). Directionality in translation: Investigating prototypical patterns in editing procedures. Translation and Interpreting, 7(1), 111–125. Fryer, L. (2019). Psycholinguistics and perception in audiovisual translation. In L. Pérez-González (ed.), Routledge Handbook of Audiovisual Translation. (pp. 209– 224). London/New York: Routledge. Gambier, Y. (2006). Pour une socio-traduction. In J. Ferreira Duarte, A. Assis Rosa, and T. Seruya (eds.), Translation Studies at the Interface of Disciplines (pp. 29–42). Amsterdam/Philadelphia: John Benjamins. Genette, G. (1987). Seuils. Paris: Editions du Seuil. Genette, G. (1997). Paratexts: Thresholds of Interpretation, trans. J. E. Lewin. Cambridge: Cambridge University Press. Goffman, E. (1974). Frame Analysis. New York: Harper Row. Goffman, E. (1981). Forms of Talk. Philadelphia: University of Pennsylvania Press. Gumperz, J. 1992). Contextualization and understanding. In A. Duranti and C. Goodwin (eds.), Rethinking Context: Language as an Interactive Phenomenon (pp. 229–253). Cambridge: Cambridge University Press. Gutt, E. A. (2000). Translation and Relevance: Cognition and Context, 2nd ed. Manchester: St. Jerome Publishing. Halverson, S. (2017). Developing a cognitive semantic model: Magnetism, gravitational pull, and questions of data and method. In G. de Sutter, M.-A. Lefer, and I. Delaere (eds.), Empirical Translation Studies: New Methods and Theoretical Traditions (pp. 9–45). Berlin: Mouton de Gruyter. Harding, S.-A. (2012). Beslan: Six Stories of the Siege. Manchester: University of Manchester Press. Hatim, B., and Mason, I. (1990). Discourse and the Translator. London: Longman. Hatim, B., and Mason, I. (1997). The Translator as Communicator. London/New York: Routledge. Hill, H. (2007). The Bible at Cultural Crossroads: From Translation to Communication, Manchester: St. Jerome. Hermans, T., ed. (1985). The Manipulation of Literature. Beckenham: Croom Helm. Hermans, T. (1996). The translator’s voice in translated narrative. Target, 8(1), 23–48. House, J. (1977/1981). A Model for Translation Quality Assessment. Tübingen: Narr. House, J. (1997). Translation Quality Assessment: A Model Revisited, Tübingen: Narr.

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

17 Context in Translation and Interpreting Studies

Hvelplund, K. T. (2017). Eye tracking in translation process research. In J. W. Schwieter and A. Ferreira (eds.), The Handbook of Translation and Cognition (pp. 248–264). Hoboken, NJ: Wiley-Blackwell. Hymes, D. H. (1964). Toward ethnographies of communicative events. American Anthropologist, 66, 12–25; repr. in P. Giglioli, ed. (1972). Language and Social Context (pp. 21–44). Harmondsworth: Penguin. Jääskeläinen, R. (1999). Tapping the Process: An Explorative Study on the Cognitive and Affective Factors Involved in Translating. University of Joensuu Publications in the Humanities 22, Joensuu: University of Joensuu. Jakobsen, A. L. (2006). Research methods in translation–translog. In P. Kirk, H. Sullivan, and E. Lindgren (eds.), Computer Keystroke Logging: Methods and Applications (pp. 95–105). Oxford: Elsevier. Jones, H. (2019). Shifting characterizations of the “common people” in Modern English retranslations of Thucydides’ History of the Peloponnesian War: A corpus-based analysis, Palgrave Communications, 5(135). DOI:10.1057/s41599-019-0348-0. Kang, J.-H. (2007). Recontextualization of news discourse: A case study of translation of news discourse on North Korea. The Translator, 13(2), 219–42. Kenny, D. (1998). Equivalence. In M. Baker (ed.), Routledge Encyclopedia of Translation Studies (pp. 96–99). London/New York: Routledge. Kenny, D. (2009). Corpora. In M. Baker and G. Saldanha (eds.), Routledge Encyclopedia of Translation Studies, 2nd ed. (pp. 59–62). London/New York: Routledge. Koskinen, K. (2008). Translating Institutions: An Ethnographic Study of EU Translation, Manchester: St. Jerome. Kruger, J.-L. (2012). Making meaning in AVT: Eye tracking and viewer construction of narrative. Perspectives, 20(1), 67–86. Kübler, N., and Volanschi, A. (2012). Semantic prosody and specialized translation, or how a lexicogrammatical theory of language can help with specialized translation. In A. Boulton, S. Carter-Thomas, and E. Rowley-Jolivet (eds.), Corpus-Informed Research and Learning in ESP: Issues and Applications (pp. 105–135). Amsterdam: John Benjamins. Kung, S.-W. (2010). Network and cooperation in translating Taiwanese literature into English. In A. Fawcett, K. L. Guadarrama García, and R. Hyde Parker (eds.), Translation: Theory and Practice in Dialogue (pp. 164–180). London: Continuum. Lacruz, I. (2017). Cognitive effort in translation, editing, and post-editing. In J. W. Schwieter and A. Ferreira (eds.), The Handbook of Translation and Cognition (pp. 248–264). Hoboken, NJ: Wiley Blackwell. Lattimore, S., trans. (1998). Thucydides: The Peloponnesian War. Indianapolis: Hackett Publishing Company. Laviosa, S. (2002). Corpus-Based Translation Studies: Theory, Findings, Applications. Amsterdam/New York: Rodopi. Lee, S. (2021). Translating YouTube vlogs for a global audience: Innovative subtitling and community building. International Journal of Cultural Studies. DOI:10.1177/ 1367877920979717. Leppihalme, R. (1997). Culture Bumps: An Empirical Approach to the Translation of Allusions. London: Multilingual Matters. Lindstrom, L. (1992). Context contests: Debatable truth statements on Tanna (Vanuatu). In A. Duranti and C. Goodwin (eds.), Rethinking Context: Language as an Interactive Phenomenon (pp. 101–124). Cambridge: Cambridge University Press.

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

389

390

LUIS PÉREZ-GONZÁLEZ

López-Rodríguez, C. I. (2016). Using corpora in scientific and technical translation training: Resources to identify conventionality and promote creativity. Cadernos de Tradução, 36(1), 88–120. Lörscher, W. (1991). Translation Performance, Translation Process, and Translation Strategies: A Psycholinguistic Investigation. Tübingen: Gunter Narr. Marín Lacarta, M. (2020). Research methodologies, translation. In M. Baker and G. Saldanha (eds.), Routledge Encyclopedia of Translation Studies, 3rd ed. (pp. 479– 484). London/New York: Routledge. Marinetti, C. (2011). Cultural Approaches. In Y. Gambier and L. van Dorsaaler (eds.), Handbook of Translation Studies, Vol. II (pp. 26–30). Amsterdam/Philadelphia: John Benjamins. Mason, I. (1999). Introduction. The Translator, 5(2), 147–160. Mason, I. (2001). Introduction. In I. Mason, ed., Triadic Exchanges: Studies in Dialogue Interpreting (pp. i–vi). Manchester: St. Jerome. Mason, I. (2004). Projected and perceived identities in dialogue Interpreting, Paper delivered at the I IATIS Conference: Translation and the Construction of Identity, Seoul, August 12–14, 2004. Mason, I. (2006). On mutual accessibility of contextual assumptions in dialogue interpreting, Journal of Pragmatics, 38, 359–373. Mason, I. (2012). Gaze, positioning and identity in interpreter-mediated dialogues. In C. Baraldi and L. Gavioli (eds.), Coordinating Participation in Dialogue Interpreting (pp. 177‒199). Amsterdam/Philadelphia: John Benjamins. Mason, I., and Stewart, M. (2001). Interactional pragmatics, face and the dialogue interpreter. In I. Mason (ed.), Triadic Exchanges: Studies in Dialogue Interpreting (pp. 51–70). Manchester: St. Jerome. McDonough Dolmaya, J. (2011). A window into the profession: What translation blogs have to offer translation studies. The Translator, 17(1), 77–104. McDonough Dolmaya, J. (2012). Analysing the crowdsourcing model and its impact on public perceptions of translation, The Translator, 18(2), 167–191. Morris, R. (1999). The gum syndrome: Predicaments in court interpreting, Forensic Linguistics, 6(1), 6–29. Munday, J. (2009). Issues in translation studies. In J. Munday (ed.), Routledge Companion to Translation Studies (pp. 1–19). London/New York: Routledge. Nida, E. A. (1964). Toward a Science of Translating, with Special Reference to Principles and Procedures Involved in Bible Translating. Leiden: E. J. Brill. Nida, E. A., and Taber, C. (1969). The Theory and Practice of Translation. Leiden: E. J. Brill. Nord, C. (1991). Text Analysis in Translation. Amsterdam/Atlanta: Rodopi. Olohan, M. (2004). Introducing Corpora in Translation Studies. London: Routledge. Orero, P. (2008). Three different receptions of the same film. European Journal of English Studies, 12, 179–193. Orero, P., and Vilaró, A. (2012). Eye tracking analysis of minor details in films for audio Description. MonTI, 4, 295–319. Pasquandrea, S. (2011). Managing multiple actions through multimodality: Doctors’ involvement in interpreter-mediated interactions, Language in Society, 40(4), 455-481. Pérez-González, L. (2006). Interpreting strategic recontextualization cues in the courtroom: Corpus-based insights into the pragmatic force of non-restrictive relative clauses. Journal of Pragmatics, 38, 390–417.

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

17 Context in Translation and Interpreting Studies Pérez-González, L., and Susam-Saraeva, Ş. (2012). Non-professionals translating and interpreting: Theoretical and methodological perspectives. The Translator, 18(2), 149–165. Pérez-González, L. (2014). Audiovisual Translation: Theories, Methods and Issues. London/New York: Routledge. Qi, L. (2021). Source text readers as censors in the digital age: A paratextual examination of the English translation of Wuhan Diary. Perspectives. DOI:10.1080/ 0907676X.2021.1939741. Rabin, C. (1958). The linguistics of translation. In A. H. Smith (ed.), Aspects of Translation: Studies in Communication (Vol. II, pp. 123–145). London: Secker & Warburg. Reiss, K. (1971). Möglichkeiten und Grenzen der Übersetzungskritik. Munich: Hueber. Rodríguez-Inés, P. (2017). Corpus-based insights into cognition. In J. W. Schwieter and A. Ferreira (eds.), The Handbook of Translation and Cognition (pp. 248–264). Hoboken, NJ: Wiley-Blackwell. Roy, C. (2000). Interpreting as a Discourse Process. Oxford: Oxford University Press. Saldanha, G. (2009). Linguistic approaches. In M. Baker and G. Saldanha (eds.), Routledge Encyclopedia of Translation Studies, 2nd ed. (pp. 148–152). London/ New York: Routledge. Saldanha, G. (2014). Style in, and of, translation. In S. Berman and C. Porter (eds.), A Companion to Translation Studies (pp. 95–106). Chichester: Wiley-Blackwell. Sapiro, G., ed. (2014). Translation and globalization. Special issue of Bibliodiversity, Journal of Publishing in Globalization, February. Schäffner, C. (2003). Third ways and centres: Ideological unity or difference? In M. Calzada Pérez (ed.), Apropos of Ideology: Translation Studies on Ideology. Ideologies in Translation Studies (pp. 23–41). London/New York: Routledge. Schäffner, C. (2012). Unknown agents in translated political discourse. Target, 24(1), 103–125. Schweda Nicholson, N. (1994). Professional ethics for court and community interpreters. In D. L. Hammond (ed.), Professional Issues for Translators and Interpreters (pp. 79–97). Amsterdam/Philadelphia: John Benjamins. Schweda Nicholson, N., and Martinsen, B. (1997). Court Interpretation in Denmark. In S. E. Carr, R. Roberts, A. Dufour, and D. Steyn (eds.), The Critical Link: Interpreters in the Community (pp. 259–270). Amsterdam/Philadelphia: John Benjamins. Sharrock, W., and Anderson, B. (1986). The definition of alternatives: Some sources of confusion in interdisciplinary discussion. In G. Button and J. Lee (eds.), Talk and Social Organization (pp. 290–321). Clevendon: Multilingual Matters. Slettebakk Berge, S. (2018). How sign language interpreters use multimodal actions to coordinate turn-taking in group work between deaf and hearing upper secondary school students. Interpreting, 20(1), 96–125. Soffritti, M. (2019). Multimodal corpora and audiovisual translation studies. In L. Pérez-González (ed.), Routledge Handbook of Audiovisual Translation (pp. 334– 349). London/New York: Routledge. Sperber, D., and Wilson, D. (1995). Relevance: Communication and Cognition, 2nd ed. Oxford: Blackwell.

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

391

392

LUIS PÉREZ-GONZÁLEZ

Szymor, N. (2017). Translation: Universals or cognition? A usage-based perspective. Target, 30(1), 53–86. Tagnin, S. E., and Teixeira, E. D. (2012). Translator-oriented, corpus-driven technical glossaries. Corpora, 7(1), 51–67. Tesseur, W. (2014). Transformation through translation: Translation policies at Amnesty International. Unpublished doctoral thesis, Aston University. Torikai, K. (2009). Voices of the Invisible Presence: Diplomatic Interpreters in Post-World War II Japan. Amsterdam/Philadelphia: John Benjamins. Tymoczko, M. (2013). Translation theory. In C. Chapelle (ed.), Encyclopedia of Applied Linguistics (pp. 5928–5937). Chichester: Wiley Blackwell. Van Dijk, T. A. (2001a). Text and context revisited. In Proceedings of the First Seoul International Conference on Discourse and Cognitive Linguistics: Perspectives for the 21st Century, Seoul, June 9–10. Seoul: The Discourse and Cognitive Linguistics Society of Korea. Van Dijk, T. A. (2001b). Discourse, ideology and context. Folia Linguistica, 35, 11–40. Venuti, L. (1995). The Translator’s Invisibility. London/New York: Routledge. Vermeer, H. J. (1989). Skopos and commission in translational action. In A. Chesterman (ed.), Readings in Translation Theory (pp. 173–187). Helsinki: Oy Finn Lectura Ab; repr. in L. Venuti, ed. (2000), The Translation Studies Reader (pp. 221–232). London/New York: Routledge. Wadensjö, C. (1998). Interpreting as Interaction. London/New York: Longman. Watts, R. (2005). Packaging Post/Coloniality: The Manufacture of Literary Identity in the Francophone World. Lanham: Lexington Books. Wilson, D., and Sperber, D. (2004). Relevance theory. In L. R. Horn and G. Ward (eds.), The Handbook of Pragmatics (pp. 607–632). Oxford: Blackwell. Yu, C. (2017). Collaborative translation in online communities of practice: An ethnographic study of Yeeyan. Unpublished doctoral thesis, Hong Kong Baptist University. Zwischenberger, C. (2020). Research methodologies, interpreting. In M. Baker and G. Saldanha (eds.), Routledge Encyclopedia of Translation Studies, 3rd ed. (pp. 474–479). London/New York: Routledge.

https://doi.org/10.1017/9781108989275.018 Published online by Cambridge University Press

18 The Role of Context in Clinical Linguistics Louise Cummings

18.1 Introduction Clinical linguistics is the branch of linguistics that characterizes and attempts to explain the many ways in which language may be impaired. It is a foundational discipline of study for those who seek to practice the profession of speech-language pathology (in the US) or speech and language therapy (in the UK). Yet, not much is known or written about the different ways in which clinical linguistics, and the language disorders it studies, intersects with context, a core concept in the study of language. In this chapter, I plan to put clinical linguistic applications of context center stage. In an important sense, this is long overdue. While disciplines like pragmatics and sociolinguistics have always explicitly examined context, the concept has been somewhat hidden in the background of work in clinical linguistics. And yet no language disorder can be adequately characterized, assessed, or treated apart from a range of contexts. The language-impaired child who talks to a parent at home encounters contextual variables in this setting that are quite different from those that he/she must navigate when communicating with a teacher at school or with a friend in the playground. Context weaves its way through each of these spoken interactions, making some aspects of communication challenging while facilitating other aspects. The clinical linguist can no more afford to shun context than the sociolinguist can afford to overlook the influence of social class, gender, and age on language. This will be my starting point in the discussion to follow. The chapter will unfold along the following lines. In Section 18.2, we examine the scope of clinical linguistics and consider its relationship to the related profession of speech-language pathology. This section will also examine language disorders, a prominent category of communication disorders, as well as other main categories of communication disorder that can be found in children and adults.

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

394

LOUISE CUMMINGS

In Section 18.3, the concept of context is explored as it relates to clinical linguistics and speech-language pathology. Five themes are introduced to facilitate the discussion. Children and adults with language disorder can make nonnormative use of context, with implications for both their interpretation and their expression of language. We examine some populations of language-disordered children and adults where this is most evident (theme 1). Some individuals with language disorder can usefully exploit context to compensate for their impaired language skills, while for others context overwhelms their language processing abilities, leading to characteristic anomalies in their use of language. We consider these opposing responses to context in people with language disorder (theme 2). The language disorders clinic is the context in which language skills are most often evaluated and treated. And yet this environment does not represent how children and adults use language across educational, social, and work contexts in their everyday lives (I refer to this as the “clinic paradox”). We consider the limitations inherent in using a clinical setting to understand something as dynamic as language disorder (theme 3). To address the clinic paradox, speech-language pathologists must consider the ecological validity of the instruments they use to assess language skills (theme 4) and how intervention can be tailored to contexts that are salient in the lives of clients (theme 5). The chapter concludes in Section 18.4 with some reflections on how clinical linguists and speechlanguage pathologists may integrate context more fully into their work.

18.2 The Scope of Clinical Linguistics Many of the “prefixed” disciplines in linguistics (e.g., sociolinguistics, psycholinguistics, and neurolinguistics) have arisen as it became apparent to linguists that new terms and concepts were needed to characterize certain linguistic phenomena. The same is true of clinical linguistics. It is difficult to mark the exact starting point of any linguistic discipline. But we can do no better than look to the work of British linguist David Crystal for the first book bearing the title of this new field of linguistic study (Cummings 2014a). Published in 1981, Crystal’s book defined clinical linguistics as “the application of linguistic science to the study of communication disability, as encountered in clinical situations” (1981: 1). Crystal’s definition made it clear that clinical linguistics was not a purely academic discipline, and that the purpose of the field was to understand “communication disability . . . in clinical situations.” From this early starting point, the connection between clinical linguistics and individuals with communication disability in speech therapy clinics was explicitly forged. This connection remains as strong today. But before we examine this relationship in more detail, it is worth considering what else Crystal has said about clinical linguistics. In a later definition, he teases apart the linguistic aspects of the discipline. Clinical linguistics is “the application of linguistic theories and methods to the analysis of disorders of spoken, written, or signed language”

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

18 The Role of Context in Clinical Linguistics

(Crystal 1997: 418). This definition helpfully moves us beyond the common misconception that all language disorders involve spoken language. They do not. When an adult develops aphasia (an acquired language disorder) as a result of a stroke or a cerebrovascular accident (to give it its medical term), the ability to produce and comprehend written language is as likely to be disrupted as the ability to produce and comprehend spoken language. If this same adult is a user of a signed language like American Sign Language prior to his or her stroke, the ability to produce and comprehend manual signs may also be impaired. Language disorders compromise the understanding (reception) and production (expression) of linguistic symbols in all modalities, spoken, written, and signed. Let us examine some of these disorders in more detail by considering three distinctions used by clinical linguists to classify language disorders. The first distinction concerns the difference between a developmental and an acquired language disorder. For most children, the acquisition of language in the developmental period is a process that requires no effort or special instruction. However, for a sizable minority of children, first language acquisition is anything but straightforward or effortless, resulting in a developmental language disorder. These children may be born with a genetic syndrome and have intellectual disability that makes it difficult or, in severe cases, impossible to acquire language. Alternatively, children may be born with an anatomical defect of their articulators (e.g., cleft lip and palate) that results in deviant speech sound substitutions such as the replacement of oral stops with a glottal stop (e.g., cat /kæt/ → [ʔæt]). The collapse of distinction between the oral plosives – they cannot be distinguished when they are all realized as a glottal stop – leads to a phonological disorder, as important contrasts in the child’s sound system are lost. As a result, the child’s language disorder is developmental in nature as the acquisition of phonology is compromised. For other children and adults, they can acquire language along normal lines, only for an illness, injury, or disease to lead to its impairment. This gives rise to an acquired language disorder. For example, the adult with previously normal language skills who has a stroke or develops a brain tumor and cannot form and understand questions has an acquired language disorder. Similarly, a 16-year-old child may sustain a head injury in a road traffic accident and lose the ability to produce well-formed, grammatical utterances. The impairment of grammar in this case, too, is an acquired language disorder even though the impairment occurs in a child. This is because the distinction between a developmental and an acquired language disorder rests ultimately on how much language has been acquired by the time an injury or illness occurs. In a 16-year-old child, many aspects of language acquisition, including the acquisition of phonology and grammar, are complete. In the same way that children can have acquired language disorders, adults may have developmental language disorders. For example, an individual who is born with a genetic syndrome like Down syndrome will continue to experience into adulthood the same impairment of language skills that arose in the developmental period as

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

395

396

LOUISE CUMMINGS

a result of intellectual disability. This individual has a developmental language disorder even as an adult. Clinical linguists also draw a distinction between an expressive and a receptive language disorder. In an expressive language disorder, the ability to formulate a sentence or an utterance is compromised. For example, the child with intellectual disability who cannot form sentences containing a relative clause or a passive voice construction has an expressive language disorder. Similarly, the adult with Alzheimer’s dementia who cannot retrieve the words needed to produce a spoken utterance also has an expressive language disorder. Expressive language disorders often occur alongside deficits in receptive language. In a receptive language disorder, the ability to comprehend or understand language is compromised. For example, a child with intellectual disability may also be unable to understand utterances that contain a relative clause or a passive voice construction. Similarly, the comprehension of words and their meanings may be disrupted in an adult who sustains a stroke. It is important to recognize that the impairment of language comprehension is not related to a sensory deficit (e.g., hearing loss) – the child or adult with an expressive language disorder can adequately hear, for example, the spoken utterance but cannot decode its linguistic structure and interpret its meaning. In the same way, an impairment of expressive language is not related to a loss of movement of the speech articulators, although a speech disorder may also be present. Rather, it results from an inability to select and encode the linguistic structures that are required to express an utterance. A third distinction that is acknowledged by clinical linguists is the distinction between a speech disorder and a language disorder. Speech and language are frequently treated as one and the same thing. However, for clinicians and clinical linguists, they represent quite separate components of communication. Speech production is a complex motor activity that requires the integration of several biophysical processes such as nerve impulse transmission and muscle contraction. If any part of this complex, highly integrated motor system is disrupted, a speaker can have a speech disorder such as dysarthria or apraxia of speech. For example, a child may be exposed to a viral infection in utero that damages the development of the motor centers in the brain, causing a condition called cerebral palsy. As well as impaired voluntary movement of the limbs, head, and torso, a child with cerebral palsy may experience impaired movement of the speech articulators, resulting in developmental dysarthria. The impairment of speech production may be mild, moderate, or severe, leading to varying levels of unintelligibility. But even in the child with cerebral palsy who has severe dysarthria, possibly necessitating the use of an alternative communication system, expressive and receptive language skills may nonetheless be intact. A quite different situation arises in the adult with non-fluent aphasia who may struggle to produce even one- or two-word utterances. For this adult, the problem with the production of spoken utterances relates solely to difficulty encoding language and is not in any way related to the production of speech. This adult has a language disorder in the absence of a speech disorder.

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

18 The Role of Context in Clinical Linguistics

It can be seen from the above discussion that no account of language disorders is possible in the absence of the children and adults who have these disorders. To understand their impairments of language, we must know something of the developmental disorders and other conditions that are the aetiology of these impairments. This requires us to look beyond language and engage with medical and scientific disciplines that, like clinical linguistics, are also foundational to speech-language pathology. And indeed, what we find is that clinical linguistics is one of several disciplines that speechlanguage pathologists must study in order to assess and treat children and adults with language disorders (Cummings 2018). But speech-language pathology extends more widely than clinical linguistics in another important respect. For the clinicians who practice this profession are concerned with the assessment and treatment of all communication disorders, and not just disorders of language. Communication disorders include fluency disorders like stuttering, voice disorders like alaryngeal communication following laryngectomy (surgical removal of the larynx), and hearing disorders like conductive hearing loss. Communication disorders also include speech disorders such as dysarthria, which, as we have already seen in Section 18.2, are properly set apart from language disorders (see Figure 18.1). But even the broad grouping of communication disorders does not exhaust the scope of practice of speech-language pathology. For some years, the assessment and treatment of swallowing disorders (dysphagia) in children and adults have also been part of the professional remit of speech-language pathologists. It emerges that speech-language pathology

Voice disorders

Language disorders

Speech disorders COMMUNICATION DISORDERS

Fluency disorders

Hearing disorders

Figure 18.1 Communication disorders

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

397

398

LOUISE CUMMINGS

is a broad, interdisciplinary area of clinical practice that draws on the concepts and theories of clinical linguistics in much the same way that it draws on disciplines like neurology, psychology, and anatomy. The mutual dependence of speech-language pathology and clinical linguistics will be integral to the discussion of context in the next section.

18.3 Context in Clinical Linguistics It should be apparent by now to readers that when we discuss clinical linguistics, we must also consider the related profession of speech-language pathology. This will provide the backdrop for the discussion of the present section. Speech-language pathology weaves its way through each of the five themes that will be used to address context in clinical linguistics in this section. The first theme concerns what is known about how children and adults with language disorders process context. This processing deviates markedly from normative uses of context, with terms such as “context insensitivity” used to capture clients’ difficulties. The second theme examines how for some children and adults with language disorder context can be a barrier to communication, while for others it can serve to facilitate communication. In the latter use, context can be a powerful strategy in the compensation of impaired receptive and expressive language skills. The third theme examines the role of context in the language disorders clinic. Clinics in speechlanguage pathology present something of a paradox. They aim to equip clients with language skills that will serve them in their daily lives, but they are constrained to do so from within a setting that is quite unlike most natural language contexts. We examine how clinicians attempt to resolve this paradox. The fourth theme examines context in relation to language assessment and considers how clinicians have increasingly moved beyond word- and sentence-testing formats to adopt assessments with greater ecological validity. The fifth theme considers context from the point of view of language intervention and considers how this concept is embedded in a client’s therapy goals, among other aspects of intervention.

18.3.1 Theme 1: Nonnormative Use of Context in Language Disorder To the extent that there is a normative use of context, it may be characterized in the following terms. Context can be overwhelming for our cognitive and sensory resources. The reason it does not actually overwhelm our mental resources, even though it has the potential to do so, is that we are particularly adept at attending to certain aspects of context and suppressing other aspects. Our skill in navigating context often only becomes apparent when that ability momentarily breaks down and we misinterpret a written or spoken utterance.

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

18 The Role of Context in Clinical Linguistics

This happened to me recently when reading the following headline in an online news article: Husband of teacher accused of having sex with pupil says they were trying for a baby

When I read this headline, I understood – or as it turned out, misunderstood – that it was the husband who was accused of having sex with the pupil. What led me to interpret the past participle clause as relating to the noun phrase husband instead of teacher? Maybe it was my background belief – or some might say my biased thinking – that men are more likely than women to engage in an illicit sexual relationship with a school pupil. I had possibly allowed this belief to have greater sway in my interpretation of the utterance than was strictly warranted. I had maybe also not attributed enough significance to the fact that the woman in this scenario was a teacher and, as such, was more likely than her husband to have contact with the pupil in question. If I had done so, I may have avoided my error of interpretation altogether. Whatever was the source of my error, it demonstrates very clearly the fine line that we tread with context during the interpretation of any utterance. The question now is whether children and adults with language disorder can tread this same line. The difficulties that many children and adults have with context can be characterized in one of two ways. Firstly, children and adults may be unable to inhibit or suppress aspects of context so that a part of context that might not ordinarily be prominent comes to dominate interpretation. This is a more exaggerated form of the behavior that I displayed when one of my background beliefs became unduly salient, leading to an erroneous interpretation of a news headline. This lack of context inhibition is supported by clinical studies. For example, Titone et al. (2000) examined use of context during a semantic priming task in eighteen patients with schizophrenia and twenty-four nonpsychiatric controls. When sentences moderately biased the subordinate (less common or nondominant) meanings of words (e.g., the animal enclosure meaning of pen), controls showed priming only of subordinate target meanings, while patients with schizophrenia showed priming of subordinate and dominant target meanings (in the case of pen, the writing implement meaning of the word). In other words, while controls were able to inhibit dominant target meanings, a similar inhibitory effect was not observed in patients with schizophrenia. Wiener et al. (2004) reported an impairment in inhibition at the lexical-semantic level of language processing in five individuals with Wernicke’s aphasia. This impairment correlated with significant reductions in auditory comprehension, revealing that a failure to inhibit automatically evoked, distracting stimuli was integral to the comprehension deficits of these aphasic individuals. What might a failure to inhibit aspects of context look like in children and adults with language disorder? In terms of receptive language, we might expect to see a predominance of the dominant meanings of words during

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

399

400

LOUISE CUMMINGS

interpretation even when these meanings should be inhibited. This could see idioms, metaphors, and other figurative forms of language interpreted literally, as the literal meanings of the words in these expressions are typically their dominant meanings (Cummings, 2009, 2014b). For example, the English expression to kick the bucket means ‘to die or to pass away.’ The literal (dominant) meaning of bucket makes no contribution whatsoever to the meaning of this idiom. Yet, there are innumerable examples of children and adults with language disorder failing to inhibit the dominant meaning of words in idiomatic and metaphoric expressions and interpreting them in a literal fashion in consequence. In the examples below, an adult with right-hemisphere damage (RHD) is explaining the meaning of figurative expressions in the Metaphors subtest in the RICE-3 (Harper et al. 2010). The speaker with RHD fails to inhibit the dominant meanings of the words in these figurative expressions, leading to a literal interpretation in each case: A stitch in time saves nine. “If you have a hole in your sock, sew it up before it gets to be a great big sock and one stitch will fix it early on but later it will take nine stitches.” It takes two to tango. “It takes two to dance, it’s not much fun if you’re just dancing by yourself, so it takes two to tango.”

When asked to explain the meaning of the metaphor in the utterance My friend’s mother-in-law is a witch, a male patient with RHD examined by Abusamra et al. (2009) replied as follows: “It means being tied down to religious sects, to religions, to umbanda.”

Yet again, the failure to inhibit the dominant meaning of the word witch leads the speaker toward a literal interpretation of the meaning of this utterance. In terms of expressive language, reduced inhibition of context might result in the intrusion of irrelevant information in response to questions. The following exchanges between a researcher (R) and a child participant (P) with autism spectrum disorder (ASD) were recorded as part of a study by De Villiers et al. (2012). In each exchange, the child with ASD is unable to suppress information that is not relevant to the researcher’s question. This occurs immediately after a response to a question has been given. It gives the appearance in each exchange that there is a highly activated wider context that the child is then unable to inhibit: Exchange 1: R: who’s in your family? P: hm I don’t know. R: are there five of you?

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

18 The Role of Context in Clinical Linguistics

P: yes. P: my cat. Exchange 2: R: Do you have a sister? P: Yes and she won! R: What did she win? The failure to inhibit context is strikingly evident in the following response of an adult with schizophrenia to a question from a doctor (Thomas 1997). The extended reply addresses several irrelevant topics and suggests widespread activation of context that the adult is unable to inhibit: Then I left San Francisco and moved to . . . where did you get that tie? It looks like it’s left over from the 1950s. I like the warm weather in San Diego. Is that a conch shell on your desk? Have you ever gone scuba diving? (1997: 41)

Each of the above difficulties relates to a failure of contextual inhibition. Too much context is primed or salient, and the child or adult with language disorder is unable to suppress this activated information. It then intrudes into an individual’s responses to questions or leads to literal interpretation of figurative utterances. But children and adults with language disorder make nonnormative use of context in a second way. This arises when context is insufficiently activated. Alternatively, context may be activated, but the child or adult with language disorder is unable to attend to it during their processing of utterances. There is also evidence from clinical studies that this type of context insensitivity pervades language processing in individuals with language disorder. To illustrate this point, we need only consider research on social cognition and theory of mind in ASD (Cummings 2013, 2014c, 2017). Children and adults with ASD have a significant impairment of theory of mind (ToM). This manifests as a failure to attribute cognitive mental states like knowledge and beliefs and affective mental states like happiness and anger to the minds of others. The ToM deficit in ASD is vividly illustrated in a study by Loukusa et al. (2007). These researchers read the following scenario to a 7-year-old boy with Asperger’s syndrome, a form of ASD. The boy was then asked a question: The researcher shows a picture of a boy sitting on the branch of a tree, with a wolf underneath the boy at the bottom of the tree. The wolf is growling at the boy. A man with a gun is walking nearby. The researcher reads the following verbal scenario aloud and then asks a question: “The boy sits up in the tree and a wolf is at the bottom of the tree. How does the boy feel?” Boy: Fun because he climbs up the tree. I always have fun when I climb up a tree.

The boy’s response reveals a deficit in affective ToM in that he is unable to detect the fear that the boy in the picture will experience in the presence of

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

401

402

LOUISE CUMMINGS

a wolf. This failure to perceive the mental state of the child in the picture is a type of context insensitivity that has implications for the interpretation of language. For if we cannot establish the mental states of a speaker, then we cannot establish the communicative intentions that motivate a speaker to produce utterances. As a result, many of the implicatures, speech acts, and other forms of language that we regularly encounter and take for granted cannot be adequately interpreted. By way of illustration, let us return to another example from Loukusa et al. (2007). On this occasion, a 9-year-old boy with Asperger’s syndrome is presented with a scenario and asked a question: The researcher shows the child a picture with a mother and a girl. The girl has a dress on and she is running. There are muddy puddles on the road. The girl has just stepped in the puddle and the picture shows the mud splashing. The researcher reads the following verbal scenario aloud and then asks a question: “The girl with her best clothes on is running on the dirty road. The mother shouts to the girl: ‘’Remember that you have your best clothes on!’’ What does the mother mean?” Boy: You have your best clothes on.

Clearly, the mother is using her utterance to warn the girl to keep her clothes clean. This speech act is only understood when the communicative intention that motivated the mother to produce it is established. However, because of his ToM difficulties, the boy with Asperger’s syndrome cannot attribute this intention to the mother. Instead, he merely reasserts part of the mother’s utterance, with no appreciation of what it is intended to achieve. The mother’s mental states are part of the boy’s context for the interpretation of the utterance. However, this part remains inaccessible to him on account of his ToM difficulties. This same insensitivity to context is also seen in children and adults with other clinical conditions (see, for example, Champagne-Lavau et al. (2018) for RHD and Whiting et al. (2005) for Parkinson’s disease). Colle et al. (2013) investigated the production and comprehension of a range of pragmatic aspects of language in seventeen adults with schizophrenia. Participants were required to process linguistic and nonlinguistic features of context to establish the meaning of utterances including direct and indirect speech acts and irony. One scenario used in the study is presented below: The subject is shown a videotaped scenario in which a boy and a girl are eating a disgusting soup. The boy smacks his lips with a gesture meaning “It’s very good!” Test question and subject’s response: What did the boy mean by that? He meant to say that she cooked a delicious soup.

The boy is clearly being ironic – the soup is anything but delicious. However, the adult with schizophrenia in this scenario does not appreciate the irony in

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

18 The Role of Context in Clinical Linguistics

the boy’s statement. His failure is related to his apparent difficulty in attributing significance to the boy’s gesture alongside explicit information that the soup is delicious. One way to conceive of this difficulty is that the adult’s context for the interpretation of the boy’s utterance is not fully activated. Alternatively, the adult’s context may be fully activated, but he fails to attend adequately to it. Either way, it is the adult’s insensitivity to context that leads to his failure to interpret the irony of the boy’s statement.

18.3.2 Theme 2: Context as a Barrier to, and Facilitator of, Communication We have already seen how context can be a barrier to communication when children and adults with language disorder fail to inhibit context or are insensitive to context. But we have not addressed the ways in which context can facilitate communication in individuals with language disorder. To illustrate what is at issue, let us consider how context normally facilitates the interpretation of utterances. I must draw on context to understand each of the following utterances: (1) The body was discovered next to the bank. (2) What a delightful child! (3) We lived here 20 years ago.

The lexical ambiguity in (1) is resolved by means of context. I can use my knowledge of what has already been said in a conversation, my memory of a news report on the television, or a picture in the local newspaper to establish that the bank in utterance (1) is the bank of the local river and not the bank on the main street in the center of town. The irony of (2) is apparent to anyone who is present when the speaker produces the utterance and observes a boisterous child creating havoc and destruction. To interpret (3) requires that I know the referents of the deictic expressions we and here. Knowledge of the speaker of the utterance and other people present in a conversation might get me some of the way in identifying the referent of we, while I must know the speaker’s location to establish the referent of here. Context weaves its way seamlessly through my interpretation of each of the above utterances. But imagine that I have language disorder and I can only understand part of the linguistic utterance that I hear. How does my relationship to context change under these circumstances? In the presence of a degraded linguistic code, I have no option but to rely more heavily on wider context to establish the meaning of the speaker’s utterance. On many occasions, my reliance on context pays off and I can establish what the speaker means – that the utterance in (2) is an example of irony based only on the speaker’s exasperated facial expression and the presence of a destructive child. On other occasions, my reliance on context lets me down and I misinterpret what the speaker is saying – that the speaker of the utterance in (1) uses bank to mean financial institution because I have just seen him exiting the bank in

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

403

404

LOUISE CUMMINGS

town. In each case, context is filling the void that has been created by my impaired receptive language skills. The role of context has moved beyond one of facilitation, typical of normal utterance interpretation, to become one of compensation. I can also use context to compensate for impaired expressive language skills, such as when I point to objects that I cannot name, or when I rely on my conversational partner to supplement my spoken message. The use of context to compensate for impaired language skills is well attested in clinical research and practice. I investigate the impact of neurodegeneration on language in my research (Cummings 2020, 2021). I am repeatedly struck by how well my study participants appear to manage the demands of conversation, only for their linguistic performance to drop off markedly during structured tasks when they cannot draw on context so easily to compensate for their impaired language skills. Early studies in aphasiology also highlighted the role of context in compensating for impaired receptive language skills in people with aphasia (Waller and Darley 1978; Pierce and Wagner 1985; Cannito et al. 1986; Hough et al. 1989). Linguistic and extralinguistic context has been shown, for example, to facilitate comprehension of specific lexical items and reversible passive sentences (e.g., The cat was chased by the mouse) in aphasic individuals with low comprehension skills on standard tests of auditory comprehension, but not in aphasic subjects with higher-level auditory comprehension skills (Pierce and Beekman 1985). The use of context to compensate for impaired receptive language has also been demonstrated in children with Down syndrome (Levorato et al. 2009) and adults with schizophrenia (Chakrabarty et al. 2014). The same compensation is evident during language expression. People with aphasia have been shown to use pantomime to compensate for information that they cannot convey in speech (van Nispen et al. 2018). Young children with language delay have been found to make use of communicative gestures to compensate for their small oral expressive vocabulary (Thal and Tobias 1992). Children’s reliance on context to compensate for impaired expressive language skills is vividly demonstrated in the following conversational exchange between a boy known as JB and an examiner (E) (McCardle and Wilson 1993). JB has FG syndrome and agenesis of the corpus callosum (partial or complete absence of the nerve fibres that link the two cerebral hemispheres of the brain). He is developmentally delayed. JB sat at 15 months, walked at 26 months, and used phrases at 3 years (typically developing children start using phrases between 18 months and 2 years of age). He was 5 years and 7 months old at the time of the recording: E: Tell me about your dog. JB: It go woof woof. I have a doggie, yep. E: What’s your doggie’s name? JB: Spot. Spot doggie puppy dog.

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

18 The Role of Context in Clinical Linguistics

They go pee-pee. Go pee-pee (pointing to the floor) Smell (holding nose, laughing) I go fight doggie (kicking the air) Puppy dog go bite. JB has very limited expressive language skills. He produces short utterances between one and four words in length. He omits the suffixes of verbs (e.g., goes) and uses onomatopoeia (woof-woof) instead of lexical verbs (bark). His use of subject pronouns is inconsistent, with pronouns used on some occasions when they are needed (they go pee-pee), but not used on other occasions (smell). However, FB effectively compensates for these expressive language difficulties through use of pointing, manual gestures (holding his nose), and full body movements (kicking the air).

18.3.3 Theme 3: The Role of Context in the Language Disorders Clinic Children and adults with language disorder are assessed and treated by speechlanguage pathologists in a range of different contexts. Typically, these contexts involve clinics in hospitals, schools, and residential care settings. Less commonly, clinicians visit clients in their own homes. Apart from home visits, the contexts in which most speech-language pathology takes place often deviate markedly from communication in the mundane contexts of daily life. A clinical encounter seldom involves the number and type of participants with whom we routinely communicate or reflects the social relationships that exist between speakers and hearers in work and leisure settings. These factors have a direct impact on the type of language that is used in clinics. For example, clients seldom feel empowered to pose questions in a clinical setting, let alone contest the validity of an assessment or intervention. Restricted use of these speech acts by clients in a language disorders clinic reflects the unequal power that exists between patients and clinicians in many medical and health interactions (Nimmon and Stenfors-Hayes 2016). Speech-language pathologists must address a pressing “clinic paradox”: they are constrained to assess and treat children and adults with language disorders in a clinical context that bears little similarity to the reallife contexts in which most communication takes place. Some reflection on this paradox and its implications for the language disorders clinic is in order. To help us conceive of the different ways in which context relates to the language disorders clinic, it is helpful to introduce a distinction between micro-context and macro-context. As these terms suggest, these different types of contexts relate largely to scale. A micro-context captures features of a task, exercise, or interaction that have the potential to influence how these activities are performed. A therapist’s question is a linguistic micro-context for a client’s answer in conversation. A picture is a visual micro-context for an auditory comprehension activity. A set of objects or toys is a physical micro-context for an instruction to put the spoon on top of the box. Micro-context operates

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

405

406

LOUISE CUMMINGS

alongside macro-context in the language disorders clinic. The latter type of context captures the setting in which an interaction between a therapist and a client takes place, the people who are present in an interaction (spouse or carer in addition to the therapist and client), the duration of an interaction, and the time of day that it takes place. Macro-context can extend more widely still to include a client’s social network, leisure activities, and occupational role. Features of macro-context can also influence how a client performs a range of activities. The presence of a spouse may facilitate or hinder conversation in a client with aphasia. A spouse can elicit target words with skilled use of cues or can create frustration and reduced conversational participation through use of sequences of test questions to which the answers are already known (Beeke et al. 2013). Speech-language pathologists are increasingly reflecting on the implications of both types of context for how they assess and treat clients with language disorders. Formal language tests were once the dominant approach to language assessment. These assessments adopt a tightly constrained micro-context consisting of instructions to point to objects or pictures that correspond to spoken words or sentences. If such a thing as “pure” linguistic competence exists, the scores on these tests are presumed to represent it. We will see below how speech-language pathologists have expanded the micro-context of language evaluation through use of techniques like conversation analysis and discourse analysis. Embedded within a conversation or a narrative, a client’s linguistic utterance relates to a much wider micro-context, consisting of the speaker’s and the hearer’s mutual expectations and shared knowledge about the world. The macro-context in which speech-language pathology is practiced has also been substantially revised. Speech-language pathologists are now as likely to address in therapy reduced participation in activities outside of the home as they are a goal such as the ability to use three-word utterances to communicate daily needs in response to pictures with 75 percent accuracy. A significant driver of this change in speech-language pathology and other health disciplines has been the introduction of the International Classification of Functioning, Disability and Health (ICF; World Health Organization, 2001). The ICF framework and its impact on the macro-context of the language disorders clinic are discussed further below.

18.3.4 Theme 4: Context and the Assessment of Language in Speech-Language Pathology Formal language tests have long been used by speech-language pathologists to assess language skills in children and adults. It is easy to see why this is the case. When they are standardized and norm-referenced, formal tests permit an objective assessment to be made of a person’s language skills at different points in time. This may be before and after an intervention has taken place so that we can measure the effects of treatment. Alternatively, formal testing can be conducted at different points in time in order to gauge a client’s language recovery following a stroke or traumatic brain injury, for example. Formal

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

18 The Role of Context in Clinical Linguistics

tests can also be used to compare a client’s language skills to those of a healthy population or a population of other people who have the same condition as the client (e.g., aphasia). The administration and scoring of formal tests and interpretation of their results are explicitly set out in instruction manuals and are not subject to individual judgments. Formal tests would appear to be an effective, accurate way in which to assess a client’s language skills. However, these advantages of formal tests are achieved at the expense of ecological validity. As defined by Dawson and Marcotte (2017), ecological validity refers to “generalisability (veridicality, or the extent to which assessment results relate to and/or predict behaviours outside the test environment) and representativeness (verisimilitude, or the degree to which assessments resemble everyday life contexts in which the behaviours will be needed)” (p. 617). There are many examples of the low generalizability and representativeness of formal language tests. By way of illustration, consider how language skills are assessed in the Clinical Evaluation of Language Fundamentals – Fifth Edition Metalinguistics (CELF-5 Metalinguistics; Wiig and Secord 2014). This assessment is suitable for individuals aged 9;0 to 21;11 years who have adequate linguistic knowledge, can understand basic concepts, and can produce grammatically correct sentences, but who lack the higher-level linguistic skills that are needed to master grade-level curriculum for Grades 3 and up. There are five stand-alone tests: a Metalinguistics Profile; two tests of meta-pragmatic skills (Making Inferences and Conversation Skills); and two tests of meta-semantic skills (Multiple Meanings and Figurative Language). Let us examine how this assessment evaluates a client’s conversation skills, one of the tests of metapragmatic skills. The examiner presents a pictured scene that creates a conversational context and says two or three words to the student. The same words are printed above the scene. The student is required to formulate a sentence based on the picture, using the words in the exact form (tense, number, etc.) in which they appear above the picture. How ecologically valid is this task? Does it reflect the conversation skills that speakers use in their daily lives? Clearly not. When we participate in conversation, we do not construct sentences based on a predetermined set of words and using a predetermined context. Rather, we actively construct context based on what has already been said in the conversation, what a hearer may reasonably be expected to know, and what purpose we want our utterance to fulfill in the exchange. The purpose of my utterance may be to persuade a friend not to buy a dress that is two sizes too small for her. I am aware of the potential of my utterance to upset my friend and am keen to avoid any threat to our friendship. And so, I decide that some form of indirect expression is required. I know my friend likes polka dots and that purple is her favorite color. I also know that we saw a dress that matched these requirements in Bella’s Boutique, a store we visited earlier in our shopping trip. I turn to my friend and say, “I think you will look stunning in that dress we saw in Bella’s Boutique.” My friend returns the small dress to its hanger and we both leave the shop still on good terms. The active construction of context that I have just described is what actual conversation involves and what the CELF-5 subtest on Conversation Skills

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

407

408

LOUISE CUMMINGS

fails to assess. To address the issue of ecological validity, speech-language pathologists also use a range of informal assessments of language. These assessments, which include analysis of naturally occurring conversation and discourse, examine a broadly construed, authentic micro-context that is in stark contrast to the tightly constrained, artificial micro-context of formal language assessments. By viewing context as evolving over consecutive turns in a conversation or utterances in a story, speech-language pathologists can examine pronominal reference, topic management, and other dynamic aspects of conversation and discourse. By way of illustration, consider the following extract from the Cinderella narrative produced by a 51-year-old man with alcohol-related brain damage. The author and client jointly examined pictures in a wordless Cinderella picture book, whereupon the book was closed, and the client narrated the story from memory: well she’s out (1:58) with a horse (.) and him (1:37) I feel so stupid so I do now, and ah (3:35) she wants to go to the ball she meets the old woman ends up the fairy godmother (1:09) sh, sh, she turns a pumpkin into a (0:93) a carriage (1:89) takes her to the ball and she has a lovely gets a lovely dress glass shoes [. . .] (Cummings 2021)

This client performs two illicit shifts in the reference of the pronoun she. The first three instances of the pronoun refer to Cinderella. However, the speaker then introduces the fairy godmother into the discourse context, so that there are now two potential referents of the pronoun she. In the absence of any attempt to make one of these referents salient, the speaker uses the pronoun she on a fourth occasion to refer to the fairy godmother (she turns a pumpkin into a carriage) and then on a fifth occasion to refer once again to Cinderella (she has a lovely gets a lovely dress). We can track these uses of pronominal reference across extended discourse and use these illicit shifts in reference to confirm our impression that this speaker is difficult to follow. But no sense can be made of this speaker’s use of reference if we do not engage fully with the broad, dynamic concept of context that is integral to our everyday communication.

18.3.5 Theme 5: Context and Intervention in Speech-Language Pathology Speech-language pathologists must also attend to macro-context in their work with children and adults with language disorder. This requires clinicians to think about the setting of intervention, the participants in intervention, and the goals an intervention should strive to attain. Intervention in speech-language pathology has moved beyond its once exclusive focus on language impairments to include rehabilitation goals that address the impact of a language disorder on a client’s conversational participation and social integration. A key driver in this change of focus has been the World Health Organization’s (2001) ICF framework. This framework classifies functioning and disability associated with health conditions. It is intended to complement the ICD-11 (World Health

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

18 The Role of Context in Clinical Linguistics

Organization 2019), which is an aetiological framework for the classification of diseases, disorders, and other health conditions based on diagnosis. The ICF framework has two main parts: Functioning and Disability and Contextual Factors. Under functioning and disability, the framework draws a distinction between Body Functions and Structures and Activity and Participation. Under contextual factors, Environmental Factors and Personal Factors are distinguished. To illustrate how these components of the ICF framework are interrelated, let us use the example of aphasia. An adult with strokeinduced aphasia has a neurological impairment (an impairment of body functions and structures). Aphasia may limit this adult’s ability to read novels for leisure or deliver scriptures in church (an activity limitation). Owing to his/her communication difficulties, the adult with aphasia may increasingly withdraw from social interaction with others. He/she may cease church attendance altogether, for example (a participation restriction). Factors that contribute positively to this client’s rehabilitation are his/her high level of motivation (a personal factor) and the support of his/her family and friends (an environmental factor), while right hemiparesis (weakness of the right arm and leg) and limited ambulance transport are personal and environmental factors, respectively, that might hinder rehabilitation. The professional body for speech-language pathologists in the United States – the American Speech-Language-Hearing Association (ASHA) – has used the ICF framework to set person-centered functional goals of intervention for aphasia and other communication disorders (ASHA 1997–2020). One of the ways in which speech-language pathologists have enlarged the macro-context of intervention in the last twenty years is to engage partners of adults with conditions like aphasia and dementia directly in the therapeutic process. Partners may be familiar or unfamiliar to the person with a communication disorder and can include spouses, carers, and volunteers. Several conversation partner training programs exist, including Supporting Partners of People with Aphasia in Relationships and Conversation Analysis (Lock et al. 2001) and Supported Conversation for Adults with Aphasia (Kagan 1998) for partners of clients with aphasia, and the Conversation Analysis Profile for People with Cognitive Impairments (Perkins et al. 1997) for partners of clients with dementia (for a review of these programs for aphasia and dementia, respectively, the reader is referred to Turner and Whitworth 2006 and Kindell et al. 2017). Notwithstanding differences of approach, these programs all share the same starting point, namely, that partners can be taught how to facilitate conversation with the person with aphasia or dementia, often through adjustments to their own style of communication. The need for such intervention can be vividly illustrated by the case of Harry, a 72-year-old man with progressive supranuclear palsy (PSP) and cognitive impairment who was studied by the author (Cummings 2021). Harry’s wife reported a complete lack of social participation by her husband: Wife: what happens when we’re out with people for meals? Harry: we’re out with people, yes, I just sit there, don’t engage in conversation or anything.

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

409

410

LOUISE CUMMINGS

It was clear that, although Harry had cognitive and language problems related to his PSP, he had adequate language skills to participate in conversation. It was not long after meeting Harry and his wife that I discovered the source of his lack of social participation. Harry’s wife had a very dominant style of communication. She assumed total control during conversation, responding to questions that I directed to Harry and only permitting him entry to the conversation in a very regulated way. Harry was prompted by his wife to speak through her use of several utterance types: (i) (ii) (iii) (iv) (v)

direct questions (e.g., What countries did we pass through?) explicit commands (e.g., List the other places you went) sound and syllable cues (e.g., mu, mu, Germany, Munich) letter cues (e.g., Begins with an “R”) sentence completion prompts (e.g., From there you went to . . .)

In addition, when Harry did attempt to contribute to conversation, his wife explicitly corrected what he said on many occasions (e.g., Not the church choir). The combination of these conversational moves had disempowered Harry in conversation to the point where he had become passive and had all but completely “opted out” of social participation with others. Although I was not providing intervention to Harry and his wife, it was evident that this conversational dyad was not working well, and that Harry’s wife could benefit from conversation partner training. It is a feature of macro-context – the communication style of Harry’s wife – that was responsible for his lack of social participation with others. This same feature must be the starting point for any intervention in speechlanguage pathology.

18.4 Future Directions This discussion has demonstrated the different ways in which context plays a role in people with language disorders, and how it permeates the work of speech-language pathologists. Not only must clinicians understand the nonnormative use of context by children and adults with language disorder, but they must also be attentive to the reach of context when they assess and treat clients in clinic. But is there something more that speech-language pathology can do to integrate context into its work? I believe that there is. In this section, I outline briefly what I think those future developments for the field should be. The clinical education of speech-language pathologists must include an explicit focus on context. This central concept is left in the background of educational efforts as if it were simple and unproblematic and, hence, not worthy of discussion, or as if students could somehow naturally assimilate it. Neither of these scenarios is true. We all recognize the need to train students in phonetics, neurology, and child language development. We must now

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

18 The Role of Context in Clinical Linguistics

recognize the same need in relation to context. If students are encouraged to think about every interaction with clients in terms of context, this will provide an important impetus to the development of assessments and interventions that can reflect the actual communicative challenges and needs of clients. One context into which many clients with language and communication disorders are still not able to achieve successful (re)integration is the workplace (Meulenbroek et al. 2016). As clinicians, we must design assessments and interventions that can address the unique challenges of this context. To a limited extent, some speech-language pathologists are already doing exactly that. For example, Isaki and Turkstra (2000) have established that communication measures that use impairment- and disability-based tasks are better able to predict work reentry following traumatic brain injury than measures that use impairment-level tasks alone. Speech-language pathologists who have been directly trained to be aware of context and who are guided by it in their clinical practice are best placed to undertake these developments in the service of their clients. Another future development relating to context concerns a greater focus on long-term outcomes in children and adults with language disorder. These outcomes are not realized immediately after an episode of therapy but often occur many years after language intervention has taken place. For a child with developmental language disorder, a long-term outcome may include an individual’s literacy and language skills, social networks, mental health, or vocational status as a young adult (Whitehouse et al. 2009a, 2009b). It is these outcomes that speech-language pathologists must increasingly use to demonstrate the effectiveness of their interventions to clients, their families, and the health providers who commission clinical language services. In most service delivery models, language intervention is delivered in one or two sessions per week, with an episode of therapy completed in a certain number of weeks. An intervention is judged to be effective if language performance, often measured in test scores, has improved by the end of an episode of care. But when the context of language intervention is expanded beyond the language disorders clinic to include a range of participants and settings, it also becomes necessary to use longer-term outcomes to assess the effectiveness of intervention. The use of long-term outcomes in language intervention research has been hampered by issues such as cost – it is expensive to follow clients over a long period of time – and small sample sizes, with many clients lost to follow-up. But these outcomes are a more reliable criterion against which to assess the impact of intervention on the lives of clients who receive speech-language pathology. There is a final area in which I believe context can be more fully integrated into the work of speech-language pathology. Notwithstanding the significant societal costs of language disorder, there has been little effort to characterize language disorder as a public health issue (for further discussion, see section 1.6 in Cummings 2018). This is the case even though language disorder exerts the same population-level health and economic effects as many other conditions. There is lost productivity when people with language disorder are unable to

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

411

412

LOUISE CUMMINGS

participate in the workforce and family members must leave the workforce in order to support and care for them. Economically inactive people must be supported through the receipt of welfare benefits. There are also additional healthcare costs associated with language disorder (Cronin et al. 2017). These costs are accrued through the provision of support services like speechlanguage pathology and the management of poor mental health associated with language difficulties. By conceiving of language disorder as a public health issue, there is also fresh impetus to investigate the epidemiology of language and communication disorders. There has been little discussion of, and a dearth of research into, the epidemiology of communication disorders (Byles 2005; Enderby and Pickstone 2005). This has resulted in a lack of knowledge of the prevalence and incidence of these disorders, which is essential for workforce planning in speech-language pathology. By using context to revise how we conceive of language disorder in the same way that we have used context to revise how we assess and treat children and adults with language disorder, substantial progress can also be made in the provision of clinical language services.

References Abusamra, V., Côté, H., Joanette, Y., and Ferreres, A. (2009). Communication impairments in patients with right hemisphere damage. Life Span and Disability, 12(1), 67–82. American Speech-Language-Hearing Association (ASHA). (1997–2020). The International Classification of Functioning, Disability, and Health (ICF). www .asha.org/slp/icf/ (accessed September 29, 2020). Beeke, S., Beckley, F., Best, W., Johnson, F., Edwards, S., and Maxim, J. (2013). Extended turn construction and test question sequences in the conversations of three speakers with agrammatic aphasia. Clinical Linguistics & Phonetics, 27(10– 11), 784–804. Byles, J. (2005). The epidemiology of communication and swallowing disorders. International Journal of Speech-Language Pathology, 7(1), 1–7. Cannito, M. P., Jarecki, J. M., and Pierce, R. S. (1986). Effects of thematic structure on syntactic comprehension in aphasia. Brain and Language, 27(1), 38–49. Chakrabarty, M., Sarkar, S., Chatterjee, A., Ghosal, M., Guha, P., and Deogaonkar, M. (2014). Metaphor comprehension deficit in schizophrenia with reference to the hypothesis of abnormal lateralization and right hemisphere dysfunction. Language Sciences, 44, 1–14. Champagne-Lavau, M., Cordonier, N., Bellmann, A., and Fossard, M. (2018). Context processing during irony comprehension in right-frontal brain-damaged individuals. Clinical Linguistics & Phonetics, 32(8), 721–738. Colle, L., Angeleri, R., Vallana, M., Sacco, K., Bara, B. G., and Bosco, F. M. (2013). Understanding the communicative impairments in schizophrenia: A preliminary study. Journal of Communication Disorders, 46(3), 294–308. Cronin, P., Reeve, R., McCabe, P., Viney, R., and Goodall, S. (2017). The impact of childhood language difficulties on healthcare costs from 4 to 13 years: Australian longitudinal study. International Journal of Speech-Language Pathology, 19(4), 381–391.

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

18 The Role of Context in Clinical Linguistics

Crystal, D. (1981). Clinical Linguistics. Vienna: Springer-Verlag. Crystal, D. (1997). The Cambridge Encyclopedia of Language, 2nd ed. Cambridge: Cambridge University Press. Cummings, L. (2009). Clinical Pragmatics. Cambridge: Cambridge University Press. Cummings, L. (2013). Clinical pragmatics and theory of mind. In A. Capone, F. Lo Piparo, and M. Carapezza (eds.), Perspectives on Linguistic Pragmatics (Perspectives in Pragmatics, Philosophy & Psychology, Vol. 2) (pp. 23–56). Cham: Springer International Publishing AG. Cummings, L. (2014a). Clinical linguistics. In M. Aronoff (ed.), Oxford Bibliographies in Linguistics. New York: Oxford University Press. www.oxfordbibliographies.com /view/document/obo-9780199772810/obo-9780199772810-0079.xml (accessed January 26, 2021). Cummings, L. (2014b). Pragmatic Disorders. Cham: Springer International Publishing AG. Cummings, L. (2014c). Pragmatic disorders and theory of mind. In L. Cummings (ed.), Cambridge Handbook of Communication Disorders (pp. 559–577). Cambridge: Cambridge University Press. Cummings, L. (2017). Pragmatic disorders in forensic settings. In F. Poggi and A. Capone (eds.), Pragmatics and Law (Perspectives in Pragmatics, Philosophy & Psychology, Vol. 10) (pp. 349–377). Cham: Springer International Publishing AG. Cummings, L. (2018). Speech and Language Therapy: A Primer. Cambridge: Cambridge University Press. Cummings, L. (2020). Language in Dementia. Cambridge: Cambridge University Press. Cummings, L. (2021). Language Case Files in Neurological Disorders. New York: Routledge. Dawson, D. R., and Marcotte, T. D. (2017). Editorial: Special issue on ecological validity and cognitive assessment. Neuropsychological Rehabilitation, 27(5), 599–602. De Villiers, J., Myers, B., and Stainton, R. J. (2012). Differential pragmatic abilities and autism spectrum disorders: The case of pragmatic determinants of literal content. In M. Macaulay and P. Garcés-Conejos (eds.), Pragmatics and Context. Toronto, CA: Antares. Enderby, P., and Pickstone, C. (2005). How many people have communication disorders and why does it matter? Advances in Speech-Language Pathology, 7(1), 8–13. Harper, A. S., Cherney, L. R., and Burns, M. A. (2010). Rehabilitation Institute of Chicago Evaluation of Communication Problems in Right Hemisphere Dysfunction (RICE-3). Chicago: Rehabilitation Institute of Chicago. Hough, M. S., Pierce, R. S., and Cannito, M. P. (1989). Contextual influences in aphasia: Effects of predictive versus nonpredictive narratives. Brain and Language, 36(2), 325–334. Isaki, E. and Turkstra, L. (2000). Communication abilities and work re-entry following traumatic brain injury. Brain Injury, 14(5), 441–453. Kagan, A. (1998). Supported conversation for adults with aphasia: Methods and resources for training conversation partners. Aphasiology, 12(9), 816–830. Kindell, J., Keady, J., Sage, K., and Wilkinson, R. (2017). Everyday conversation in dementia: A review of the literature to inform research and practice. International Journal of Language & Communication Disorders, 52(4), 392–406. Levorato, M.C., Roch, M., and Beltrame, R. (2009). Text comprehension in Down syndrome: The role of lower and higher level abilities. Clinical Linguistics & Phonetics, 23(4), 285–300.

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

413

414

LOUISE CUMMINGS

Lock, S., Wilkinson, R., and Bryan, K. (2001). SPPARC: Supporting Partners of People with Aphasia in Relationships and Conversations. Bicester: Winslow Press. Loukusa, S., Leinonen, E., Jussila, K., Mattila, M.-L., Ryder, N., Ebeling, H., and Moilanen, I. (2007). Answering contextually demanding questions: pragmatic errors produced by children with Asperger syndrome or high-functioning autism. Journal of Communication Disorders, 40(5), 357–381. McCardle, P., and Wilson, B. (1993). Language and development in FG syndrome with callosal agenesis. Journal of Communication Disorders, 26(2), 83–100. Meulenbroek, P., Bowers, B., and Turkstra, L. S. (2016). Characterizing common workplace communication skills for disorders associated with traumatic brain injury: A qualitative study. Journal of Vocational Rehabilitation, 44(1), 15–31. Nimmon, L., and Stenfors-Hayes, T. (2016). The “handling” of power in the physicianpatient encounter: Perceptions from experienced physicians. BMC Medical Education, 16(114), 1–9. doi: 10.1186/s12909-016-0634-0. Perkins, L., Whitworth, A., and Lesser, R. (1997). Conversation Analysis Profile for People with Cognitive Impairments (CAPPCI). London: Whurr. Pierce, R. S., and Beekman, L. A. (1985). Effects of linguistic and extralinguistic context on semantic and syntactic processing in aphasia. Journal of Speech and Hearing Research, 28(2), 250–254. Pierce, R. S., and Wagner, C. M. (1985). The role of context in facilitating syntactic decoding in aphasia. Journal of Communication Disorders, 18(3), 203–213. Thal, D. J., and Tobias, S. (1992). Communicative gestures in children with delayed onset of oral expressive vocabulary. Journal of Speech and Hearing Research, 35(6), 1281–1289. Thomas, P. (1997). What can linguistics tell us about thought disorder? In J. France and N. Muir (eds.), Communication and the Mentally Ill Patient: Developmental and Linguistic Approaches to Schizophrenia (pp. 30–42). London: Jessica Kingsley Publishers. Titone, D., Levy, D. L., and Holzman, P. S. (2000). Contextual insensitivity in schizophrenic language processing: Evidence from lexical ambiguity. Journal of Abnormal Psychology, 109(4), 761–767. Turner, S., and Whitworth, A. (2006). Conversational partner training programmes in aphasia: A review of key themes and participants’ roles. Aphasiology, 20(6), 483–510. Van Nispen, K., Mieke, W. M. E., van de Sandt-Koenderman, E., and Krahmer, E. (2018). The comprehensibility of pantomimes produced by people with aphasia. International Journal of Language & Communication Disorders, 53(1), 85–100. Waller, M. R., and Darley, F. L. (1978). The influence of context on the auditory comprehension of paragraphs by aphasic subjects. Journal of Speech and Hearing Research, 21(4), 732–745. Whitehouse, A. J. O., Watt, H. J., Line, E. A., and Bishop, D. V. M. (2009a). Adult psychosocial outcomes of children with specific language impairment, pragmatic language impairment and autism. International Journal of Language & Communication Disorders, 44(4), 511–528. Whitehouse, A. J. O., Line, E. A., Watt, H. J., and Bishop, D. V. M. (2009b). Qualitative aspects of developmental language impairment relate to language and literacy outcome in adulthood. International Journal of Language and Communication Disorders, 44(4), 489–510.

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

18 The Role of Context in Clinical Linguistics

Whiting, E., Copland, D., and Angwin, A. (2005). Verb and context processing in Parkinson’s disease. Journal of Neurolinguistics, 18(3), 259–276. Wiener, D., Connor, L. T., and Obler, L. (2004). Inhibition and auditory comprehension in Wernicke’s aphasia. Aphasiology, 18(5–7), 599–609. Wiig, E. H., and Secord, W. A. (2014). Clinical Evaluation of Language Fundamentals. Fifth Edition. Metalinguistics. San Antonio, TX: Pearson. World Health Organization (2001). International Classification of Functioning, Disability and Health. Geneva: World Health Organization. World Health Organization (2019). International Classification of Diseases, 11th ed. Geneva: World Health Organization.

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

415

https://doi.org/10.1017/9781108989275.019 Published online by Cambridge University Press

Part V

Advances in Multimodal and Technological Context-Based Research

Published online by Cambridge University Press

Published online by Cambridge University Press

19 Nonverbal Communication and Context: Multimodality in Interaction Pauline Madella and Tim Wharton

19.1 Introduction For obvious reasons, the study of linguistics has focused on verbal modality. Phonologists study the sound structures of language, syntacticians the grammatical structures, semanticists the meaning inherent in language. There is, however, a bigger picture, and those of us working in the sub-discipline of linguistic pragmatics find ourselves quite regularly faced with a dilemma. In contrast to semantics, which studies context-independent sentences, the aim of a pragmatic theory is to explain how utterances are understood, and utterances, of course, incorporate both verbal and nonverbal modalities. We do not deliver our words in a monotonic, robot-like voice. We impose quasi-musical contours on the words we say and move our faces and hands – in sometimes highly idiosyncratic ways – in order to guide the hearer to our intended meanings. Sometimes, such behaviors tell us more about the emotional state of a friend or colleague than the words they utter. Whether we are researchers, or simply trying to help a friend, nonverbal modalities cannot be ignored. This chapter is an introduction to the major role nonverbal modes play in communication and, in so doing, promotes a “multimodal” view of human communication and looks at how the different modes identified not only constrain the construction of context but effectively guide our interlocutors to our intended meaning. We begin by addressing two questions: (1) What is multimodality? (2) How do nonverbal behaviors communicate?

In the next section we address the first question and explore the relationship between the terms “paralanguage,” “multimodality,” and “nonverbal.” In Section 19.3 we introduce the different categories of nonverbal behavior and Tim Wharton would like to thank Mengyang Qiu and Chara Vlachaki for their help in preparing this chapter.

https://doi.org/10.1017/9781108989275.020 Published online by Cambridge University Press

420

PAULINE MADELLA AND TIM WHARTON

sketch an answer to the second question from the perspective of Relevance Theory (Wilson and Wharton 2006; Wharton 2009). In Section 19.4 we address a third and final question: (3) How can we move toward a principled pragmatic account of multimodality in interaction?

We argue that a pragmatic account based on Relevance Theory can be carried over to a wide range of multimodal phenomena, requires none of the intermediate levels of description that necessarily form part of less pragmatic accounts, and falls out naturally from general communicative principles.

19.2 What Is Multimodality? The argument that nonverbal modalities play a major role in accounts of human communication is not new. Insights from early pragmatics into the significance of paralinguistic behaviors in interaction certainly give us a taste of the nonverbal dominance view of communication. While Birdwhistell (1970: 158) reports that “probably no more than 30 to 35 percent of the social meaning of a conversation or an interaction is carried by the words,” Mehrabian’s 7%–38%–55% rule (1971) stipulates that 55 percent of the information that we convey in a given exchange is communicated via our body language, 38 percent via our tone of voice or how we say what we say, and 7 percent from what is said (i.e., the words themselves). The importance of acknowledging the role of paralinguistic phenomena in face-to-face interaction is articulated by Abercrombie (1968: 65): “The conversational use of spoken language cannot be properly understood unless paralinguistic elements are taken into account.” While Stevick (1982: 163) describes nonverbal communication as providing “the surface on which the words are written and against which they must be interpreted.” In this chapter, we view multimodality as close to Claudel’s notion of L’oeil ecoute (1946; Madella 2021), encompassing both visible and audible modes of communication and interpretation, which are to “converge to contribute to a final meaning or message of which they are an intrinsic part” (Riley 1979: 143). “Multimodal” inputs – what Adam Kendon refers to as “poly-modalic” (2014a: 67) – invite the eyes to be as alert as the ears. Linguistic communication is, beyond any reasonable doubt, multimodal. Pragmatic accounts of linguistic communication must be multimodal also. The above accounts coincided with the development of the term “paralanguage” as first introduced by Trager in 1958. The term was originally introduced to refer to the way something is said, rather than what is said (Trager 1958). In Paralanguage: A First Approximation (1958), Trager described paralanguage as centered around modifications

https://doi.org/10.1017/9781108989275.020 Published online by Cambridge University Press

19 Nonverbal Communication and Context

of the voice and other vocal features, to the exclusion of gestural ones. As Poyatos puts it (1983: 129): What truly gives the spoken words their total meaning (which at any rate is not contained only in them) are a series of vocal/narial voice modifications and independent sounds and meaningful silences which today we subsume under paralanguage; and that, if visually perceived, those verbal expressions are accompanied by a great number of facial, manual and bodily gestures, gaze activities, manners, postures, postural shifts and stills, which constitute kinesics.

At the same time, the study of kinesics was beginning to take on a life of its own and that term came to be used to refer to all bodily movements, including those that are not intentionally communicative. A subtly different stance was then proposed by Abercrombie (1968: 68), who rejected what he described as the “unfortunate separation of the visible and the audible components.” Abercrombie integrated both audible and visible paralinguistic elements into his definition of paralinguistic phenomena. However, he made a point of distinguishing elements that are conversational in nature from those that are not, and only those that are intentionally communicative or “consciously controllable” qualify as paralinguistic according to his definition (p. 65). In this chapter, we adopt Abercrombie’s inclusive definition of paralinguistic behavior, on the grounds that the pragmatics of those paralinguistic modes of communication is to be found in the way that they interact with each other. We equate paralanguage with nonverbal or multimodal communication. However, it is worth noting immediately that Abercrombie’s exclusion of spontaneous, uncontrollable behaviors is controversial. According to Wharton (2009) spontaneous behaviors might be deliberately and overtly shown in a way that makes it in a sense “intentional.” Such behaviors might also be shown while giving my interlocutor the impression that it was not done intentionally. We touch on these later. Paralanguage functions to constrain the context and guide the hearer to the speaker’s intended meaning. This is precisely what makes these inputs “multimodal” or, as Kendon (2014a: 67) terms them, “poly-modalic.” The parallels between all these inputs are so strong that a unified, homogeneous account of para-/nonlinguistic behaviors seems to be required, one that embraces the fact that they are, for the most part, closely interlinked. Indeed, an inclusive definition of paralinguistic phenomena is precisely what allows us to argue for an inferential model of (verbal and nonverbal) communication from the perspective of Relevance Theory. In the next section we turn to prosody, one element of paralanguage that falls under Abercrombie’s broad definition, and one that has received a great deal of attention in the pragmatics literature.

https://doi.org/10.1017/9781108989275.020 Published online by Cambridge University Press

421

422

PAULINE MADELLA AND TIM WHARTON

19.3 The Relevance of Prosody: How Do Nonverbal Behaviors Communicate? When it comes to prosody, one point upon which there is broad agreement in the literature is that it comes in a range of different flavors. Some aspects of prosody feel entirely natural, in the sense that the link between form and meaning is far from arbitrary (or, to use Grice’s 1957 term, nonnatural). For example, a person speaking to a young child will typically adopt a higher pitch, which I perceived to be nonthreatening;1 when people are angry, the volume of their voice tends to increase. Others are clearly language-specific (Gussenhoven 2002). However, drawing a line between the natural and the nonnatural is notoriously difficult. According to Bolinger (1964: 282) prosody exists, at best, at “the edge of language,” and, he observes, even those fully nonnatural elements of prosody elements betray a degree of naturalness. Early work in prosody did not refer much to ideas in pragmatics, largely because pragmatic theories were perceived as underdeveloped. So while historically there is a huge phonological literature providing detailed analyses of English prosody and how it relates to meaning (Halliday 1963, 1967; O’Connor and Arnold 1973; Brazil 1975; Ladd 1978, 1996; Bolinger 1983a, 1983b; Ward and Hirschberg 1988; Hirschberg and Ward 1995; Pierrehumbert and Hirschberg 1990; Gussenhoven 1984, 2002, 2006; Chen and Gussenhoven 2003), little of this work took place within a clearly articulated theoretical pragmatic framework. However, since the 1990s, there has been an increase in work on the semantics and pragmatics of prosody thanks largely to advances in how pragmatic inference is understood (Vandepitte 1989; Clark and Lindsey 1990; House 1990, 2006; Escandell-Vidal 1998; Imai 1998; Fretheim 2002; Wichmann 2002; Wilson and Wharton 2006; Clark 2007; Madella 2021; Scott 2021). Recent work in psychology and cognitive science has built on insights gained in the 1940s and 1950s on “bounded rationality”, the idea that an individual’s thought processes are limited by both the finite nature of cognition and the time available to make decisions. It has been proposed that evolution has left humans with economical rules-of-thumb, which enable us to make the most of our finite cognitive capacity: “simple, intelligent heuristics capable of making near optimal inferences” (Kahneman 2011: 168). Work within cognitively oriented approaches to pragmatics has allowed us to suggest new ways of thinking about prosody, and new ways of framing original hypotheses as to how it works. According to Relevance Theory, cognition is geared toward the maximization of relevance, where information is relevant to the extent that it provides communicative effects for relatively minimal processing effort. Other things being equal, the more communicative effects gained, and the less processing effort expended in gaining those effects, the greater the relevance of the 1

This is a fairly universal phenomenon, presumably based on the fact that a higher-pitched voice equals a shorter larynx, equals a smaller creature, equals less danger.

https://doi.org/10.1017/9781108989275.020 Published online by Cambridge University Press

19 Nonverbal Communication and Context

information to the individual who processes it. This linguistic theory chimes harmoniously with theories developed in cognitive science. Indeed, cognitive scientists such as Gerd Gigerenzer, Peter Todd, and the ABC Research Group could be seen to be echoing one of the fundamental assumptions of Relevance Theory when they write: “Cognition is the art of focusing on the relevant and deliberately ignoring the rest” (Gigerenzer et al. 1999: 21). The disposition to search for relevance is routinely exploited in human communication, which gives rise to fairly precise expectations about how relevant an interpretation will be. Speakers know that listeners will pay attention only to stimuli that are relevant enough, and so in order to attract and hold an audience’s attention they should make their communicative stimuli appear at least relevant enough to be worth processing. This motivates the relevancetheoretic comprehension heuristic below (Wilson and Sperber 2004: 613): (a) Follow a path of least effort in computing communicative effects. (b) Stop when your expectation of relevance is satisfied. According to the framework developed in Wilson and Wharton (2006) and Wharton (2009), pragmatic inference and relevance play central roles in the interpretation of prosody. Prosodic inputs to the comprehension process fall at various points along a continuum between natural and conventional and can be divided into four categories (see Figure 19.1). Some of the natural elements of prosody are signs – interpreted entirely by inference – and others are signals, interpreted via decoding and inferential processes. The difference between the two varieties draws on the evolutionary (and ethological) notion of adaptive function (see Hauser 1996). To use a nonverbal behavior as an example, if Peter shivers, Mary might infer that he is cold. But the function of shivering is not to convey information: That is not the reason that shivering continues to proliferate among humans. Shivering is therefore a sign. If Peter smiles, Mary will understand that he is happy (or at least not a threat). Work in the psychology has shown that, in contrast to shivering, the

Prosodic inputs

Nonnatural

Natural

Signs

Signals

Inference

Coding

Linguistic

Cultural

Coding (plus inference)

Figure 19.1 Prosodic inputs

https://doi.org/10.1017/9781108989275.020 Published online by Cambridge University Press

423

424

PAULINE MADELLA AND TIM WHARTON

adaptive function of smiling is to convey information: it is therefore a signal (Hooff 1972; Fridlund 1994; Ekman 1999). Conventionalized prosodic elements are either linguistic signals or nonlinguistic/cultural signals. The paradigm archetypal linguistic signal is the word, but there are other conventional communicative signals exploited by humans according to culture-specific rules: consider, for example, the British two-fingered insult (McNeill 2000). These categories can be carried over to prosody unproblematically. It is not difficult to think of prosodic counterparts to the examples above. A speaker’s mental or physical state may affect the prosodic properties of her utterance, and a hearer with the appropriate experience or background knowledge will therefore be able to infer whether a speaker is sick or healthy, tired or alert. As with shivering, it is not the function of these prosodic properties to carry information about the speaker’s mental or physical state: they are natural signs, interpreted by inference rather than decoding. On the other hand, certain affective tones of voice, in a manner analogous to smiles, may well be natural signals, interpreted by innately determined codes. While the focus in this chapter is the “natural” side of multimodal communication, a word is necessary about how these innately determined codes (characterized in Figure 19.1 as natural signals) might work. In this regard, Wharton (2009) develops proposals made originally by Diane Blakemore (1987, 2002) and applied to different aspects of prosody by Vandepitte (1989), Clark and Lindsey (1990), House (1990), Escandell-Vidal (1998, 2002), Imai (1998), and Fretheim (2002). If, as relevance theorists take it, linguistic communication involves a combination of both decoding and inference, then coded linguistic signals might be expected to encode information of two distinct types. On the one hand there is regular conceptual encoding, where a word (e.g., bird) encodes a concept (e.g., BIRD) which plays a role as constituent of the logical form of sentences in which that word occurs. On the other, we might expect to find a form of procedural encoding, where an expression encodes information that – rather than sitting in the logical form – has the function of guiding the hearer’s inferential search for relevance. Fully linguistic expressions that have been analyzed as encoding procedural information include discourse connectives and particles and mood indicators and discourse particles (cf. Blakemore 1987, 2002; König 1991; Wilson and Sperber 1993; Hall 2004). Wharton (2003a) provides a survey of the literature and concludes that interjections – expressions such as wow, yuk, and ouch – are best analyzed as falling on the natural rather than the properly linguistic side. However, in terms of the distinctions presented above, he also argues that interjections are signals rather than signs, and that they encode procedural rather than conceptual information. The line of argument is taken further in Wharton (2003b), where it is proposed that all natural signals should also be analyzed as encoding procedures rather than conceptual information. The idea can then be extended to natural prosody, such as affective tone of voice, as mentioned above. On this approach, the function of affective tone of voice – a natural signal – would be

https://doi.org/10.1017/9781108989275.020 Published online by Cambridge University Press

19 Nonverbal Communication and Context

to facilitate the retrieval of similar propositional-attitude descriptions to those activated by interjections. In contrast, prosodic distinctions such as the ones between im’port and ‘import, and ex’port and ‘export, are clearly best analyzed as properly linguistic. However, parallel to the gestures referred to above, other conventionalized prosodic phenomena might be best analyzed as cultural. Such phenomena might include the prosodic inputs discussed by Ladd (2008). These “calling” or “hailing” contours, which are highly stylized intonation patterns, consist of two sustained notes, moving from high to low.2 Sperber and Wilson (1986/1995, 2015) have acknowledged that developing an adequate theory of utterance interpretation necessarily involves going beyond Grice’s notion of speaker’s meaning and including cases of both “showing” and “meaningNN.” As a result, Relevance Theory not only allows for a much wider range of communicative acts – including, for example, the intentional showing of otherwise spontaneous natural behaviors – but also accommodates easily cases of vague communication, where what a speaker intends to convey is not a single, definite, and salient interpretation, but rather a “diffuse impression” (Wilson and Wharton 2006) in the form of an array of weakly communicated assumptions (Sperber and Wilson 2015). As Wilson and Carston (2019: 34) remind us: Relevance theorists set out from the start to look for a set of pragmatic principles and mechanisms that can deal with the full range of overtly intentional communicative acts: verbal and non-verbal, showing and telling, determinate and indeterminate, literal and figurative, propositional and non-propositional.

Cases of showing and weakly communicated interpretations are often associated with nonverbal modes of communication and Relevance Theory fully acknowledges the role of multimodal input in both context construction and utterance interpretation. Multimodal behaviors guide the utterance interpretation process by altering the salience of possible interpretations of utterances (including possible disambiguations, reference resolutions, contextual assumptions, implicatures, speech-act descriptions, etc.). What Relevance Theory adds to this intuitive description is the idea that the salience of interpretations can be affected not only by altering processing effort but also by manipulating expected cognitive effects. This, then, is the answer to the “how” question. Jill House (2007: 369) is talking about prosody when she writes that it “guide[s] the listener in how to proceed: how to access the relevant cognitive context within which to interpret the speaker’s contribution, how to evaluate that contribution, and how to construct the interaction itself, to enable the communication to take place.” She could, however, easily have been writing about nonverbal behaviors in general. We argue, with her, for a more natural

2

The archaic English coo-eee call is an example.

https://doi.org/10.1017/9781108989275.020 Published online by Cambridge University Press

425

426

PAULINE MADELLA AND TIM WHARTON

account of prosody, which falls out from general pragmatic principles and can also be applied to all the many modes of communication.

19.4 Prosody as Gesture: Toward an Account of Multimodality in Interaction As we have noted, intonation has been described as ranging along a continuum from natural to purely linguistic accounts. As a rigorous (and notorious) defender of the natural view, Bolinger (1983a) notes: Intonation . . . assists grammar – in some instances may be indispensable to it – but it is not ultimately grammatical . . . If here and there it has entered the realm of the arbitrary, it has taken the precaution of blazing a trail back to where it came from.

Adopting this radical position, Bolinger rejects the idea of normal stress and rule-based accent placement, advocated by structuralists, on the basis that accent is only predictable if you are a mind-reader of speakers’ intentions (1972). As summarized in Ladd’s words (1996: 167), Bolinger believed that “[w]hat speakers decide to highlight is . . . a ‘matter of what they are trying to say on a specific occasion in a specific context’. While both accounts regard intonation as meaningful, only the natural view sees intonational meaning as essentially a reflection of speakers’ intentions and speakers’ choices as to what features of an utterance should be made more salient.” Bolinger’s description of a possibly pre-linguistic, almost biological, highlighting function of intonational contours used for the reading of speakers’ mental states and intentions has been disputed (Ladd 1996). However, his description seems to suit an arguably less controversial pre-linguistic, and certainly biological, universal of human communication: pointing (Madella 2021). Indeed, the perception and interpretation of contrastive stress in English seems to be facilitated by its coexistence and interaction with cospeech gestures, including co-pointing gestures, head movements, and facial expressions. Madella (2021, forthcoming) draws on previous accounts of contrastive stress which define it as a natural highlighting device or a vocal form of pointing (Sperber and Wilson 1986/1995; Grice 1989; Wilson and Wharton 2006; Wharton 2009; Scott 2017a, 2017b, 2021), which achieves its effects via the automatic working of the relevance-theoretic comprehension procedure. In this chapter, we take this approach further and argue that prosody needs integrating into a relevance-theoretic analysis of multimodal input. The argument for adopting a natural view of prosody indeed also coincides with a need to approach prosody as one tool in a set of broader gestural ones, and one cue in a set of contextual gestural ones. Madella (2021, forthcoming) makes a case for acknowledging our biological predisposition to use not only prosody and gesture together, but also and more crucially, prosody as gesture. These intimate connections between vocal and

https://doi.org/10.1017/9781108989275.020 Published online by Cambridge University Press

19 Nonverbal Communication and Context

gestural paralinguistic behaviors have been thoroughly investigated by Bolinger himself. In “Where does intonation belong?” (1983a) and “Intonation and gesture” (1983b), not only does Bolinger demonstrate the co-occurring of intonation and gesture, he also, and importantly, emphasizes the overlapping and blending of sounds and movements in one complex of intonational and physical gestures. Reflecting this, he makes a crucial point that we “read intonation the same way as we read gesture” (1983b: 157). This is observed in cases where “nods, hand gestures, and eye contact coincide very precisely with events in the spoken message” (Ladd 1996: 34; Kendon 1972) or what Bolinger (1983c: 98) describes as modalities working in parallel: If intonation is part of a gestural complex whose primitive and still surviving function is – however elaborated and refined – the signalling of emotions and their degrees of intensity, then there should be many obvious ways in which visible and audible gesture are coupled to produce similar and reinforcing effects. This kind of working in parallel is easiest to demonstrate with exclamations. An ah! of surprise, with a high fall in pitch, is paralleled by a high fall on the part of the eyebrows . . . A similar coupling of pitch and head movement can be seen in the normal production of a conciliatory and acquiescent utterance such as “I will” with the accent at the lowest pitch – we call this a bow when it involves the head, but the intonation bows at the same time.

The “bow” phenomenon described by Bolinger can be applied to that of a nodding gesture. A nod involves a head movement, but the intonation can often be seen to nod at the same time, as if to agree, as seen in (4): (4) He \ did

Madella (2021) follows up on this and demonstrates that we read contrastive stress or “vocal pointing” in the same way as we read a pointing gesture. (5) \ He liked it.

In (5), the chin-, finger-, and eye-pointing toward “he” and the use of contrastive stress on “he” is said to coincide. All modalities point in the same direction. This difficulty to unravel intonation contours from their paralinguistic context (Ladd 1996) precisely shows that intonation is an integral part of the paralinguistic context. These multimodalities co-contribute to the speaker’s meaning. This is even more obvious when we look at a case of mismatch, or integrative interaction (Madella 2021), as in (7): (6) Is he trying to lose weight? (7) He \/ was (forward head movement, frown, and grimace; the speaker maintains eye contact; her face shows that this is not the whole story. The use of a fall-rise (“He \/ was”) also indicates that her answer means more than “he is no longer trying.”

https://doi.org/10.1017/9781108989275.020 Published online by Cambridge University Press

427

428

PAULINE MADELLA AND TIM WHARTON

Example (7) provides a good illustration of those fuzzy intuitions about the speaker’s meaning due to the speaker not committing to one particular interpretation (Sperber and Wilson 2015). This second case further shows how modalities are interpreted by virtue of their interaction; thus, they are inferentially interpreted. Encoded concepts carried by individual modes of communication are to be “adjusted or modulated, in the course of the interpretation process” (Sperber and Wilson 2015: 145). A nod and an affirmative statement, “he\was,” have to be understood in relation to the accenting of “was” and the use of the past tense. We suggest that in such cases, attention to prosody can often help the hearer bridge the gap between the linguistic message and the attitude of the speaker. In the above example, the intonational input communicates that there is more to be inferred, and the speaker’s face may lead to more than one possible interpretation: he was but has given up, he was but something went wrong, she is not supposed to tell . . . In English, mismatches are possible between linguistic meaning, an auxiliary verb form marked positively (e.g., “is”), a nodding gesture that apparently shows “yes,” and a frown which comes to confound the apparent positivity of the message. The paralinguistic input will be used to work out how the word is to be interpreted (Stevick 1982). By paying attention to prosodic information, the audience understands that the speaker is disconfirming assumptions, correcting information, and that behind the nodding face disapproval, disagreement, or unhappy feelings are to be found. Madella’s experimental study (2021) shows that this has particular implications for Chinese learners of L2 English, for whom “is” and “yes” can only mean agreement. Learning to integrate paralinguistic input in their interpretation and moving beyond encoded concepts is found to be challenging yet crucial, as it plays a central role in enhancing their oral inferential abilities and in preventing L2 fossilization. Scott (2021) argues that contrastive stress does not encode anything but, rather, disconfirms the hearer’s expectation and acts as a cue to ostension. As such, it prompts and guides his search for extra effects for putting the hearer to more effort as a result of its unexpected pattern. One possibility is sketched below. Consider the utterances in (8). The subtly different shades of meaning of these utterances stem solely from the contrasting position of the emboldened syllable, which is emphatically stressed and pitch prominent. (8) a. I don’t like The Beatles. b. I don’t like The Beatles. c. I don’t like The Beatles.

There are a range of approaches in which the contrastive use of stress, as typified here, is analyzed as part of language proper. These include the postulation of layers of information structure to accommodate it or appeal to syntactically determined “unmarked” and “marked” accent positions (Calhoun 2009; Zubizarreta 2016). Although many of these semantic analyses are intuitively appealing, it is hard to see how what are, in essence, largely descriptive

https://doi.org/10.1017/9781108989275.020 Published online by Cambridge University Press

19 Nonverbal Communication and Context

proposals can be integrated into the kind of cognitively informed account of prosody we are aiming to develop. And there are numerous problems. The unmarked/marked distinction is controversial (Bolinger 1972), and the notion of linguistic defaults is extremely problematic (Sperber and Wilson 1986/1995; Noveck 2004): The default form in one context may not be the default form in another. A pragmatic account based on Relevance Theory requires no intermediate levels of description, falls out naturally from general communicative principles, and can be carried over to a wide range of prosodic phenomena. Since, in English, it is the final strong syllable of an intonation phrase that is accented, a hearer of (8b) or (8c) will have expectations about the prosodic contour disconfirmed. This will automatically lead them to seek an alternative interpretation, which in turn will involve extra processing effort. Given his expectations of relevance, the hearer knows that the speaker would not put him to this extra effort unless some other communicative effects were intended, and so the hearer is prompted to seek out an alternative interpretation. The placement of the accent naturally highlights what these alternatives might be. In (8b) the deaccenting of the syllable /biːt/ results in the preceding strong syllable being naturally highlighted. The extra processing effort the hearer has been put to will result in their forming new hypotheses about the speaker’s intended meaning: in this case, that the speaker is perhaps refuting an accusation that they don’t like The Beatles. As another example of the strength of this kind of account, consider the proposal by Chen and Gussenhoven (2003). They argue that the interpretation of paralanguage is governed by biological codes. An example of one such code is the effort code, which links the amount of energy expended in the production of speech to a range of interpretive effects. An increase in effort may, for example, lead to increased articulatory precision, creating an impression of “helpfulness,” or “obligingness”; or it may result in a wider pitch range, creating an impression of “forcefulness” or “certainty” or conveying affective meanings such as “agitation” or “surprise.” But is clear articulation a natural signal, interpreted by an innately determined biological code? We suggest that it is much better treated as a natural sign of the speaker’s desire to be understood and interpreted via inference rather than decoding. This particular element of prosody belongs under the left-most node (see Figure 19.1). A great deal can be communicated by using clear articulation – that you intend to convey helpfulness, or that you are being obliging, or that you want to convey any one of a wide range of other impressions – but nothing is encoded by it at all. The account developed by Madella (2021, forthcoming) lends further support to Scott’s account (2021) by suggesting that the co-occurring modalities reinforce the presumption of extra effects which, in turn, are retrieved by attending to contrastive stress and how it relates to its co-occurring counterparts. It is precisely because it is never interpreted in isolation but, rather, in the context of its co-occurrence with other modalities that contrastive stress can be said to be a non-encoded phenomenon. It is by virtue of its interaction with

https://doi.org/10.1017/9781108989275.020 Published online by Cambridge University Press

429

430

PAULINE MADELLA AND TIM WHARTON

co-pointing modalities, either pointing in the same direction or to the fact that there is more to infer, that it is inferentially interpreted. It is, as we argue here, one prosodic tool in a larger gestural complex, or what Madella calls prosodic pointing (Madella and Romero-Trillo 2019; Madella 2021; Madella, forthcoming). This forms the biological argument for a multimodal view of communication. Gesture researchers describe those concomitant paralinguistic elements as “parts of a single psychological structure” (McNeill 1985: 351). Gesture researchers (Kendon 1980, 2004, 2014a, 2014b; McNeill 1992; Clark 1996) agree that vocal and visual modalities together form the act of communication, namely the utterance. It is in line with the immediacy assumption (Hagoort and van Berkum 2007: 808), according to which we, communicators, take “all communicative acts, including eye gaze, gestures, smiles and pointing, into consideration” and process them immediately. The immediacy assumption is based on the observation that the mind follows the eye (Hagoort and van Berkum 2007). This argument also resonates with Claudel’s description of total and immediate receptivity in his essay L’oeil ecoute (1946). This is also in line with the integration of paralanguage into a general view of communication. For example, conversation analysts Streeck et al. (2011: 7) describe multimodal interaction along these lines: “Talk and embodied behaviours occur as interdependent phenomena, not separable modes of communication.” It follows that it is by virtue of their interaction that those interlinked verbal and nonverbal behaviors constrain and guide the hearer in reaching an interpretation. It is also by virtue of their interaction that encoded concepts carried by individual modes of communication are to be “adjusted or modulated, in the course of the interpretation process” and for the purpose of that one particular interpretation (Sperber and Wilson 2015: 145). An inclusive definition of paralinguistic phenomena is also and precisely what allows us to argue for an inferential model of (verbal and nonverbal) communication from the perspective of Relevance Theory.

19.5 Conclusion Wharton (2009: 1) begins as follows: Sentences are rarely uttered in a behavioural vacuum. We colour and flavour our speech with a variety of natural vocal, facial and bodily gestures, which indicate our internal state by conveying attitudes to the propositions we express or information about our emotions or feelings. Though we may be aware of them, such behaviours are often beyond our conscious control: they are involuntary or spontaneous. Almost always, however, understanding an utterance depends to some degree on their interpretation. Often, they show us more about a person’s mental/physical state than the words

https://doi.org/10.1017/9781108989275.020 Published online by Cambridge University Press

19 Nonverbal Communication and Context

they accompany; sometimes, they replace words rather than merely accompany them.

The statement is not less true now than it was then. People working in pragmatics cannot, as many theoretical linguists do, abstract away from these behaviors. In this chapter we have introduced the major role nonverbal modes play in communication and, in so doing, attempted to promote a “multimodal” view of human communication and show how certain behaviors guide our interlocutors to our intended meaning by constraining the construction of context. We have answered three questions: (1) What is multimodality? (2) How do nonverbal behaviors communicate? (3) How can we move toward a principled pragmatic account of multimodality in interaction? We argue that a pragmatic account based on Relevance Theory can be carried over to a wide range of multimodal phenomena, requires none of the intermediate levels of description that necessarily form part of less pragmatic accounts, and falls out naturally from general communicative principles. There are many different accounts of multimodality (Kress and van Leeuwen 2001; Forceville 2021), but the blade of Occam’s Razor, we argue, favors the one presented here.

References Abercrombie, D. (1968). Paralanguage. International Journal of Language & Communication Disorders, 3(1), 55–59. Birdwhistell, R. L. (1970). Kinesics and Context: Essays on Body Motion and Communication. Philadelphia, PA: University of Pennsylvania Press. Blakemore, D. (1987). Semantic Constraints on Relevance. Oxford: Blackwell. Blakemore, D. (2002). Relevance and Linguistic Meaning: The Semantics and Pragmatics of Discourse Markers. Cambridge: Cambridge University Press. Bolinger, D. (1964). Around the edge of language: Intonation. Harvard Educational Review, 34, 282–296. Bolinger, D. (1972). Accent is predictable (if you’re a mind-reader). Language, 48(3), 633–644. Bolinger, D. (1983a). Where does intonation belong? Journal of Semantics, 2(2), 101–120. Bolinger, D. (1983b). Intonation and gesture. American Speech, 58(2), 156–174. Bolinger, D. (1983c). The inherent iconism of intonation. In J. Haiman (ed.), Iconicity in Syntax (pp. 97–109). Amsterdam: John Benjamins. Brazil, D. (1975). The Communicative Value of Intonation in English. Birmingham: University of Birmingham. Calhoun, S. (2009). What makes a word contrastive: Prosodic, semantic and pragmatic perspectives. In D. Barth-Weingarten, N. Dehé, and A. Wichmann (eds.), Where Prosody Meets Pragmatics: Research at the Interface (pp. 53–78). Bingley: Emerald.

https://doi.org/10.1017/9781108989275.020 Published online by Cambridge University Press

431

432

PAULINE MADELLA AND TIM WHARTON

Chen, A., and Gussenhoven C. (2003). Language-dependence in signalling of attitude in speech. In N. Suzuki and C. Bartneck (eds.), Proceedings of Workshop on the Subtle Expressivity of Emotion, CHI 2003 Conference on Human and Computer Interaction. Clark, B. (2007). Blazing a trail: Moving from natural to linguistic meaning in accounting for the tones of English. In R. A. Nilsen, N. A. A. Amfo, and K. Borthen (eds.), Interpreting Utterances: Pragmatics and Its Interfaces: Essays in Honour of Thorstein Fretheim (pp. 69–81). Oslo: Novus. Clark, B., and Lindsey, G. (1990). Intonation, grammar and utterance interpretation. University College London Working Papers in Linguistics, 2, 32–51. Clark, H. (1996). Using Language. Cambridge: Cambridge University Press. Claudel, P. [1946] (1990). L’oeil ecoute. Collection Folio essais (127), Gallimard Education. Ekman, P. (1999). Emotional and conversational nonverbal signals. In L. Messing and R. Campbell (eds.), Gesture, Speech and Sign (pp. 45–57). Oxford: Oxford University Press. Escandell-Vidal, V. (1998). Intonation and procedural encoding: The case of Spanish interrogatives. In V. Rouchota and A. Jucker (eds.), Current Issues in Relevance Theory (pp. 169–203). Amsterdam: John Benjamins. Escandell-Vidal, V. (2002). Echo-syntax and metarepresentations. Lingua, 112(11), 871–900. Forceville, C. (2021). Multimodality. In W. Xu and J. R. Taylor (eds.), The Routledge Handbook of Cognitive Linguistics (pp. 676–687). New York/London: Routledge. Fretheim, T. (2002). Intonation as a constraint on inferential processing. In B. Bel and I. Marlien (eds.), Speech Prosody 2002: Proceedings of the First International Conference on Speech Prosody, Aix-en-Provence, 11–13 April 2002 (pp. 59–64). Aixen-Provence: Laboratoire Parole et Langage. Fridlund, A. (1994). Human Facial Expression: An Evolutionary View. San Diego: Academic Press. Gigerenzer, G., Todd P., and the ABC Research Group (1999). Simple Heuristics That Make Us Smart. Oxford: Oxford University Press. Grice, H. P. (1957). Meaning. Philosophical Review, 66, 377–388. Grice, H. P. (1989). Studies in the Way of Words. Cambridge, MA: Harvard University Press. Gussenhoven, C. (1984). On the Grammar and Semantics of Sentence Accents. Dordrecht: Foris. Gussenhoven, C. (2002). Intonation and interpretation: phonetics and phonology. In B. Bel and I. Marlien (eds.), Speech Prosody 2002: Proceedings of the First International Conference on Speech Prosody, Aix-en-Provence, 11–13 April 2002 (pp. 47–57). Aix-en-Provence: Laboratoire Parole et Langage. Gussenhoven, C. (2006). Semantics of prosody. In K. Brown (ed.), Encyclopedia of Language and Linguistics, 2nd ed. (Vol. XI, pp. 170–172). Oxford: Elsevier. Hall, A. (2004). The meaning of but: a procedural reanalysis. UCL Working Papers in Linguistics, 16, 199–236. Halliday, M. (1963). Explorations in the Function of Language. London: Arnold. Halliday. M. (1967). Intonation and Grammar in British English. The Hague: Mouton. Hagoort, P., and Berkum, J. van (2007). Beyond the sentence given. Philos. Trans. R. Soc. B, 362, 801–811. Hauser, M. (1996). The Evolution of Communication. Cambridge, MA: MIT Press.

https://doi.org/10.1017/9781108989275.020 Published online by Cambridge University Press

19 Nonverbal Communication and Context

Hirschberg, J., and Ward, G. (1995). The interpretation of the high-rise question contour in English. Journal of Pragmatics, 24, 407–412. Hooff van, J. (1972). A comparative approach to the phylogeny of laughter and smiling. In R. Hinde (ed.), Non-verbal Communication (pp. 209–238). Cambridge: Cambridge University Press. House, J. (1990). Intonation structures and pragmatic interpretation. In S. Ramsaran (ed.), Studies in the Pronunciation of English (pp. 38–57). London: Routledge. House, J. (2006). Constructing a context with intonation. Journal of Pragmatics, 38(10), 1542–1558. House, J. (2007). The role of prosody in constraining context selection: A procedural approach. Nouveaux Cahiers de Linguistique Française, 28, 369–383. Imai, K. (1998). Intonation and relevance. In R. Carston and S. Uchida (eds.), Relevance Theory: Applications and Implications (pp. 69–86). Amsterdam: John Benjamins. Kahneman, D. (2011). Thinking Fast and Slow. New York: Farrar, Straus and Giroux. Kendon, A. (1972). Some relationships between body motion and speech. In A. Siegman and B. Pope (eds.), Studies in Dyadic Communication (pp. 177–210). New York: Pergamon Press. Kendon, A. (1980). Gesticulation and speech: Two aspects of the process of utterance. In M. R. Key (ed.), The Relationship of Verbal and Nonverbal Communication (pp. 207–227). The Hague: Mouton and Co. Kendon, A. (2004). Gesture: Visible Action as Utterance. New York: Cambridge University Press. Kendon, A. (2014a). The poly-modalic nature of utterances and its relevance for inquiring into language origins. In D. Dor, C. Knight, and J. Lewis (eds.), The Social Origins of Language (pp. 67–76). Oxford: Oxford University Press. Kendon, A. (2014b). Semiotic diversity in utterance production and the concept of “language.” Philos. Trans. R. Soc. Lond. B Biol. Sci., 369(1651), 20130293. doi: 10.1098/rstb.2013.0293. König, E. (1991). The Meaning of Focus Particles: A Comparative Perspective. London: Routledge. Kress, G., and Leeuwen, T. van (2001). Multimodal Discourse: The Modes and Media of Contemporary Communication. Oxford: Oxford University Press. Ladd, R. (1978). The Structure of Intonational Meaning. London: Indiana University Press. Ladd, R. (1996). Intonational Phonology. Cambridge: Cambridge University Press. 2nd ed. 2008. Ladd, R. (2008). Intonational Phonology, 2nd ed. Cambridge: Cambridge University Press. Madella, P. (2021). Prosodic pointing: From pragmatic awareness to pragmatic competence in Chinese hearers of L2 English. Ph.D. thesis, University of Brighton. Madella, P. (forthcoming). Relevance and multimodal prosody: implications for L2 teaching and learning. In T. Wharton, D. Maillat, C. Jagoe, and K. Scott, Relevance in Mind, Front. Psychol. Madella, P., and Romero-Trillo, J. (2019). Prosodic pointing in inferential comprehension: The application of relevance theory to L2 listening instruction. Letrônica, 12(4), 1–17. McNeill, D. (1992). Hand and Mind. Chicago: University of Chicago Press.

https://doi.org/10.1017/9781108989275.020 Published online by Cambridge University Press

433

434

PAULINE MADELLA AND TIM WHARTON

McNeill, D. (1985). So you think gestures are nonverbal? Psychological Review, 92(3), 350–371. McNeill, D., ed. (2000). Language and Gesture. Cambridge: Cambridge University Press. Mehrabian, A. (1971). Silent Messages: Implicit Communication of Emotion and Attitudes. Belmont, CA: Wadworth. Noveck, I. A. (2004). Pragmatic inferences related to logical terms. In I. A. Noveck and D. Sperber (eds.), Experimental Pragmatics (pp. 301–321). London: Palgrave Macmillan. O’Connor, J., and Arnold, G. (1973). Intonation of Colloquial English. Harlow: Longman. Pierrehumbert, J., and Hirschberg, J. B. (1990). The meaning of intonational contours in the interpretation of discourse. In P. Cohen. J. Morgan, and M. Pollack (eds.), Intentions in Communication (pp. 271–311). Cambridge MA: MIT Press. Poyatos, F. (1983). New Perspectives in Nonverbal Communication. Oxford: Pergamon Press. Riley, P. (1979). Viewing comprehension: L’oeil écoute. In The Teaching of Listening Comprehension: Papers Presented at the Goethe Institut Colloquium Held in Paris in 1979. London: The British Council. Scott, K. (2017a). Ostension, Expectations and Non-encoded Meaning [Presentation]. International Pragmatics Association (IPrA) Conference, Belfast, UK. Scott, K. (2017b). Prosody, procedures and pragmatics. In I. Depraetere and R. Salkie (eds.), Semantics and Pragmatics: Drawing a Line. Logic, Argumentation & Reasoning, 11 (pp. 323–341). Berlin: Springer. Scott, K. (2021). Contrastive stress in English meaning, expectations, and ostension. In E. Ifantidou, L. de Saussure, and T. Wharton (eds.), Beyond Meaning, Pragmatics & Beyond New Series 324 (pp. 29–42). Amsterdam: John Benjamins. Sperber, D. and Wilson, D. (1986/1995). Relevance: Communication and Cognition. Oxford: Blackwell. Sperber, D., and Wilson, D. (2015). Beyond speaker’s meaning. Croatian Journal of Philosophy, 15(44), 117–149. Stevick, E. W. (1982). Teaching and Learning Languages. Cambridge: Cambridge University Press. Streeck, J., Goodwin, C., and Lebaron, C. (2011). Embodied Interaction: Language and Body in the Material World. Cambridge: Cambridge University Press. Trager, G. L. (1958). Paralanguage: A first approximation. Studies in Linguistics, 13, 1–12. Vandepitte, S. (1989). A pragmatic function of intonation. Lingua, 79, 265–297. Ward, G., and Hirschberg, J. (1988). Intonation and propositional attitude: The pragmatics of L*+H L H%. In Proceedings of the Fifth Eastern States Conference on Linguistics (pp. 512–522). Columbus: Ohio State University. Wharton, T. (2003a). Interjections, language and the “showing”/”saying” continuum. Pragmatics and Cognition, 11(1), 39–91. Wharton, T. (2003b). Natural pragmatics and natural codes. Mind and Language, 18(5), 447–477. Wharton, T. (2009). Pragmatics and Non-verbal Communication. Cambridge: Cambridge University Press. Wichmann, A. (2002). Attitudinal intonation and the inferential process. In B. Bel and I. Marlien (eds.), Speech Prosody 2002: Proceedings of the First International Conference

https://doi.org/10.1017/9781108989275.020 Published online by Cambridge University Press

19 Nonverbal Communication and Context

on Speech Prosody, Aix-en-Provence, 11–13 April 2002 (pp. 11–16). Aix-en-Provence: Laboratoire Parole et Langage. Wilson, D., and Carston, R. (2019). Pragmatics and the challenge of “non-propositional” effects. Journal of Pragmatics, 145, 31–38. Wilson, D., and Sperber D. (1993). Linguistic form and relevance. Lingua, 90(1), 1–25. Wilson, D., and Sperber, D. (2004). Relevance theory. In L. R. Horn and G. Ward (eds.), The Handbook of Pragmatics (pp. 607–632). Oxford: Blackwell. Wilson, D., and Wharton, T. (2006). Relevance and prosody. Journal of Pragmatics, 38, 1559–1579. Zubizarreta, M. L. (2016). Nuclear stress and information structure. In C. Féry and S. Ishihara (eds.), The Oxford Handbook of Information Structure (pp. 163–184). Oxford: Oxford University Press.

https://doi.org/10.1017/9781108989275.020 Published online by Cambridge University Press

435

20 AI, Human–Robot Interaction, and Natural Language Processing Ian McLoughlin and Nitin Indurkhya You can’t connect the dots looking forward; you can only connect them looking backwards. – Richardson and Arthur (2013)

20.1 Introduction If I speak to a computer dialogue system such as Apple’s Siri, or Tencent’s Xiaowei, I am, to some extent, expecting it to understand my utterance much as a human would. In future, these systems will probably respond as well as a human would (and may even transcend beyond that to responding better than a human, whatever that entails), but at the time of writing there is a gap, an uncanny valley (Mori et al. 2012), between human and computer dialogue. In the past, the gap was blamed on the inability of automatic speech recognition (ASR) technology to accurately identify the words being uttered. Conversational speech word error rates (WER) improved from about 80% in 1996 to around 40% in 2003 and maybe 25% in 2010, with human WER being around 4%.1 Fortunately the power of machine learning, in particular deep neural networks (McLoughlin 2016), has driven computer ASR performance to reach human-like levels of performance in the 2020s. So if ASR systems can recognize speech as well as humans, the question remains as to why the experience of talking to speech dialogue systems does not currently match that of human-to-human conversation. Much of the answer relates to context, and much of that is the technology domain of Natural Language Processing (NLP). Imagine a simple utterance to a smart home dialogue system Tell Jane I will be late. A modern ASR system might find the captured (or recorded) sound 1

All of these are approximate figures assuming clear English speech of general vocabulary, uttered by an average speaker in a quiet background. More on the interrelationship between vocabulary, noise, and speaker can be found in McLoughlin and Sharifzadeh (2008).

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

20 AI, Human–Robot Interaction and NLP

relatively easy to convert to text. But the scale of the technological difficulties faced by a dialogue system are immense. These might include; • • • •

Who am I? Who is Jane, or which Jane? What event will I be late for? How should this be communicated to Jane?

Furthermore, we must consider that emphasis placed on parts of the sentence such as “Tell,” “I,” or “will be” can subtly alter the meaning, and it is of course all context sensitive to some degree. Human communication appears simple to the average speaker but is actually rather deep and complex. For further reference, Crystal (2003) provides a fascinating overview of it all. The authors believe that as computational dialogue systems become more popular and prevalent, we will begin to see evidence that humans are adapting their speech and language to communicate more effectively (just as we all do in any new context in which we are faced with communications difficulties). However, at the time of writing, these technologies are fortunately still nascent and evolving, and therefore expected to adapt to us rather than vice versa. In this chapter we will examine the challenges that these technologies need to overcome, specifically at four very different context scales. The extent to which these challenges are met over the next decade will very much define whether or not we can all look forward to JARVIS2-like computational assistants in future.

20.2 Front-End and Audio and Speech Processing When we interact with a computational speech dialogue system, whether that is a smart home, vehicle assistant, robot, or automated call center, there is a general structure to such technology (Baker et al., 2009). Figure 20.1 illustrates this from a high-level systems perspective. We will separately examine the speech input and speech output parts of the front end and then specifically consider the role and levels of context.

20.2.1 Speech Input Speech from a user (on the right in Figure 20.1) impinges on a microphone, or increasingly often, a set of microphones – since more microphones generally mitigates better against interfering noise. The electrical signal from the microphone transducers is amplified and then converted to the digital domain

2

Just A Rather Very Intelligent System, a view of what the smart home of the future could become, from Marvel’s Iron Man.

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

437

438

IAN MCLOUGHLIN AND NITIN INDURKHYA

external information Dialogue System

loudspeaker

NLP

TTS

DAC

ASR

ADC

LM

Digital Analogue domain domain

local data & memory Semantics Utterances Dialogue & meta-level

Words NLP

microphone amplifier

Phones Sounds Front end

Figure 20.1 High-level diagram of a dialogue system showing major components and context level.

through an analogue-to-digital converter (ADC). In the digital domain it is represented as one vector of audio samples per transducer, and typically blocked into a set of samples – a frame – which is 10 to 30 ms long. Frames are then converted into representations such as mel frequency cepstral coefficients (MFCCs) or perceptual linear prediction (PLP) coefficients (McLoughlin 2016), and clumped together with a few preceding and following frames into an acoustic feature vector which is ready for input to an ASR system. ASR architectures typically rely on an internal, pre-trained understanding, encoded in an acoustic model and language model (LM). This helps to convert input feature vectors into sub-phonetic units (e.g., senones) (Jin et al. 2018). These are then strung together into phones and, with the help of a dictionary, into words. The conversion from features to senones is generally a classification process in which the underlying sounds captured in those features are matched to the most likely senone that they encode – although there is a lot of variability in the research literature, and many different approaches exist to do this (Cai et al. 2018). Senones, sub-phonetic units, phones, or whatever the intermediate unit of speech used in a particular system, are then gathered into token strings to be formed into words. While the printed page conveniently places a space between words, human speakers don’t always leave nice gaps between words. Indeed, some words might have longer gaps within them than between them. It is therefore sometimes difficult to detect whether a particular subphonetic unit at the end of one word should instead belong at the beginning of the next. Furthermore, the individual sub-word units from the first stage of the ASR system often contain one or more errors, meaning that the information we are working with is inherently imperfect. Systems may need to look at the second or even the third highest probability senones too. The task needed to form the series of tokens into words is now a sequencing problem. It aims to produce the sequence which has the highest probability of being correct. A simple example would be the token string he-ee-lo-w-wo-or-rl-d being formed into the words “hello world.” Most ASR systems do not make use of

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

20 AI, Human–Robot Interaction and NLP

written grammar rules because speakers do not comply to such rules – if they do they begin to sound more like Mr Spock3 than like a normal human. However, there are still some constraints on spoken grammar, including physiological restrictions on speech formation.4 These restrictions can be useful in ASR. Beyond grammar and physiology, if systems are able to be trained for a particular user, they would be able to exploit knowledge of user-specific characteristics, so that the highest probability senone and word choices are those which are most likely for the current speaker. The same input tokens for a different speaker might result in a different set of words and would almost certainly be spoken in a different way. The ability to tune to a particular user can very significantly improve ASR performance in the real world. At this point in our description of ASR, the system has hopefully managed to convert spoken language into text; just like the text on this page. We will resume this discussion in Section 20.2.3, but first we consider the logical opposite of ASR.

20.2.2 Speech Output A text-to-speech (TTS) system operates in the reverse direction, taking a list of words and then converting them to speech. There are a number of TTS methods of varying quality and flexibility which we will not discuss here (Taylor 2009). However, the structure in Figure 20.1 outlines the general principle that a list of words is input to a TTS block which in turn produces a sequence of samples which, when decoded through a digital-to-analogue converter (DAC), amplified and replayed through a loudspeaker should sound like intelligible speech. Like the ASR block, TTS makes use of language rules and possibly grammar models to replay speech. Rules and models are important because there is quite a lot of nontrivial knowledge beyond simply pronunciation, which affects how something is spoken. For example, in English, “€124” should be read as “one hundred and twenty-four Euro” rather than “Euro one hundred and twenty-four.” Likewise for times (e.g., “12.05pm”), and many more common abbreviations, such as “Oxford St.” (“St.” read as “street”). Furthermore, in many languages, notoriously so for English, pronunciation can be irregular (e.g., proper nouns like “Darby,” which is pronounced “dar-bee,” “Leicester,” which is pronounced “less-ter,” and even “Cholmondeley,” which is pronounced “chum-lee”5). Pronunciation dictionaries, exception dictionaries, and grammar rules can all aid the TTS system in avoiding strange and sometimes hilarious attempts at speech recreation. Figure 20.1 shows the TTS and ASR systems sharing 3 4 5

A reference to a character from Star Trek, original series by Gene Rodenberry, 1966–1969. For example, the need to draw breath. Serious language research texts should use the international phonetic alphabet (IPA), which is invariant to personal quirks of pronunciation; however, few people can read IPA script fluently, and hence we avoid it here for simplicity of discussion.

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

439

440

IAN MCLOUGHLIN AND NITIN INDURKHYA

a language model. This is correct in a logical sense, since they both share the same language, but practical systems tend to keep ASR and TTS separate, and are unlikely to share models.

20.2.3 Speech Contains More Than Words The description of ASR and TTS given above presents a realistic, but not full picture of what is happening at the front end of a dialogue system. One realworld consideration we have skipped is the fact that human speech is not simply a sequence of letters and words. Speech is spoken with emphasis, tone, feeling, and a sprinkling of non-speech sounds. All of this is termed prosody, and it is a very important part of human-to-human speech communication (not least in that it is also the aspect of speech which personalizes it – makes it mine and not yours). A TTS system that speaks in a droning monotone is unlikely to succeed commercially, and that is why the best systems are given a kind of personality through the way they speak. Conversely though, if a computer captures my speech and stores it as words on a page, it will have lost the emphasis and emotion. Systems clearly need to work with more than just words and letters.6 Suffice it to say at this point that there is a lot more to speech than either audio samples, phonetic strings, or written words (Crystal 2003). AI speech dialogue systems, robots that communicate by speech, and technology that processes speech must, if it is to work well in real-world settings, take good account of this. One way to describe the difference between a purely mechanical, phonetic, or word-level description of speech and the spoken voice in all of its fullness, is that these differences are all elements of context.

20.2.4 The Role of Context Context generally widens from the front end, through ASR, NLP, and into the full dialogue system. The widening encompasses both scope and temporal duration. The most natural way to discuss this will be to start at the very front of the front end; the human. The sounds of speech have a complex origin (McLoughlin 2016) but ultimately are controlled by human musculature, and thus they are sequenced at a rate determined by muscle motion. Digital speech systems sample speech waveforms at a rate of 8 kHz for low audio quality systems to 16 kHz for high audio quality (by contrast, high fidelity audio is sampled at 44.1 kHz). Human speech sampled by a 16-bit ADC at 16 kHz and then replayed through a matching DAC would be indistinguishable to humans from the input.7 As noted previously, recorded speech is represented within a computer by 6

7

One attempt to encode prosody is the use of speech synthesis markup language (SSML), although this, and the whole aspect of prosody in speech technology, is still very much research that is in progress. For further information on speech and sampling, refer to chapters 1 and 3 of McLoughlin (2016).

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

20 AI, Human–Robot Interaction and NLP

a vector of individual samples, captured at the sampling rate (e.g., a sequence of 16-bit integers). Samples are grouped together into frames for analysis, transmission, and storage. These frames represent the duration over which speech is considered to be pseudo-stationary, i.e., its frequency characteristics are unchanging, which is typically around 20 to 30 ms. The lowest level of context in computer speech processing is probably a frame of speech (i.e., a short vector of sequential samples) and serves as the basic unit of most speech systems. A frame can contain a sub-phonetic unit, a small part of a phoneme, silence, non-speech sounds, or a part of a sustained utterance. Frames are then generally converted to a short vector of features such as MFCCs, as noted earlier. A sequence of frame features strung together can make a sub-phonetic unit, such as a senone (Ferrer et al. 2015). A sequence of senones can make a phoneme, a sequence of phonemes can make a word, a sequence of words makes an utterance, multiple utterances make a conversation, and so on. Sample-to-sample context is captured in a speech frame, as noted above. Frames are often overlapped, meaning that frame-to-frame correlation is extremely high, but even without overlap, context between frames is what makes up sub-phonetic and then phonetic units. Machine learning systems working on frame-based features often explicitly capture context by combining frame feature vectors together. For example, when analyzing or processing one speech feature, it is normal to concatenate the previous and subsequent few frames, e.g., a context size of eleven frame features is arranged as 5 − 1 − 5 where the current frame feature is the center one (Jiang et al. 2014). Shifted delta cepstra (Kohler and Kennedy 2002) is a method used to increase context size in speech systems and may span a significantly wider context than eleven frames8. Context is often exploited within machine learning architectures too, as well as represented in the features. Long Short-Term Memory (LSTM), Recurrent Neural Networks (RNN), and many novel variants of both, maintain an internal context (Greff et al. 2016). This is designed to exploit feature-tofeature and even feature group-to-feature group correlations (and hence context). In many cases, the additional context that LSTM provides can yield significant performance boost (Miao et al. 2021). The LSTM and RNN units will also be discussed in the next section. Machine learning systems handling input speech often have a two-stage architecture, where the first stage transforms features to sub-phonetic units or tokens, and the second stage transforms these to a final target. The target differs depending on the language used, for example Mandarin Chinese systems typically output characters, whereas English systems may output phones (auditory representations of phonemes) or words. 8

If a frame is represented by n features, each of length m, then the system input might be a matrix of dimension [n,m], or perhaps a vector of length n×m. Typical values may be [11, 36] or [21, 24].

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

441

442

IAN MCLOUGHLIN AND NITIN INDURKHYA

The sequence matching task to arrange tokens into words or utterances is almost always probabilistic and makes heavy use of context. It seeks the most likely arrangement given rules of language, grammar, what has gone before, and what is coming next – best performance is obtained by considering lookahead and look-behind, aided by the use of hidden Markov models and the Viterbi algorithm (Frankel and King 2001). The output of the front-end input pathway, and the input to the front-end output pathway, is a sequence of words plus (hopefully) prosodic markup information. Considering just the input pathway, we have noted quite a lot of context being exploited so far: • • • • •

Sample-to-sample context within an auditory frame feature. Frame-to-frame context within a senone. Senone-to-senone context within a phonetic unit (e.g., a phone). Phone-to-phone context within a syllable or character. Perhaps also syllable-to-syllable context within a word.

The reverse direction of text-to-speech likewise makes use of the same levels of context to ensure that natural-sounding speech results. Now we have words, perhaps with prosodic markup, it is then the role of NLP to make some sense of those words; to actually do something interesting with them . . . and that requires even more contextual awareness than we have identified so far.

20.3 Natural Language Processing (NLP) An important reason for converting speech into words, besides making it human-readable, is that this enables the use of methods in the field of NLP. The field is as old as computing itself, and many of the earliest attempts at building intelligent systems involved some form of language processing. This has been a focus of Artificial Intelligence since the beginning. For example, to pass the Turing test (Turing 2009) a computer needs to behave in a manner that is indistinguishable from a human being – with all interactions to be done in written natural language. In 2011, when IBM’s Watson program convincingly won the Jeopardy tournament, it was seen as a major triumph for NLP. In general, NLP methods take written text as input, and this fact alone adds an interesting context to the task, since a written language is characterized by its script. There are thousands of languages in the world, and hundreds of scripts. Many languages tend to share scripts. For example, the Romance languages (the major ones being French, Italian, Portuguese, and Spanish) use script derived from Latin. Most research in NLP is done with English, the lingua franca of the research community, although the methods and technologies developed for English can often be applied to other languages. But

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

20 AI, Human–Robot Interaction and NLP

inevitably, there are subfields of NLP for different languages. For example, Chinese Language Processing, Korean Language Processing, etc. When dealing with input to speech systems, it is usually possible to identify the language at an acoustic level (Jin et al. 2018), or from the characteristics of the string of words or characters itself. Identifying, or confirming, language is a critical step because without this context the words cannot be effectively processed further. Even the task of finding words can be quite different depending on the language. This procedure can be extremely dynamic because in many countries it is not uncommon to mix languages during communication. In Asian countries, for example, casual conversation has a mix of languages even within the same sentence – referred to as code-switching (Çetinoğlu et al. 2016). The computational aspects of NLP borrow heavily from the paradigms in the more established and broader field of linguistics where the study of language is broken down into a sequence of stages such that the output of one stage becomes the input of the next. Two important stages are syntax and semantics. One key stage at the start of any NLP system, which relies heavily on contextual information, is morphological analysis. This describes how words should be understood. For instance, the word bores can be understood as “bore” with the plural tag, but this would only be the case if the context of the word suggested it was being used as a noun. In a different context the same word would have a completely different meaning. While human beings are able to disambiguate quite routinely, there are times when words are confused, because the input is deliberately intended as such, for humor or for the sake of art. So at the generative end, an automated system needs to be equally careful about any such ambiguity and consider whether it should be included or not. For example, in dealing with a customer, who is classified as “stressed” based on the ongoing conversation, might not a chatbot be more effective if it made a different choice of words than it normally would? In some languages, such as Japanese, the morphology can change depending on the level of familiarity, the social standing, and even the age difference between the participants in the conversation. For a professional one-off conversation one can easily choose the “safest” level (most polite) for a robot to use. But imagine a robot as a social companion, or a chatbot in a social-service setting with interactions across a range of age groups and cultures. Can the context help the robot figure out the most appropriate vocabulary and communication style? Generally speaking, though, assuming the lexical/morphological issues are sorted out and words are clearly identified, the role of syntax is to add a layer of understanding to the sequence of words. Language is temporal and the sequence of words follows a timeline. While considerable effort was expended in predicting future words based on past words (with a window of past words being used as context for the prediction), dramatic improvements have recently been made by allowing future words to be included in the context for the purpose of building the model. Depending on the application, the future words may or may not be available at model deployment time. For

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

443

444

IAN MCLOUGHLIN AND NITIN INDURKHYA

simultaneous translation (i.e., in real time), a window of future words might be made available by allowing the translation to lag slightly behind. Syntactic analysis in NLP closely follows the linguistic model where the sequence of words is first augmented with part-of-speech labels and then parsed into structures that highlight the relationships between the words. Earlier systems used tree representations, whereas more recent ones are graphs – enabling better semantic processing, which is the next stage in NLP. Graphs of entities, that the words represent, can be used to retrieve more relevant information using contextual cues (Voskarides et al. 2018). If an AI system or computer can be deemed to have understood some input text via NLP, it typically involves the system demonstrating the meaning of the text. In linguistics this is known as semantics. The reader can refer to the earlier chapters in Part III of this handbook for a more detailed description of semantics and the issues it involves. Here, we shall instead focus on how AI systems involved in human–robot communication might operationalize semantics and the role context plays in this process. A key driver for this is the availability of large databases of text and speech that have enabled the development of pure data-driven approaches to directly acting on the meaning of text. The dominance of data-based approaches can be somewhat attributed to the increasingly powerful role of the Internet in facilitating communications, not just between humans and robots but also between human beings. Digital communications can be recorded, saved, replayed, and used by machine learning algorithms to build models that can participate in meaningful (and useful) ways in similar communications. Hence, we can have chatbots that appear to understand what a human being says or types – the evidence of the understanding is in the appropriateness of the response given and its linguistic form (Bender and Koller 2020). How is the availability of data making all this possible? Modern neural language models require extremely large amounts of data to learn and generalize from. These are able to encode (and replay) very complex context-dependent relationships between words. The architectures have mechanisms specifically tailored to the sequential nature of language and customized for generalizing on arbitrarily large contexts of words. The models are typically classification models — with an extremely large number of classes (the entire vocabulary, in the case of an open response generation) or a smaller set of fixed classes (for instance, a set of sentiments, in the case of sentiment analysis). The earliest models were based on Recurrent Neural Networks (RNN) to capture context within hidden units, essentially as compressed representations of the entire sequence up to that point. The output of each hidden unit could be seen as an abstract view of the preceding subsequence. More complex models then used layers of such hidden units (essentially viewing the sequence at varying levels of abstraction) and also considered looking into the future. Because of the compression, longdistance relationships can be outweighted by local ones, and in such scenarios another architecture involving special units, referred to as Long Short-Term Memory (LSTM) unit, is helpful. A major advantage of an LSTM-network

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

20 AI, Human–Robot Interaction and NLP

over an RNN is that LSTMs allow better management of context, since anything that is not useful can be removed more easily. This helps the LSTM network adapt better to a variety of scenarios. A more efficient version of LSTM is the Gated Recurrent Unit (GRU). It essentially combines the best of the LSTM and the RNN units. A fundamental issue that remains is that the context information is processed sequentially, so the training process can be extremely slow. In the last few years, a new type of architecture, referred to as Transformers, has revolutionized the use of Neural Language Processing (Hu and Singh 2021). The basic idea behind a transformer is to eliminate the recurrent connections that enforced the sequential training of networks involving RNN or LSTM units. Units are simply stacked in layers with customized connections between the layers. The entire network is highly parallel, both for execution and training. A key aspect of Transformers is a self-attention layer (Miao et al. 2021). The units in this layer allow the network to directly capture and store context information (of arbitrary length). The attention units can be compared to each other and the best ones can be kept. Tasks such as co-reference resolution, where the context can be quite long range as the referent of a pronoun may be far away, are especially suited to this sort of architecture (Joshi et al. 2020). Data-based NLP approaches are not without their problems (Akiwowo et al. 2020) – hateful speech, misinformation, privacy issues have been just some of the, often unexpected, issues witnessed even with big pre-trained models (Carlini et al. 2020). Some of the issues can be attributed to lack of quality control and to inadequate/poor sampling. But a more fundamental issue here is that the data is viewed as providing all of the context for the task of understanding. When this is not the case, problems arise. Throwing more data into the mix won’t solve these problems; it will only make them worse. A general issue with most data approaches is their inability to integrate human insights about language. Important extensions such as intention and grounding to the real-world simply cannot be done simply from samples of language usage (Thomason et al. 2016). The issue is often mistakenly seen as one of lack of the right data rather than the wrong architecture for solving these problems. One key reason for the popularity of data-based approaches is that the availability of data enables a clear way to evaluate a system and compare it to others. Without data, evaluation in language-processing systems is typically expensive and unreliable, especially in systems where some form of automatic generation is involved (Clark et al. 2021). However, many NLP tasks are too subjective to evaluate. A good example is argumentation. In its most basic form, it is seen in stylized formal debates, often between two parties. In a more implicit way, argumentation is part of all conversations as the exchange of ideas is a social activity. There are very good economic reasons for being interested in understanding and automating argumentation. It underpins almost all advertising, public policy creation, education, and law. A system that participates in an argument or debate needs to be able to do several tasks, few of which are grounded in right-or-wrong or true-and-false. The system

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

445

446

IAN MCLOUGHLIN AND NITIN INDURKHYA

must be able to identify the relevant arguments in a debate and detect the stance and their quality. In crafting a response, it must organize the arguments into a compelling narrative that is effective in the ongoing argument. This requires the use of competitive debating methods such as rhetoric and framing (to favor its position), and it has commonalities with game-playing where the objective is not to optimize but to just be better than one’s opponent, sometimes by identifying and exploiting their strategic weaknesses. Each of these skills can be data-based. For example, the ability to take a set of arguments and craft a narrative is abstractly a summarization task. The important caveat is that the tasks must be done with a high degree of correctness for the debate to work properly. If the model identifying an opponent’s stance has low accuracy, the system may end up arguing in support of the opponent! Furthermore, all these tasks must fit together in a manner that is adaptive and experiential. Basically, the debater system must develop its own style and evolve within the context of the specific debates it encounters (Slonim et al. 2021). At the time of writing, by far the most analyzed NLP system is GPT-3 (General Pre-trained Transformer) from OpenAI (Brown et al. 2020). Released in 2020, with 175B parameters, it is also one of the largest and about ten times the size of GPT-2. There are even larger ones, such as Microsoft NVIDIA’s Megatron-Turing NLG 530B, which has 530B parameters. The bragging rights for the largest language model in the world are actually with Alibaba’s M6-10 T which has 10 trillion parameters (Lin et al. 2021). However, it is trained on as much graphic data as text data and is more multimodal, so not directly comparable to GPT-3. Related to it are the switch transformer language models, which are also in the trillion-plus parameter range (Fedus et al. 2021). Coming soon, at the time of writing, is GPT-4, expected to be 500 times larger than GPT-3 and have 100 trillion parameters. The race for bigger and bigger models is not likely to end there. Since its release in 2020, GPT-3 has been extensively reported in the media and in the research community. It can participate in written dialogue, reply to questions, and even write essays. An intriguing example is an article that it was asked to write for the Guardian newspaper arguing that robots have essentially peaceful intentions toward humans (GPT-3 2020). The resulting article is astonishingly similar to something that could be written by a human; however, many researchers have pointed out that it is the output of a postediting process by a human journalist, and GPT-3 may simply be an excellent mimic, making use of deep contextual cues to reproduce segments of prior learned content, albeit rearranged in a novel fashion. But isn’t that how many children learn? GPT-3 is text only – confined to the written word as input, and as output. That text is also well curated (hence GPT-3’s response to real-world errors and operation “in the wild” seems to have not been fully established). It is thus currently unable to speak, or to listen to speech, despite the continual performance improvements of modern TTS and ASR engines. Unfortunately, the behavior of NLP systems in the real world often leaves a lot to be desired, amply demonstrated by the embarrassing fate of Microsoft’s Tay chatbot

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

20 AI, Human–Robot Interaction and NLP

(Wolf et al. 2017). Recent improvements (Zhou et al. 2020) notwithstanding, translation of working systems from the laboratory into daily general use is likely to be still some time away as we learn more about how humans build relationships with such systems (Skjuve et al. 2021).

20.4 Dialogue and Meta-processing Just as real human speech is much more than simply letters and words, real dialogue is much more than just utterances or continuous speech (McTear 2002). Dialogue is conversation, a responsive communication between two parties with a two-way flow of information, concepts, or instruction. Both parties adjust what they are saying, and the way that they say it, to respond to the other party (Crystal 2003). They may naturally accept points that the other party has made, layer upon those ideas, or complete the thoughts of the other party. They may refute, twist the words, or negate what the other party has said. Often one party will quote, or deliberately and identifiably misquote, the other party. Good dialogue may be like poetry where the tempo, meter, and rhyme is shared. Utterances may alternate, but then again may not, and cutting into a monologue can be an art – especially as some of the most talented speakers of monologue seemingly have no need to draw breath. Contemporary computer dialogue systems need to improve a long way before they can match human dialogue abilities. Apart from issues of miscommunications (discussed in Section 20.5), current systems exhibit unnaturally long delays between query and response. Such delays in a conversation are annoying and – in humans at least – lead to a stilted dialogue that both parties may wish to end quickly. It is possible that we will learn to expect less of our computers, conversationally speaking, than our fellow humans. However, it is more likely that computer dialogue systems will only achieve wide-scale adoption once such issues have been addressed. This stands in contrast to computer speech systems which have been adopted widely, such as ASR alone (e.g., dictation systems or automatic closed captioning of a video), or TTS alone (e.g., GPS navigation system instructions or verbal announcements in a lift). Even simple query–response systems (e.g., “Alexa, turn on the TV” or “Siri, set an alarm”) are increasingly common. But these can’t really have a dialogue with us yet, partly because they are unable to understand much realworld dialogue, and partly because the time delay between a spoken query and a response is excessive as mentioned above. Moore’s law and ever more powerful computers, faster communications links, and more responsive architectures (McLoughlin 2018) will clearly continue to improve the responsiveness of future systems, but it is the sheer amount of potential information that must be dealt with in dialogue systems which causes them to usually be slow (the scope of this context information is discussed more in the following section).

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

447

448

IAN MCLOUGHLIN AND NITIN INDURKHYA

Performance can be improved in terms of both accuracy and speed by restricting the scope that a system needs to deal with. This includes restricting conversation to predefined topics such as airline bookings, legal discussion, tax matters (Juang and Rabiner 2005). It also includes use of restricted grammars, fixed usage, or query terms, or needing to respond to a single user only (for which it has been very well trained to interact with). A rather fruitful line of research has examined dialogue as a sequence of actions called dialogue acts (Bunt 2011). This makes participation in a dialogue similar to planning. Except here the plan is being made in conjunction with another independent participant, who may or may not be cooperating. Imagine a dialogue in a computer-aided tutoring session between an AI-tutor and an unwilling student! One advantage of treating dialogue in this fashion is that it automatically provides a lot of contextual information and limits the choices, both for understanding and generation, to make the entire process more tractable. One might argue that all participants in the dialogue are working with plans (to achieve some physical or cognitive goal). The plan behind the dialogue may not always be apparent to all the participants. This is often the case due to cultural norms. Consider, for example, a society where it is rude to ask a total stranger their age. A person might initiate a dialogue where the objective is to build enough of a rapport and familiarity so that the age can be asked. Or the conversation might be steered toward topics that cause various age-related proxies to be revealed. The plan to know the age might be part of a bigger plan to sell a number of age-related items. Psychotherapy has often relied on conversation and dialogue to probe a person’s mind and uncover root causes of mental health disorders. Indeed, one of the simplest and earliest success stories of AI was in this domain (Weizenbaum 1966). More recent developments in the construction of persuasive artificial agents have involved integrating domain information and borrowing ideas from the philosophy of argument – which accords a larger role for beliefs, attitudes, feelings, and personal intuitions and a considerable diminished role for logic. There are valuable applications of such techniques, for example, in preventative medicine. Beyond content-level considerations of dialogue is the more mechanical aspect of the ebb and flow of a conversation, specifically its timing. Humans are generally9 adept at knowing when to speak and when not to. In a two-way conversation, a speaker may expect the other person to wait for their pause before beginning to speak – although there are times and occasions when they would not wait. Pauses by a speaker can be left deliberately for emphasis, to solicit feedback (for example, okay, uh-huh, oh my goodness, a tongue cluck, a nod or smile), to provide an opportunity to acknowledge comprehension or otherwise (for example, in Japanese it is common for one of the interlocutors to constantly repeat “はい、はい” as an affirmation). Computers talking to humans need to understand the intent of the speaker and respond accordingly. One of the currently most glaring aspects of talking to a computer is when and 9

Note: the first author’s wife disagrees with this statement.

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

20 AI, Human–Robot Interaction and NLP

how it is able to indicate non-comprehension, or ask for a speaker to repeat what they have said. This is a very active research area at present, and likely to be at the forefront of efforts to improve the usability and acceptance of computer speech dialogue systems for real- world use over the coming decade (Truong et al. 2020). Finally, we should acknowledge something hinted at above, which is the role of non-speech sounds in conversation. Robot–human interfaces that are able to utilize vision could benefit from nonverbal communication cues, but all speech-based systems could benefit from nonverbal auditory cues. Pausing when a listener is coughing, sneezing, or weeping is only “polite,” and understanding the meaning of cues ranging from a sharp intake of breath or a toothwhistle, would be needed for truly human-level natural communications.

20.5 Multi-Level Context Few people like talking to a brick wall, and being misunderstood brings frustration to many. Yet these are common problems experienced with contemporary computer dialogue systems. Better microphones, improved sensors, enhanced de-noising algorithms, and more ubiquitous embedded sensors can endow more “brick walls” with ears in future (helping alleviate the problem of computer dialogue systems not realizing that they are being spoken to, or not picking up speech sufficiently clearly to process it). Yet these advances alone cannot solve problems of misunderstanding. Misunderstanding can occur at many levels – just as they can in human-tohuman communications. Communications difficulties range from pronunciation issues (e.g., “I thought you said bone not phone”) through word errors (e.g., “Hot tea or not tea”) up to semantic errors and beyond (e.g., “why do you wish I would kick the bucket”10). Low-level phonetic or word error rates have become very good in modern ASR systems, matching or even exceeding the abilities of a human in some circumstances. But high-level errors remain. To examine these, let us consider the influence of context in the human-to-human communications scenario. 1. Spatial context – both react to the situation and events that they are currently experiencing. This includes pertinent information, but also distractions, interfering noise, and cross-sensory information. It might be local (e.g., things they see, smell, feel, sounds they hear) but could also include the wider environment, such as neighborhood, city, region, nationality. 2 Temporal context refers to the past content of the current conversation (note that I include past conversations under “historical context” below). Temporal context encompasses the meaning behind summary phrases such as “yes, let’s do that” or “I agree” as well as anything that builds on the foundation of earlier parts of a conversation. 10

English slang meaning ‘to die.’

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

449

450

IAN MCLOUGHLIN AND NITIN INDURKHYA

3 Historical context – if there is a shared history between two parties, then they are naturally layering their communications upon that foundation. Many matters may be understood but remain unsaid. This includes past interactions that they have had, connections in common, or even news items and events that both have knowledge of. 4 Professional, familial, vocational context – clearly two lawyers would be expected to communicate in a different way from two doctors, as would members of two different families. Not only do professions like to coin unique phrases, but families and sometimes friendship groups do too. 5 Linguistic, cultural, and dialect context – this may again derive from shared history, but regional or cultural history rather than personal history. It would include the way language has developed due to geopolitical differences over many years, although there are many other reasons which lead to cultures and sub-cultures forming, including reasons of religion, generational speech patterns, gangs, and race. 6 Absent context – by which we mean the often quite important aspect of what has remained “unsaid” in a conversation. There are many situations in which humans convey the most important information in a way that requires the listener to “read between the lines.” This could range from a conversational elephant in the room which is not addressed directly but which everyone knows is there while conversation dances around it, to something which is implied but never directly specified, to a deliberate omission whose absence speaks volumes. A more detailed discussion is beyond the scope of this chapter – the interested reader can refer to (Zerubavel 2006) for an in-depth analysis. 7 Social context11 – somewhat overlapping with the others, but in a more dynamic way, it refers to a single speaker constantly adjusting their language to fit their social goals. When we make a presentation, our speech is clearer, is more deliberate with well-formed sentences; when we speak to a toddler we may use “baby talk.” These are fairly obvious adjustments. More subtle are adjustments when we want to push someone away, or “pull rank” on someone, or empathize with someone in trouble, or seek cooperation, or “fit in” with a certain social group. It is also interesting how we can often recognize these contextual cues from others purely from their language, particularly if the social context is recognized. When a computer is expected to process speech, hold a conversation, or conduct dialogue in a manner which is human-like, then these levels of context all need to be addressed. Failure to maintain one or more levels of the above context in human communications would clearly lead to some kind of difficulty. At worst it could lead to complete miscommunication, at best it may lead to the need to clarify, explain, or rephrase something (which humans 11

Note: we coin the term “social context” here as a convenient description, but are aware that the term normally conveys a different set of meanings in other research areas, and in fact may be more usefully described as a register in linguistics.

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

20 AI, Human–Robot Interaction and NLP

would do naturally, but contemporary computer dialogue systems have not yet mastered). Returning to the fictional JARVIS, futurologists and science fiction writers alike have tended to portray positive12 instances of human–computer or robot– human vocal interaction as something like a well-trained butler or valet. This involves a conversational partner who knows our current situation, history, preferences. One who is familiar with the way we speak, unusual words or terms we might use, and the meaning of coded phrases. Furthermore, one who knows how we connect to the world around us, who we talk with, and what we are likely to say. A good butler would also take initiative in undertaking actions that he knows we would wish (e.g., “I told the caller that you would be busy until next week” or “I have taken the liberty to book your favorite table for lunch”). The multifaceted and deeply layered contextual awareness required to implement such a system means that, although it remains a target for many researchers, it will remain out of reach for some time yet.

20.6 Conclusion In this chapter we have examined the architecture of a robot–human or AI dialogue system, considered speech input, speech output, and the complex realm of NLP. We have imagined the future and assessed the capabilities of the present. At all levels, from acoustic, through sub-phonetic, phonetic, word, sentence, utterance, and an entire conversation, we have explored the role that context plays in spoken communications. The importance of context extends beyond the specifics of phonetics, language, mere word-level meaning, or even the depths of semantic understanding. It extends further and in reality embraces shared experience, culture, environment, and an understanding of the wider world. In order for future humanity to benefit from workable, reliable, and comfortable robot–human communications, technological solutions will need to embrace and exploit context over the full range of those levels, and perhaps – if future researchers get this right – herald the emerging and uniquely new context of human–computer relations.

References Akiwowo, S., Vidgen, B., Prabhakaran, V., and Waseem, Z., eds. (2020). Proceedings of the Fourth Workshop on Online Abuse and Harms. Association for Computational Linguistics. https://aclanthology.org/volumes/2020.alw-1/. Baker, J. M., Deng, L., Glass, J., Khudanpur, S., Lee, C.-H., Morgan, N., and O’Shaughnessy, D. (2009). Developments and directions in speech recognition and 12

Negative examples such as killer robots, by contrast, tend to be eerily threatening, often palpably devoid of emotion in speech and humanity in content.

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

451

452

IAN MCLOUGHLIN AND NITIN INDURKHYA

understanding, Part 1 [DSP Education]. IEEE Signal Processing Magazine, 26(3), 75–80. Bender, E. M, and Koller, A. (2020). Climbing towards NLU: On meaning, form, and understanding in the age of data. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (pp. 5185–5198). Association for Computational Linguistics. https://aclanthology.org/2020.acl-main.463. Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., Agarwal, S., Herbert-Voss, A., Krueger, G., Henighan, T., Child, R., Ramesh, A., Ziegler, D. M., Wu, J., Winter, C., Hesse, C., Chen, M., Sigler, E., Litwin, M., Gray, S., Chess, B., Clark, J., Berner, C., Mc-Candlish, S., Radford, A., Sutskever, I., and Amodei, D. (2020). Language models are few-shot learners. CoRR, abs/2005.14165. Bunt, H. (2011). The semantics of dialogue acts. In Proceedings of the Ninth International Conference on Computational Semantics (IWCS 2011). SIGSEM. https://aclanthology.org/W11-0100. Cai, W., Chen, J., and Li, M. (2018). Exploring the encoding layer and loss function in endto-end speaker and language recognition system. arXiv preprint arXiv:1804.05160. Carlini, N., Tramer, F., Wallace, E., Jagielski, M., Herbert-Voss, A., Lee, K., Roberts, A., Brown, T., Song, D., Erlingsson, U., et al. (2020). Extracting training data from large language models. arXiv preprint arXiv:2012.07805. Çetinoğlu, Ö., Schulz, S., and Vu, N. T. (2016). Challenges of computational processing of code-switching. arXiv preprint arXiv:1610.02213. Clark, E., August, T., Serrano, S., Haduong, N., Gururangan, S., and Smith, N. A. (2021). All That’s “human” is not gold: Evaluating human evaluation of generated text. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Vol. I: Long Papers) (pp. 7282–7296). Association for Computational Linguistics. https:// aclanthology.org/2021.acl-long.565/. Crystal, D. (2003). The Cambridge Encyclopedia of English Language. Cambridge: Cambridge University Press. Fedus, W., Zoph, B., and Shazeer, N. (2021). Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity. CoRR, abs/2101.03961. Ferrer, L., Lei, Y, .McLaren, M., and Scheffer, N. (2015). Study of senone-based deep neural network approaches for spoken language recognition. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 24(1), 105–116. Frankel, J., and King, S. (2001). ASR-articulatory speech recognition. In Proceedings of Eurospeech-2001 (pp. 599–602). Aalborg: International Speech Communication Association. GPT-3. (2020 )(Sep). A robot wrote this entire article. Are you scared yet, human? – GPT-3. The Guardian, September 8, www.theguardian.com/commentisfree/2020/ sep/08/robotwrote-this-article-gpt–3. Greff, K., Srivastava, R. K., Koutník, J., Steunebrink, B. R., and Schmidhuber, J. (2016). LSTM: A search space odyssey. IEEE Transactions on Neural Networks and Learning Systems, 28(10), 2222–2232. Hu, R., and Singh, A. (2021). Transformer is all you need: Multi-modal multitask learning with a unified transformer. arXiv e-prints, arXiv– 2102. Jiang, B., Song, Y., Wei, S., Liu, J.-H., McLoughlin, I. V., and Dai, L.-R. (2014). Deep bottleneck features for spoken language identification. PLoS ONE, 9(7), e100795.

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

20 AI, Human–Robot Interaction and NLP

Jin, M., Song, Y., McLoughlin, I., and Dai, L.-R. (2018). LID-senones and their statistics for language identification. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 26(1), 171–183. Jobs, S. (2005). Steve Jobs connect the dots [video file]. www.youtube.com/watch? v=5BSbOc5VYY8. Accessed June 22, 2023. Joshi, M., Chen, D., Liu, Y., Weld, D. S., Zettlemoyer, L., and Levy, O. (2020). Spanbert: Improving pre-training by representing and predicting spans. Transactions of the Association for Computational Linguistics, 8, 64–77. Juang, B.-H., and Rabiner, L. R. (2005). Automatic speech recognition: A brief history of the technology development. Georgia Institute of Technology. Atlanta Rutgers University and the University of California. Santa Barbara, 1, 67. Kohler, M. A., and Kennedy, M. (2002). Language identification using shifted delta cepstra. In Proceedings of the 45th IEEE International Midwest Symposium on Circuits and Systems (pp. III–69). doi:10.1109/MWSCAS.2002.1186972. Lin, J., Yang, A., Bai, J., Zhou, C., Jiang, L., Jia, X., Wang, A., Zhang, J., Li, Y., Lin, W., Zhou, J., and Yang, H. (2021). M6-10 T: A Sharing-Delinking paradigm for efficient multi-trillion parameter pretraining. https://arxiv.org/abs/2110.03888. McLoughlin, I. V. (2016). Speech and Audio Processing: a MATLAB-Based Approach. Cambridge: Cambridge University Press. McLoughlin, I. V. (2018). Computer Systems: An Embedded Approach. Singapore: McGraw Hill Education. McLoughlin, I. V., and Sharifzadeh, H. R. (2008). Speech recognition for smart homes. In Vasile-Florian Păiș (ed.), Speech Recognition: Technologies and Applications (pp. 477–494). Rijeka: IntechOpen. https://doi.org/10.52305/BKWM8996. McTear, M. F. 2002. Spoken dialogue technology: Enabling the conversational user interface. ACM Computing Surveys (CSUR), 34(1), 90–169. Miao, X., McLoughlin, I., Wang, W., and Zhang, P. (2021). D-MONA: A dilated mixed-order non-local attention network for speaker and language recognition. Neural Networks, 139, 201–211. Mori, M., MacDorman, K. F., and Kageki, N. (2012). The Uncanny Valley [From the Field]. IEEE Robotics Automation Magazine, 19(2), 98–100. Richardson, J., and Arthur, M. B. (2013). Just three stories: The career lessons behind Steve Jobs’ Stanford University Commencement Address. Journal of Business & Management, 19(1), 45–57. Skjuve, M., Følstad, A., Fostervold, K. I., and Brandtzaeg, P. B. (2021). My Chatbot companion: A study of human–chatbot relationships. International Journal of Human-Computer Studies, 149, 102601. Slonim, N., Bilu, Y., Alzate, C., Bar-Haim, R., Bogin, B., Bonin, F., Choshen, L., Cohen-Karlik, E., Dankin, L., Edelstein, L., et al. (2021). An autonomous debating system. Nature, 591(7850), 379–384. Taylor, P. (2009). Text-to-Speech Synthesis. Cambridge: Cambridge University Press. Thomason, J., Sinapov, J., Svetlik, M., Stone, P., and Mooney, R. J. (2016). Learning multimodal grounded linguistic semantics by playing “I Spy.” In Proceedings of the 25th International Joint Conference on Artificial Intelligence (IJCAI) (pp. 3477–3483), July 2016. Palo Alto: AAAI Press / International Joint Conferences on Artificial Intelligence.

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

453

454

IAN MCLOUGHLIN AND NITIN INDURKHYA

Truong, M., Fast, N. J., and Kim, J. (2020). It’s not what you say, it’s how you say it: Conversational flow as a predictor of networking success. Organizational Behavior and Human Decision Processes, 158, 1–10. Turing, A. M. (2009). Computing machinery and intelligence. In R. Epstein, G. Roberts, and G. Beber (eds.), Parsing the Turing Test (pp. 23–65). Dordrecht: Springer. https://doi.org/10.1007/978-1-4020-6710-5_3. Voskarides, N., Meij, E., Reinanda, R., Khaitan, A., Osborne, M., Stefanoni, G., Kambadur, P., and de Rijke, M. (2018). Weakly-supervised contextualization of knowledge graph facts. In The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval (pp. 765–774). New York: Association for Computing Machinery. Weizenbaum, J. 1966. ELIZA: A computer program for the study of natural language communication between man and machine. Communications of the ACM, 9(1), 36–45. Wolf, M. J., Miller, K. W., and Grodzinsky, F. S. (2017). Why we should have seen that coming: Comments on Microsoft’s Tay “Experiment,” and wider implications. The ORBIT Journal, 1(2), 1–12. Zerubavel, E. (2006). The Elephant in the Room: Silence and Denial in Everyday Life. Oxford: Oxford University Press. Zhou, L., Gao, J., Li, D., and Shum, H.-Y. (2020). The design and implementation of xiaoice, an empathetic social chatbot. Computational Linguistics, 46(1), 53–93.

https://doi.org/10.1017/9781108989275.021 Published online by Cambridge University Press

21 Social Media and Computer-Mediated Communication Francisco Yus

21.1 Introduction: Social Media Social media, together with messaging apps, are nowadays the most active and dynamic areas of internet-mediated communication. As described in Yus (2017a: 555), the generation of content by users, the availability of areas for its publication and sharing, and the options for commenting on other users’ uploaded information crosscut many popular sites such as Facebook, Twitter, Instagram, YouTube, and so forth (see boyd 2010). Social networking sites are often equated to social media, but the latter is a broader umbrella term (also covering blogs, for instance). Besides, these media have become highly imbricated and hybridized with the users’ offline interactions, to the extent that a certain level of online–offline congruence is nowadays expected on these sites (Bolander 2019; Yus 2022). An important aspect of social media that pertains to any analysis of context on these sites is the fact that social media interactions take place through interfaces, whose design offers users different options for contextualization, ranging from cues-filtered text-based messaging to richer media such as videomediated interactions. Furthermore, through these interfaces different types of discourse are exchanged, from plain text to multimodal discourses (e.g., memes; see Eisenlauer 2017: 229). This chapter addresses key issues regarding context (and contextualization) on the Internet and specially on social media. The chapter starts with general issues regarding context and the static/dynamic pragmatic approaches to explaining context. Next, an account is provided of the important online– offline dichotomy and how contextualization involves both settings. The main analysis of the chapter is provided in the next section, which includes a preliminary account of online discourses and so-called contextual constraints influencing the users’ options for contextualization, and then moves on to three basic areas of context on social media: personal, interactive, and social.

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

456

FRANCISCO YUS

21.2 Context: From Typologies to a Set of Assumptions There is general agreement within pragmatics that context (or contextual information) plays a major role during interpretation. However, there has been an important evolution regarding how context is pictured in pragmatics and its role in comprehension, from an initial stage that can be labeled static view to today’s dynamic view. In the first – static – stage, context was somehow taken for granted and the main interest lay in proposing types or varieties of context, assuming, as Jones (2004) correctly remarks, that context is given beforehand, a sort of static “theatre-stage backdrop.” By contrast, today’s dynamic view entails a picture of context as not given, but “sought” as part of the inferential strategies devoted to obtaining an adequate interpretation of the utterance (or text, or multimodal discourse, among other possibilities). At this stage, all kinds of context are, simply, information (a set of assumptions in relevance-theoretic terminology, see Sperber and Wilson 1995), mentally accessed and represented, and whose accessibility is both predicted by the addresser and carried out by the addressee in order to turn the schematic string of words into a relevant proposition. Which information from context is retrieved and processed in parallel to the interpretation of the utterance depends on the demands of the particular utterance, and accessibility to that information directs the addressee in one inferential direction or another (see Blommaert et al. 2020: 53). Crucially, this pairing of contextual information and the utterance being processed takes place when inferring any stimulus, regardless of whether the utterance being processed is produced in a face-to-face environment or through an interface. In this sense, one of the claims of cyberpragmatics (Yus 2011) is that the qualities of the medium for internet-mediated communication (app, platform, interface, website, etc.) impact the options for contextualization on the Internet. Social media are a good example of how evolutions in interface design constrain or favor strategies of contextualization (see Section 21.4.2 below). Within the dynamic view of context, relevance theory proposes the term manifestness in relation to context. In a nutshell, as summarized in Yus (2016a), people construct different concepts and representations of the world, just as their personal experiences are different. This array of information is called cognitive environment, which is made up of manifest assumptions, the ones that a person is capable of representing mentally on a particular occasion. The sum of all the assumptions that are manifest to a person makes up their cognitive environment. Crucially for contextualization, when a speaker produces an utterance, he/she expects that certain information will be manifest to the interlocutor so as to interpret the utterance correctly (this information then becoming mutually manifest to both interlocutors, a term that entails a rewriting of more traditional notions such as shared knowledge).

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

21 Social Media and Computer-Mediated Communication

Manifest information may be retrieved from different sources, most importantly from encyclopedic background knowledge, but also from the physical surroundings where the interlocutors are located, among others. Needless to say, the same applies to utterances produced and interpreted on social media. In this sense, Zappavigna (2017: 450) correctly remarks that social media research has to dismiss the idea that context is stable and assume its dynamism instead: Any public or mass media communication integrates a range of meanings from different contexts to which any particular audience may not necessarily have unmediated access. It is nevertheless possible for a media text to target specific groups through modulating particular communicative choices.

21.3 Context Accessibility across the Online–Offline Divide If context is information that is accessed in parallel to the interpretation of an utterance, then the capacity of the channel (website, interface, app, portal, etc.) to convey (i.e., provide interlocutors accessibility to) this contextual information is crucial to guarantee a successful interpretive outcome. In this sense, some research claims that the internet is unable to convey all the contextual information that is sometimes important to arrive at the intended interpretation, especially in cues-filtered text-based communication such as the one carried out through messaging apps, compared to the contextually rich environment of a face-to-face scenario. Besides this critical position on internet communication, in Yus (2011) other areas in this debate were highlighted: (a) Some studies claim that online and offline communication are simply different, one major quality in this respect being the “mutual monitoring possibilities” that these technologies favor, “the different ways in which they allow us to be present to one another and to be aware of other peoples’ presence” (Jones 2004: 23). (b) Other studies propose a picture of complementarity or extension regarding online and offline scenarios (Jones 2004: 24; Bolander and Locher 2020). (c) Finally, in Yus (2011) the picture is that of hybridization between online and offline scenarios, the user turning nowadays into a node of intersecting interactions of a physical, virtual, and mostly mixed online–offline quality. Furthermore, it should be added that users often make the most of an apparently limited channel of virtual communication and manage to access (and make accessible) contextual information. As remarked in Yus (2022: 35) regarding the communication of nonpropositional information attached to a text (important as part of contextualization strategies), users actually manage to convey as many feelings and emotions online as in face-toface communication, and with a similar (if not higher) intensity. The key lies in how users detect and exploit the linguistic and expressive resources (i.e., the affordances) that the interface provides for the communication of these affective attitudes and effects. A good example is to be found in messaging apps, whose users rely on text alteration and the use of stickers, GIFs, and emojis to

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

457

458

FRANCISCO YUS

transfer their attitudes, feelings, and emotions more accurately (and hence to aid interlocutors in a more fine-grained contextualization of their communicative intentions). In sum, the picture that should be concluded regarding the online–offline divide in terms of contextualization is that, obviously, a scale of contextualization may be arranged depending on how rich the interface is in its capacity to convey and allow access to contextual information, but internet users make the most of these possibilities and often obtain as much contextual information as in a richer face-to-face scenario. Users even act upon the affordances of the interface so as to obtain greater richness and options to manage their interactions. As stressed in Yus (2022: 57), users adapt and reshape technologies to their own communicative needs. This is what happened with Twitter users, who were dissatisfied with its interface options and started to label tweets by using hashtags, as well as creating the retweet nomenclature. These end-user innovations were eventually incorporated into the Twitter interface by the company. Figure 21.1 represents some of the issues that interests us regarding context and the online–offline divide and which will be developed in the next sections. In the center of Figure 21.1, “discourse” is located. In offline environments, discourse is typically conveyed in an oral, written, or multimodal form, often taking advantage of parallel vocal and visual nonverbal behavior that normally aids interlocutors in the right inferential direction. Offline discourse may also exhibit community-bound features whose proper contextualization builds up barriers of discursive specificity. Finally, individuals also exhibit their own linguistic features as unique idiolects. These may also constrain contextualization in offline scenarios. These offline features are correlated with those found online. Discourse on the Internet is often typed, but users rely on emojis, GIFs, stickers, and text alteration (repetition of letters, creative use of punctuation, etc.) to achieve their communicative needs. Richer means of interaction include recorded audio files and phone/video calls through messaging apps. Multimodal discourses (e.g., memes) also abound in online communication, and the aforementioned community- and individual-based linguistic features are also found online (Yus 2015). Online and offline discourses favor or limit the kind of contextualization that may be performed during interpretation in both settings. In this chapter, context will be studied by preliminarily distinguishing between the labels of personal, interactive, and social. Offline personal context includes the person’s name, physical appearance, and their body as anchoring sources of identity. Online personal context retains these attributes but adds the possibility of interacting via nicks and avatars and neutralizing the user’s body, a clear anchorage of identity offline. Most importantly, in both offline and online personal contexts, there is a certain amount of information that is manifest to the person/user when interpreting discourses. This information may vary radically from person to person (or user to user) and affect inferential strategies. On other occasions,

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

PERSONAL CONTEXT (ONLINE) Nick Avatar SNS user-centered image Identity / self-concept

INTERACTIONAL CONTEXT (ONLINE) Messaging conversation Locative media Video-mediated dialogue Avatar-mediated chat

SOCIAL CONTEXT (ONLINE) Messaging group Internet fora Virtual community Avatar community

CONTEXTUAL CONSTRAINTS ONLINE DISCOURSE Typed + text alteration Visual discourse (images) Emoji / sticker / GIF Multimodal discourse Audio/video file Virtual community lang. User-specific idiolect

OFFLINE DISCOURSE Oral discourse Written discourse Multimodal discourse Vocal NVC (paralanguage) Visual NVC (kinesics) Community language Person-specific idiolect CONTEXTUAL CONSTRAINTS

Figure 21.1 Discourse and contexts across the online–offline divide.

PERSONAL CONTEXT (OFFLINE) Name Physical appearance Body identity / self-concept

INTERACTIONAL CONTEXT (OFFLINE) Face-to-face conversation Phubbing Phone call Offline identity roles

SOCIAL CONTEXT (OFFLINE) Offline gathering Offline social space Offline community Work-centered gathering

460

FRANCISCO YUS

though, much of this information is of a common-sense, encyclopedic quality, and its accessibility is expected by the addresser and almost unconsciously retrieved by the addressee when turning coded inputs (oral utterances, typed messages, audio files, etc.) into meaningful interpretations. This is why contextual information is not only predicted between interlocutors but also revealed as the interaction unfolds successfully. An often-cited example is (1) below: (1) Max: How was the party? Did it go well? Amy: There wasn’t enough drink, and everyone left early. (2) Parties are not successful if people leave early and they run out of drink. (3) The party was awful and did not go well.

It is clear that Amy’s main intended interpretation of her utterance is the implicature in (3). In this case, Max’s access to common-sense background information such as (2) will easily lead to that implicature. It should also be noted that the explicit interpretation of Amy’s utterance also demands much contextualization, again retrieving general background knowledge from context. Firstly, drink is narrowed into specifically “alcoholic drinks.” Secondly, everyone is also narrowed to “everyone at the party.” Finally, the specific time frame of early depends on what is commonly assumed to be “early” for parties in the age group to which Max and Amy belong. Continuing with Figure 21.1, interactive context entails the default form of contextualization typically addressed in pragmatics, namely the one in which two interlocutors exchange utterances and hold expectations of context accessibility (and assume this accessibility, sometimes wrongly) so that utterances are correctly inferred, and the conversation runs smoothly into an optimal outcome. In relevance-theoretic terms, the speaker/user in an interaction not only makes manifest information to the interlocutor but makes it mutually manifest to both interlocutors, the intersecting area of manifest information of the interlocutors, which becomes larger as a consequence of this mutual manifestness. An example is the irony in (4) (Yus 2016a: 210–211): (4) Tom:

[Tom and John on a cold, wet, windy English day in London]. [Smiling, with a distinctively ironic tone of voice] When a man is tired of London, he is tired of life.

In this example, a great deal of contextual information invalidates any chance that the speaker intended to communicate the explicit interpretation of that utterance, and Tom expects his interlocutor to access specific information from context that will allow him to get the irony. Clearly, Tom is confident that John will be aware of the miserable weather, that he will also draw further contextual information including general encyclopedic information (nobody could possibly like to live in such a place), mutuality of information with the speaker (perhaps Tom has repeatedly complained about how miserable life in

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

21 Social Media and Computer-Mediated Communication

London is), and nonverbal communication (probably Tom has produced his utterance with an ironic tone of voice and a smile). This contextual information becomes mutually manifest as a result of the successful communication of the irony. Contextualization via interactions takes place across the online–offline divide and is only differentiated by the interface constraints that may favor or limit the accessibility to contextual information between interlocutors. In other words, there is only one way in which the human cognitive system pairs information from utterances with contextual information to yield adequate or meaningful interpretations, but the act of contextualization itself may be constrained, to a greater or lesser extent, by the interfaces used for communication (in online scenarios) and other negative features such as ambient noise or bad pronunciation (in offline scenarios). This will be addressed in Section 21.4.2 below. Finally, social context in Figure 21.1 refers to information that is taken for granted within a community of individuals/users in both online and offline scenarios. This contextual information is enacted and successfully managed by those who belong to the group or community, fostering feelings of group membership and creating barriers of discursive specificity, as already mentioned. Needless to say, the aforementioned context categories (personal, interactive, social) are not mutually exclusive and in fact they overlap in most of today’s interactions.

21.4 The Contexts of Social Media Discourse Social media are a good example of the contexts briefly mentioned above. On these sites and platforms, thousands of textual, visual, and multimodal discourses are exchanged among users, all of them demanding contextualization. Similarly, users hold a personal array of information being manifest to them and making up their personal cognitive environment (personal context). This manifest information becomes mutually manifest through the whole array of interactions that can take place on social media, ranging from synchronous messaging chats to asynchronous commenting–replying on users’ posts (interactive context). Finally, as proposed in Yus (2007), information of social or communal assumptions is typically intended but may also leak from these interactions, especially if these entail particular uses of language, jargons, vocabulary, or registers, making up a store of group-specific information (social context). A good example is the social networking site Facebook. As depicted in Figure 21.2, its desktop interface differentiates areas for these types of context. The list of Facebook groups indicates the kind of social gathering and sources of information that the user is keen on, with an offset of communal bonding and feelings of group membership. The Messenger area is devoted to (mainly synchronous) chats with other users. Finally, the user’s profile includes the main picture of the user, the profile photo, and the content made manifest by the user

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

461

462

FRANCISCO YUS

Figure 21.2 Personal, interactive, and social contexts on a Facebook profile.

as entries or ephemeral stories. The profile also includes personal information about the user (studies, current job, relatives, etc.). All of this information exchanged on a unique site is one of the most outstanding features of social networking sites (see Spilioti 2015). Bolander and Locher (2020) illustrate this with the inclusion of text comments on photo-sharing sites; text (and video) responses to YouTube videos; text (and voice) chat during multiplayer online games; and text messages from mobile phones posted to interactive TV programs. They add the trend to engage with these sites and technologies and with other users via a growing range of mobile devices (see Yus 2022), adding a layer of information regarding whether the users are physically immobile or on the move in different physical, material spaces while using these devices.

21.4.1 Online Discourse As any type of discourse, online discourse is coded and has to be contextualized inferentially, with a lesser or greater gap between what is coded and what is meant (and eventually interpreted). How this task is accomplished depends on the interface affordances, as already pointed out. Twitter, with its 280character limit, tends to provoke greater gaps between “coded” and “meant.” Texts on messaging apps are also cues-filtered, but users manage to convey their intentions more neatly thanks to “accompanying props” such as text alteration, emojis, GIFs, and stickers, among others. Finally, video-mediated interactions online exhibit, on paper, the richest medium in terms of options for contextualization, even if the medium itself is still limited compared to face-toface communication. As argued in Yus (2022: 112–113), authentic expressions of emotion (mostly in face-to-face scenarios) are essential to our understanding.

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

21 Social Media and Computer-Mediated Communication

But those nonverbal hints disappear on pixelated video or are even frozen, smoothed over, or delayed. Indeed, to recognize emotion, we have to embody it, which makes mirroring essential to empathy and connection, and “when we can’t do it seamlessly, as happens during a video chat, we feel unsettled because it’s hard to read people’s reactions and, thus, predict what they will do” (Murphy 2020: n.p.). Besides, online discourses on social media are often not delimited, with a clear structure and expected reading patterns. Instead, they are often linkmediated and the user has to apply cohesion and coherence to these discourses, often taking the responsibility for which path is taken and what eventual interpretation is obtained, all that through dynamic contextualizing tasks. Crucially, as already mentioned in passing, interface affordances (with parallel options for contextualization) are imposed on the social media users, but users can act on the interface in order to achieve their communicative needs, also when aiding interlocutors in an adequate contextualization (Gruber 2019: 71). Eisenlauer (2017: 56) points in the same direction when remarking that the function of a particular social medium “rests on the interplay of the platform’s communicative exigencies and the ways in which individual users adapt to these affordances in different ways. As shown, the functions that users associate with a particular service may or may not correspond with the initial functions of the developers.” Regarding contextualization, this interplay of site affordances and user innovation impacts the eventual options for contextualization that users are granted and, hence, also impacts the final interpretation obtained from discourses on social media. A similar interplay occurs in the case of messaging areas within social media interfaces. Initially, these were mainly text-based, but both users and companies increased the options for contextualization through these interfaces. Companies gradually incorporated emojis, GIFs, stickers, audio files, text type (bold, italics, and so on) and images into their interfaces, again increasing the options for contextualization (see Yus 2011; 2017a: 556; 2022; Pano Alamán and Mancera Rueda 2014: 245).

21.4.2 Contextual Constraints/Affordances As already highlighted in passing, online communication in general and strategies of contextualization in particular are constrained by the qualities of the interface through which online exchanges take place, that is, its affordances, defined in Eisenlauer (2014: 73) as the possibilities “afforded by an object or an environment in relation to social actors and their individual capabilities.” Contextual constraints may be defined as aspects which underlie acts of communication and users’ interactions (i.e., they exist prior to the act of communication) and determine their eventual (un)successful outcome. They frame, as it were, communication and have an impact both on the quality of interpretation and on the willingness to engage in future interactions.

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

463

464

FRANCISCO YUS

Specifically, regarding social media interfaces, as described in Yus (2022), the profile provides users with different affordances that restrict or allow for the kinds of discourse that may be used therein. By making the interface more user-friendly, designers expect to provide an effortless experience while using their social media interfaces, what Costa (2018: 3651) labels affordances-inpractice, a term that highlights the idea that “social media affordances are the results of the repeated interactions between humans and platforms. They are the results of social and material enablement. As such, they are not fixed and stable properties but are implicated in different ongoing processes of constitution, which may radically vary across social and cultural contexts” (see also Eisenlauer 2017: 232–234). To achieve user satisfaction, the information presented on the interface has to be clearly arranged, with suitable links that allow users to find, create, and process social information easily. Other researchers, not within pragmatics, have also addressed this layering of relevance in multimodal discourses (see Kress and van Leeuwen 1996; Hassan and van Leeuwen 2008; van Leeuwen 2020). A term proposed in this direction is salience, decided by viewers’ instinctive judgment of the importance of various elements in a composition. The greater the importance of an element, the greater is its salience. This salience cannot be measured objectively and stems from complex interaction among many factors including size, sharpness of focus, color contrast, perspective, placement of visual elements, and specific cultural factors. The degree of salience results in a hierarchy of attentiondegree among the visual elements of images. Another useful term is framing, this time related to whether there are devices that separate or connect visual elements in an image and impact their eventual interpretation (see Chen and He (2015). In any case, what matters most from the (cyber)pragmatics point of view is whether users’ interactions are favored or limited depending on site design. Constraints become crucial especially regarding smartphones, where users show less willingness to expend effort in managing their interactions on small touchscreens. Desktop profiles and their smartphone counterparts may be similar, but some features inevitably need to be discarded or rearranged from the main desktop interface to the smartphone app in order to simplify the items displayed on the screen and make app management as effortless as possible. The most radical example of this variation between desktop and app interface is Facebook (Yus 2022), whose desktop site provides a visual display of several relevant items which become simultaneously available to the user (see Figure 21.2). By contrast, using the Facebook app entails two separate apps for the kind of activity that users perform on a single desktop screen. This may seem like a trivial alteration to some readers, but having to switch between apps to manage communication and interaction on Facebook discourages many users, who bitterly complain about the effort involved. Interface constraints are further constrained by the site’s algorithmic procedures, which select who appears on the user’s feed or which content is presented to them, among other filtering activities (supposedly predicting

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

21 Social Media and Computer-Mediated Communication

the eventual relevance of these automatic choices for the end user; see my proposal of algorithmic manifestness in Section 21.4.3 below). Unlike many studies of social media that conceptualize its interface affordances as technological issues separated from its users (Costa 2018: 3650), now these are viewed as intertwined with users with implications for interaction, contextualization, and sociability on these sites. In any case, and despite their limitations compared to contextually rich faceto-face interaction, social media allow for meaningful and rewarding interactions whose contextualization need not necessarily be thwarted by interface design. As summarized in Quinn (2016: 593), social media offer many options for rewarding relationship maintenance, including their broadcast capabilities (dissemination of news and relevant information), and their messaging and chat functions (for one-to-one or group communication). Social media also preserve information that can subsequently be retrieved and reenacted in the future. Persistence of publications and interactions also translates into a stable and dedicated social address, “which gives others the ability to locate and contact their connections over time and across geographic, temporal, and social distance, further extending the ability to tap resources” (p. 593). Below, two examples are provided of interface constraints. The first one of a negative quality (messaging areas embedded in social networking sites such as Facebook or Instagram), and the other of a positive quality (the use of hashtags on Twitter to allow for better contextualization). In this sense, it should be noted that although it might sound as if constraints necessarily have to be negative (in the sense of “limiting” choices and options), in reality, within cyberpragmatics these are also regarded as possibly exhibiting positive qualities. For example, in the just-mentioned example, hashtags are useful in constraining the wide variety of interpretations that the accompanying text might generate, thus reducing mental effort and increasing eventual relevance. Some qualities of messaging areas embedded in social networking sites are a burden for optimal contextualization of the utterances typed therein. One of them is disrupted turn allocation. This refers to the situation in which the system mixes up the conversational turns that are produced in the messaging conversation (see Yus 2017c). For example, in Yus (2016c) the following WhatsApp conversation was reproduced. It took place between a female (A) and a male (B) user: (5)

A: B: A: A:

La voy a facturar en el aeropuerto [I’m going to check her in at the airport]. Eso [That’s it]. Egipto, que allí los idolatran [To Egypt, since they idolise them there]. O a Marruecos pa q aprenda lo que vale un peine [Or to Morocco, so that she learns the tough way]. Q está muy mimadita [because she’s too spoiled].

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

465

466

FRANCISCO YUS

B: A: B: A: B: A:

Yeah Yastan aqui mis padres [My parents are here already]. Que vea que la vida no es solo hacer trastadas [She has to realise that life is not all about playing tricks around]. [emoji of anguish]. Ohhhh. Planazo Ohhhh. Great plan]. Total [Totally].

In this conversation, the initial topic is how angry A is with her naughty cat. Halfway through this dialogue, A informs B that her parents have just arrived (Yastan aquí mis padres), but B’s next message is still related to the naughty cat, since the system has reproduced messages in strict order of arrival. Similarly, A’s next message, an emoji of anguish which codes a whole proposition (roughly “my parents’ visit depresses me”), does not refer to the cat either, although it follows B’s cat-related message. These mixed-up threads in messaging conversations may be a potential source of misunderstanding or increased processing effort. On the other hand, an example of positive interface constraint on social media is the Twitter hashtag and its role in aiding contextualization. Its communicative role was a discursive innovation by Twitter users and not initially integrated into the interface, as already mentioned. When attached to the tweet, the hashtag may perform a number of interesting functions aiding the audience in contextualization and in the right inferential direction. For example, Puschmann’s (2015: 30) and Dayter’s (2016) study uses of hashtags for varied identity-related pragmatic strategies. Another example can be found in Scott (2015), for whom the hashtag performs several roles when it comes to obtaining the intended interpretation of the tweet next to which they are typed. For instance, the hashtag in (6a) allows the reader to locate the referent for “it” and obtain the derivation of an optimal interpretation like (6b). And the hashtag in (7a) allows for the inference of the user’s propositional attitude toward the tweet, interpreted as (7b) (see also Wikström 2014; Zappavigna 2017: 445–446): (6)

a. She’s done it! An amazing amazing effort. Please txt FIVE to 70510 #davina #windermere. b. British television presenter Davina McCall has finished her charity swim across Lake Windermere.

(7)

a. One week from today I can start throwing again. #finally. b. The user is relieved that she can start throwing one week from today.

21.4.3 Personal Context on Social Media The user’s personal context is made up of all the information (the set of assumptions in relevance-theoretic terminology) that is manifest to the user, making up their cognitive environment. This information is activated and

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

21 Social Media and Computer-Mediated Communication

updated through repeated interactions and associated contextualization. In this sense, an important part of this manifest information refers to the user’s identity, both offline and online, and how identity-centered strategies (e.g., self-presentation) allow for a more specific self-concept and self-awareness. In fact, in Yus (2011, 2014, 2016b, 2022) it is claimed that most of the users’ communicative activity on social media is related to identity claims and identity shaping, with an expectation of online–offline congruence, as already stated (see Davidson and Joinson 2021). Therefore, the user’s personal context includes not only the information that is manifest to them but also the information that they decide to make manifest to other users, often with an identity-related communicative strategy in mind. Manifest information is relevant to the user in three main cases, according to relevance theory: (a) when it leads to previously held assumptions being corroborated; (b) when these assumptions are contradicted and eventually erased; and (c) when it combines with existing assumptions to yield relevant conclusions. In the case of personal information strategically made manifest to other users on social media, audience validation is essential to reinforce the validity of the user’s personal assumptions regarding their identity. Besides this information strategically made manifest to others, the user also has access to a lot of information from their peers, friends, and acquaintances, some of which combines (i.e., is contextualized), in a relevant way, with the user’s previously held assumptions to yield a more accurate and updated personal cognitive environment. In this sense, it should be noted that much of the information that is manifest to the user is not as a consequence of the user’s decision to find it but is presented to the user through the system’s algorithms that decide which information is likely to be relevant to the user by judging from the user’s previous activity online. This tracking activity may be labeled algorithmic manifestness. As described in Yus (2017b) regarding Facebook, the site automatically filters and makes manifest information to the user in three different ways: (1) By affinity: the more one contacts somebody and the more someone visits the profile and engages in conversations with him/her, the more likely that Facebook will show updates from that user. (2) By the relative value of content: for example, updates on user status, e.g., on the user being no longer married etc., are more valuable and the system emphasizes them, but other types of content are also underlined by the system if it detects the user’s tendency to see that kind of content. (3) By time: obviously, the most recently published entries have prominence over the oldest. In short, content is also monitored and managed by the site’s software and algorithms (see Lomborg 2014: 2).

21.4.4 Interactive Context on Social Media If information manifest to the user (or made manifest by the user to other users) is at the heart of personal context, mutual manifestness of information is at the heart of the interactive context on social media, since they “facilitate

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

467

468

FRANCISCO YUS

a distinctive form of social interaction online, creating a constantly expanding network of social relationships characterized by varying degrees of familiarity and tenuousness and by the exchange of symbolic content in multiple formats and modalities that is made available to others with varying degrees of openness and restrictiveness” (Thompson 2020: 6–7). During the various types of interactions that users may carry out on these sites (synchronous text-based, asynchronous multimodal, etc.), users expect their audience to have access to the information that will allow them to understand their utterances correctly (i.e., that this information will be manifest to that audience) and, as a consequence of successful contextualization, that information becomes part of their mutual cognitive environment, made up of mutually manifest assumptions (as vividly revealed by successful comprehension). An example is the audience’s background store of information regarding the user, which is crucial for communicative strategies such as irony. Indeed, ironists usually rely on information supposedly shared (mutually manifest) between the interlocutors or a specific group of friends or acquaintances. Such a mutuality is not only assumed or predicted but also revealed by the irony itself. This desire to foreground mutuality via irony is also present in online communication. In Yus (forthcoming), the Facebook post in (8a) below is proposed, which foregrounds information on marriage and sex roles. It assumes a mutuality (mutual manifestness) of information with other female Facebook users regarding marriage and their husbands’ lack of cooperation in taking care of their children. The male user’s reply to the post in (8b) holds similar expectations of mutuality with certain male users and ironically counters that prior criticism. He echoes wives’ habit of pulling the blanket to their side of the bed, leaving husbands uncovered and cold. Both posts rely on the mutual manifestness of this specific mutuality of information, likewise enhanced and revealed through such ironic remarks. (8)

a. Facebook post: Voy a acostarme esta noche en el lado de la cama de mi marido. Por lo visto en ese lado no oyes a los niños cuando se despiertan por la noche. [I am going to sleep on my husband’s side of the bed tonight. Apparently, on that side you can’t hear children when they wake up in the middle of the night]. b. Reply to post: Yo en el de mi mujer. Al parecer allí sí hay manta. [I will sleep on my wife’s. Apparently, on that side there is a blanket].

Again, social media interactions are influenced by interface-related contextual constraints. On sites such as Facebook, interactions normally comprise chains of successive messages related to an initial post which are sequentially uploaded by the system. This attribute does not necessarily mean that these interaction messages are disorganized or lack cohesion and coherence. In fact, users tend to build on one another’s contributions and these turns are thematically organized by the system, that is, the system arranges the messages in such a way that the threads can be visually identified and do not get mixed up. For

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

21 Social Media and Computer-Mediated Communication

example, those which may be labeled as comment-centered interactions on Facebook are visually distinguished from entry-centered interactions, the former appearing with a smaller letter size and being placed by the system immediately below the comment that triggered the post (see Yus, forthcoming). Furthermore, users typically post information on social media (as entries, ephemeral video stories, etc.) with three types of expectation in mind: (a) an expectation that this information will be relevant enough to arouse the audience’s attention with an expectation of relevance; (b) an expectation of mutuality of information allowing the audience to contextualize this information correctly; and (c) an expectation that this information will provoke the audience’s reaction to it and prospective interactions around it in terms of “likes,” comments, etc., what in Yus (2014) was called interactivity trigger. The problem is that the user’s audience is made up of a very heterogeneous mixture of intimate friends, acquaintances, colleagues, former students, etc. who have different accessibility to background contextual information regarding the user and the information posted on the site, and therefore will react differently to this information, ranging from cases of “joy of mutual manifestness” revealed from correct interpretation and subsequent dialogues based on the information posted, to cases of inability to contextualize the information properly, leading to misunderstanding and puzzlement. The heterogeneous quality of users making up the audience on social media was called context collapse by Marwick and Boyd (2011). Therefore, as highlighted in Yus (forthcoming), despite the possibility of interacting with hundreds of users on the site in question, users tend to interact with only a handful of users on the friends list, others being left in a permanent state of “to be contacted if ever required or desired.” Within this environment of context collapse, with varied users and heterogeneous audiences, many users do not really know how many people make up their audience on the site, and many can only identify a smaller percentage of their network correctly, so this “interactive narrowing” comes as no surprise (Szabla and Blommaert 2020: 252). In this environment, audiences are, rather, imagined audiences of which the user has no clear picture (see Beam et al. 2018: 2299) or invisible audiences that escape user control (Vitak 2012: 453). Besides, as was mentioned above, both friends and their entries on the site are managed by the site’s algorithm that decides which users are visible and which of their content can be accessed by other users. Context collapse is an interesting notion that affects interactions not only on social media but also on other types of online interaction (Androutsopoulos 2014: 63), and has an impact on contextualization and (un)successful inferential outcomes too. As Marwick and boyd (2011: 122) comment regarding Twitter, there is a requirement to present a verifiable, singular identity to the user’s audience, but this makes it impossible to differentiate selfpresentation strategies depending on group variables such as degree of closeness or job relatedness, thus creating tension as diverse groups of people flock to social network sites under the same unifying label of “audience.” In other

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

469

470

FRANCISCO YUS

words, users must hold expectations of relevance and expect that their entries will interact fruitfully with the cognitive environments of groups of users that would normally not be brought together as a unique virtual group. In order to navigate these tensions arising from reactions of (ir)relevance toward the same entry addressed to this myriad of users, a variety of tactics may be adopted, such as using multiple accounts, pseudonyms, and nicknames, among others, to obscure their real identities. Of course, these tactics depend on the purposes and agreed uses of the site itself. In certain networking sites such as Facebook or LinkedIn, there is an expectation of what in Yus (2022) is called offline–online congruence, that is, the expectation that the user remains more or less the same across these environments (see van Dijck 2013). In this sense, one of the most frequent techniques is so-called lowestcommon denominator effect, when individuals only post information that they believe their broadest group of acquaintances will find relevant and nonoffensive (see Duguay 2016: 894). In relevance-theoretic terms, users tend to make manifest information that will probably be found relevant (will interact positively with the heterogeneous audience of the user) via contextualization with the audience’s background store of contextual information against which this new piece of information is inferred. Other strategies include users negotiating “multiple, overlapping audiences by strategically concealing information, targeting tweets to different audiences and attempting to portray both an authentic self and an interesting personality” (p. 894; see also Gil-Lopez et al. 2018: 128). It should be noted that context collapse may occasionally be strategically used. This is the foundation of Davis and Jurgenson’s (2014) differentiation between context collisions and context collusions. The former occur when people’s manifest information and overall online behavior accidentally jumps to non-predicted audiences, with different social environments unintentionally and unexpectedly crashing into each other (p. 480). This process could produce negative communicative outcomes, such as disagreement (for instance, manifest information contradicting the background beliefs of some users in that audience) or that audience’s inability to contextualize the information properly. On the other hand, collusions occur when people strategically use various contexts of their social network in exchange for social utility, that is, “the process whereby social actors intentionally collapse, blur, and flatten contexts, especially using various social media” (p. 480). Potential benefits include increased bonding through engagement.

21.4.5 Social Context on Social Media Social context is made up of assumptions that are manifest to a whole collectivity within a virtual social media group, community, or gathering. The users in these groups are aware of the specificity that these assumptions entail and also of the discursive barriers generated out of this communal access to specific sets of assumptions. Very often, the users’ (in)ability to access these background

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

21 Social Media and Computer-Mediated Communication

assumptions during contextualization signals whether these users belong to the social group or are outside, as already mentioned above in passing. Normally, background social assumptions are assumed and hence not coded, which makes inference more difficult or impossible by outsiders. Clearly, contextualization acquires importance not only at a purely discursive level but also socially. It helps interlocutors frame utterances within their social and collective environment, and aids users in shaping their identities, their position in groups, and the state of their networks of friends and acquaintances. Group specificity is identified when contextualizing and inferring discourses exchanged within the bounded virtual group, but it may also be signaled when producing discourses therein. For example, certain genres are essential to understand how to participate in the collective actions of a community within the trend of participatory culture. Whether textual or social, genres are important framing devices, especially as generic conventions set up expectations (Erickson 1997; Bou-Franch and Garces-Conejos Blitvich 2018; Yus 2018: 119). An example is the LOLCat genre. Miltner (2014) uses this family of memes to illustrate how the form and structure of the LOLCat are distinct, the proper execution of the generic conventions being essential to its appeal. Participants repeatedly refer to font, text placement, image subject, and animal characterization as integral to the enjoyment of a LOLCat. In other words, adherence to specific design elements is considered paramount to LOLCats enjoyment and also reinforces collective bonding and awareness of group membership.

21.5 Concluding Remarks and Future Projections Context is information that is accessed in order to interpret utterances correctly, that is, to turn schematic coded discourses into meaningful interpretations. Since context is information stored, expected, and retrieved, it is also the array of assumptions making up the individual’s personal context, the assumptions made mutually manifest between interlocutors during interactions and the assumptions crosscutting the members of certain groups or communities. In the case of internet-mediated communication and specifically social media, this varied contextual information is also at work, but interfaces play a major role in how this information is presented, activated, processed, and combined with online discourses in the accomplishment of ordinary virtual interactions. In this case, as hinted in Yus (forthcoming), when applying pragmatic parameters such as context and contextualization, we are faced with two apparently contradictory statements. On the one hand, internet makes no difference, in the sense that online users also interpret other users’ utterances with the aid of contextual information, just like interlocutors in offline scenarios. Therefore, applications of pragmatics to this virtual environment should be straightforward. However, on the other hand internet makes all the difference, since virtual communication takes place through interfaces, often in

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

471

472

FRANCISCO YUS

a cues-filtered environment, typically text based (even nowadays), and with fewer options and resources for contextualization (e.g., lack of nonverbal communication, of physical co-presence, etc.). In future research regarding context online, some trends are bound to remain (or become) important, some of which were sketched in Yus (2019): 1. What is coded versus what is meant. Social media discourses exhibit a great variability of possibilities for contextualization, but the inferential “filling up” of coded content to be turned into propositional interpretations will remain an essential object of future research. Users will continue to use text-enriching strategies to get a more reliable interpretation of their discourses as well as to convey the feelings and emotions associated with them. 2. The pervasiveness of multimodal discourses. (Cyber)pragmatics will be particularly interested in those instances of multimodal discourses whose contextualization entails the combination of the partial meanings of the modes making up the discourse to yield relevant implicated conclusions (see Kress and van Leeuwen 1996; Jewitt 2004; van Leeuwen 2014, 2015). 3. Typed discourses with additional support. Many users still prefer to resort to cues-filtered discourses such as typed texts, and will continue to “color” these texts with emojis, stickers, text alteration, and GIFs. Contextualization may be aided by these enriching strategies applied to plain text, an interesting research area. 4. Varying sources of mutuality in the physical–virtual interface. As claimed in Yus (forthcoming), smartphones are used continuously by people in physical scenarios, and even in offline social gatherings where oral communication should be the norm. In this sense, there are prospects of research for (cyber)pragmatics regarding sources of mutuality of information and how online and offline sources of information overlap, complement, or contradict each other (Yus 2021). An example is the mixed face-to-face and virtual interactions, for example the situation of interlocutors who chat face to face and at the same time text other physically absent interlocutors with their smartphones. 5. The importance of contextual constraints. These have been highlighted as impacting the quality of contextualization strategies in this chapter. These “framing qualities” will remain an important issue in future (cyber)pragmatic research. 6. Big data and their impact on users’ interactions. As also remarked in this chapter, algorithms decide which content is offered to the user and which user may interact with whom. As such, it is an interesting area for future (cyber)pragmatic research on context and contextualization. This automatic assessment of sources of interest by algorithms does play a part in what content and users are presented to the user, and therefore also on contextualization strategies.

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

21 Social Media and Computer-Mediated Communication

7. Pragmatics of virtual agents. Nowadays, computer assistants, bots, and agents (e.g., Cortana, Siri, Alexa) are increasingly entering our daily lives. These have an amazing capacity to access information on the Internet and come up with supposedly relevant answers to the user’s queries. In the near future, researchers in computer science will be interested in how to “train” these devices to become more “pragmatic” and move closer to a human way of processing information, and also closer to the way humans combine in-coming information with context to yield relevant interpretations. This is especially noticeable in the case of online bots, which can interact through oral or typed dialogues with users (especially customers in shopping websites) and access a lot of background information available online in order to tailor the conversation to the user’s needs. 8. Online polylogues. Nowadays, interactions are increasingly devoid of the traditional dyadic quality, that is, utterances are no longer exchanged between one single sender and one single receiver with a single intended interpretation. Interactions are now more frequently multiparty and form different layers or polylogues, either text-based or with the aid of visual and multimodal content. This constitutes an interesting area of (cyber)pragmatic research regarding how these partial contributions are contextualized and contrasted with the overall interpretation of the thread or interaction as a whole (see Bou-Franch et al. 2012; Bou-Franch and Garcés-Conejos Blitvich 2014). 9. Media convergence and multiple simultaneous areas of interaction. New interfaces for interaction are becoming popular, for example on smartphones in the shape of either dedicated apps or interfaces that “migrate” from the desktop environment to the new screen-size realm of the smartphone ecosystem, with changes in the presentation and management of the content made available through these apps having an impact on how information is contextualized.

References Androutsopoulos, J. (2014). Languaging when contexts collapse: Audience design in social networking. Discourse, Context and Media, 4–5, 62–73. Beam, M. A., Child, J. T., Hutchens, M. J. and Hmielowski, J. D. (2018). Context collapse and privacy management: Diversity in Facebook friends increases online news reading and sharing. New Media & Society, 20(7), 2296–2314. Blommaert, J., Smits, L. and Yacoubi, N. (2020). Context and its complications. In A. De Fina and A. Georgakopoulou (eds.), The Cambridge Handbook of Discourse Studies (pp. 52–69). Cambridge: Cambridge University Press. Bolander, B. (2019). Social media research. In Handbook of Pragmatics, vol. 22. Amsterdam: John Benjamins, pp. 31–48. Bolander, B., and Locher, M. A. (2020). Beyond the online offline distinction: Entry points to digital discourse. Discourse, Context & Media 35.

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

473

474

FRANCISCO YUS

Bou-Franch P., and Garces-Conejos Blitvich, P. (2014). Conflict management in massive polylogues: A case study from YouTube. Journal of Pragmatics, 73, 19–36. Bou-Franch P., and Garces-Conejos Blitvich, P. (2018). Relational work in multimodal networked interactions on Facebook. Internet Pragmatics, 1(1), 134–160. Bou-Franch P., Lorenzo-Dus, N., and Garces-Conejos Blitvich, P. (2012). Social interaction in YouTube text-based polylogues: A study of coherence. Journal of Computer-Mediated Communication, 17, 501–521. boyd, D. (2010). Social Network Sites as networked publics: Affordances, dynamics, and implications. In Z. Papacharissi (ed.), A Networked Self: Identity, Community, and Culture on Social Network Sites. New York: Routledge, pp. 39–58. Costa, E. (2018). Affordances-in-practice: An ethnographic critique of social media logic and context collapse. New Media & Society, 20(10), 3641–3656. Chen, Y., and He, W. (2015). Interpretation of the compositional meaning of movie posters from the perspective of multimodal discourse analysis. Paper presented at International Symposium on College Foreign Languages Education Reform and Innovation. Wuhan (China). Davidson, B. I., and Joinson, A. N. (2021). Shape shifting across social media. Social Media + Society 7(1), 1–11. Davis J. L., and Jurgenson, N. (2014). Context collapse: Theorizing context collusions and collisions. Information, Communication & Society, 17(4), 476–485. Dayter, D. (2016). Discursive Self in Microblogging: Speech Acts, Stories and Self-Praise. Amsterdam: John Benjamins. Duguay, S. (2016). “He has a way gayer Facebook than I do”: Investigating sexual identity disclosure and context collapse on a social networking site. New Media & Society, 18(6), 891–907. Eisenlauer, V. (2014). Facebook as a third author – (Semi-)automated participation framework in social network sites. Journal of Pragmatics, 72, 73–85. Eisenlauer, V. (2017). Social network sites/Facebook. In C. R. Hoffmann and W. Bublitz (eds.), Pragmatics of Social Media (pp. 225–242). Berlin: De Gruyter Mouton. Erickson, T. (1997). Social interaction on the Net: Virtual community as participatory genre. In R. H. Sprague (ed.) Proceedings of the Thirtieth Hawaii International Conference on System Sciences (Vol. VI, pp. 13–21). Washington, DC: IEEE Computer Society Press. Gil-Lopez, T., Shen, C., Benefield, G. A., Palomares, N. A., Kosinski, M., and Stillwell, D. (2018). Journal of Computer-Mediated Communication, 23, 127–145. Gruber, H. (2019). Genres, media, and recontextualization practices: Re-considering basic concepts of genre theory in the age of social media. Internet Pragmatics, 2(1), 54–82. Hassan, H., and van Leeuwen, T. (2008). Semiotic analysis: Compositional meaning of corporate web pages. In H. Habil and H. Hassan (eds.), New Perspectives in Language and Communication Research (pp. 39–52). Skudai: UTM. Jewitt, C. (2004). Multimodality and new communication technologies. In Philip LeVine and Ron Scollon (eds.), Discourse and Technology. Multimodal Discourse Analysis (pp. 184–195). Georgetown: Georgetown University Press. Jones, R. (2004). The problem of context in computer-mediated communication. In P. LeVine and R. Scollon (eds.), Discourse and Technology: Multimodal Discourse Analysis (pp. 20–33). Washington: Georgetown University Press.

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

21 Social Media and Computer-Mediated Communication

Kress, G., and van Leeuwen, T. (1996). Reading Images: The Grammar of Visual Design. London: Routledge. Lomborg, S. (2014). Social Media, Social Genres: Making Sense of the Ordinary. Abingdon: Routledge. Marwick, A. and boyd, D. (2011). I tweet honestly, I tweet passionately: Twitter users, context collapse, and the imagined audience. New Media & Society, 13(1), 114–33. Miltner, K. M. (2014). “There’s no place for lulz on LOLCats”: The role of genre, gender, and group identity in the interpretation and enjoyment of an internet meme. First Monday, 19(8). https://doi.org/10.5210/fm.v19i8.5391. Murphy, K. (2020). Why Zoom is terrible. The New York Times, April 29. www .nytimes.com/2020/04/29/sunday-review/zoom-video-conference.html. Pano Alamán, A., and Mancera Rueda, A. (2014). La “conversación” en Twitter: Las unidades discursivas y el uso de marcadores interactivos en los intercambios con parlamentarios españoles en esta red social [The “conversation” on Twitter: The discursive units and the use of discourse markers in exchanges with Spanish members of parliament in this social networking site]. Estudios de Lingüística del Español, 35, 243–277. Puschmann, C. (2015). The form and function of quoting in digital media. Discourse, Context and Media, 7, 28–36. Quinn, K. (2016). Contextual social capital: Linking the contexts of social media use to its outcomes. Information, Communication & Society, 19(5), 582–600. Scott, K. (2015). The pragmatics of hashtags: Inference and conversational style on Twitter. Journal of Pragmatics, 81, 8–20. Sperber, D., and Wilson, D. (1995). Relevance: Communication and Cognition, 2nd ed. Oxford: Blackwell. Spilioti, T. (2015). Social media discourse. In K. Tracy, C. Ilie, and T. Sandel (eds.), The International Encyclopedia of Language and Social Interaction. Chichester: Wiley & Sons. Szabla, M., and Blommaert, J. (2020). Does context really collapse in social media interaction? Applied Linguistics Review, 11(2), 251–279. Thompson, J.B. (2020). Mediated interaction in the digital age. Theory, Culture & Society, 37(1), 3–28. van Dijck, J. (2013). “You have one identity”: Performing the self on Facebook and LinkedIn. Media, Culture & Society, 35(2), 199–215. van Leeuwen, T. (2014). About images and multimodality: A personal account. In S. Norris and C. D. Maier (eds.), Interactions, Images and Texts: A Reader in Multimodality (pp. 19–24). Berlin: de Gruyter. van Leeuwen, T. (2015). Multimodality. In D. Tannen, H. E. Hamilton, and D. Schiffrin (eds.), The Handbook of Discourse Analysis, 2nd ed. (pp. 447–465). Chichester: Wiley. van Leeuwen, T. (2020). The new visuality of writing. In H. Stöckl, H. Caple, and J. Pflaeging (eds.), Shifts toward Image-Centricity in Contemporary Multimodal Practices (pp. 64–85). Abingdon: Routledge. Vitak, J. (2012). The impact of context collapse and privacy on social network site disclosures. Journal of Broadcasting & Electronic Media, 56(4), 451–470. Wikström, P. (2014). #Srynotfunny: Communicative functions of hashtags on Twitter. SKY Journal of Linguistics, 27, 127–152.

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

475

476

FRANCISCO YUS

Yus, F. (2007). Virtualidades reales: Nuevas formas de comunidad en la era de Internet [Real virtualities. New forms of community in the Internet age]. Alicante: University of Alicante, Servicio de Publicaciones. Yus, F. (2011). Cyberpragmatics: Internet-Mediated Communication in Context. Amsterdam: John Benjamins. Yus, F. (2014). El discurso de las identidades en línea: El caso de Facebook [The discourse of online identities: The case of Facebook]. Discurso & Sociedad, 8(3), 398–426. Yus, F. (2015). Discourse and identity. In J. D. Wright (ed.), International Encyclopedia of the Social and Behavioral Sciences, 2nd ed. (Vol. VI, pp. 498–502). Oxford: Elsevier. Yus, F. (2016a). Humour and Relevance. Amsterdam: John Benjamins. Yus, F. (2016b). Discourse, contextualization and identity shaping. The case of social networking sites and virtual worlds. In M. L. Carrió-Pastor (ed.), New Tools, New Approaches: Technology Implementation in Higher Education for Second Language Teaching and Translation Studies (pp. 71–88). Berlin: Mouton de Gruyter. Yus, F. (2016c). Towards a cyberpragmatics of mobile instant messaging. In J. RomeroTrillo (ed.), Yearbook of Corpus Linguistics and Pragmatics 2016: Global Implications for Culture and Society in the Networked Age (pp. 7–26). Berlin: Springer. Yus, F. (2017a). Social media. In A. Barron, Y. Gu, and G. Steen (eds.), The Routledge Handbook of Pragmatics (pp. 550–562). Abingdon: Routledge. Yus, F. (2017b). Putting relevance at centre stage in research on human activity on the Internet. In A. Piskorska and E. Wałaszewska (eds.), From Discourse to Morphemes: Applications of Relevance Theory (pp. 86–102). Newcastle-upon-Tyne: Cambridge Scholars Publishing. Yus, F. (2017c). Contextual constraints and non-propositional effects in WhatsApp communication. Journal of Pragmatics, 114, 66–86. Yus, F. (2018). Identity-related issues in meme communication. Internet Pragmatics, 1(1), 113–133. Yus, F. (2019). An outline of some future research issues for internet Pragmatics. Internet Pragmatics, 2(1), 1–33. Yus, F. (2021). Cyberpragmatics in the age of locative media. In C. Xie, F. Yus, and H. Haberland (eds.), Approaches to Internet Pragmatics: Theory and Practice (pp. 75–105). Amsterdam: John Benjamins. Yus, F. (2022). Smartphone Communication: Interactions in the App Ecosystem. Abingdon: Routledge. Yus, F. (forthcoming). Irony and the Internet. In R. W Gibbs and H. L Colston (eds.), The Cambridge Handbook of Irony and Thought. Cambridge: Cambridge University Press. Zappavigna, M. (2017). Evaluation. In C. R. Hoffmann and W. Bublitz (eds.), Pragmatics of Social Media (pp. 435–458). Berlin: De Gruyter Mouton.

https://doi.org/10.1017/9781108989275.022 Published online by Cambridge University Press

Index

Abercrombie, D., 420–421, 431 action theory see also philosophy of action, 2, 95 Adaptive Context, 319–323, 326 adjacency pair, 16–19, 23 Aijmer, K., 4, 135, 289, 291, 294, 298, 302, 304–305, 307–313 Allan, K., 222, 247, 263–264, 309 ambiguity, 53, 108, 151, 155, 157–158, 188, 193, 202, 319, 354–356, 359–363, 365, 368, 403, 414, 443 analogization see also analogy, 62, 64 analogy see also analogization, 62, 64, 127, 233 Andersen, G., 301, 307–308, 312 Archer, D., 212–213, 221, 292, 308 Ariel, M., 252, 264, 294, 309 artificial intelligence (AI), 5, 130–131, 184, 285, 357, 436, 440, 444 assessment, 4, 19, 24, 41, 72, 205, 331–335, 338–342, 345–348, 373, 377, 388, 397–398, 405–407, 413 Atkinson, J. M., 12, 32–34, 37–38, 40–41, 44–45, 326 Attardo, S., 354–355, 357, 364, 369 Austin, J. L., 9, 37, 87, 95–98, 104, 113, 178, 183, 188, 201, 207, 221, 290, 309, 337, 348 automatic speech recognition (ASR), 436, 438–440 background knowledge, 188, 214, 252, 269–272, 277–278, 280–281, 284, 290, 294, 424 See also sociocultural background knowledge, 269–272, 277–278, 280–281, 284 Baker, M., 27, 157, 202, 274, 281, 283, 285, 372–375, 379, 383, 385–386, 389–392, 437 Barron, A., 301–303, 309–311, 313 Beeching, K., 294–297, 309 Bernstein, B., 128, 133 Berry, M., 76, 87, 116, 120–121, 123–124, 133 Bialystok, E., 332, 348 Biber, D., 294–295, 309–310 Blakemore, D., 60, 66, 191, 201, 424, 431 Blochowiak, J., 194, 197, 201, 264 Blommaert, J., 78–79, 82–85, 88 Bolinger, D., 25, 37, 314–316, 324, 422, 426–427, 429, 431 Brazil, D., 144, 156, 315, 324, 422, 431

Brezina, V., 309, 311 Brinton, J.L., 292, 294, 303, 309, 349 Brown, P., 20, 31, 37, 39, 42, 45, 47, 89, 157, 200, 202, 222, 297, 309, 321, 324, 432 Bühler, K., 118, 153, 157 Butler, C., 116, 118, 131, 134, 297, 309 Buysse, L., 306, 309 Bybee, J., 49–50, 65–66, 172, 176 Byram, M., 337, 348 Carston, R., 60, 66, 157, 194, 202, 222, 247, 249, 251–252, 258, 260, 263–264, 266–267, 425, 433, 435 Chafe, W., 145–146, 157, 316, 318, 324, 354, 369 Chang, Y.-F., 210–211, 221–222 Chiaro, D., 4, 353, 356, 362, 369 Chomsky, N., 9, 37, 141, 143, 149, 157, 161–162, 164, 176, 248, 264, 268, 285–286, 336–337, 340–341, 348 Clancy, B., 293, 295, 299, 309 Claudi, U., 51, 67 clinical linguistics, 5, 184, 393–394, 397–398, 412–413 cognition, 9, 37, 40, 46, 66–67, 69, 78, 86, 90, 131, 138, 158–159, 161, 176, 178, 201, 203, 223, 226, 245, 247, 254, 256, 266, 268, 275–276, 285–287, 312, 325, 327, 332, 388–389, 391–392, 401, 422, 434 cognitive linguistics, 3, 65, 161, 173, 176, 178, 271, 273, 287 cognitive episode, 104–105 cognitive significance, 100–101, 104, 108–109, 113–114 cohesion, 16, 23, 40, 89–90, 160, 358 common ground, 147, 153–154, 184, 188–192, 195, 197–198, 252–254, 262, 266, 270, 273–274, 278, 284, 325, 381, 387 communication disorder, 393, 397, 409, 411–413 competence, 140, 143, 146, 149–150, 164, 171, 184, 208, 210, 221–222, 282, 287, 321, 323, 332, 337–338, 340–341, 348–349, 354, 365, 368, 406, 433 computer-mediated communication, 5

https://doi.org/10.1017/9781108989275.023 Published online by Cambridge University Press

478

INDEX

conceptual knowledge, 271–274, 276–278, 282, 284–285 Conde-Silvestre, C., 67 conflict, 15, 197, 240, 326, 343, 374, 382, 384, 386 constituency, 140, 143, 343 Construction Grammar, 66–70, 150–151, 157, 177, 206, 215, 221 context of situation, 72–73, 116–117, 119, 121, 124–125, 127, 130 contextualization cue, 74–75, 86–87, 213, 271, 294, 380 Conversation Analysis (CA), 10–12, 14–15, 21–22, 28, 32, 34–35, 37–40, 43, 45, 47, 68, 177, 264, 336–340, 349, 406, 413, 434 corpus linguistics, 4, 135, 289–291, 301, 303–304, 307–309, 311–312, 374 corpus pragmatics, 4, 289–294, 296–297, 299–300, 302–304, 306–313, 326 co-text, 49, 60, 64–65, 69, 73, 77, 117, 212–214, 220 Couper-Kuhlen, E., 11, 20, 25, 38–39, 47 Critical Discourse Analysis (CDA), 73, 77, 82, 91, 136, 138 Critical Discourse Studies see also Critical Discourse Analysis, 71, 81, 349 Croft, W., 50–51, 65–66, 173, 176, 272, 276, 285 Cruse, D. A., 272, 279, 285 Cruttenden, A., 315–316, 324 Crystal, D., 316, 324, 394–395, 413, 437, 440 Culpeper, J., 205, 212–213, 215, 221, 292, 297, 308–309 cultural keyword, 226, 234–236 cultural script, 226, 234, 239–240, 242–243, 245–246 Cummings, L., 5, 393–394, 397, 400–401, 404, 408–409, 411, 413 Davidse, K., 59, 66, 69 Davitti, E., 382, 387 De Fina, A., 79, 81, 87, 90 De Ponte, M., 2, 95, 100, 104, 109, 113 de Saussure, F., 95, 113 Deschrijver, C., 2, 71, 83, 86, 88 Desilla, L., 378, 387–388 diachronic construction grammar, 51–52, 57, 62 dialogue, 5, 42–43, 89, 91, 141, 149, 153–154, 156, 158–159, 292, 311, 360–361, 378, 384–385, 387–390, 436–437, 440 Dirven, R., 49, 67, 118, 134, 166, 173, 176–177 Discourse Analysis (DA), 2, 39–40, 42, 71, 73, 88, 90–91, 138 discourse marker (DM) see also pragmatic marker, 46, 50, 65–67, 74–76, 86–87, 90, 147, 213, 223, 294, 300, 304, 310–312, 326, 336, 341, 431 Donnellan, K.S., 98, 100, 113 Drew, P., 13, 20, 22–23, 25, 28–34, 37–39, 42, 46, 48, 223, 380, 388 Duranti, A., 35, 39, 72, 78, 88–89, 319, 325, 388–389 Dynamic Model of Meaning (DMM), 319–320 Ellis, R., 65–66, 119, 129, 131, 134, 331, 349 encyclopedic knowledge, 4, 167, 268, 271–272, 274–278, 280, 283–286, 363 Enfield, N., 25–26, 31, 38–39, 41, 46–47 entailment, 192, 198 Escandell-Vidal, V., 422, 424, 432

explication see semantic explication, 228–231, 233, 235–238 explicature, 186, 258–260, 264 Fairclough, N., 71, 82, 88, 345–346, 349–350, 352 Fauconnier, G., 160, 166, 176, 273, 286 Fawcett, R., 117, 123, 129, 134, 389 Fetzer, A., 50, 67, 76, 88, 206, 213, 222–223, 319, 324 field, 19, 22, 30, 116, 122, 133, 138 Fillmore, C.J., 163, 167, 176, 218, 222, 272, 280, 286 first language (L1), 210, 270, 272–274, 280, 283, 285–286, 319, 323, 331–336, 340–342, 346–347, 395 Fishman, J., 342, 349 Fodor, J., 249, 265 Fontaine, L., 3, 116, 121, 123, 133–138 Fox, B. A., 23, 25, 39, 46–47, 184, 198–199, 202, 309, 366, 369 Frege, G., 95–97, 99–102, 104–105, 109, 111, 113, 115, 188, 202 Freire, P., 338, 349 Freudian slip, 358, 363 Functional Discourse Grammar (FDG), 142, 146–147, 149, 153, 156, 158 functional models see also functionalism, 118 functionalism, 118–119, 134, 156–157 Garfinkel, H., 10, 38–40, 89, 325 Gazdar, G., 183, 193–194, 202 Geeraerts, D., 3, 52, 67, 160–161, 165, 173–174, 176–179, 273, 279, 286 Geluykens, R., 318, 324 generative grammar, 141, 143, 158, 160–164 Generative Semantics see also generative grammar, 162–163 genre, 120, 122, 137, 206–207, 212–213, 215–217, 219–223, 290, 304, 307–308, 373–376, 378 Georgakopoulou, A., 80–81, 84–85, 87–88, 90 Gibbs, R. W., 248, 265–266, 275–276, 286, 290, 310 Giora, R., 319, 324 Gironzetti, E., 4, 331, 345–346, 348–349 Givón, T., 51, 67, 145–146, 157, 287 Gladkova, A., 3, 225–226, 237, 239–241, 244, 246 Goddard, C., 226–231, 233, 239–240, 243–246, 271, 286 Goffman, E., 9–10, 14, 21, 35, 40–41, 73, 84, 89, 380, 388 Goldberg, A., 30, 52, 57, 67–68, 150, 157, 211, 222 Gonzálvez-García, F., 118, 131, 134 Goodwin, C., 25, 30–32, 40, 47, 72, 78, 88–89, 319, 325, 388–389, 434 grammaticalization, 50–56, 59–60, 62–63, 66–69, 304, 309 Grice, H.P., 56, 67, 95–98, 102, 104, 113–114, 183–184, 186–189, 192, 194, 201–202, 207, 222, 247, 249, 251–252, 258, 265, 290, 310, 362, 422, 425–426, 432 Grisot, C., 194, 201, 264 Gumperz, J.J., 74–76, 78, 86–89, 269, 282, 286, 294, 310–311, 380, 388 Gussenhoven, C., 316, 325, 422, 429, 432 Halliday, M. A. K., 16, 23, 40, 73, 89, 116–128, 132–136, 138, 315–319, 321, 325, 422, 432

https://doi.org/10.1017/9781108989275.023 Published online by Cambridge University Press

Index

Hasan, R., 16, 23, 40, 73, 89, 119–120, 123–124, 127, 129–130, 134, 136, 138, 319, 325 Hatch, E., 338, 349 Hatim, B., 373, 376, 388 Haugh, M., 36, 38, 41, 211, 222, 224 Head-Driven Phrase Structure Grammar, 142, 158 Heine, B., 51, 53–54, 56, 67–69 Hengeveld, K., 142, 145–147, 149, 153, 158 heritage language (HL), 4, 331, 334, 342–347, 349–352 Heritage, J., 2, 9–10, 12–17, 19–21, 23–26, 30, 32–35, 37–42, 44–48, 74, 89, 223, 319, 325–326, 342–344, 349–352, 380, 388 Hernández-Campoy, J.M., 67 Hirschberg, J., 249, 265, 315, 326, 422, 433–434 historical linguistics, 2, 49, 51, 64, 303 Hockett, C., 356–357, 369 holophrase see also holophrastic utterances, 144–146, 148, 156 holophrastic utterances, 144–145, 154 Hopper, P.J., 51, 61, 68 Horn, L.R., 56, 61, 68, 183, 191, 194, 196, 199, 202, 247, 249, 265, 267, 327, 392, 435 House, J., 37, 281, 286, 349–350, 372–373, 376, 387–388, 422, 424–425, 433 Huddleston, R., 142, 158 humor, 4, 211, 251, 353–358, 360, 362–370, 443 Hünnemeyer, F., 51, 67 Hymes, D., 76, 89, 377, 389 Ifantidou, E., 211, 222, 434 implicature, 57, 63, 68, 95, 113, 184, 186–191, 193–200, 202, 207, 212, 214, 218, 224, 249–251, 258–260, 262, 264–265 conventional implicature, 184, 189–191, 201–202 conversational implicature, 3, 52–53, 56, 183–184, 187–192, 194–195, 201, 203, 264–265 generalized implicature, 249–251 particularized implicature, 249, 251 scalar implicature, 184, 189, 197–203, 250, 264 incrementality, 148 incremental syntax, 143 indexicality, 74, 79, 82, 85, 300 Indurkhya, N., 5 inference, 53, 68, 97, 107, 124, 145, 158, 196–202, 211–214, 220, 224, 248, 254, 264–265, 310, 326, 422–424, 429 institutional talk, 32–34 Intercultural Pragmatics, 114, 210, 245, 286–287 interlanguage, 209–210, 221, 223, 305, 307 intonation, 29, 48, 213, 271, 314–318, 321–327, 332, 425–427, 429, 431–434 irony, 211, 251, 355, 359, 366, 368, 402–403, 412 Jackendoff, R., 142, 157–158, 268, 287 Jaszczolt, K., 222, 247, 263–264, 309 Jefferson, G., 14–15, 25, 29–31, 36, 42, 45–46, 312 Johnson, M., 166, 177, 253, 262–263, 280, 286, 334, 349, 366, 412 Johnson-Laird, P. N., 280, 286 Joke, 353–355, 357, 359, 364–365, 367–369 Jucker, A. H., 49, 68, 205, 212, 222, 291, 298, 303, 305, 307–308, 310–312, 432

Kasper, G., 209, 222 Kecskes, I., 4, 210, 213, 222, 268–271, 274, 276, 278, 280–281, 286, 319–320, 325 Kendon, Al, 9, 42, 420–421, 427, 430, 433 Koike, D., 4, 331 Korta, K., 2, 95, 100, 104, 109, 113–114, 184–186, 202 Kövecses, Z., 166, 169, 173, 177 Krashen, S., 331, 336, 348–349 Kress, G., 431, 433 Kristiansen, G., 173, 176–178 Ladd, R., 314, 316, 325, 422, 425–427, 433 Lakoff, G., 160, 163, 165–166, 177, 218, 222, 280, 287 Langacker, R. W., 59, 68, 123, 136, 160, 163, 166, 168, 171–172, 177, 187, 202, 273, 287 language change, 2, 49–51, 59, 309 language disorder, 5, 393–399, 401, 403, 405–406, 408, 410–411 language learning, 4, 184, 223, 276, 331–332, 336–338, 342, 346–347, 349–352 LaPolla, R. J., 284, 287 Leech, G. N., 205–208, 213, 215, 220, 222, 290–291, 309, 311 Leibniz, G. W., 226, 245, 278 Lepore, M., 249, 251–252, 262, 264–265 Levinson, S. C., 11, 21, 31–32, 36–37, 39, 42–43, 45–47, 56, 68, 89, 144, 158, 183, 191, 196–197, 200, 202, 208, 212, 223, 249–252, 265, 282, 286, 297, 304, 309–311, 314, 319, 321, 324–325, 361 Levisen, C., 234–236, 245–246 Lewis, D., 111–112, 114, 252–253, 265–266, 304, 307–308, 433 Lexical Functional Grammar, 142, 157 lexicogrammar, 117, 121–123, 125–127, 133 linguistic anthropology, 74, 375 linguistic creativity, 4, 353, 355 linguistic knowledge, 4, 162, 271–272, 274–278, 284, 365, 407 Logic, 67, 96–99, 103, 107–108, 113–115, 144, 162–163, 183, 185, 200, 202–203, 222, 245, 264–265, 267, 310, 385, 434 Long, M., 51, 68, 337, 350, 441, 444 Love, R., 293, 309, 311, 387 Mackenzie, J. L., 3, 140, 142, 145–147, 149, 153, 157–158 Madella, P., 5, 318, 325, 327, 419–420, 422, 426–429, 433 Malinowski, B., 72–73, 90, 118–119, 319, 325 Manner, see Maxim of Manner, 187, 196 Marmaridou, S., 3, 205–207, 213, 219, 223 Martin, J., 67, 73, 90, 120–121, 131, 137, 157–158, 178, 354, 369 Mason, I., 371, 373, 376, 380–382, 388, 390 Matthiessen, C. M., 121–122, 124, 126, 132, 134, 136–137, 316, 318, 325 Mauranen, A., 144, 159 Maxim of Minimization, 144 Maxim of Quantity, 144, 199 McCarthy, M., 299, 309 McEnery, T., 311 McLoughlin, I., 5, 436, 438, 440 metaphor, 52, 166, 169–170, 173, 176–178, 211, 222, 234, 251, 264, 266–267, 286, 359, 400, 412 conceptual metaphor, 161, 166, 173, 177

https://doi.org/10.1017/9781108989275.023 Published online by Cambridge University Press

479

480

INDEX

metonymy, 52, 166, 169–170, 176–177, 216 Mey, J.L., 290, 311 minimal language, 226, 243–244 misunderstanding, 18, 154, 252–253, 280, 283, 320–321, 323–324, 354, 359 mode, 116, 122, 138 Moeschler, J., 3, 183, 194, 199–200, 202–204, 249, 267 molecule, see semantic molecule, 228, 233 Montague, R., 163, 178 Morris, C. W., 95, 114, 207, 223, 380, 390 multimodality, 30, 43, 83, 90, 117, 128–130, 133, 137, 341, 419–420, 426, 431–432 gaze, 15, 22, 31, 42, 45, 146, 156, 168, 362, 382, 387, 390, 421, 430 gesture, 22, 31–32, 40, 50, 107, 129, 146–147, 154, 353, 402–403, 426–428, 430–434 Munday, J., 373, 390 natural language, 3, 5, 96–98, 103, 163, 165, 187, 193, 201, 225–226, 275, 279, 398, 436, 442 Natural Language Processing (NLP), 5, 436, 440, 442–444 Natural Semantic Metalanguage (NSM), 3, 225–229, 231, 233–236, 239, 242–243, 245–246 Nida, E. A., 372, 390 Nikiforidou, K., 206, 215, 217–219, 221, 223 nonverbal communication, 5, 251, 419–420, 433–434 Norrick, N. R., 222–223, 295, 311 Noveck, I., 183, 197, 199–200, 203, 249–251, 266–267, 429, 434 Nuyts, J., 161, 178 Ochs, E., 33, 43, 46, 75, 90, 300, 311 ostensive communication, 256, 258 Östman, J. O., 206, 216, 222–223 paralanguage, 419–421, 429–431, 434 Peeters, B., 226, 246, 287 Peirce, C. S., 95, 114, 207 Pérez-González, L., 5, 371, 376, 381, 385–386, 388, 390–391 performance, 10, 78, 87, 136, 140, 143, 146, 164, 172, 209, 219, 237, 268, 315, 337, 340–341, 348, 367, 370, 375, 390, 404, 411, 436, 439, 441–442 Perry, J., 2, 95, 99–100, 103–104, 108–109, 111–114, 184–186, 201–202 philosophy of action see also Action theory, 95, 98, 103–104 philosophy of language, 2, 95–98, 103–104, 106–107, 112, 159, 183, 188, 203, 205, 265, 312 Pierce, R. S., 404, 412–414 Pinker, S., 268, 287, 352, 354, 370 pointing, 31, 40, 75, 82, 107, 153, 218, 256, 318, 325, 345, 405, 426–427, 430, 433 politeness, 37, 200, 202, 208–210, 291, 294, 297, 300, 303, 307, 309–310, 321, 324 Pomerantz, A. M., 20–21, 25, 44, 103, 319, 326 Poyatos, F., 421, 434 Prabhu, N. S., 337, 351 pragmalinguistics, 205, 207–210, 213, 220, 223 pragmatic marker (PM) see also discourse marker, 185, 253, 262, 291–292, 294–297, 300–302,

304–305, 307–309, 311–312, 317–318, 321–323, 326 pragmatic punting, 317, 326 preference organization, 19–21, 44 pre-sequence see also sequence, 20, 26–27 presupposition, 74, 98–99, 145–146, 190, 193, 202, 265 prime, see semantic primitive, 72, 226–227, 247, 253, 262–263, 274, 366 Prince, E. F., 318, 326 Principle of Informativeness, 144 progressivity, 28–30, 41, 47 proper names, 55, 99–100, 109, 113, 217 proposition, 75, 95, 99, 101, 103, 105–106, 109, 111–112, 163, 184–186, 188, 190–192, 194, 249, 258, 275–276 prosodic pragmatics, 314–315, 323, 326 prosody, 4, 50, 75, 90, 154, 156, 294, 314–315, 317–320, 322–323, 326, 389, 421–426, 428–429, 431–435, 440 prototype theory, 165, 171–173, 176, 286, 352 psycholinguistics, 3, 143, 147, 149–150, 156, 355, 388, 394 Pullum, G. K., 142, 158 Quality, see Maxim of Quality, 42, 187, 195, 199, 388 Quantity, 56, 187, 195–196, 198–199, 264–265 see Maxim of Quantity Rampton, B., 283, 287 Raskin, V., 354, 357, 364, 369 Reboul, A., 183, 195, 199–201, 203–204, 249, 265, 267 Recanati, F., 183, 203, 252, 266, 319, 326 recursivity, 76, 83 referentialism, 100–101 register, 65, 73, 119, 122, 136–138, 222, 325, 347, 356, 373, 376, 379, 450 Relation, see Maxim of Relation, 135, 187, 196 Relevance Theory, 4, 60, 183, 187, 191, 247, 254–256, 263–264, 267, 291, 325, 392, 420–422, 425, 429–433, 435 repair, 15, 18, 22, 28–30, 32, 36, 38–39, 43, 45–46, 155, 321 Romero-Trillo, J., 1, 4, 213, 223, 237, 239, 244, 291, 305, 309, 311–312, 314–318, 320–323, 325–327, 430, 433 Rosch, E., 165, 173, 178, 280 Rose, K. R., 209, 222–223 Roy, C., 200, 203, 380, 391 Rühlemann, C., 291–292, 308–313 Russell, B., 96–98, 100–102, 104, 114, 114, 121, 125, 139 Sacks, H., 9–11, 14–17, 20–21, 24–26, 30, 32, 39, 45–46, 299, 312 Sadeghi, Y., 4, 314, 327 salience, 165, 172, 217, 236, 257, 259–260, 270, 279, 324–325, 425 Schegloff, E. A., 9–10, 12, 14–20, 23, 25–26, 28–31, 33, 35, 45–46, 71, 73, 80, 82, 90, 312 Schieffelin, B.B., 33, 43 Schiffrin, D., 16, 46, 48, 68, 75–76, 87, 90, 202, 265, 294, 312, 327

https://doi.org/10.1017/9781108989275.023 Published online by Cambridge University Press

Index

Scott, K., 66, 157, 222, 263–266, 422, 426, 428–429, 433–434 Searle, J., 95, 103, 114, 146, 159, 183–184, 203, 207, 223, 252, 266, 268, 287, 290, 312, 319, 327, 337, 351 second language (L2), 11, 50, 134, 142, 206–211, 222–223, 273–274, 276, 278–279, 281, 283–287, 312, 319, 323–326, 331–333, 335–342, 344–349, 351, 428, 433 semantic molecule, 225, 228–229, 243, 245 semantic primitive, see prime, 225–226, 228–229, 246 semantic template, 229, 231, 233 semantic-pragmatic change see also language change, 50–52 Semino, E., 173, 178 Sequence, 10, 12–15, 17–19, 23, 25, 27, 29–32, 34–35, 39, 41–42, 46, 73, 76, 105, 126, 141–143, 146–147, 150, 184, 259, 317, 337, 375, 438–444 Sharifian, F., 173, 178 Simpson, P., 366, 370, 386 Sinclair, J., 65, 144, 159 Slobin, D. I., 273, 276, 287 small stories, 72, 79, 86–88 social action, 9–11, 20, 30, 40, 44, 78, 84, 130 social media, 2, 5, 85–86, 89, 139, 184, 296, 367 sociopragmatics, 3, 205–214, 220, 222–224, 307 speech act theory, 11, 39, 43, 60, 95, 101, 104, 114, 121, 159, 183, 207–210, 214–215, 223, 226, 234, 249, 286–287, 290–292, 298–304, 307, 309–310, 312–313, 319, 337, 362, 402, 405 speech community, 50, 144, 149, 270, 274, 278, 281, 309 speech disorder, 396–397 speech processing, 5, 350, 437, 441 speech input, 437 speech output, 437, 439 speech-language pathology, 393–394, 397–398, 405–406, 408, 410–411 Sperber, D., 60, 69, 183, 186–187, 203, 247, 249–255, 257–263, 266–267, 290, 312, 319, 327, 378, 391–392, 423–426, 428–430, 434–435 Spolsky, B., 334, 351 stance, 11, 14, 24–25, 29, 38–41, 43–44, 47, 67, 75, 79, 89, 135, 294, 300, 310, 322, 373, 383–384, 421 Stivers, T., 11, 14, 25–26, 30, 35, 37, 39, 41, 43–45, 47 Stone, M., 251–252, 262, 264–265 storytelling, 15, 32, 47, 80–81, 341, 348 Strawson, P. F., 96–99, 101–103, 113–114, 190, 203 structuralism, 91, 160, 164, 336, 341 subjectification, 58–61, 66, 68–69 Svartvik, J., 135, 293, 312 Swan, M., 340, 351 Sweetser, E. E., 61, 69

Systemic Functional Grammar (SFG), 73, 134, 137, 160 Szczepek Reed, B., 47 Talmy, L., 67, 157, 160, 166, 168, 178 Taylor, J. R., 160, 165, 171, 177–178, 245, 274–275, 287, 432, 439 Tenor, 41, 116, 122, 124, 138 Terkourafi, M., 200, 203, 214, 224 theory of descriptions, 98 Thomas, J. A., 68, 177, 205–208, 220, 224, 361, 370, 389, 401, 414 Thompson, S. A., 25, 39, 46–47, 123, 133–135, 137–138 tonality, 317 tone, 315–317, 323 tone unit, 316–317, 321 tonicity, 317 Trager, G. L., 420, 434 Traugott, E., 2, 49, 51–52, 54, 56–57, 59–60, 62, 68–70, 289, 296, 312 turn-taking, 14, 17, 30, 32, 34, 36, 47, 49, 65, 299, 307, 309, 312, 382, 391 van Dijk, T. A., 78–79, 83, 90, 131, 138 van Leeuwen, T., 431 variational pragmatics, 300, 302, 304, 307, 309 Verschueren, J., 43, 72, 74–75, 87, 90, 158, 206, 213, 222, 224, 290, 312 Vygotsky, L., 68, 338, 352 Ward, G., 202, 247, 265, 267, 316, 327, 392, 422, 433–435 Wegener, R., 3, 116–117, 119–121, 123, 130–132, 134, 136–138 Weigand, E., 153, 156, 159, 369 Wharton, T., 5, 263, 267, 419–426, 430, 433–435 Whitehouse, A. J. O., 411, 414 Whorf, B. L., 118, 282, 288 Wichmann, A., 66, 422, 431, 434 Wierzbicka, A., 225–231, 234–235, 238–239, 243–246, 274, 276, 288, 321 Wilson, D., 4, 10, 48, 58, 60, 69, 157, 183, 186–187, 203, 247, 249, 251–252, 254–264, 266–267, 290, 312, 319, 327, 378, 391–392, 404, 414, 420, 422–426, 428–430, 434–435 Wittgenstein, L., 9, 48, 96, 98, 207 Wodak, R., 77, 88, 90–91, 138 Yus, F., 5 Zappavigna, M., 131, 139 Zufferey, S., 199, 203–204, 249, 267

https://doi.org/10.1017/9781108989275.023 Published online by Cambridge University Press

481

https://doi.org/10.1017/9781108989275.023 Published online by Cambridge University Press