Patricia Bou - Analyzing Digital Discourse PDF

Edited by Patricia Bou-Franch and Pilar Garcés-Conejos Blitvich
Analyzing Digital Discourse
New Insights and Future Directions

Analyzing Digital Discourse
“With contributions by established as well as new scholars, this volume pulls

together research on discourse and communication in various languages and
digital spaces, from consumer reviews and health forums to social networking
and video interaction. All papers examine issues at the forefront of current
research, focusing on multimodality, identities in discourse, relationship work,
and language ideologies. A kaleidoscope of contemporary research in language
and digital communication, offering new insights in global patterns and local
details of digital discourse.”
—Jannis Androutsopoulos, Professor in German and Media Linguistics,
Universität Hamburg, Germany
“Analyzing Digital Discourse includes an exciting range of studies that are at the
cutting edge of the field. Going beyond the foci of many earlier studies, this col-
lection interrogates examples of digital discourse that range from parody Amazon
reviews, profiles on LinkedIn to multi-semiotic data such as sexting messages,
memes and emoji. Each of these studies is interesting in their own right and
together demonstrates the importance of analysing online interactions both at
the micro and macro level.”
—Ruth Page, Senior Lecturer, University Birmingham
“This book brings together original research in the field of digital discourse anal-
ysis, offering new critical insights and exciting avenues for research. While
engaging with widely debated issues such as face and identity online, the volume
also rewards the reader with a range of well-informed and nuanced approaches
to the study of multimodality, as well as language and media ideologies—highly
recommended to anyone who wants to understand and analyse digital discourse
in a networked world.”
—Tereza Spilioti, Senior Lecturer, Cardiff University
“Taking a critical and situated perspective on social media platforms and com-
munities ranging from Amazon reviewers to Greek Twitter and British sexters,
this cutting-edge volume lays down exciting new paths for future research which
embrace three important aspects of contemporary digital communication: its
multimodal nature; the mediated co-construction of identity and sociability; and
the discursive (re)construction of ideologies online. An absolute must-read for
anyone interested in the development of the field of digital discourse studies.”
—Dr Caroline Tagg, Lecturer in Applied Linguistics, Open University, UK
Patricia Bou-Franch
Pilar Garcés-Conejos Blitvich
Editors
Analyzing Digital
Discourse
New Insights and Future Directions
Editors
Patricia Bou-Franch Pilar Garcés-Conejos Blitvich
IULMA - Department of English and Department of English
German Philology University of North Carolina at Charlotte
Universitat de València Charlotte, NC, USA
Valencia, Spain
ISBN 978-3-319-92662-9 ISBN 978-3-319-92663-6 (eBook)

https://doi.org/10.1007/978-3-319-92663-6
Library of Congress Control Number: 2018948842
© The Editor(s) (if applicable) and The Author(s) 2019

This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and
transmission or information storage and retrieval, electronic adaptation, computer software, or by similar
or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, express or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
Cover design by Tom Howey
This Palgrave Macmillan imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Contents
Part I Introduction 1
1 Introduction to Analyzing Digital Discourse: New

Insights and Future Directions 3
Pilar Garcés-Conejos Blitvich and Patricia Bou-Franch
Part II Past, Present and Future 23
2 The Coevolution of Computer-Mediated Communication

and Computer-Mediated Discourse Analysis 25
Susan C. Herring
Part III Multimodality 69
3 “Of course I’m married!” Communicative Strategies and

Transcription-Related Issues in Video-Mediated
Interactions 71
Maria Grazia Sindoni
v
vi Contents
4 Multimodality in Memes: A Cyberpragmatic Approach 105

Francisco Yus
5 Digital and Written Quotations in a News Text:

The Hybrid Genre of Political Opinion Review 133
Marjut Johansson
6 Emoticons in Relational Writing Practices on WhatsApp:

Some Reflections on Gender 163
Carmen Pérez-Sabater
Part IV Face and Identity 191
7 From “My Manly Husband…” to “… Sitting Down

to Take a Pee”: The Construction and Deconstruction
of Gender in Amazon Reviews 193
Camilla Vásquez and Addie Sayers China
8 Linguistic Expert Creation in Online Health Practices 219

Marie-Thérèse Rudolf von Rohr, Franziska Thurnherr, and
Miriam A. Locher
9 How Social Media Shape Identities and Discourses

in Professional Digital Settings: Self-Communication
or Self-Branding? 251
Sandra Petroni
10 Losing Face on Facebook: Linguistic Strategies to Repair

Face in a Spanish Common Interest Group 283
Carmen Maíz-Arévalo
Contents vii
Part V Language and Media Ideologies 311
11 Sexting and Hegemonic Masculinity: Interrogating Male

Sexual Agency, Empowerment and Dominant Gendered
Norms 313
Antonio García-Gómez
12 Twitter, Politeness, Self-Presentation 341

Maria Sifianou and Spiridoula Bella
13 Pedagogy, Audience, and Attitudes: Influencing University

Students’ Metalinguistic Awareness About Texting
Practices 367
Rebecca Roeder, Elizabeth Miller, and Pilar Garcés-Conejos
Blitvich
Index391
Notes on Contributors
Spiridoula Bella is Professor of Linguistics at the National and Kapodistrian

University of Athens. Her research interests include pragmatics, linguistic
(im)politeness, second language acquisition and second language teaching. Her
research output in these areas has appeared in international journals (Journal of
Pragmatics, Journal of Politeness Research, Multilingua, Pragmatics) as well as a
number of collected volumes both in English and in Greek. She is the author of
three books on second language acquisition and teaching and one book on prag-
matics and language teaching published in Greek.
Patricia Bou-Franch is Professor of English and Director of the Institute of
Applied Modern Languages (IULMA) at the Universitat de València. She is
interested in social media discourse, gender, identity and im/politeness, and has
published in international journals like Journal of Pragmatics, Journal of
Computer-Mediated Communication, Intercultural Pragmatics, Journal of Politeness
Research, Journal of Language and Politics, Gender and Language, Pragmatics and
Society and Journal of Language Aggression and Conflict. She has also published
edited volumes and special issues on social media and gender and language.
Pilar Garcés-Conejos Blitvich is Professor of Linguistics in the Department of
English at the University of North Carolina at Charlotte (USA). She is inter-
ested in im/politeness models, genre theory, identity construction, and tradi-
tional and social media on which she has published and lectured extensively. She
sits on the editorial board of various journals and is co-editor in chief of the
Journal of Language of Aggression and Conflict (John Benjamins). https://clas-
pages.uncc.edu/pilar-garces-conejos-blitvich/.
ix
x Notes on Contributors
Antonio García-Gómez is Associate Professor of Linguistics at the University

of Alcalá (Spain), where he teaches discourse analysis. He holds a PhD in
Linguistics from Universidad Complutense de Madrid. His expertise lies in dis-
course analysis and discursive psychology. A main strand of his research has
focused on gender, identity and language use in new media. He has published
numerous articles and authored two books. His latest publications are “Teen
Girls and Sexual Agency: Exploring the Intrapersonal and Intergroup Dimensions
of Sexting” (Media, Culture and Society, 2017) and “Dragons’ Den: Enacting
Persuasion in Reality Television” (Discourse, Context and Media, forthcoming).
Susan C. Herring is Professor of Information Science and Linguistics at Indiana
University Bloomington, where she directs the Center for Computer-Mediated
Communication. Trained in linguistics, she was one of the first scholars to apply
discourse analysis methods to computer-mediated communication (CMC), ini-
tially with a focus on gender issues. She has published numerous works on CMC,
is editor of the online journal Language@Internet, and is a past editor of the
Journal of Computer-Mediated Communication. Her current research interests
include multilingual CMC, multimodal CMC, and telepresence robotics.
Marjut Johansson is Professor at the Department of French Studies, the School
of Languages and Translation Studies, University of Turku. Her recent research
interest covers a variety of topics on digital interaction and discourse, and it
includes work that examines online news discourse, social media, digital discus-
sions and videos. At the moment, she is interested in multimodal, digital
humanities and mixed method approaches. She has co-edited several volumes
pertaining to these topics: Special issue on Digital Agora (Johansson, Kleinke
and Lehti 2017) and Language in the Net—Perspectives to Digital Interaction
(Helasvuo, Tanskanen and Johansson 2014, in Finnish).
Miriam A. Locher works as Professor of the Linguistics of English at the
University of Basel, Switzerland. She works on interpersonal pragmatics, linguis-
tic politeness, relational work, the exercise of power, disagreements, advice-
giving and computer-mediated communication. Her publications comprise
monographs, edited collections, and special issues, as well as numerous articles
in journals and collections.
Carmen Maíz-Arévalo is Associate Professor of Pragmatics and English at the
Complutense University of Madrid, having obtained her Ph.D. in English
Linguistics in 2001. Her fields of interest are mainly pragmatics and intercul-
tural pragmatics. Her most recent publications include articles in Computer
Notes on Contributors xi
Assisted Language Learning (2017), the Journal of Politeness Research (2015), the
Journal of Pragmatics (2012 and 2013) or Discourse Studies (2013). She also acts
as a reviewer for different journals such as the Journal of Pragmatics, GIST, Sage
Open or Revista Iberoamericana de Tecnologías del Aprendizaje.
Elizabeth Miller is Associate Professor in the Department of English at the
University of North Carolina at Charlotte (USA). Her research has explored the
role of identity, ideology, agency and power relations in the learning of English
among adult immigrants to the US. Her most recent work examines the inter-
relationship of agency and emotion in language teacher practice. Along with
numerous journal articles and chapters in edited volumes, she has published The
Language of Adult Immigrants: Agency in the Making (2014) and the co-edited
volume Theorizing and Analyzing Agency in Second Language Learning:
Interdisciplinary Approaches (2015), both by Multilingual Matters.
Carmen Pérez-Sabater, Ph.D., is Associate Professor at the Universitat
Politècnica de València (Spain), Department of Applied Linguistics. She is inter-
ested in the analysis of computer-mediated communication through a sociolin-
guistic perspective as well as in the study of language learning through technology
in English for Specific Purposes. Her research has been published in prestigious
journals such as Ibérica, Journal of Pragmatics, Linguistik Online, RESLA, Revista
Signos, and Written Communication. She has also been actively involved in dif-
ferent government-funded research projects. From 2014 to 2017, she led the
UPV team of the European project CoMoViWo (Communication in Mobile
and Virtual Work). https://scholar.google.es/citations?user=M2Smsq8AAAAJ&
hl=es; https://upv-es.academia.edu/CarmenPerezSabater.
Sandra Petroni is Associate Professor at the Department of Humanities of the
University of Rome “Tor Vergata” (Italy), where she teaches English Language
and Linguistics. Her research fields are multimodality, critical discourse analysis,
and specialized discourse—in particular information and communication tech-
nology (ICT) discourse. She is the author of several research articles, chapters
and two books, Self-Study. La multimedialità e l’apprendimento della lingua ing-
lese nel nuovo sistema universitario italiano (2004) and Language in the Multimodal
Web Domain (2011).
Rebecca Roeder is Associate Professor in the English Department at the
University of North Carolina at Charlotte (USA), where she teaches applied
linguistics. She holds a Ph.D. in Linguistics from Michigan State University.
Her area of research interest is sociophonetics, with a specific focus on systemic
vowel change over apparent time in dialects of North American English.
xii Notes on Contributors
Marie-Thérèse Rudolf von Rohr completed her Ph.D. in English linguistics at

the University of Basel, Switzerland, investigating persuasive mechanisms from
an interpersonal pragmatic perspective in public health discourse online in
2017. She was a research member of the SNF-project Language and Health
Online (143286).
Addie Sayers China is a Ph.D. Candidate in Applied Linguistics at the
University of South Florida (USA). Her research interests include discursive
constructions of gender and race in digital space and multimodal digital dis-
course analysis. In her dissertation, Sayers examines the intersection of gender
and race in social media, exploring pop culture icon Beyoncé as a visual and
linguistic semiotic resource in Twitter, Tumblr, and Pinterest. Her research about
Beyoncé as a counterhegemonic resource will be published in A Tumblr Book,
the first book devoted to the social networking platform.
Maria Sifianou is Professor Emerita, Department of English Language and
Literature, National and Kapodistrian University of Athens, Greece. Her publi-
cations include Politeness Phenomena in England and Greece, Discourse Analysis
(Hillside Press, 2006) and a number of articles in edited books and international
journals. She has co-edited Themes in Greek Linguistics (Benjamins, 1994) and
Linguistic Politeness Across Boundaries (Benjamins, 2001) among others. She is
on the editorial board of various journals and co-editor-in-chief of the Journal of
Language Aggression and Conflict. Her main research interests include politeness
phenomena and discourse analysis in an intercultural perspective.
Maria Grazia Sindoni is Associate Professor in English Linguistics and
Translation at the University of Messina (Italy). Her research interests include
webcam-mediated spontaneous interaction, multimodal studies, digital liter-
acy, systemic-functional grammar, and multimodal critical discourse analysis.
She was awarded the AIA Book Prize in 2016 for the best research monograph
in linguistics (2013–2016) for Spoken and Written Discourse in Online
Interactions. A Multimodal Approach (Routledge, 2013). She is PI for a
European-funded project on multimodality and digital literacy involving seven
European partners.
Franziska Thurnherr is a Ph.D. Candidate in English linguistics at the
University of Basel, Switzerland. Her research focuses on interpersonal pragmat-
ics in online (mental) health discourse. She was a research member of the SNF-
project Language and Health Online (143286).
Notes on Contributors xiii
Camilla Vásquez is Professor of Applied Linguistics in the Department of

World Languages at the University of South Florida (USA), where she teaches
courses such as Language of the Internet, Discourse Analysis, and Sociolinguistics.
Vásquez is the author of The Discourse of Online Consumer Reviews (Bloomsbury,
2014), and her research about online review language has been published in
journals such as Current Issues in Tourism, Discourse Context & Media, Food &
Foodways, Journal of Pragmatics, Language@Internet, Narrative Inquiry and
Visual Communication. Her current research explores linguistic creativity and
humor across various social media platforms.
Francisco Yus is Professor at the Department of English Studies of the
University of Alicante (Spain). He teaches pragmatics and has specialised in the
application of pragmatics (especially relevance theory) to media discourses (e.g.
Ciberpragmática 2.0, Ariel, 2010; Cyberpragmatics, John Benjamins, 2011)
and conversational issues. His latest research has to do with the analysis of irony
in conversation, as well as the production and interpretation of humorous dis-
courses (e.g. Humour and Relevance, John Benjamins, 2016). Yus is Head of the
Institute of Applied Modern Languages (IULMA) at the University of Alicante.
List of Figures
Fig. 2.1 CMC reconceptualized as inherently multimodal 41

Fig. 3.1 Nobita is waving his hands (intentional movement). Nobita is
the shown, and John is the showing90
Fig. 3.2 Shizuka’s personal space 92
Fig. 3.3 Nobita’s interrogative look 94
Fig. 4.1 Adjustment 118
Fig. 4.2 Reversed interpretation 123
Fig. 4.3 Adjusted referents 127
Fig. 9.1 The top-ten nouns 266
Fig. 9.2 The top-ten verbs 267
Fig. 9.3 The top-ten adjectives 267
Fig.10.1 Face-repairing strategies 293
Fig.10.2 Strategies to express support (Maíz-Arévalo & Sánchez-Moya,
2017, p. 215) 294
xv
List of Tables
Table 2.1 The computer-mediated discourse analysis (CMDA) ‘toolkit’ 27

Table 2.2 Three phases in the coevolution of CMC and CMDA 39
Table 5.1 Quotation as a dialogic object (modified from Johansson,
2002, p. 256) 140
Table 5.2 Basic information concerning the quotations in the
Huffington Post article 144
Table 5.3 Organization of news text 148
Table 5.4 Object of stances by the quoted persons 150
Table 6.1 Mean table with the results of the questionnaire 174
Table 6.2 Example 1. A group of close male friends organizing a meal
out, with English translation 175
Table 6.3 Example 2. Part one of an exchange among women
organizing a meal out with the English translation
on the right 176
Table 6.4 Example 3. Part two of an exchange among women
organizing a meal out with the English translation
on the right 176
Table 6.5 Example 4. Male coordination chat to watch a film 178
Table 7.1 In-text references to bona fide review authors’ gender and/
or relational identities 201
Table 7.2 In-text references to parody review authors’ gender and/or
relational identities 206
xvii
xviii List of Tables
Table 8.1 Overview of our four data sets, ordered according to

decreasing presence of health professionals and increasing
level of interactivity 226
Table 8.2 Three important medium and situational factors of the data
sample229
Table 8.3 List of recurring expert identity strategies 233
Table 10.1 Situations of face-loss in the corpus 292
Table 10.2 Frequency of face-repairing strategies in the corpus 293
Table 11.1 The communal layer of the self: enacting hegemonic
masculinity321
Table 11.2 The personal layer of self: attempts to maintain a masculine
identity329
Table 13.1 Distribution of study participants by class and year (N = 108)372
Table 13.2 Students’ attitudes toward appropriateness of texting features,
Spring 2015 and Spring 2016 combined results (N = 108)375
Table 13.3 Change in students’ attitudes about appropriateness of
texting features, from Spring 2015 to Spring 2016 376
Table 13.4 Change in metapragmatic awareness in relation to audience 377
Table 13.5 Self-reporting of “I would make a point of using proper
grammar with everyone” 379
Part I
Introduction
1
Introduction to Analyzing Digital
Discourse: New Insights and Future
Directions
Pilar Garcés-Conejos Blitvich and Patricia Bou-Franch
The aim of this book is to offer new insights and set future directions for
the analysis of digital discourse. The analysis of digital discourse lies at the
intersection of (non)language resources, society, and technology.
Therefore, digital researchers can draw on a range of diverse socially ori-
ented language disciplines, whose methods and research tools may be of
use in carrying out empirical research. However, some of these methods
and tools may need to be critically assessed and reflectively adapted, and
perhaps also expanded and even combined with others to suitably account
for the communicative practices that occur in the digital world and their
embeddedness within the social world at large. Discourse, in our view, is
P. Garcés-Conejos Blitvich (*)
Department of English, University of North Carolina at Charlotte,
Charlotte, NC, USA
e-mail: pgblitvi@uncc.edu
P. Bou-Franch
Department of English and German Philology, Universitat de València,
Valencia, Spain
e-mail: patricia.bou@uv.es
© The Author(s) 2019 3

P. Bou-Franch, P. Garcés-Conejos Blitvich (eds.), Analyzing Digital Discourse,
https://doi.org/10.1007/978-3-319-92663-6_1
4 P. Garcés-Conejos Blitvich and P. Bou-Franch
concerned with “social practice” (Fairclough, 1992, p. 28) rather than
language in use, as it was originally—and more narrowly—conceived in
1980s–1990s. Therefore, we view discourse analysis as the study of “the
ways people build and manage their social world using various semiotic
systems” (Jones, Chik, & Hafner, 2015, p. 3). Put differently, in our view,
digital discourse analysis is concerned with how multimodal, multisemi-
otic resources are employed to enact identities, activities, and ideologies
in the digital world, as part of a larger social world (Gee, 2005).
The field of digital discourse analysis, variously called computer-mediated
discourse, new media sociolinguistics, or language and digital communica-
tion, has been discussed in terms of three waves, since Androutsopoulos
(2006), inspired by Herring’s (1996) foundational work, called for “a shift
of focus from medium-related to user-related patterns of language use”
(p. 421). While studies within the first wave contained mainly descriptive
linguistic approaches and were carried out in the 1990s, the 2000s saw the
consolidation of a second wave of computer-mediated discourse studies
which brought into the picture socially oriented language researchers con-
cerned with linguistic variability, social diversity, issues of identity, and
community formation and maintenance: in sum, a collection of studies
more specifically concerned with the study of digital social practices
(Georgakopoulou, 2006; Herring & Androutsopoulos, 2015). Recent
research claims that a third wave should further take into consideration
issues of “translocality”, the complex ways in which diverse local practices
come together in global spaces (Tagg & Seargeant, 2014), “transmediality”,
or how users transcend different media and should move toward incorpo-
rating multimodal analyses of the sociocultural practices of computer-
mediated communication (CMC) (Androutsopoulos, 2015; Herring, this
volume). Further, Georgakopoulou and Spiliotti (2016) recently called for
research to develop critical and ethical agendas, thus placing the focus on
ideologies about the media and as enacted, challenged, and negotiated in
the digital world (Thurlow, 2017a; Thurlow & Mroczek, 2011).
Thus, the present volume is concerned with current debates on digital
practices. More specifically, these include adapting current paradigms in
view of past, present, and future research (Part II), looking at how users
employ the wealth of multimodal resources provided by digital technolo-
gies (Part III) to get things done and be certain kinds of people (Part IV)
Introduction to Analyzing Digital Discourse: New Insights… 5
and identifying the ideologies that underpin the construction of digital

texts in the social world (Part V).
When it comes to computer-mediated discourse analysis, before we
can offer new insights and future directions to move the field forward, it
is necessary to first look back and take stock of the work done in the dif-
ferent areas of interest that have emerged over the past 30+ years. This
should help us to consider where we stand and, from here, to begin to
identify trends and areas that deserve further attention. Tracing the devel-
opment of the field and indicating future venues for research is precisely
one of the main goals of Susan Herring in her chapter titled The
Coevolution of Computer-Mediated Communication and Computer-
Mediated Discourse Analysis, which constitutes Part II of this book. The
rationale behind her study is that the approach she developed for the
analysis of digital practices known as computer-mediated discourse anal-
ysis (CMDA) was devised for textual interactions. As technologies evolve,
however, research methods and paradigms need to evolve too. Thus, the
author makes the case for a research move truly concerned with multise-
miotic analyses, and she develops and expands the extant CMDA para-
digm to open the way in this direction. In doing so, Herring reviews
technological advances in the field organized around three historical
phases: pre-Web (stand-alone textual clients), Web 1.0, and Web 2.0
technological phases. Each phase is discussed alongside insights gained
from computer-mediated discourse studies that analyzed digital interac-
tions therein. In doing so, she further traces the evolution of the CMDA
paradigm, as it was modified and expanded to account for the increasing
communicative possibilities of each phase.
Along the same lines as Herring, across the board, key scholars in the
field of language and digital communication (Georgakopoulou &
Spiliotti, 2016; Herring & Androutsopoulos, 2015; Jones et al., 2015;
Thurlow, 2017a; Thurlow & Mroczek, 2011) agree that text-based stud-
ies, the traditional focus of analysis, need to move forward by incorporat-
ing other modes of communication. All human communication is
multimodal (Norris, 2004) but digital technologies are almost always,
and are becoming increasingly, vastly multimodal by combining writing,
images, sounds, and other semiotic modes. Ignoring this fact in our anal-
yses of digital genre practices makes for very partial accounts of
c ommunication therein. Thus, we should place “the concepts of multi-

modality and multisemioticity as central to our current research on lan-
guage and digital media” (Georgakopoulou & Spiliotti, 2016, p. 3).
Consequently, a chapter by Jewitt (2016), included in the handbook
these authors coedit, introduces key concepts and tools of multimodal
analysis that can be of use to scholars of digital communication and pro-
vides guidance on how to collect and transcribe data and on how to ana-
lyze single modes and carry out analyses across modes.
Multimodality entered linguistics through the groundbreaking work
of Kress and Van Leeuwen in Reading Images (1996) and Multimodal
Discourse (2001). Although much of the work in multimodal discourse
analysis (for a comprehensive review see O’Halloran, 2013) has been
based on Halliday’s systemic functional theory, combining approaches to
discourse with multimodal frameworks of communication has occupied
center stage in the work of scholars of different persuasions. Among oth-
ers, Lemke (2011, 2012) has been at the fore of these efforts by bringing
together discourse analysis and visual semiotics to provide ways of analyz-
ing the interconnecting meaning makings of discourse and images and
looking into hypertextuality and traversals as digital technology mediated
ways of creating meaning. For their part, Bateman and Wildfeuer (2014)
have put forth a framework to reengage with visual communication arti-
facts, such as visual narratives, in ways compatible with methods devel-
oped for verbal linguistic artifacts. Taking a step further into other
digitally mediated genres, Gee (2014, p. 1) proposed a unified theory of
discourse analysis to study “language, games, science and human action
and interaction in the real world and imaginary worlds”. Gee shows how
conversations, avatars (identities), affordances (the functions those iden-
tities allow/restrict us to carry out), situated meaning, as well as the build-
ing blocks of syntax and semantics can be found in the real world as well
as in games. Since the world and games share all these primordial ele-
ments, they can be analyzed by using one unified theory of meaning
making. The common thread between language and games, Gee argues,
is their multimodality and ultimate functionality.
Since the main goal of this volume is to explore the advances and new
insights in the rapidly changing field of language and digital communi-
cation, studies of multimodality play a central role in it. Four chapters
(by Sindoni, Perez-Sabater, Johansson, and Yus) and a significant part of

Herring’s contextualization chapter focus on multimodal issues.
Along similar lines as Gee, Herring also proposes a theory of multi-
modal CMC that includes graphical phenomena such as emoji, image
memes, and avatars in virtual worlds, as well as certain kinds of robots.
By doing so, Herring’s proposal extends the definition of CMC itself as
fundamentally multimodal. Since text, audio, and video CMC have been
addressed often in the literature, Herring focuses on three newer phe-
nomena: communication on interactive multimodal platforms (IMPs);
graphical communication, including avatar-mediated communication
(AMC); and robot-mediated communication (RMC). Moving forward,
and to be able to give a proper account of multimodality in digitally
meditated communication, Herring argues that CMC needs to become
highly interdisciplinary. Some of the fields that CMC can draw from are
semiotics, ethnography, human-computer interaction, and human-robot
interaction. Comic books can assist in understanding the relationship
between text and image, and scholars may also need to relearn or devise
new transcribing methods.
As Herring argues, and multimodal scholars have long held (see Norris,
2004), transcribing multimodal interaction is often a necessary step prior
to analysis. Therefore, it is necessary to devise new transcribing methods
commensurate with the vast multimodality of digital communication.
Sindoni (this volume) presents a transcription of a conversation on an
IMP, a Skype video chat, carried out by the author herself and one of the
students taking part in a research project. Sindoni’s aim is not to assess
both transcriptions comparatively but to understand how students make
sense of the video data and several semiotic modes such as language,
proxemics, kinesics, and gaze patterns. The process evidences the difficul-
ties involved in multimodal transcription and the widespread logocen-
tricity that makes students normalize, what they perceive as, irregular
linguistic patterns. Sindoni argues convincingly that this type of exercises
“should be incorporated into University curricula as heuristics for the
improvement of students’ critical skills and, more importantly, as open
windows to gain precious access into covert (educational) ideologies that
still privilege the written normativity of language”. This is an interesting
conclusion, as Sindoni crucially sees multimodal transcriptions as a way
for the analyst to relate microinteractions with societal ideologies at the

macro-level. Jewitt (2016) argues that multimodal scholars need to bring
in social theory and historical contexts in order to do so.
Graphical communication, Herring claims, including AMC, has been
around since the 1990s, became popular in the 2000s with Second Life,
and is still a common feature of virtual game worlds such as World of
Warcraft. Included in this category, we find emoticons, emoji, stickers,
GIFs, and text-in-image memes; video clips may also serve similar func-
tions. Herring refers to these as graphicons and argues that there are few
discourse studies that analyze graphicons-in-use. Perez-Sabater (this vol-
ume) does just that, however, by shedding light on the linguistic conven-
tions of use of emoticons in several WhatsApp communities interacting
in Peninsular Spanish, with a special focus on gender differences seen in
adults’ use of emoticons. Another way in which Perez-Sabater’s paper
helps advance the field is by adopting a multimethod approach based on
ethnography and case studies. Her results show that, despite the large
number of graphicons available, men include them infrequently in their
chats while women, in contrast, make abundant use of them. This can be
related, the author argues, to women conforming to standard practices
more often than men, which would entail use of graphicons in this con-
text. Furthermore, there seem to be no difficulties regarding the interpre-
tation of the meaning/functions of graphicons here, which may be due to
the graphicons themselves and/or to the fact that participants form
coherent groups regarding in/formality expectations.
Also, focusing on a different type of graphicon and applying cyber-
pragmatics, a cognitive approach to the analysis of digital phenomena
based on the fundamentals of Relevance Theory, Yus analyzes the “image
macro meme”. The application of cognitive pragmatics theories to the
study of digital, multimodal phenomena represents a clear case of the
cross-fertilization with other fields, which Georgakopoulou and Spiliotti
(2016) see as one of the signs of the language and digital field coming of
age. These memes consist of a line or two of text on top of the meme,
line(s) of text at the bottom and one picture in the middle. Yus (2016)
argues that in the same way as we have explicit interpretations (explica-
tures) and implicit or implicated interpretations (implicatures) of verbal
utterances, visual content also leads to visual explicatures and visual
implicatures. Consequently, the interpreting of a meme entails the pro-

cessing of the text, the processing of the picture, and the identification of
possible connotative meanings for text, picture, and text-picture combi-
nations for which iconic literacy is required. His purpose is to find out
what text/picture, multimodal combination occurs more frequently, why
this is so, and how these combinations relate to predictions of interpre-
tive relevance. To do so, he applies McCloud’s (1994) taxonomy of cate-
gories for comics (a genre very close to memes in its verbal-visual
multimodal quality) to the corpus. Yus’s results revealed that the language
mode acquires prominence in the final interpretation of the meme,
whereas visual information either illustrates, amplifies, elaborates, or aids
in altering what the linguistic information provided.
Important issues regarding the remediation and recontextualization of
genres and their configuring modes of communication are brought up by
Johansson who focuses her chapter on the opinion review genre and the
transformations it undergoes when migrating online. What is of more
significance here is that the genre becomes highly multimodal, as it con-
tains multiple semiotic modes of presentation, hypertextual links, and
algorithm-based technological affordances for user participation. More
importantly, language is not necessarily the primary mode of communi-
cation which may be assumed by other semiotic modes such as videos,
interactive maps, figures, slideshows, and so on. Johansson looks closely
at the nature and function of digital quotations such as tweets and videos
and how they integrate with ordinary quotations and the story line in
different meaning making ways. The multimodal complexity of the genre
under scrutiny requires a multimethod approach that combines digital
discourse analysis with sociopragmatics. The presence of the recontextu-
alized quotations mediated by different semiotic modes creates a poly-
glossic, hybrid genre in which an internal polylogic exchange unfolds
among persons and ideas that may have not interacted before.
All in all, despite its great relevance, Jewitt (2016) concludes that multi-
modality as a field in itself is in its early stages of development in terms of
theory and practices of transcription. Furthermore, we would argue, many
of the multimodal frameworks available were not natively digital and thus
present many problems when digitized in terms of their application to the
analysis of digital multimodal data (Bou-Franch & Garcés-
Conejos
Blitvich, 2018). Therefore, we see a unique opportunity here through

which the systematic application of multimodality concepts and frame-
works may help advance not only the language and digital communication
field, but also multimodality itself to its final goal of accounting for how
“contemporary societies are grappling with the social implications of the
rapid increase in sophistication and range of multimodal practices particu-
larly within interactive digital media” (O’Halloran & Smith, 2011, p. 1).
Looking at the identities we construct digitally is another staple of the
field and the focus of the four chapters included in Part IV, which are
concerned with face and identity. These chapters elaborate on notions
relevant to the sociolinguistic awareness brought about by second-wave
digital discourse studies; however, they advance the field toward its third
wave in that they report on critical and qualitative studies.
Language and social media researchers seem to agree on the centrality
of the social processes of self-presentation and relational management
when communicating in online environments. Thus, a significant num-
ber of studies focuses on media practices that aim to construct who we
are and how we relate to others (e.g. Bolander & Locher, 2015; Bou-
Franch & Garcés-Conejos Blitvich, 2014a, 2014b; Garcés-Conejos
Blitvich, 2010; Garcés-Conejos Blitvich, Bou-Franch, & Lorenzo-Dus,
2013; Georgakopoulou, 2013; Georgalou, 2016; Lorenzo-Dus, Garcés-
Conejos Blitvich, & Bou-Franch, 2011; Page, 2012; Papacharissi, 2011;
Tagg & Seargeant, 2014, 2016). Previous research in discourse analysis
sees identity practices as essentially discursive and relational, that is, as
socially constructed. Identities emerge in interaction with others, and are
constantly changing, as different aspects become salient in interaction
and individuals engage in processes of negotiation, identification, and
disidentification with others (Bucholtz & Hall, 2005; De Fina, Schiffrin,
& Bamberg, 2006). Social media create new spaces for online identity
performances and negotiations, and the study of the processes behind the
“formation of new forms of social organization and social interaction”
(Castells, 2000, p. 693) needs to pay special attention to the role played
by the social and technological affordances (Herring, 2007). The avail-
ability of multiple semiotic modes for identity construction, alongside
users’ agentive choices to employ certain resources, is an issue that will
affect identity construction and negotiation.
In their analysis of Facebook, for instance, Tagg and Seargeant (2014,

2016) focus on issues of identity and community as the two crucial
dynamics for the discursive construction of social networking sites. Self-
presentation processes and relational practices are interconnected, they
argue, through the audience design strategies that online users employ to
communicate through a collapsed audience. The influential notion of
context collapse—or the bringing together of different social groups into
the same digital space for interaction—in relation to social networking
sites (Marwick & boyd, 2011) is of importance in this respect, as users
face the need to change the way they act and interact socially in addressing
their imagined networked audiences (Page, 2012). In particular, resources
such as choice of language or topic, or strategic uses of ambiguity and
vagueness, among others, have been discussed as means of orienting to
imagined audiences, and further selecting or blocking specific (groups of )
addressees in multilingual, translocal communities where contexts col-
lapse (e.g. Androutsopoulos, 2014; Tagg & Seargeant, 2014, 2016). These
constitute important ways in which identity and relational practices are
performed and negotiated. Nevertheless, to move the field forward, fur-
ther studies of ways of doing sociability, of entextualizing identity and
relational practices in social media, are still needed. Particularly, explora-
tions of identity adopting critical perspectives are in short supply.
Incidentally, this is one of the directions toward a third wave of digital
discourse studies (Georgakopoulou & Spiliotti, 2016). For Thurlow
(2017a), a critical perspective on digital practices within the field of criti-
cal discourse analysis should examine the ways in which microlevel prac-
tices construct social worlds and how macro-level structures and ideologies
shape our communicative practices, that is, how texts and the worlds are
mutually shaping/shaped (by) each other. Indeed, we agree for the need to
fruitful interconnections between digital discourse and critical discourse
analysis. It must be noted, however, that the focus of much critical dis-
course analysis is on what is known as “elite discourses” (van Dijk, 1989)
and, therefore, on the top-down processes of ideological hegemony to the
neglect of the “‘bottom-up’ strategies of those who may contest or subvert
these ideologies” (Bucholtz, 2003, p. 58). Recent critical studies of (gen-
der) ideologies and social identity have called researchers to focus on pop-
ular digital practices, those of ordinary (as opposed to elite) individuals, as
technologies have brought about a “demotic turn” (Turner, 2010) through

which citizens have “gained access to a public sphere in which to either
contest or reinforce dominant ideologies” (Bou-Franch, 2013; Bou-
Franch & Garcés-Conejos Blitvich, 2014a, p. 229). Herring, too, argues
that one of the challenges facing digital discourse analysis is to expand
paradigms by adopting “broader qualitative and critical perspectives” (this
volume).
This challenge is taken up in the four chapters that address issues of
identity and face in this volume (by Vasquez & Sayers; Rudolf von Rohr,
Thurnherr, & Locher; Petroni; Maíz-Arévalo). Vasquez and Sayers’ exam-
ination of the intertextual and heteroglossic (Androutsopoulos, 2011)
constructions of gender identities and ideologies in a corpus of Amazon
reviews constitutes an innovative, critical perspective on identity con-
struction. While Amazon constitutes a digital space for the selling of a
variety of products, it has appropriated one of the central features of
social networking sites, that is, supporting sociability, as the corporate
site offers consumers the possibility of posting comments about the prod-
ucts on sale therein (Seargeant & Tagg, 2014). Research has demon-
strated the strength of this type of communication between consumer/
reviewers and prospective consumers/readers, as the latter are affected by
such reviews and by the extent of their social communion with the
explicit/implicit information they can gather about the social identities of
online reviewers. In their analysis of “bona fide” and parody Amazon
reviews, the authors identify the construction and circulation of dis-
courses and counter-discourses of normative gender ideologies, the
engendering of certain products, and discussions of gender politics. Thus,
the corporate site is shown to accommodate multiple voices and diverse
worldviews positioned alongside modernist/postmodernist (de)construc-
tions of gender while consumers shop, consume, and discuss products.
Rudolf von Rohr, Thurnherr, and Locher, too, adopt a qualitative
methodology in their analysis of identity alongside expertise in online
health settings, thus moving the analysis of digital discourse forward in
relation to Herring’s claim mentioned above. The construction of expert
identities is investigated in data sets compiled from four digital spaces of
interaction which include (1) static, noninteractive websites where pro-
fessionals offer their recommendations, (2) online advice columns con-
taining exchanges of letters between a professional persona and lay

participants, (3) dyadic (professional/lay) email exchanges, and (4) a final
set of polylogic, lay interactions from a health-minded forum. By focus-
ing on how professional and lay users discursively enact expert identities
Rudolf von Rohr et al. elaborate on the claim that traditional dichoto-
mous identities have become porous (Tagg & Seargeant, 2014). The
authors’ objective goes beyond the identification of expert-building prac-
tices to include a study of how each online context shapes, empowers, or
facilitates such practices. Importantly, notions of sociotechnological con-
straints and affordances (Gibson, 1986) emerge in this study. As Jones
et al. (2015) argue, the technological affordances of the media should not
be seen in any simplistic, deterministic sense and should instead be taken
as possibilities and resources made available by technologies which users
may rely on and resort to according to their communicative purposes on
specific occasions. The close qualitative analysis carried out by Rudolf
von Rohr, Thurnherr, and Locher, relying on positioning theory, crucially
revealed the complex ways in which lay and professional users employed
a range of interconnected positioning strategies for the creation of expert
identities and the ways in which said strategies were embedded in the
four digital media under scrutiny. Interestingly, the authors were able to
show how a number of visual aspects of the layout of the websites ana-
lyzed contributed to enhance the construction of expertise. In addressing
visual features, they overcome a previous shortcoming of scholarly efforts
that attempted to explain the interaction between technology and dis-
course using frameworks specifically designed for textual CMC, as we
mentioned above.
The exploration of issues of identity vis-à-vis the confluence of users’
agency and technology, which is at the heart of the debate on the socio-
technological affordances of digital communication, constitutes the pri-
mary object of inquiry of the chapter by Petroni. Her study focuses on
the construction of professional identities on LinkedIn; more specifically,
Petroni’s chapter investigates the interconnections of self-presentation
and self-branding in cases where identities are constructed by resorting to
promotional discourse strategies. A second objective of research lies in
the extent to which said marketized identities are produced through ver-
bal resources or are generated by the functionalities associated with the
architecture of the technology itself. In this way, the author addresses

questions posed by Herring and Androutsopoulos (2015, p. 143) regard-
ing what aspects of discourse a given technology shapes, “how strongly, in
what ways, and under what circumstances”. Another way in which this
chapter moves the field of digital discourse analysis forward lies in the use
of mixed methods to carry out the research. A corpus-assisted study is
designed to classify identity building strategies as leaning toward self-
branding and promotion. The second, combined method employs a criti-
cal perspective in assessing the social meanings of the technology itself,
and thus responds to the call to adopt “an altogether more critical, care-
fully theorized take on technology” (Thurlow & Mroczek, 2011, p. xxiv),
which takes into account how our digital practices are influenced and
even controlled by the social media.
The last chapter in Part IV, devoted to face and identity, is related to
group-building and community practices on Facebook. Issues of com-
munity constitute one of the main dynamics for the social construction
of social networking sites, and one of the core concepts that needs to be
reconceptualized in current language and social media debates
(Georgakopoulou & Spiliotti, 2016; Tagg & Seargeant, 2016). Maíz-
Arévalo examines ways in which users attend to each other in their media
practices by engaging in negotiations of face. Her study draws on notions
of facework (Goffman, 1955; Penman, 1990) to specifically investigate
how participants in the Facebook genre of public common interest groups
employ strategies for self and other face restoration or repair. The group
under scrutiny conforms an online community with no offline relation-
ship and can, therefore, be associated with users that come together due
to a shared interest but have very loose ties. This Facebook group, there-
fore, shares ambient affiliation in the sense of Zappavigna (2011), which
is different from the type of “node-oriented” community found in other
Facebook interactions (Tagg & Seargeant, 2014). In this context, and
due to the loose ties among members, the author argues, community
members are not expected to increase their interactional involvement by
resorting to other-repair practices. However, her analysis shows that par-
ticipants engage in different forms of face restoration which are crucially
seen as stemming from a concern to maintain intra-group cohesion. This
chapter, therefore, contributes to new insights into digital discourse by
identifying ways in which social processes underlying the formation and

maintenance of social identity and community are performed through
face-repairing strategies.
As has been mentioned above, advancing the field of language and
digital communication arguably involves for it to become more self-
reflective and to develop a critical agenda regarding discourses and ide-
ologies of digital communication (Georgakopoulou & Spiliotti, 2016).
As Thurlow and Mroczek (2011, p. xxvi) discuss, digital technologies are
inherently ideological regarding both “their political economies of access
and control and … their potential as mechanisms or resources for both
normative and resistive representations”. Jones et al. (2015) suggest a
number of areas in which the interconnections between power and ideol-
ogy can be visible, such as in the ideological agendas expressed on social
media, as well as how the interface between software and web interfaces
limit users’ options or forces them to agree to certain terms and condi-
tions. The ethics involved in researching digital data and coming to grips
with the borders between the public and the private also emerge as a
major concern (Spilioti & Tagg, 2017) as do concerns about surveillance
(Jones, 2016).
Part V of this collection is devoted to language and media ideologies
and reflects the current critical involvement of the field of digital dis-
course analysis. It does so in two different ways: by looking at discourses
about digital media (see among others Spilioti, 2016; Thurlow, 2006,
2017a) and also by analyzing the microlevel where language ideologies
are constantly constructed and reconstructed (Blackledge, 2002).
Sexting as it relates to semiotic ideologies has been the focus of recent
research (Thurlow, 2017b). Here, by analyzing young men’s understand-
ing of sexting, Garcia-Gomez looks into what the self-representations of
heterosexual young men can tell us about gendered discourses of youth
sexualities, so as to gain insights into the ideologies of sexualized youth
cyberculture. His data are made up of guided discussions and personal
interviews with 27 British young men (aged 18–21 years). The results of
Garcia-Gomez’s analysis show how participants construct their male
identities by abiding by norms of traditional hegemonic masculinity.
Furthermore, these online sexual practices seem to put additional
demands on participants to perform as strong, virile, and sexually active
men, revealing men’s confusion about sexual agency and choice as a result
of young women’s sexually active agency and disinhibition. Sexting
emerges not only as a popular digital practice, but also as an influential
mechanism for claiming and gaining social recognition and value that
allows young people to inhabit a “legitimate” subject position. These
views as especially interesting as they stand in sharp contrast with adult
ideologies about the digital practices of the youth, which have been dis-
cussed in terms of moral panics caused by the perception that such prac-
tices may impact and alter the existing social order and which are viewed
as ways of disciplining the youth and sexuality (Thurlow, 2006, 2014,
2017b).
Politeness is an essential component of social meaning and has been
widely researched within pragmatics, sociolinguistics, anthropological
linguistics, and other related fields. How to define politeness is still sub-
ject to debate among politeness scholars. There is agreement, however,
that politeness can be defined from a second-order, analyst-based, per-
spective or from a first-order, lay participant’s, perspective (Watts, 2003).
When taking the latter, politeness emerges as discursively constructed,
subject to discursive struggle, and profoundly ideological (Eelen, 2001;
Mills, 2003). In their chapter, Sifianou and Bella resort to twitter to look
into “common sense” ideologies of Greek politeness (Eelen, 1999).
Searching for instances of the keyword ευγένεια (“politeness” in Greek)
within text messages or twitter tags (hashtags #), they compiled the
Twitter Corpus of Greek Politeness (TC-GP) consisting of 345,000
words and 19,550 tweets released by Greek tweeters from February 2009
to February 2015. The results of their analysis show politeness being con-
ceptualized as both verbal and nonverbal and in broadly different terms,
which is proof of the discursive struggle to which such notions are sub-
jected. The authors related their findings to orientation to networked
audiences and the necessary brevity of the messages. Furthermore, they
saw the quoting of sources as an attempt to construct a knowledgeable
identity by imparting a sophisticated, witty view possibly aiming at a
positive self-presentation. Sifianou and Bella argue that this representa-
tion is tied to powerful ideologies associated with the positive view of
cultured individuals in the Greek culture. Twitter, in the authors’ view,
emerges as a new source of naturally occurring data which can provide
insights into the perceptions of various groups of people who may not be
accessed in other ways and into how broad social discourses are con-
structed at the microlevel.
The last chapter in the book tackles young people’s language ideologies
regarding standardization and texting (see also Thurlow, 2006, 2014;
Thurlow & Bell, 2009). In it, Roeder, Miller, and Garcés-Conejos Blitvich
report on a study on undergraduate students’ audience awareness and
attitudes about appropriate use of language and particular language forms
in text messages. This work was patterned after survey studies that have
found that students’ attitudes toward nonstandard language varieties,
such as African-American Vernacular English, can change after taking a
single linguistics class. The data for the study were collected by means of
a survey that was administered to students in three undergraduate lin-
guistics classes at a large urban research university in the South East of the
USA and were subjected to both quantitative and qualitative analyses.
Results indicated that students have significant pragmatic awareness
regarding recipiency coming into the classes and that explicit instruction
can lead to increased awareness of pragmatic norms and positively affect
language ideologies, albeit in an abstract way. While students’ voluntary
comments indicated a persistent perception that prescriptive norms of
language should generally be adhered to, students in the target class dem-
onstrated that these powerful norms can be rethought and challenged
when exposed to studies and class discussions that treat these practices as
appropriate depending on one’s communicative purpose and audience.
Moving forward, language and digital communication scholars may
want to look at what has been called “multimodal critical discourse stud-
ies” (Machin, 2013) or “critical multimodal analysis of digital discourse”
(Moschini, 2014). This line of enquiry is especially interesting as it brings
critical studies and multimodality together and responds to van Leeuwen’s
(2013) claim to the effect that, with few exceptions, there has been little
critical work done on the way that discourses are “communicated, natu-
ralized, and legitimized beyond the linguistic level” (Machin, 2013,
p. 347) and aims to investigate the ways critical discourse studies can help
in understanding meaning making in multimodal communication. The
key question is how scholars should approach the way that discourse
and ideologies are disseminated concurrently across different kinds of
semiotic modes and genres. Another interesting contribution to research

on internet and digital phenomena is Critical Technocultural Discourse
Analysis. It is framed by cultural theory—critical race, feminism, queer
theory, and so on—to avoid “deficit-based models of underrepresented
populations’ technology use” (Brock, 2016, p. 1). These new approaches,
developed within media studies, offer interesting cross-fertilization
possibilities.
In this introduction, we have embedded our overview of the contribu-
tions to this volume within a narrative that reviews past and extant
research on language and digital communication. We have taken special
care to highlight the ways in which each chapter advances the field. In
order to do so, we have carefully identified new methodological and
empirical insights put forth by the different authors. Specially, we have
highlighted the steps contributors to this volume have taken to help
establish the so-called third wave of research and how these steps point to
future directions in which to expand the field of language and digital
communication.
References
Androutsopoulos, A. (2014). Languaging when contexts collapse: Audience
design in social networking. Discourse, Context and Media, 4–5, 62–73.
Androutsopoulos, J. (2006). Introduction: Sociolinguistics and computer-
mediated communication. Journal of Sociolinguistics, 10(4), 419–438.
Androutsopoulos, J. (2011). From variation to heteroglossia in the study of
computer-mediated discourse. In C. Thurlow & K. Mroczek (Eds.), Digital
discourse: Language in the new media (pp. 277–298). Oxford: Oxford
University Press.
Androutsopoulos, J. (2015). Towards a ‘third wave’ of digital discourse studies:
Audience practices on Twitter. Unpublished Plenary talk delivered at the 1st
International Conference Approaches to Digital Discourse Analysis – ADDA,
Valencia, 18–10 November, 2015.
Bateman, J. A., & Wildfeuer, J. (2014). A multimodal discourse theory of visual
narrative. Journal of Pragmatics, 74, 180–208.
Blackledge, A. (2002). The discursive construction of national identity in mul-
tilingual Britain. Journal of Language, Identity, and Education, 1(1), 67–87.
Bolander, B., & Locher, M. (2015). ‘Peter is a dumb nut’: Status updates and
reactions to them as ‘acts of positioning’ in Facebook. Pragmatics, 25(1),
99–122.
Bou-Franch, P. (2013). Domestic violence and public participation in the media:
The case of citizen journalism. Gender and Language, 3(3), 275–302.
Bou-Franch, P., & Garcés-Conejos Blitvich, P. (2014a). Gender ideology and
social identity processes in online language aggression against women. Journal
of Language Aggression and Conflict, 2(2), 226–248.
Bou-Franch, P., & Garcés-Conejos Blitvich, P. (2014b). Conflict management
in massive polylogues: A case study from YouTube. Journal of Pragmatics, 73,
19–36.
Bou-Franch, P., & Garcés-Conejos Blitvich, P. (2018). Relational work in mul-
timodal networked interactions on Facebook. Internet Pragmatics, 1(1),
134–160.
Brock, A. (2016). Critical technocultural discourse analysis. New Media and
Society, 20(3), 1–16.
Bucholtz, M. (2003). Theories of discourse as theories of gender: Discourse
analysis in language and gender studies. In J. Holmes & M. Meyerhoff (Eds.),
The handbook of language and gender (pp. 43–68). Malden, MA: Blackwell.
Bucholtz, M., & Hall, K. (2005). Identity and interaction: A socio-cultural lin-
guistic approach. Discourse Studies., 7(4/5), 585–614.
Castells, M. (2000). Toward a sociology of the network society. Contemporary
Sociology, 29(5), 693–699.
De Fina, A., Schiffrin, D., & Bamberg, M. (Eds.). (2006). Discourse and iden-
tity. Cambridge: Cambridge University Press.
Eelen, G. (1999). Politeness and ideology: A critical review. Pragmatics, 9(1),
163–173.
Eelen, G. (2001). A critique of politeness theories. Manchester: St. Jerome.
Fairclough, N. (1992). Discourse and social change. Cambridge: Polity Press.
Garcés-Conejos Blitvich, P. (2010). The YouTubification of politics, impoliteness
and polarization. In R. Taiwo (Ed.), Handbook of research on discourse behav-
ior and digital communication: Language structures and social interaction
(pp. 540–563). Hershey, PA: IGI Global.
Garcés-Conejos Blitvich, P., Bou-Franch, P., & Lorenzo-Dus, N. (2013).
‘Despierten, Latinos’ (‘wake up, Latinos’): Latino identity, US politics and
YouTube. Journal of Language and Politics, 12(4), 558–582.
Gee, J. P. (2005). An introduction to discourse analysis: Theory and method.
London: Routledge.
Gee, J. P. (2014). Unified discourse analysis: Language, reality, virtual worlds and
video games. London: Routledge.
Georgakopoulou, A. (2006). Postcript: Computer-mediated communication in
sociolinguistics. Journal of Sociolinguistics, 10(4), 548–557.
Georgakopoulou, A. (2013). Small stories and identities analysis as a framework
for the study of im/politeness in-interaction. Journal of Politeness Research,
9(11), 55–74.
Georgakopoulou, A., & Spiliotti, T. (2016). Introduction. In A. Georgakopoulou
& T. Spilioti (Eds.), The Routledge handbook of language and digital communi-
cation (pp. 1–16). London: Routledge.
Georgalou, M. (2016). ‘I make the rules on my wall’: Privacy and identity man-
agement practices on Facebook. Discourse & Communication, 10(1), 40–64.
Gibson, J. J. (1986). The ecological approach to visual perception. Boston, MA:
Psychology Press.
Goffman, E. (1955). On face-work: An analysis of ritual elements in social
interaction. Psychiatry, 18(3), 213–231.
Herring, S. C. (1996). Computer-mediated communication: Linguistic, social and
cross-cultural perspectives. Amsterdam: John Benjamins.
Herring, S. C. (2007). A faceted classification scheme for computer-mediated
discourse. Language@Internet, 4.
Herring, S. C., & Androutsopoulos, J. (2015). Computer-mediated 2.0. In
D. Tannen, H. E. Hamilton, & D. Schiffrin (Eds.), The handbook of discourse
analysis (pp. 128–151). Chichester: Blackwell.
Jewitt, C. (2016). Multimodal analysis. In A. Georgakopoulou & T. Spilioti
(Eds.), The Routledge handbook of language and digital communication
(pp. 69–84). London: Routledge.
Jones, R. H. (2016). Surveillance. In A. Georgakopoulou & T. Spilioti (Eds.),
The Routledge handbook of language and digital communication (pp. 408–411).
London: Routledge.
Jones, R. H., Chik, A., & Hafner, C. A. (2015). Introduction: Discourse analy-
sis and digital practices. In R. H. Jones, A. Chik, & C. A. Hafner (Eds.),
Discourse analysis and digital practices: Doing discourse analysis in the digital age
Kress, G. R., & Van Leeuwen, T. (1996). Reading images: The grammar of visual
design. Oxon: Routledge.
Kress, G. R., & Van Leeuwen, T. (2001). Multimodal discourse: The modes and
media of contemporary communication. London: Arnold.
Lemke, J. L. (2011). Doing multimedia analysis of visual and verbal data: A guide.
Retrieved from http://static1.1.sqspcdn.com/static/f/694454/12422538/
1306511144627/Doing+Multimedia+Analysis+of+Visual+and+Verbal+D
ata.pdf?token=nzmJVDQ7wvyHDxuEgKljQonELow%3D
Lemke, J. L. (2012). Multimedia and discourse analysis. In J. P. Gee &
M. Handford (Eds.), The Routledge handbook of discourse analysis (pp. 79–89).
London: Routledge.
Lorenzo-Dus, N., Garcés-Conejos Blitvich, P., & Bou-Franch, P. (2011).
On-line polylogues and impoliteness: The case of postings sent in response to
the Obama Reggaeton YouTube video. Journal of Pragmatics, 43(10),
2578–2593.
Machin, D. (2013). What is multimodal critical discourse studies? Critical
Discourse Studies, 10(4), 347–355.
Marwick, A. E., & boyd, d.m. (2011). I tweet honestly, I tweet passionately:
Twitter users, context collapse, and the imagined audience. New Media &
Society, 13(1), 114–133.
McCloud, S. (1994). Understanding comics: The invisible art. New York: Harper
Collins.
Mills, S. (2003). Gender and politeness. Cambridge: Cambridge University Press.
Moschini, I. (2014). Critical multimodal analysis of digital discourse prelimi-
nary remarks. LEA – Lingue e letterature d’Oriente e d’Occidente, 3, 197–201.
Norris, S. (2004). Analyzing multimodal interaction: A methodological framework.
New York, NY: Routledge.
O’Halloran, K. (2013). Multimodal analysis and digital technology. In
F. Montagna (Ed.), Readings in intersemiosis and multimedia (pp. 35–53).
Israel: IBIS Editions.
O’Halloran, K., & Smith, B. A. (Eds.). (2011). Multimodal studies: Exploring
issues and domain. New York: Routledge.
Page, R. E. (2012). Stories and social media: Identities and interaction. Oxford:
Routledge.
Papacharissi, Z. (Ed.). (2011). A networked self: Identity, community, and culture
on social network sites. New York: Routledge.
Penman, R. (1990). Facework & politeness: Multiple goals in courtroom dis-
course. Journal of Language and Social Psychology, 9(1/2), 15–38.
Seargeant, P., & Tagg, C. (2014). Introduction: The language of social media. In
P. Seargeant & C. Tagg (Eds.), The language of social media: Identity and com-
munity on the internet (pp. 1–20). London: Palgrave Macmillan.
Spilioti, T. (2016). Digital discourses: A critical perspective. In A. Georgakopoulou
& T. Spilioti (Eds.), The Routledge handbook of language and digital communica-
tion (pp. 133–145). London/New York: Routledge.
Spilioti, T., & Tagg, C. (2017). The ethics of online research methods in applied
linguistics: Challenges, opportunities, and directions in ethical decision-
making. Applied Linguistics Review, 8(2/3), 163–168.
Tagg, C., & Seargeant, P. (2014). Audience design and language choice in the
construction and maintenance of translocal communities on social network
sites. In P. Seargeant & C. Tagg (Eds.), The language of social media: Identity
and community on the internet (pp. 161–185). Basingstoke: Palgrave Macmillan.
Tagg, C., & Seargeant, P. (2016). Facebook and the discursive construction of
the social network. In A. Georgakopoulou & T. Spilioti (Eds.), The Routledge
handbook of language and digital communication (pp. 339–353). Oxon:
Routledge.
Thurlow, C. (2006). From statistical panic to moral panic: The metadiscursive
construction and popular exaggeration of new media language in the print
media. Journal of Computer Mediated Communication, 11(3), 667–701.
Thurlow, C. (2014). Disciplining youth: Language ideologies and new tech-
nologies. In A. Jaworski & N. Coupland (Eds.), The discourse reader (3rd ed.,
pp. 481–496). London: Routledge.
Thurlow, C. (2017a). Digital discourse: Locating language in new/social media.
In J. Burgess, T. Poell, & A. Marwick (Eds.), The SAGE handbook of social
media. New York: Sage.
Thurlow, C. (2017b). “Forget about the words”? Tracking the language, media
and semiotic ideologies of digital discourse: The case of sexting. Discourse,
Context & Media, 20, 10–19.
Thurlow, C., & Bell, K. (2009). Against technologization: Young people’s new
media discourse as creative cultural practice. Journal of Computer-Mediated
Communication, 14(4), 1038–1049.
Thurlow, C., & Mroczek, K. (Eds.). (2011). Digital discourse: Language in the
new media. Oxford: Oxford University Press.
Turner, G. (2010). Ordinary people and the media: The demotic turn. London:
Sage.
van Dijk, T. A. (1989). Structures of discourse and structures of power.
Communication Yearbook, 12, 18–59.
van Leeuwen, T. (2013). Critical analysis of multimodal discourse. In C. A.
Chapelle (Ed.), The encyclopedia of applied linguistics (pp. 1–6). London and
New York: Blackwell Publishing.
Watts, R. J. (2003). Politeness. Cambridge: Cambridge University Press.
Yus, F. (2016). Humour and relevance. Amsterdam: John Benjamins.
Zappavigna, M. (2011). Ambient affiliation: A linguistic perspective on Twitter.
New Media & Society, 13(5), 788–806.
Part II
Past, Present and Future
2
The Coevolution of Computer-Mediated
Communication and Computer-Mediated
Discourse Analysis
Susan C. Herring
2.1 Introduction
Computer-mediated communication (CMC) was originally produced
and read as typed text and accessed through stand-alone clients.
Increasingly, however, textual CMC has been supplemented by graphical,
audio, and/or video channels of communication, and multiple modes1 of
CMC are available on Web 2.0 platforms and smartphones. As the tech-
nological affordances of CMC systems have evolved over time, so too
have the efforts of scholars to analyze the discourse produced using those
systems. One approach is computer-mediated discourse analysis
(CMDA), a specialization within the broader interdisciplinary study of
CMC distinguished by its focus on language and language use and by its
use of methods of discourse analysis to address that focus (Herring,
2004a). However, CMDA was developed for the analysis of textual CMC;
it has had little to say regarding, for example, the visual aspects of online
S. C. Herring (*)
Indiana University, Bloomington, IN, USA
e-mail: herring@indiana.edu

https://doi.org/10.1007/978-3-319-92663-6_2
26 S. C. Herring
discourse. It also tends to assume that online communication takes place

primarily through one semiotic mode (i.e., text).
In this chapter, I describe efforts to develop and extend CMDA over
time in order to address nontextual communication and the trend toward
convergence of multiple modes of CMC in a single platform. The exposi-
tion is structured in relation to three historical phases of CMC: pre-Web
(stand-alone textual clients), Web 1.0, and Web 2.0. For each phase,
developments in CMC technologies are juxtaposed with developments in
research on computer-mediated discourse (CMD) and the CMDA para-
digm itself, with a focus on attempts to address multimodality within the
paradigm, none of which have been entirely satisfactory to date. As an
alternative, I propose a theory of multimodal CMC that suggests a new
direction for CMDA going forward. This theory allows for the inclusion,
under the umbrella of ‘CMC’, of communication mediated by graphical
phenomena such as emoji, image memes, and avatars in virtual worlds, as
well as by certain kinds of robots; in so doing, it extends the definition of
CMC itself. Each of these phenomena can mediate human-to-human
communication, support social interaction, and co-occur with other
semiotic modes of communication, and thus their use constitutes fertile
ground for CMDA.
I conclude by challenging linguists who study CMD to move beyond
the confines of familiar methods and approaches, including traditional
CMDA, and to follow CMD where it leads, including beyond linguis-
tics, in order to analyze emergent and unprecedented discourse phenom-
ena in all their manifestations.
2.2 Computer-Mediated Discourse Analysis

CMDA was first conceptualized in 19942 and developed by the author as a
paradigm over the subsequent decade (Herring, 1997, 2001, 2004a, 2004b).
By 2004 it had evolved into a ‘methodological toolkit’ organized around
four linguistic levels: structure, meaning, interaction management, and
The Coevolution of Computer-Mediated Communication… 27
Table 2.1 The computer-mediated discourse analysis (CMDA) ‘toolkit’

Levels Issues Phenomena Methods
Structure Orality, formality, Typography, Structural/descriptive
efficiency, expressivity, orthography, linguistics, text
complexity, genre morphology, syntax, analysis, corpus
characteristics, etc. discourse schemata, linguistics, stylistics
formatting
conventions, etc.
Meaning What is intended? Meaning of words, Semantics, pragmatics
What is communicated? utterances (speech
What is accomplished? acts), exchanges, etc.
Interaction Interactivity, timing, Turns, sequences, Conversation analysis,
management coherence, repair, exchanges, threads, ethnomethodology
interaction as etc.
co-constructed, etc.
Social Social dynamics, Linguistic expressions Interactional
phenomena power, influence, of status, conflict, sociolinguistics, critical
identity, community, negotiation, face discourse analysis,
cultural differences, etc. management, play, ethnography of
discourse styles/lects, communication
etc.
social behavior, reflecting a progressive broadening of focus from the micro

(structure) to the macro (social) level.3 Associated with each level is a set of
issues, language phenomena, and methods adapted, for the most part, from
linguistics, as summarized in Table 2.1 (adapted from Herring, 2004a).
Because CMD in the 1990s was overwhelmingly textual, the CMDA
toolkit includes methods originally developed for textual analysis, such as
text analysis and corpus linguistics; but because much CMD is dynami-
cally interactive and includes ‘oral’ features, the toolkit also includes
methods traditionally applied to spoken discourse, such as conversation
analysis and interactional sociolinguistics. Regardless of the methods
used to analyze them, however, CMD data were originally produced and
read as typed text on a computer screen, and most CMDA still focuses on
interactive textual communication.
28 S. C. Herring
Over the years, modifications have been proposed to specific CMDA

methods (e.g., Stromer-Galley & Martinson, 2009). Moreover, the
CMDA paradigm as a whole was the subject of a special issue of
Language@Internet (Androutsopoulos & Beißwenger, 2008), in which
contributors proposed ways to expand the paradigm. These included
incorporating perspectives from outside linguistics, such as ethnogra-
phy (Androutsopoulos, 2008) from anthropology and ‘communities of
practice’ (Stommel, 2008) from education. ‘Multimodal’ enhancements
to CMDA were also proposed by Beißwenger (2008) and Marcoccia,
Atifi, and Gauducheau (2008), who analyzed video recordings of gaze
direction and body movements in individuals communicating via
instant messaging and (text) chat, an approach reminiscent of user stud-
ies in the field of human-computer interaction (e.g., McKinlay, Procter,
Masting, Woodburn, & Arnott, 1994). However, the focus of their
approach was on the behind-the-scenes production of textual CMD,
rather than on multimodal CMD as a joint enterprise manifest to all
participants in an interaction.
More recently, jointly produced multimodal CMD has begun to
receive attention from language scholars. Conversation analysis (e.g.,
Jenks & Firth, 2013; Licoppe & Morel, 2012) and social semiotics
(Sindoni, 2014) have been applied to audio- and videoconferencing, for
example, and pragmatic and content analysis approaches have been
adapted to analyze emerging forms of graphical communication (e.g.,
Herring & Dainas, 2017). However, these studies draw on different
methodological paradigms and, with the exception of Herring and Dainas
(2017), do not self-identify as CMDA.
Thus, new methods for analyzing online discourse have emerged over
time in response to changes in CMC technology. This raises the question
of whether CMDA is still relevant in the age of multimodal CMC. If
CMDA is expanded to include approaches such as those mentioned in
the preceding paragraphs, is it still meaningful to call it CMDA? Or is a
new overarching paradigm needed, one that draws its assumptions and
methods from ‘native’ multimodal approaches, rather than from linguis-
tic discourse analysis?
The remainder of this chapter is divided into two broad parts. In the
first part, I conduct a retrospective review of three broad stages of tech-
nological evolution that have shaped CMC from 1985 to the present:
pre-Web, Web 1.0, and Web 2.0. For each phase, I identify what new
CMC modes were available at the time, what (new) aspects of CMD
researchers focused on most, and the status of CMDA as a methodologi-
cal paradigm. The second part of the chapter is forward looking. In order
to address the increasing multimodality and convergence of CMC, I
propose a reconceptualization of CMC itself as fundamentally multi-
modal. From this vantage point, I argue that CMDA remains relevant,
not just because textual CMC continues to be important, but because
the principles at the core of the paradigm apply equally to interaction in
nontextual modes.
2.3 A
Historical Retrospective
2.3.1 Some Preliminary Remarks
My historical overview highlights connections between CMDA research

and the technological properties of CMC.4 However, I do not assume
that technology has been the only, or even necessarily the most impor-
tant, factor shaping CMC or that CMDA research has been driven
exclusively by changes in CMC technology. Rather, CMC modes are
sociotechnical constructs (Herring, 2002). Social and cultural prac-
tices, as well as shifts in intellectual fashion in, for example, discourse
analysis and sociolinguistics over the last several decades (see, e.g.,
Androutsopoulos, 2006), have also played shaping roles. Nonetheless,
it seems indisputable that the available technology sets limits on what
is possible in CMC and influences—if not strongly shapes—it and that
CMC is the medium in which CMD takes place. I assume as much in
what follows.
Moreover, I assume that the influence of CMC on CMD research and
the CMDA paradigm has been primarily unidirectional. CMDA research-
ers, like researchers of CMC in general, have tended to follow current
30 S. C. Herring
developments, choosing to study newer modes more often than older

ones (Herring, 2004c). This makes sense when one considers the rapid
rate at which the internet has evolved and at which new CMC modes
have emerged over a relatively short time span, as well as the heightened
interest that tends to surround the latest developments. Thus, there has
been a general directionality of influence from CMC technology, to com-
munication via CMC, to research that addresses that communication. As
a consequence, there is often a time lag between when a CMD phenom-
enon emerges and when it becomes the subject of published research, not
least because of delays caused by the publishing process. For this reason,
I sometimes include research with a later publishing date as part of an
earlier evolutionary stage.
Finally, the following overview is selective and necessarily subjective,
although I have tried to make it broad. In developing it, I drew on my
extensive personal knowledge of CMDA as the originator of the para-
digm, as well as my firsthand experience as a user and observer of CMC
since 1985. The works I cite are focused on language and language use in
CMC and written in English, excluding works that primarily concern
applied domains. In addition to my own work, I consulted a list of 220
‘top articles’ on computer-mediated language that I had previously com-
piled with input from 20 leading CMD scholars from 10 countries; this
was supplemented by focused Google Scholar searches. It is impossible in
this chapter to credit all the sources I considered, and I apologize in
advance to scholars whose important work there was not enough space to
mention.
2.3.2 Phase I: Pre-Web (1983–1993)
I begin my history around 1983, when the internet per se emerged from
its precursor, the Arpanet, which had been created in the late 1960s by
the United States Defense Advanced Research Projects Agency (DARPA)
(Hafner & Lyon, 1996). By 1983, the term ‘computer-mediated
communication’ had been in use for several years (e.g., Arnold, 1978),
and a number of modes of CMC were already in existence—email, group
conferencing, Usenet newsgroups, game Multi-User Dimensions or

Dungeons (MUDs), and an early form of synchronous one-to-one chat,
Unix ‘talk’. The end of Phase I would see the further introduction of the
first public group chat platform, internet Relay Chat (IRC); the internet
service provider AOL; and social MUDs and MOOs (MUDs, Object
Oriented).
These early modes of CMC were text-only—that is, users typed char-
acters on a keyboard that appeared as words and symbols on their and
their readers’ screens. Even users of MUDs and MOOs created and navi-
gated virtual spaces—rooms in houses, entire geographies—entirely in
typed text. Another notable feature of early CMC is that the different
modes were accessed through separate stand-alone clients—email at first
through telnetting to a server using a line-by-line interface and later
through specialized email clients, and IRC and MUDs through telnet-
ting to different dedicated servers, while newsgroups required specialized
‘reader’ software, and AOL required proprietary software as well as a
monthly subscription fee. These clients were not interoperable; perhaps
for this reason, few internet users in the 1980s and early 1990s regularly
used multiple modes of CMC (if they even knew about them). Moreover,
accessing the internet during this period required a dial-up modem and
was often painfully slow.
A small number of pioneering scholars first began addressing language
use in CMC in the mid-1980s. Their studies focused on disparate phe-
nomena ranging from abbreviated spelling and grammar to dialog struc-
ture, in limited-access CMC systems such as the Swedish COM system
(Severinson Eklundh, 1986), an IBM intranet (Murray, 1985, 1988), and
a ‘wizard of oz’ chat system designed to test a travel service prototype
(Ferrara, Brunner, & Whittemore, 1991). In a forward-looking early
essay, Baron (1984) speculated on the effects of CMC on the future of the
English language.
A second wave of study began around 1993, as internet access spread
and CMC started to come into more popular use; this wave continued
through the rest of the decade. This period saw the emergence of threads
of research focused around themes such as orality, creativity, and play in
typography, orthography, and morpho-syntax (Cherny, 1999(1994);
32 S. C. Herring
Danet, 1995; Werry, 1996); the influence of CMC systems on message

sequences and turn-taking (Cherny, 1999(1994); Condon & Čech, 1996;
Harrison, 1998; Herring, 1996b); and the classification of CMC in relation
to speech and writing (Baron,1998; Collot & Belmore, 1996; Ko, 1996;
Yates, 1996). Additionally, gender styles and gender power dynamics (e.g.,
Hall, 1996; Herring, 1993, 1994, 1996a; Kramarae & Taylor, 1993; Yates,
1993) emerged as a theme in response, in part, to the male-dominated
culture of the internet at the time. (Early CMC adopters were mostly
white, middle-class males). With a few exceptions, the data for most of
these studies came from public internet forums such as newsgroups, dis-
cussion lists, IRC, and MOOs. Although a number of second-wave stud-
ies were published after 1993, they are included with this set either
because their data were collected by 1993 or because the communication
they analyzed involved earlier modes of CMC which remained popular
through the mid-1990s.
Much of the language-focused research in Phase I can retroactively be
considered CMDA, even though CMDA as such did not yet exist.
Indeed, it was the blossoming of such research that generated the per-
ceived need for a language-focused paradigm such as CMDA. The earliest
pre-CMDA collection of language-focused CMC research was a panel I
co-organized with Brenda Danet for the 1993 International Pragmatics
Conference in Kobe, Japan. The papers presented in that panel later
formed the core of a book (Herring, 1996c). The CMDA paradigm itself
was emergent during this phase; it was conceived in 1994 for a 1995
workshop (see Note 2). That workshop would seek to bring together
researchers with different CMC language-related interests under a single
umbrella in order to assess what was known about online language and
language use, set future agendas, and start to establish a presence in the
broader area of CMC studies.
2.3.3 Phase II: Web 1.0 (1994–2004)
The World Wide Web was proposed by British physicist Tim Berners-Lee
in 1989 and first implemented publicly in 1991; at that time, it was
accessible only through a line-by-line browser. It quickly attracted general
notice, however, after a browser that integrated text and graphics, Mosaic,
was introduced in 1993 (Wikipedia, 2016). By 1994, the Web was being
hailed as a dramatic advancement in internet technology, both for its
graphical capabilities and for its ability to link documents in different
formats through ‘hyperlinks’ (Mitra, 1999).
Initially, Web content was not considered to be CMC by most CMC
researchers, in that it tended to be static. Moreover, the Web was used
more as a display (or advertising) medium than as a platform for recipro-
cally interactive communication. But gradually that changed, as previ-
ously stand-alone CMC modes such as chat and discussion forums were
integrated into Web platforms, and new modes of CMC emerged that
were native to the Web, such as wikis and weblogs. A second develop-
ment followed an increase in the bandwidth, or throughput capacity of
the internet in the mid-1990s. Nontextual modes of CMC were intro-
duced that did not reside on the Web, but rather on the internet itself;
these included 2D and 3D graphical virtual worlds, audio chat (Voice
over Internet Protocol (VoIP)), and video chat. Phase II also saw the rise
and fall of the chat client ICQ and the popularization of instant messag-
ing (IM) and SMS, or text messaging on mobile phones, which can be
considered a mode of CMC due to its similarities with IM. With the
exception of virtual worlds and audio chat and video chat, these new
modes remained predominantly textual.
During this period, the demographics of internet users diversified as a
result of wider access, and the number of female users increased, reaching
parity with—and even slightly surpassing—the number of male users in
2000, according to self-reported Web use (Pastore, 2000). In this same
period, starting in the mid-1990s, the internet—and with it, CMC—
began spreading rapidly from the USA and the UK to other countries in
the world. These demographic developments had implications for dis-
course and language use that were taken up in Phase II CMC research.
Notably, many researchers focused on identity issues relating to race,
gender, age, and/or sexuality (e.g., Greenfield & Subrahmanyam, 2003;
Herring & Martinson, 2004; Nakamura, 1995; Shaw, 1997; Tynes,
Reynolds, & Greenfield, 2004). With regard to gender, postmodern-influ-
enced analyses were proposed (e.g., Bucholtz, 1996; Danet, 1998; Rodino,
1997) to address phenomena such as gender switching and identity play,
34 S. C. Herring
which were claimed to break down traditional gender binaries. On the

(supposedly) anonymous internet, nobody would know even if you were
a dog.5 Another set of themes emerged in response to the growing number
of people engaging in CMC in languages other than English: language
choice/code-switching and internet multilingualism, drawing on principles
and methods from sociolinguistics (e.g., Androutsopoulos & Ziegler,
2004; Danet & Herring, 2003; Georgakopoulou, 1997; Paolillo, 1996;
Warschauer, El Said, & Zohry, 2002).
Online community also emerged as a popular theme during Phase II,
both in CMD and in CMC research more generally. From a discourse
perspective, CMD researchers addressed issues such as criteria for
community-hood, norms, and insider/outsider language (e.g., Baym,
1995; Cherny, 1999; Herring, 2004a; Paolillo, 1999). Other themes
emerged from the direct application of language-focused paradigms to
CMD. Interaction management is among the phenomena most directly
affected by the technological properties of CMC. Adapting methods
from conversation analysis, discourse-focused scholars started address-
ing issues such as openings and closings, turn-taking, topic develop-
ment, disrupted adjacency, and repair (e.g., Anderson, Beard, &
Walther, 2010 (written in 1996); Condon & Čech, 2001; Garcia &
Jacobs, 1998, 1999; Gruber, 1998; Herring, 1999; Rintel, Mulholland,
& Pittam, 2001; Rintel & Pittam, 1997; Schönfeldt & Golato, 2003).
Finally, several works addressed intertextuality, considering as intertex-
tual both hyperlinks on the Web and forms of reference more generally
(Gruber, 2000; Hodsdon Champeon, 2010 (written in 1996); Jucker,
2002; Mitra, 1999). With few exceptions, the data for these studies
came from traditional textual modes of CMC, both public and pri-
vate, which flourished during this period.
Discourse analysts did not immediately embrace Web-based or multi-
modal CMD. The first modes of Web-based CMC to be analyzed as
CMD were Web chat and Web forums, but these were initially treated
like IRC and mailing lists or newsgroups, respectively (e.g., Mauntner,
2005). Weblogs, when their communication was first studied empirically,
were addressed using methods of content analysis rather than discourse
analysis (e.g., Herring, Scheidt, Bonus, & Wright, 2004). Research on
the discourse of the new nontextual modes (graphical virtual worlds and
audio chat and video chat) would also await a later period.
The CMDA paradigm was actively under development in Phase II. I
organized a panel at the 1996 International Pragmatics Association confer-
ence on ‘Computer-Mediated Conversation’, guest edited a special issue of
the Electronic Journal of Communication on ‘Computer-Mediated
Discourse Analysis’ in 1997, and together with Tom Erickson of IBM
Research co-organized the ‘Persistent Conversation’ minitrack at the
Hawai’i International Conference on System Sciences (HICSS) in 1999,
which would run for 11 years. But while I had been using some form of
CMDA in my own research since the early 1990s and teaching it to stu-
dents since 1998, there was no published guide for others outside my
immediate sphere on how to do CMDA. Phase II saw the publication of
two chapters intended to address that gap: ‘Computer-Mediated Discourse’
in the Handbook of Discourse Analysis (2001) described a range of discourse
phenomena and what was known about them from existing research
through approximately 1999, and I laid out the CMDA approach system-
atically, including the methodological toolkit, in a chapter for a collection
on Designing for Virtual Communities in the Service of Learning (2004a).
One of my goals in constructing the toolkit was to provide an overview of
discourse phenomena that might be studied using CMDA, in the hopes of
inspiring research on thus-far-neglected phenomena. In that sense, the
2004 chapter set a broad agenda for CMD research going forward.
No sooner was the paradigm formally articulated, however, when it
faced a challenge. By the end of Phase II the Web had become increasingly
multimodal, and I began thinking about how to extend CMDA to analyze
interactive multimodal online discourse (e.g., Herring, 2004b, p. 73), espe-
cially still and moving images, since speech presents no problems for lin-
guistic approaches, in principle. Initially, I thought of extracting parameters
of graphical communication that would be analogous to principles of
grammar in verbal language, but I abandoned that idea when it led to con-
ceptualizations that were overly abstract. Instead, I turned to developing a
set of methods for analyzing Web content, including graphical elements,
based on content analysis (Herring, 2010). However, I would return to the
CMDA challenge several years later, taking a different approach.
36 S. C. Herring
2.3.4 Phase III: Web 2.0 (2004–2017)6
In 2004, Web entrepreneur Tim O’Reilly used the term ‘Web 2.0’ as the
name of a conference for “leaders of the internet Economy [to] gather to
debate and determine business strategy” (O’Reilly, 2005). The term has
come to refer, on the one hand, to changing trends in, and new uses of,
Web technology and Web design, such as participatory information shar-
ing, user-generated content, an ethic of collaboration, and use of the Web
as a social platform and, on the other hand, to the kinds of websites
where such activity takes place: blogs, microblogs, wikis, social network
sites, media-sharing sites, and so forth.
Two other important characteristics of the Web during this phase are
greatly increased bandwidth, which increases transmission speed and
supports video, audio, and graphics to a much greater extent than in
Phase II, and a tendency for different modes of CMC—including textual
modes—to converge on a single platform. As part of this trend, most
previously stand-alone CMC applications—email, chat, and forums, as
well as the various Web 2.0 platforms—became accessible through a stan-
dard Web browser.
Web communication now being pervasive and fully interactive, there
is no longer any question as to whether or not it qualifies as CMD.
Language-focused CMD research in Phase III has become more popular,7
and it has diversified into new areas. Some are natural areas for inclusion
in CMDA that for whatever reason were late to be taken up, such as prag-
matics (e.g., Atifi, Mandelcwaijg, & Marcoccia, 2011; Dresner & Herring,
2010; Herring, Stein, & Virtanen, 2013; Nastri, Peña, & Hancock,
2006; Yus, 2010) and variationist sociolinguistics (e.g., Bamman,
Eisenstein, & Schnoebelen, 2014; Hinrichs & White-Sustaíta, 2011;
McDonald, 2007; van Compernolle, 2008). Convergent media CMC—
CMC in which text co-occurs with other channels of communication on
the same platform—has been studied (e.g., Jucker, 2010; Zelenkauskaite
& Herring, 2008). Related to multimodality, researchers are also starting
to address the methodologically challenging topic of online and offline
communication, including how CMC is integrated with offline activities
(e.g., Aarsand, 2008; Danby, Butler, & Emmison, 2013; Harris, Danby,
Butler, & Emmison, 2012; Jones, 2011; Leppännen, Pitkänen-Huhta,
Piirainen-Marsh, Nikula, & Peuronen, 2011). Finally, discourse analysis

has been applied to analyze deceptive online practices such as lying
(Hancock et al., 2008), email hoaxes and frauds (Blommaert & Omoniyi,
2006; Heyd, 2013), and spam (Barron, 2006).
In Phase III, CMDA itself is being challenged to expand and adapt in
various ways. Automated corpus analysis challenges the definition of dis-
course analysis, which has traditionally been done manually by human
researchers, rather than by machines, although automated techniques are
increasingly being used to address questions of interest to discourse ana-
lysts (e.g., Danescu-Niculescu-Mizil, West, Jurafsky, Leskovec, & Potts,
2013; Dürscheid & Stark, 2011; Emigh & Herring, 2005; Herring &
Paolillo, 2006; Huffaker & Calvert, 2005; Niederhoffer & Pennebaker,
2002). From the opposite direction, researchers of social and cultural con-
text (e.g., Androutsopoulos, 2011; Chun & Walters, 2011; Jones, 2004;
Thurlow & Mroczek, 2011) challenge CMDA researchers to adopt
broader qualitative and critical perspectives. Methodological expansions of
CMDA have been proposed (e.g., Androutsopoulos, 2008; Beißwenger,
2008; Garcia, Standlee, Bechkoff, & Cui, 2009; Marcoccia et al., 2008),
as have competing theories of online communication (e.g., Weininger &
Shield, 2004; Yus, 2010). These are healthy developments that indicate
that CMDA is an active area of scholarship.
Before concluding this retrospective survey, three additional per-
spectives on CMD that have been researched intermittently over time
should be mentioned. These are participation in CMC (e.g., Herring,
1993; Herring, Johnson, & DiBenedetto, 1995; Joyce & Kraut,
2006; Marcoccia, 2004), humor (e.g., Danet, Ruedenberg-Wright, &
Rosenbaum-Tamari, 1997; Hubler & Bell, 2003; Nishimura, 2012;
Su, 2003), and language change (e.g., Baron, 1984; Berdicevskis, 2013;
Danescu-Niculescu-Mizil et al., 2013; Gao, 2006; Herring, 1998; Rowe,
2011). Moreover, a fourth theme, (im)politeness, deserves special mention
in that it has been a research focus—and an issue of concern—since Kiesler
et al. (1984) reported finding disinhibitory effects, including ‘flaming’,
in experimental studies involving CMC in the 1980s. Numerous stud-
ies have addressed online politeness over the years (e.g., Bou-Franch &
Garcés-Conejos Blitvich, 2014; Darics, 2010; Garcés-Conejos Blitvich,
2010; Graham, 2007; Haugh, 2007; Herring, 1994; Locher & Watts,
38 S. C. Herring
2005; Pihlaja, 2011; Morand & Ocker, 2003; Rice & Love, 1987), and
recent concerns about worsening incivility on the internet in the era of
trolling, political polarization, and ‘fake news’ have led to a further uptick
in research on this topic (e.g., Hardaker, 2010; Lange, 2007; Phillips,
2015; Phillips & Milner, 2017; Rowe, 2015; Santana, 2014).
With the exception of the above-mentioned themes, which are not
associated predominantly with a single phase, Table 2.2 summarizes the
CMC technology, the CMD research, and developments in the CMDA
paradigm that were new in each of the three phases surveyed in this and
the previous two sections.
Phase III has seen increasing recognition for, and use of, the CMDA
paradigm (e.g., Androutsopoulos & Beißwenger, 2008; Darics, 2010;
Koteyko, Jaspal, & Nerlich, 2013; Kushin & Kitchener, 2009). For my
part, I followed my programmatic 2004 chapter (Herring, 2004a) with a
2007 article laying out a classification scheme for CMD according to two
dimensions, or sets of facets: medium and situation variables. However,
CMD has evolved in the era of Web 2.0; it now features previously unat-
tested phenomena such as ‘likes’ and graphical ‘reactions’ on social net-
work sites, dynamically collaborative-authored content on wikis, and
asynchronous conversational exchanges via videos and images. The
CMDA paradigm is confronted not only with the challenge of develop-
ing methods to analyze these new ‘medium’ phenomena, but also with
accounting for them within its overall conceptual framework. I made
another attempt to integrate methods and issues for multimodal analysis
into CMDA in Herring (2013a), which involved proposing the addition
of a fifth, multimodal, level to the CMDA toolkit. However, ‘multimo-
dality’ is not analogous to ‘structure’, ‘meaning’, and so on, in that it
refers to the channel of communication rather than a linguistic level of
analysis, so the proposal was conceptually problematic. In fact, all the
other levels in the toolkit can be studied for (fit within) multimodal
CMD. It also lumps all multimodal CMD into a single category, without
differentiating among audio, video, graphics, and so forth, and thus is
descriptively inadequate. An alternative, and I believe more satisfactory,
solution to the problem of multimodality and CMDA is laid out in the
remainder of this chapter.
Table 2.2 Three phases in the coevolution of CMC and CMDA
Phase CMC characteristics CMD research themes CMDA paradigm
I: Pre-Web • Email, mailing lists, Usenet • Typography and • Pre-CMDA: 1993 IPrA
(1983–1993) newsgroups, IRC; AOL chat, orthography (‘oral’ panel w/Brenda Danet,
MUDs/MOOs, etc. features) which led to published
• Stand-alone clients, not • Message exchange/ volume (Herring, 1996c)
interoperable turn-taking • Name first used in 1994
• Text only (+ emoticons) • Gender styles and gender Call for Abstracts for a
• Users mostly white males in and power dynamics 1995 GURT workshop
the USA and UK • Word frequency/registers
• (Morpho)syntax
• Message/sequence structure
II: Web 1.0 • Web chat, Web forums, • Interaction management • First CMDA collection:
(1994–2003) blogs, wikis, graphical virtual • Online community Herring (1997)
worlds, IM, ICQ, SMS, • Identity • CMD and CMDA approach
graphical worlds, VoIP, video • Language choice/ formally laid out (Herring,
chat, etc. code-switching 2001, 2004a) as a broad
• Convergence of CMC modes • Internet multilingualism agenda for CMD research
on the Web • Intertextuality • Contents of static websites
• Spread of internet to other • Postmodern-influenced not considered CMD
countries gender research
• Increase in female users
(continued)
The Coevolution of Computer-Mediated Communication…
39
Table 2.2 (continued)
40
Phase CMC characteristics CMD research themes CMDA paradigm

III: Web 2.0 • Media-sharing sites, social • Pragmatics • Faceted classification
(2004–2017) network sites, microblogs, • Sociolinguistic variation scheme (Herring, 2007)
etc. • Critical language ideology • Growing recognition and
• More bandwidth means • Accommodation use of CMDA, including
more use of video, audio, • Discourse communities identification of its limits
S. C. Herring
and graphics • Deception, hoaxes, and • Critiques by

• Convergent Media CMC— spam Androutsopoulos and
textual CMC included in • Online social support Beißwenger (2008)
multimodal platforms (e.g., • Multilingualism; language • Applied to an expanding
Zelenkauskaite & Herring, choice/code-switching range of linguistic
2008) • Identity, race, ethnicity, age domains
• Web communication is fully • Language use in individual • Web communication is
interactive (distinction modes (e.g., IM, text considered CMD
between Web and non-Web messaging; blogs, wikis; • Attempts to expand to
virtually disappears) Twitter, Facebook; YouTube) multimodal CMD
• Nontextual modes unsatisfactory
• Convergent Media CMC • Identification of new
• Online and offline challenges, for example
communication how to deal with
• Automated corpus analysis changing nature of links
• Social and cultural context (and ‘likes’) in social media
• Methodological expansions
of CMDA
• Other theoretical
approaches
2.4 R
econceptualizing CMC
The current conceptualization of CMC, which has not been updated sub-
stantially since the term first started appearing in print nearly 40 years
ago,8 retains connotations of textual, one-mode-at-a-time transmission.
Thus, as a first step, I propose reconceptualizing CMC as fundamentally
multimodal. Indeed, if the proverbial Martian scholars were to come to the
Earth and encounter CMC for the first time in 2018, they would undoubt-
edly perceive its transmission via multiple semiotic modes to be inherent
in its nature—as, I assume, do young people who have never known a
world in which CMC involved only textual exchanges. A consequence of
conceptualizing CMC as multimodal is that non-multimodal CMC ceases
to exist, except historically. This reconceptualization does not exclude text-
only CMC; rather, text is one of a number of possible modes of transmis-
sion that also include voice, audio, video, and—I suggest—graphics and
certain kinds of robotic devices, as represented schematically in Fig. 2.1.9
Text, audio, and video CMC have been around for decades and have
been addressed often in the literatures on CMC and human-computer
Computer-Mediated Communication
Text Audio Video Robot Graphics
Dynamic Static
email chat etc. Internet Skype
telephony Navigable Not Drawing Photograph
Video navigable
conferencing Offline Online emoji
GIFs Modified Unmodified
avatars
image profile
Avatar-mediated memes pics
communication
Interactive
multimodal Robot-mediated
platforms communication
Fig. 2.1 CMC reconceptualized as inherently multimodal

42 S. C. Herring
interaction (see, for example, Jenks & Firth, 2013 and Jepson, 2005 for
voice chat; O’Connail, Whittaker, & Wilbur, 1993 for video conferenc-
ing). Therefore, in the following subsections, I focus on three newer phe-
nomena from Fig. 2.1: communication on interactive multimodal
platforms (IMPs); graphical communication, including avatar-mediated
communication (AMC); and robot-mediated communication (RMC).
Like other CMC modes, these emergent phenomena involve verbal lan-
guage (as well as other semiotic systems); mediate human-to-human
communication; and support social interaction, and thus the discourse
that is produced through them constitutes CMD. I then consider the
ramifications of these phenomena—and the multimodal model of CMC
more generally—for CMDA.
2.4.1 Interactive Multimodal Platforms
IMPs are digital platforms on which two or more semiotic modes—

typically, text plus audio, video, and/or graphics—are available to
support interactive human-to-human communication (Herring,
2015). IMPs differ from multimedia platforms or convergent media
CMC platforms in that IMPs support multiple forms of CMC,
whereas the other platforms may have only text commenting or chat-
ting on a site with other media, such as games, news stories, or prod-
ucts for sale, that serve primarily entertainment, informational, or
commercial purposes. An early example of an IMP is YouTube, which
supports both video exchanges and text comments. Other examples
are multiplayer online games where players can communicate with
each other via both text chat and voice—and in some games, player
avatar movements—and videoconferencing systems that support syn-
chronous video, audio, and text-based communication (and some-
times collaborative drawing on a whiteboard). The list of IMPs now
includes Skype, Google+, and Facebook. As social media sites con-
tinue to add CMC affordances such as video chat, IMPs are becoming
increasingly common.
IMPs, like convergent media CMC, are characterized by media
coactivity; that is, users engage in multiple activities involving differ-
ent semiotic modes on the same platform (e.g., Herring, Kutz, Paolillo,
& Zelenkauskaite, 2009). Especially when synchronous communica-
tion is involved, media coactivity raises issues such as competition for
users’ attention and the effects of such competition on message pro-
duction and discourse processing. Moreover, both IMPs and conver-
gent media CMC call for theories and methods of analysis that can
address the interplay between text and other modes of mediated com-
munication. IMPs differ, though, in that they provide contexts that
allow for direct comparison of communication styles and strategies
across different modes on the same platform. Studies that have focused
on IMP discourse to date include Newon (2011, World of Warcraft),
Sindoni (2014, videoconferencing), and Herring and Demarest (2017,
VoiceThread).
2.4.2 Graphical Communication
Internet users have been chatting online via graphical avatars in virtual
worlds since the mid-1990s, when 2D and 3D virtual worlds such as the
Palace and AlphaWorlds were introduced. Second Life popularized this
type of CMC in the mid-2000s, and it remains a common feature of
virtual game worlds such as World of Warcraft. While AMC usually
involves verbal communication via text and/or voice chat, some studies
have analyzed avatar movement, gaze direction, and other semiotic sig-
nals as part of the avatars’ communicative repertoire, including how their
physical behaviors interact with other modes such as text and speech
(e.g., Garau, Slater, Bee, & Sasse, 2001; Newon, 2011; Yee, Bailenson,
Urbanek, Chang, & Merget, 2007).
More recently, graphics on social media sites have evolved from cute or
funny images or videos that people share for their entertainment value to
semiotic devices that are used to convey propositional content, in lieu of,
or in conjunction with, text. As such, they can function as propositions
within messages or as stand-alone turns in conversational exchanges.
These devices include emoticons, emoji, stickers, GIFs, and text-in-image
memes; video clips may also serve similar functions. Herring and Dainas
(2017) refer to these devices as graphicons.
44 S. C. Herring
Emoji, especially, are attracting growing scholarly attention (e.g., Al

Rashdi, 2015; Dürscheid & Siever, 2017; Nishimura, 2015), as well as
speculation that they are becoming a new ‘language’ (e.g., Stockton,
2015). Examples can be found on the Web of text message exchanges and
song lyrics written entirely or almost entirely in emoji; an emoji phrase-
book also exists, as well as a ‘translation’ of Herman Melville’s classic
novel, Moby Dick, titled ‘Emoji Dick’ (Radford, Chisholm, Hachey, &
Han, 2016). However, the intended meanings of emoji are not always
clear (e.g., Miller et al., 2016), and their use raises issues of ambiguity,
ambiguity resolution, miscommunication, and repair. Moreover, most
emoji depict concrete entities, including facial expressions, or physical
actions, such as running. Emoji can express emotions (via facial expres-
sions), but emoji for abstract concepts such as ‘concept’, ‘analysis’, and
‘freedom’ are largely missing, and symbols that express syntactic relations
are also rare (the + sign, indicating coordination, is an exception).
Furthermore, one cannot easily embed propositions or indicate events
out of temporal sequence via emoji. These present limitations notwith-
standing, there is anecdotal evidence that some users are adapting emoji
whose literal referents are ‘things’ to express more abstract notions (e.g., a
rightward-pointing finger for the English directional preposition ‘to’; an
eye-rolling face as a deictic pointing to content posted earlier (‘above’) in
an exchange). In addition, emoji-only exchanges occur, typically as a
form of playful communication, in which some users construct elaborate
emoji sequences. These exchanges constitute intriguing data for analysis
in terms of their emergent structure, intended (vs. understood) mean-
ings, conversational dynamics, and social signification.
Thus far, there are relatively few discourse studies that analyze
graphicons-in-use, although this is changing in the future, as CMD schol-
arship catches up with contemporary CMC usage. Studies that have
focused on graphical conversation to date include McDonald (2007, text-
in-image exchanges on a community blog), Pihlaja (2011, video exchanges
on YouTube), Dresner and Herring (2010, emoticons), Amaghlobeli
(2012, emoticons in SMS), Bourlai and Herring (2014, Tumbler GIFs),
Voida and Mynatt (2005, photographs), Katz and Crocker (2015, selfies),
Herring and Dainas (2017, graphicons in Facebook comment threads),
Nishimura (2015, emoticons and emoji in Japanese blogs), Dürscheid

and Siever (2017, emoji), and Lim (2015, stickers).
2.4.3 Telepresence Robot-Mediated Communication
Telepresence robots are robotic devices controlled by remote users that

can be navigated through physical environments and that include two-
way audio and video conferencing capabilities. Sometimes characterized
as ‘videoconferencing on wheels’ (Desai, Tsui, Yanco, & Uhlik, 2011), the
telepresence robot is also an avatar or representation of the user that can
be remotely manipulated, similar to a graphical avatar in a virtual world.
The robot can also be thought of as a technology or channel that mediates
communication directly, analogous to text, audio, video, and graphics,
and whose properties potentially shape the nature of the communication
that occurs through it. These three conceptual relationships are repre-
sented by separate lines connecting to ‘robot-mediated communication’
in Fig. 2.1. RMC is communication in which at least one party is telepre-
sent via, and remotely controlling, such a robot (Herring, 2015).
RMC raises numerous issues as regards social interaction, many of
which relate to the technological properties and limitations of the current
generation of telepresence robots and are resulting in the emergence of
new interactional norms. For example, the limited mobility and range of
visibility of robot ‘pilots’ affect their ability to attract attention, gain and
hold the conversational floor, and time turn-taking appropriately. The
pilot may misgauge social distance due to a lack of depth perception and
position the robot too close or too far away from an interlocutor, may
talk too loudly due to a lack of audio feedback, or may linger too long
after a conversation due to missed social cues (Lee & Takayama, 2011).
‘Locals’ interacting with the robot must learn to understand that the
robot’s behaviors reflect technological issues, rather than social inappro-
priateness or intentional rudeness on the pilot’s part.
As yet little research has addressed discourse in RMC, and nonlinguis-
tic studies of RMC tend to be experimental or based on interviews. As
telepresence robots come into more common use, however, and corpora
of naturally occurring conversations between robot pilots and their local
46 S. C. Herring
interlocutor(s) become available, discourse analysis methods can fruit-

fully be applied to such data. In Herring (2016), I propose a number of
directions that such analysis could take.
2.5 Implications for CMDA

In the preceding subsections, I identified discourse and social interaction
issues raised by communication through IMPs, graphics, and telepresence
robots. To address these, methods and insights will be required that extend
beyond linguistic discourse analysis. For example, the literatures of semiot-
ics, ethnography, human-computer interaction, and human-robot interac-
tion could be instructive to consult. To analyze the interplay of text and
images, scholarship on comic books, or even ancient illuminated manu-
scripts, could potentially provide insight. To analyze videos, CMD analysts
might need to (re)learn spoken discourse analysis techniques and (re)
acquire the patience—or devise automated methods—to transcribe speech,
which has been unnecessary with textual CMD.
More generally, the inclusion of these new phenomena in the defini-
tion of CMC challenges CMDA to evolve. First, it obliges the paradigm
to fully embrace not just emergent forms of mediated discourse, but also
audio chat and video chat/conferencing, which until now have been
included in principle, but less often in practice, as part of CMD (but see
Jenks & Firth, 2013; Licoppe & Morel, 2012; Sindoni, 2014). Second,
methods need to be identified, modified, or innovated for analyzing the
discourse produced through each semiotic mode, as has been done for
textual CMD (see Herring, 2004a). The expanded conceptualization also
calls for cross-mode analysis methods, as both CMC modes (e.g., discus-
sion forums, instant messaging, and video chat) and semiotic modes
(e.g., text, graphics, and video) converge on a single platform and as
social media users produce content, such as tweets, that is reposted and
shared on other, including mass media, platforms (e.g., Squires, 2010).
Finally, at the theoretical level, the inclusion of graphics as discursive
devices broadens the definition of CMD beyond verbal language. Holistic
understandings are required to understand what behaviors, characteris-
tics, and/or assumptions unite these potentially disparate phenomena.10
Multimodal CMC, as reconceptualized in Fig. 2.1, has several charac-

teristics that are arguably shared by all modes of CMC. For example,
although motivations for use vary by individual and according to con-
text, all modes afford a facilitating social distance and an ability to be
more selective in one’s self-presentation than face-to-face communica-
tion. Norms of language and social interaction have evolved or can be
expected to evolve in all modes of CMC with use over time. Communities
of users also tend to form around CMC platforms. In these and other
regards, earlier textual CMC modes and practices can shed light on cur-
rent and future nontextual developments, as, for example, the evolution
of ASCII emoticons suggests a trajectory for emoji use. More generally,
the literatures on better-studied modes can guide the identification of
research questions and the interpretation of results from investigations
into newer modes. It appears, for example, that research on both video
conferencing and AMC has relevance for RMC (Herring, 2015).
Last but not least, it is possible to analyze CMC in each semiotic
mode—even graphics—on the levels of structure, meaning, interaction
management, and social behavior. All have structural properties; all signify
meanings in context; all can be used to hold conversational exchanges; and
social behavior can be enacted through every mode. In other words, all the
modes mediate human-human discourse, and thus discourse analysis as an
approach and a set of methods is applicable to all.
2.6 B
roader Implications and Conclusions
Over the course of its 25-year history, CMDA has faced a number of chal-
lenges from technological advances in CMC, including increased band-
width, increasing multimodality, and media convergence. In order to
address these challenges, I have suggested here that it is necessary to recon-
ceptualize CMC itself. I have proposed a unified view of multimodal CMC
that includes graphics and robotic devices as mediating channels, along
with text, audio, and video. This reconceptualization is a first, theoretical
step toward equipping CMDA and CMD researchers with the tools they
need to analyze multimodal, convergent CMC. The advantages of this view
include that it enables emergent CMC modes to be understood, in part, in
48 S. C. Herring
terms of familiar modes. It also highlights where new methods are needed
for analyzing multimodal CMD, be it in individual modes, co-occurring
modes (IMPs), or interaction among/across modes. An important next step
will be to develop specific sets of methods appropriate for analyzing each
nontextual mode, especially for less traditional phenomena.
One implication of this reconceptualization is that CMDA as a para-
digm remains relevant. Regardless of the technology that mediates it,
CMD can be analyzed in terms of its structure, its pragmatic meanings,
its interactional properties, and the kinds of social behavior it supports.
This is as true for RMC and communication via emoji as it is for tradi-
tional email and chat.
Another implication is that linguistic methods alone are insufficient to
address the range of phenomena that are currently attested in CMD (see
Gee, 2014). This leads to a broader question. As CMC continues to evolve,
linguists who study CMD have a choice: Do they remain within the
borders of known linguistics methods and approaches, including tradi-
tional CMDA? Or do we (for I am included in this category) follow the
technology where it leads, including beyond linguistics, to study CMD in
all its forms? This is a question that we all must ask and answer for ourselves
as CMC technologies become ever more rich, complex, and multimodal.
Notes
1. A ‘mode’ is a specific communication type within a medium such as the
computer (Murray, 1988, p. 353). ‘CMC modes’ in this chapter are
sociotechnical constructs that combine online messaging protocols with
the social and cultural practices that have evolved around, or are emer-
gent through, their use; examples include email, instant messaging, and
virtual worlds (cf. Herring, 2002). I use the term ‘semiotic mode’, follow-
ing Kress and van Leeuwen (2001), for the ‘modes’ implicit in the term
‘multimodal’, for example, text, audio, video, graphics, and music. When
my focus is on the pathway or mediating technology that transmits mul-
timodal communication, the term ‘channel’ may also be used. Thus, for
example, instant messaging (CMC mode) is textual (semiotic mode),
transmitted via typed text on a keyboard or mobile device (channel).
2. The name ‘Computer-Mediated Discourse Analysis’ was first used in a

1994 Call for Papers circulated by the author for a workshop by the same
name at the 1995 Georgetown University Roundtable on Languages and
Linguistics (GURT).
3. The toolkit also includes a nonlinguistic level, Participation (see Herring,
2004a).
4. The history of the internet has been written about extensively; rather
than go into detail about it here, I refer the reader to other sources (e.g.,
Hafner & Lyon, 1996). CMC modes have also been described in numer-
ous works, some of which present, or at least allow for, comparison of
the modes’ chronological emergence (e.g., Herring, 2002).
5. A reference to a 1993 cartoon in the New Yorker (see, e.g., Herring,
2003).
6. Some argue that we are no longer in the era of Web 2.0 but rather
in a new phase, Web 3.0, characterized by technological advances
in the semantic Web, artificial intelligence, machine learning, and natural
language search—the so-called Intelligent Web (Spivack, n.d.). It is pos-
sible that these technological developments are already affecting CMD,
but research themes associated with them have yet to emerge clearly, it
seems to me. Thus, I have provisionally defined Phase III as continuing
up to the time of this writing.
7. Many themes from earlier phases have continued to attract CMD
scholarship, including nonstandard typography and orthography (e.g.,
Androutsopoulos, 2007; Anis, 2007; Bieswanger, 2007; Herring &
Zelenkauskaite, 2009; Sebba, 2003; Shortis, 2007; Vaisman, 2013);
interaction management (e.g., Bou-Franch et al., 2012; Herring,
2013b; Riordan et al., 2013; Simpson, 2005; Stromer-Galley &
Martinson, 2009); multilingualism and language choice/code-switching
(e.g., Androutsopoulos, 2013; Deumert & Masinyana, 2008; Fung
& Carter, 2007; Georgakopoulou, 2011; Hinrichs, 2006; Lee, 2007,
2016; Lenihan, 2011; Morel, Bucher, Pekarek-Doehler, & Siebenhaar,
2012; Siebenhaar, 2006; Spilioti, 2009; Themistocleous, 2013); iden-
tity (e.g., Campbell, 2004; Marwick, 2013; Milani, 2013; Sargeant
& Tagg, 2014); race and ethnicity (e.g., Byrne, 2008; Heyd, 2014;
Hughey & Daniels, 2013; Walton & Jaffe, 2011) and adolescence (e.g.,
Buckingham, 2008; Kapidzic & Herring, 2011; Leppännen et al.,
2011; Subrahmanyam, Smahel, & Greenfield, 2006; Tagliamonte &
Denis, 2008; Tynes et al., 2004). Also, research that describes the dis-
course characteristics of a specific mode of CMD, such as email (Baron,
1998), MUDs and MOOs (Cherny, 1999), blogging (Herring, Scheidt,
50 S. C. Herring
Bonus, & Wright, 2004; Peterson, 2011; Puschmann, 2010), IM (Ling

& Baron, 2007), text messaging (Ling, 2005; Tagg, 2012), and Twitter
(Honeycutt & Herring, 2009; Page, 2012; Zappavigna, 2012) contin-
ues to be a popular genre.
8. In what is possibly one of the earliest uses, George Arnold used the term
‘computer-mediated communication’ in his 1978 Columbia University doc-
toral dissertation, explaining that he preferred the ‘more general’ term to
others then in use, such as ‘computer conferencing’, which he found ‘overly
restrictive’ (p. 5). Many of Arnold’s observations on ‘the nature of computer-
mediated communication’ (pp. 5–13) seem familiar 40 years later.
9. This conceptualization is expandable. In theory, additional branches
could be added to the top node in Fig. 2.1 as new technologies that
mediate human-human communication are introduced in the future.
10. See, for example, Gee (2014) for a unified view of discourse that includes
video games, among other phenomena.
References
Aarsand, P. A. (2008). Frame switches and identity performances: Alternating
between online and offline. Text & Talk, 28(2), 147–165. https://doi.
org/10.1515/TEXT.2008.007
Al Rashdi, F. (2015). Forms and functions of emojis in WhatsApp interaction
among Omanis. Doctoral dissertation, Georgetown University.
Amaghlobeli, N. (2012). Linguistic features of typographic emoticons in SMS
discourse. Theory and Practice in Language Studies, 2(2), 348–354.
Anderson, J. F., Beard, F. K., & Walther, J. B. (2010). Turn-taking and the local
management of conversation in a highly simultaneous computer-mediated
communication system. Language@Internet, 7, article 7. Retrieved from
http://www.languageatinternet.org/articles/2010/2804/index_html#d57e622
Androutsopoulos, J. (2007). Style online: Doing hip-hop on the German-
speaking web. In P. Auer (Ed.), Style and social identities: Alternative approaches
to linguistic heterogeneity (pp. 279–317). Berlin/New York: de Gruyter.
Androutsopoulos, J. (2008). Potentials and limitations of discourse-centred
online ethnography. Language@Internet, 5, article 8. Retrieved from http://
www.languageatinternet.org/articles/2008/1610

computer-mediated discourse. In C. Thurlow & K. R. Mroczek (Eds.),
Digital discourse: Language in the new media (pp. 277–298). New York:
Oxford University Press.
Androutsopoulos, J. (2013). Networked multilingualism: Some language prac-
tices on Facebook and their implications. International Journal of Bilingualism,
19(2), 185–205. https://doi.org/10.1177/1367006913489198
Androutsopoulos, J., & Beißwenger, M. (2008). Introduction: Data and meth-
ods in computer-mediated discourse analysis. Language@Internet, 5, article 2.
Retrieved from http://www.languageatinternet.org/articles/2008/1609
Androutsopoulos, J., & Ziegler, E. (2004). Exploring language variation on the
internet: Regional speech in a chat community. In B.-L. Gunnarsson et al.
(Eds.), Language variation in Europe (pp. 99–111). Uppsala: Uppsala
University Press.
Anis, J. (2007). Neography: Unconventional spelling in French SMS text mes-
sages. In B. Danet & S. C. Herring (Eds.), The multilingual internet: Language,
culture, and communication online (pp. 87–115). New York: Oxford University
Press.
Arnold, G. W. (1978). A computer-mediated structured communications system.
Doctoral dissertation, Columbia University School of Engineering Science.
Retrieved from http://www1.cs.columbia.edu/~hgs/papers/others/1978/
Arnold_ComputerMediated.pdf
Atifi, S., Mandelcwaijg, S., & Marcoccia, M. (2011). The co-operative principle
and computer-mediated communication: The maxim of quantity in news-
group discussions. Language Sciences, 33(2), 330–340. http://dx.doi.org.
proxyiub.uits.iu.edu/10.1016/j.langsci.2010.10.011
Bamman, D., Eisenstein, J., & Schnoebelen, T. (2014). Gender identity and
lexical variation in social media. Journal of Sociolinguistics, 18(2), 135–160.
https://doi.org/10.1111/josl.12080
Baron, N. S. (1984). Computer mediated communication as a force in language
change. Visible Language, 18(2), 118–141.
Baron, N. S. (1998). Letters by phone or speech by other means: The linguistics
of email. Language and Communication, 18, 133–170.
Barron, A. (2006). Understanding spam: A macro-textual analysis. Journal of
Pragmatics, 38(6), 880–904. https://doi.org/10.1016/j.pragma.2005.04.009
Baym, N. (1995). The emergence of community in computer-mediated com-
munication. In S. Jones (Ed.), Cybersociety: Computer-mediated communica-
tion and community (pp. 138–163). Thousand Oaks, CA: Sage.
52 S. C. Herring
Beißwenger, M. (2008). Situated chat analysis as a window to the user’s perspec-

tive. Language@Internet, 5, article 6. Retrieved from http://www.lan-
guageatinternet.org/articles/2008/1532
Berdicevskis, A. (2013). Language change online: Linguistic innovations in Russian
induced by computer-mediated communication. Doctoral thesis, University of
Bergen. Retrieved from http://hdl.handle.net/1956/6847
Bieswanger, M. (2007). 2 abbrevi8 or not 2 abbrevi8: A contrastive analysis of
different space and time-saving strategies in English and German text mes-
sages. Texas Linguistics Forum, 50, 1–12. Retrieved from http://studentorgs.
utexas.edu/salsa/proceedings/2006/Bieswanger.pdf
Blommaert, J., & Omoniyi, T. (2006). Email fraud: Language, technology, and
the indexicals of globalisation. Social Semiotics, 16(4), 573–605. https://doi.
org/10.1080/10350330601019942
Bou-Franch, P., & Garcés-Conejos Blitvich, P. (2014). Conflict management in
massive polylogues: A case study from YouTube. Journal of Pragmatics, 73,
19–36. https://doi.org/10.1016/j.pragma.2014.05.001
Bou-Franch, P., Lorenzo-Dus, N., & Garcés-Conejos Blitvich, P. (2012). Social
interaction in YouTube text-based polylogues: A study of coherence. Journal
of Computer-Mediated Communication, 17, 501–521. https://doi.
org/10.1111/j.1083-6101.2012.01579.x
Bourlai, E., & Herring, S. C. (2014). Multimodal communication on Tumblr: “I
have so many feels!”. Proceedings of WebSci’14, June 23–26, Bloomington,
IN. Retrieved from http://ella.slis.indiana.edu/~herring/tumblr.pdf
Bucholtz, M. (1996). Geek the girl: Language, femininity and female nerds. In
N. Warner, J. Ahlers, & L. Bilmeset al (Eds.), Gender and belief systems
(pp. 119–132). Berkeley, CA: Berkeley Women and Language Group.
Buckingham, D. (Ed.). (2008). Youth, identity, and digital media. Cambridge,
MA: MIT Press.
Byrne, D. N. (2008). The future of (the) ‘race’: Identity, discourse, and the rise
of computer-mediated public spheres. In A. Everett (Ed.), Learning race and
ethnicity: Youth and digital media (pp. 15–38). Cambridge, MA: The MIT
Press. https://doi.org/10.1162/dmal.9780262550673.015
Campbell, J. E. (2004). Getting it on online: Cyberspace, gay male sexuality, and
embodied identity. London: Routledge.
Cherny, L. (1999). Conversation and community: Chat in a virtual world.
Stanford: CSLI Publications.
Chun, E., & Walters, K. (2011). Orienting to Arab orientalisms: Language, race
and humor in a YouTube video. In C. Thurlow & K. R. Mroczek (Eds.),
Digital discourse: Language in the new media (pp. 252–273). New York:
Collot, M., & Belmore, N. (1996). Electronic language: A new variety of English.
In S. C. Herring (Ed.), Computer-mediated communication: Linguistic, social
and cross-cultural perspectives (pp. 13–28). Amsterdam: John Benjamins.
Condon, S. L., & Čech, C. G. (1996). Functional comparisons of face-to-face
and computer-mediated decision making interactions. In S. C. Herring
(Ed.), Computer-mediated communication: Linguistic, social and cross-cultural
perspectives (pp. 65–80). Amsterdam: John Benjamins.
Condon, S. L., & Čech, C. G. (2001). Profiling turns in interaction: Discourse
structure and function. In Proceedings of the Thirty-Fifth Hawai’i International
Conference on System Sciences. Los Alamitos, CA: IEEE Press.
Danby, S., Butler, C. W., & Emmison, M. (2013). When ‘listeners can’t talk’:
Comparing active listening in opening sequences of telephone and online
counselling. Australian Journal of Communication, 36(3), 91–113.
Danescu-Niculescu-Mizil, C., West, R., Jurafsky, D., Leskovec, J., & Potts, C.
(2013). No country for old members: User lifecycle and linguistic change in online
communities. Proceedings of the 22nd International Conference on World Wide
Web (WWW ‘13), 307–318. https://doi.org/10.1145/2488388.2488416
Danet, B. (Ed.). (1995). Play and performance in computer-mediated commu-
nication. Special issue, Journal of Computer-Mediated Communication, 1(2).
Retrieved from http://209.130.1.169/jcmc/vol1/issue2/
Danet, B. (1998). Text as mask: Gender, play, and performance on the internet.
In S. Jones (Ed.), Cybersociety 2.0: Revisiting computer-mediated community
and technology (pp. 129–158). Thousand Oaks, CA: Sage.
Danet, B., & Herring, S. C. (Eds.). (2003). The multilingual internet: Language,
culture, and communication in instant messaging, email, and chat [Special
issue]. Journal of Computer Mediated Communication, 9(1).
Danet, B., Ruedenberg-Wright, L., & Rosenbaum-Tamari, Y. (1997). Smoking
dope at a virtual party: Writing, play and performance on Internet Relay
Chat. Journal of Computer-Mediated Communication, 2(4), I–II. Retrieved
from http://jcmc.indiana.edu/vol2/issue4/danet.html
Darics, E. (2010). Politeness in computer-mediated discourse of a virtual team.
Journal of Politeness Research. Language, Behaviour, Culture, 6(1), 129–150.
https://doi.org/10.1515/jplr.2010.007
Desai, M., Tsui, K. M., Yanco, H. A., & Uhlik, C. (2011). Essential features of
telepresence robots. In Proceedings of the IEEE International Conference on
Technologies for Practical Robot Applications (TePRA ’11) (pp. 15–20). Los
Alamitos, CA: IEEE.
54 S. C. Herring
Deumert, A., & Masinyana, S. O. (2008). Mobile language choices: The use of
English and isiXhosa in text messages (SMS). Evidence from a bilingual
South African sample. English World-Wide, 29(2), 117–147. https://doi.
org/10.1075/eww.29.2.02deu
Dresner, E., & Herring, S. C. (2010). Functions of the non-verbal in CMC:
Emoticons and illocutionary force. Communication Theory, 20(3), 249–268.
https://doi.org/10.1111/j.1468-2885.2010.01362.x
Dürscheid, C., & Siever, C. M. (2017). Beyond the alphabet – Communication
with emojis. Zeitschrift für germanistische Linguistik, 45(2), 256–285.
Dürscheid, C., & Stark, E. (2011). SMS4science: An international corpus-based
texting project and the specific challenges for multilingual Switzerland. In
C. Thurlow & K. Mroczek (Eds.), Digital discourse: Language in the new
media (pp. 299–320). New York: Oxford University Press.
Emigh, W., & Herring, S. C. (2005). Collaborative authoring on the web: A
genre analysis of online encyclopedias. In Proceedings of the Thirty-Eighth
Hawai’i International Conference on System Sciences (HICSS-38). Los Alamitos:
IEEE Press.
Ferrara, K., Brunner, H., & Whittemore, G. (1991). Interactive written dis-
course as an emergent register. Written Communication, 8(1), 8–34.
Fung, L., & Carter, R. (2007). New varieties, new creativities: ICQ and English-
Cantonese e-discourse. Language and Literature, 16(4), 345–366.
Gao, L. (2006). Language contact and convergence in computer-mediated com-
munication. World English, 25(2), 299–308.
Garau, M., Slater, M., Bee, S., & Sasse, M. A. (2001). The impact of eye gaze on
communication using humanoid avatars. In Proceedings of the SIGCHI
Conference on Human Factors in Computing Systems (pp. 309–316). New York:
ACM.
Garcés-Conejos Blitvich, P. (2010). The YouTubification of politics, impolite-
ness and polarization. In R. Taiwo (Ed.), Handbook of research on discourse
behavior and digital communication: Language structures and social interaction
Garcia, A., Standlee, A. I., Bechkoff, J., & Cui, Y. (2009). Ethnographic approaches
to the internet and computer-mediated communication. Contemporary
Ethnography, 38(1), 52–84. https://doi.org/10.1177/0891241607310839
Garcia, A. C., & Jacobs, J. B. (1998). The interactional organization of com-
puter mediated communication in the college classroom. Qualitative
Sociology, 21(3), 299–317.
Garcia, A. C., & Jacobs, J. B. (1999). The eyes of the beholder: Understanding
the turn taking system in quasi-synchronous computer mediated communi-
cation. Research on Language and Social Interaction, 32(4), 337–367.
Gee, J. P. (2014). Unified discourse analysis: Language, reality, virtual worlds, and
video games. New York: Routledge.
Georgakopoulou, A. (1997). Self-presentation and interactional alliances in
e-mail discourse: The style and code-switches of Greek messages. International
Journal of Applied Linguistics, 7, 141–164.
Georgakopoulou, A. (2011). “On for drinkies?”: Email cues of participant
alignments. Language@Internet, 8, article 4. Retrieved from http://www.
languageatinternet.org/articles/2011/Georgakopoulou
Graham, S. L. (2007). Disagreeing to agree: Conflict, (im)politeness and iden-
tity in a computer-mediated community. Journal of Pragmatics, 39, 742–759.
Greenfield, P. M., & Subrahmanyam, K. (2003). Online discourse in a teen
chatroom: New codes and new modes of coherence in a visual medium.
Journal of Applied Developmental Psychology, 24(6), 713–738. http://dx.doi.
org.proxyiub.uits.iu.edu/10.1016/j.appdev.2003.09.005
Gruber, H. (1998). Computer-mediated communication and scholarly dis-
course: Forms of topic initiation and thematic development. Pragmatics,
8(1), 21–45.
Gruber, H. (2000). Theme and intertextuality in scholarly e-mail messages.
Functions of Language, 7(1), 79–115.
Hafner, K., & Lyon, M. (1996). Where wizards stay up late: The origins of the
internet. New York: Simon & Schuster.
Hall, K. (1996). Cyberfeminism. In S. C. Herring (Ed.), Computer-mediated
communication: Linguistic, social, and cross-cultural perspectives (pp. 147–170).
Amsterdam/Philadelphia: John Benjamins.
Hancock, J., Curry, L. E., Goorha, S., & Woodworth, M. (2008). On lying and
being lied to: A linguistic analysis of deception in computer-mediated com-
munication. Discourse Processes, 45(1), 1–23.
Hardaker, C. (2010). Trolling in asynchronous computer-mediated communi-
cation: From user discussions to academic definitions. Journal of Politeness
Research. Language, Behaviour, Culture, 6(2), 215–242. https://doi.
org/10.1515/jplr.2010.011
Harris, J., Danby, S., Butler, C. W., & Emmison, M. (2012). Extending client-
centered support: Counselors’ proposals to shift from e-mail to telephone
counseling. Text & Talk, 32(1). https://doi.org/10.1515/text-2012-0002
Harrison, S. (1998). E-mail discussions as conversation: Moves and acts in a
sample from a listserv discussion. Linguistik Online, 1(1). Retrieved from
https://bop.unibe.ch/linguistik-online/article/view/1083/1772
56 S. C. Herring
Haugh, M. (2007). The discursive challenge to politeness research: An interac-

tional alternative. Journal of Politeness Research, 3, 295–317.
Herring, S. C. (1993). Gender and democracy in computer-mediated commu-
nication. Electronic Journal of Communication, 3(2). Retrieved from http://
ella.slis.indiana.edu/~herring/ejc.doc
Herring, S. C. (1994). Politeness in computer culture: Why women thank and
men flame. In Cultural performances: Proceedings of the Third Berkeley Women
and Language Conference (pp. 278–294). Berkeley: Berkeley Women and
Language Group. Retrieved from http://ella.slis.indiana.edu/~herring/
politeness.1994.pdf
Herring, S. C. (1996a). Posting in a different voice: Gender and ethics in
computer-mediated communication. In C. Ess (Ed.), Philosophical perspec-
tives on computer-mediated communication (pp. 115–145). Albany, NY:
SUNY Press.
Herring, S. C. (1996b). Two variants of an electronic message schema. In S. C.
Herring (Ed.), Computer-mediated communication: Linguistic, social, and
cross-cultural perspectives (pp. 81–108). Amsterdam: John Benjamins.
Herring, S. C. (Ed.). (1996c). Computer-mediated communication: Linguistic,
social and cross-cultural perspectives. Amsterdam: John Benjamins.
Herring, S. C. (Ed.). (1997). Computer-mediated discourse analysis. Special issue
of the Electronic Journal of Communication, 6(3). Retrieved from http://www.
cios.org/www/ejc/v6n396.htm
Herring, S. C. (1998). Le style du courrier électronique: variabilité et
changement. Revue d’aménagement linguistique (formerly Terminogramme),
84–85(March), 9–16.
Herring, S. C. (1999). Interactional coherence in CMC. Journal of Computer-
Mediated Communication, 4(4). Retrieved from http://jcmc.indiana.edu/
vol4/issue4/herring.html
Herring, S. C. (2001). Computer-mediated discourse. In D. Schiffrin,
D. Tannen, & H. Hamilton (Eds.), The handbook of discourse analysis
(pp. 612–634). Oxford: Blackwell Publishers.
Herring, S. C. (2002). Computer-mediated communication on the internet.
Annual Review of Information Science and Technology, 36, 109–168. Retrieved
from http://ella.slis.indiana.edu/~herring/arist.2002.pdf
Herring, S. C. (2003). Gender and power in online communication. In
J. Holmes & M. Meyerhoff (Eds.), The handbook of language and gender
(pp. 202–228). Oxford: Blackwell Publishers.
Herring, S. C. (2004a). Computer-mediated discourse analysis: An approach to

researching online behavior. In S. A. Barab, R. Kling, & J. H. Gray (Eds.),
Designing for virtual communities in the service of learning (pp. 338–376).
New York: Cambridge University Press.
Herring, S. C. (2004b). Online communication: Through the lens of discourse. In
M. Consalvo, N. Baym, J. Hunsinger, K. B. Jensen, J. Logie, M. Murero, & L. R.
Shade (Eds.), Internet research annual (Vol. 1, pp. 65–76). New York: Peter Lang.
Herring, S. C. (2004c). Slouching toward the ordinary: Current trends in com-
puter-mediated communication. New Media & Society, 6(1), 26–36.
Herring, S. C. (2007). A faceted classification scheme for computer-mediated dis-
course. Language@Internet, 4. Retrieved from http://www.languageatinternet.
org/articles/2007/761
Herring, S. C. (2010). Web content analysis: Expanding the paradigm. In
J. Hunsinger, M. Allen, & L. Klastrup (Eds.), The international handbook of
internet research (pp. 233–249). Berlin: Springer Verlag.
Herring, S. C. (2013a). Discourse in Web 2.0: Familiar, reconfigured, and emer-
gent. In D. Tannen & A. M. Tester (Eds.), Georgetown University Round Table
on languages and linguistics 2011: Discourse 2.0: Language and new media
(pp. 1–25). Washington, DC: Georgetown University Press.
Herring, S. C. (2013b). Relevance in computer-mediated conversation. In S. C.
Herring, D. Stein, & T. Virtanen (Eds.), Handbook of pragmatics of computer-
mediated communication (pp. 245–268). Berlin: de Gruyter.
Herring, S. C. (2015). New frontiers in interactive multimodal communication.
In A. Georgapoulou & T. Spilloti (Eds.), The Routledge handbook of language
and digital communication (pp. 398–402). London: Routledge.
Herring, S. C. (2016). Robot-mediated communication. In R. A. Scott,
M. Buchmann, & S. M. Kosslyn (Eds.), Emerging trends in the social and
behavioral sciences. Hoboken, NJ: John Wiley & Sons. Prepublication ver-
sion. Retrieved from http://ella.slis.indiana.edu/~herring/rmc.pdf
Herring, S. C., & Dainas, A. R. (2017). “Nice picture comment!” Graphicons in
Facebook comment threads. In Proceedings of the Fiftieth Hawai’i International
Conference on System Sciences (HICSS-50). Los Alamitos, CA: IEEE. Retrieved
from http://ella.slis.indiana.edu/~herring/hicss.graphicons.pdf
Herring, S. C., & Demarest, B. (2017). “I’m the first video Voicethread – It’s
pretty sweet, I’m pumped”: Gender and self-expression on a multimodal
interactive website. Apprentissage des langues et systèmes d’information et de
communication, 20(1). https://alsic.revues.org/3007.
58 S. C. Herring
Herring, S. C., & Martinson, A. M. (2004). Assessing gender authenticity in

computer-mediated language use: Evidence from an identity game. Journal of
Language and Social Psychology, 23(4), 424–446.
Herring, S. C., & Paolillo, J. C. (2006). Gender and genre variation in weblogs.
Journal of Sociolinguistics, 10(4), 439–459.
Herring, S. C., & Zelenkauskaite, A. (2009). Symbolic capital in a virtual het-
erosexual market: Abbreviation and insertion in Italian iTV SMS. Written
Herring, S. C., Johnson, D. A., & DiBenedetto, T. (1995). ‘This discussion is
going too far!’ Male resistance to female participation on the Internet. In
M. Bucholtz & K. Hall (Eds.), Gender articulated: Language and the socially
constructed self (pp. 67–96). New York: Routledge.
Herring, S. C., Stein, D., & Virtanen, T. (Eds.). (2013). Handbook of pragmatics
of computer-mediated communication. Berlin: Mouton.
Herring, S. C., Kutz, D. O., Paolillo, J. C., & Zelenkauskaite, A. (2009). Fast
talking, fast shooting: Text chat in an online first-person game. In Proceedings
of the Forty-Second Hawai’i International Conference on System Sciences
(HICSS-42). Los Alamitos, CA: IEEE Press. Preprint. Retrieved from http://
ella.slis.indiana.edu/~herring/hicss.bzflag.pdf
Herring, S. C., Scheidt, L. A., Bonus, S., & Wright, E. (2004). Bridging the gap:
A genre analysis of weblogs. In Proceedings of the 37th Hawai’i International
Conference on System Sciences (HICSS-37). Los Alamitos: IEEE Computer
Society Press. Retrieved from http://ella.slis.indiana.edu/~herring/herring.
scheidt.2004.pdf
Heyd, T. (2013). Email hoaxes. In S. C. Herring, D. Stein, & T. Virtanen (Eds.),
Handbook of pragmatics of computer-mediated communication (pp. 387–410).
Berlin: de Gruyter.
Heyd, T. (2014). Doing race and ethnicity in a digital community: Lexical labels
and narratives of belonging in a Nigerian web forum. Discourse, Context &
Media, 4–5(June–September), 38–47. http://dx.doi.org.proxyiub.uits.iu.
edu/10.1016/j.dcm.2013.11.002
Hinrichs, L. (2006). Codeswitching on the web: English and Jamaican Creole in
e-mail communication. Amsterdam/Philadelphia, PA: John Benjamins.
Hinrichs, L., & White-Sustaíta, J. (2011). Global Englishes and the sociolin-
guistics of spelling: A study of Jamaican blog and email writing. English
World-Wide, 32(1), 46–73.
Hodsdon Champeon, C. B. (2010). Conversations within conversations:
Intertextuality in racially antagonistic dialogue on Usenet. Language@
Internet, 7, article 10. Retrieved from http://www.languageatinternet.org/
articles/2010/2820
Honeycutt, C., & Herring, S. C. (2009). Beyond microblogging: Conversation

and collaboration via Twitter. In Proceedings of the Forty-Second Hawai’i
International Conference on System Sciences (HICSS-42). Los Alamitos, CA:
IEEE Press.
Hubler, M. T., & Bell, D. C. (2003). Computer-mediated humor and ethos:
Exploring threads of constitutive laughter in online communities. Computers
and Composition, 20(3), 277–294.
Huffaker, D., & Calvert, S. (2005). Gender, identity, and language use in teen-
age blogs. Journal of Computer-Mediated Communication, 10(2). https://doi.
org/10.1111/j.1083-6101.2005.tb00238.x
Hughey, M., & Daniels, J. (2013). Racist comments at online news sites: A
methodological dilemma for discourse analysis. Media Culture Society, 35(3),
332–347. https://doi.org/10.1177/0163443712472089
Jenks, C., & Firth, A. (2013). Multi-party voice-based chat room interaction. In
S. C. Herring, D. Stein, & T. Virtanen (Eds.), Handbook of pragmatics of
computer-mediated communication (pp. 217–241). Berlin: de Gruyter.
Jepson, K. (2005). Conversations – and negotiated interaction – in text and
voice chat rooms. Language Learning & Technology, 9(3), 79–98.
Jones, R. H. (2004). The problem of context in computer mediated communica-
tion. In P. LeVine & R. Scollon (Eds.), Discourse and technology: Multimodal
discourse analysis (pp. 20–33). Washington, DC: Georgetown University Press.
Jones, R. H. (2011). C me Sk8: Discourse, technology and “bodies without
organs”. In C. Thurlow & K. Mroczek (Eds.), Digital discourse: Language in
the new media. New York: Oxford University Press.
Joyce, E., & Kraut, R. (2006). Predicting continued participation in news-
groups. Journal of Computer-Mediated Communication, 11(3), 723–747.
https://doi.org/10.1111/j.1083-6101.2006.00033.x
Jucker, A. H. (2002). Hypertextlinguistics: Textuality and typology of hyper-
texts. In A. Fischer, G. Tottie, & H. M. Lehmann (Eds.), Text types and cor-
pora: Studies in honour of Udo Fries (pp. 29–51). Tübingen: Gunter Narr
Verlag.
Jucker, A. H. (2010). “Audacious, brilliant!! What a strike!” Live text commen-
taries on the internet as real-time narratives. In C. Hoffmann (Ed.), Narrative
revisited: Telling a story in the age of new media (pp. 57–77). Amsterdam/
Philadelphia: John Benjamins.
Kapidzic, S., & Herring, S. C. (2011). Gender, communication, and self-
presentation in teen chatrooms revisited: Have patterns changed? Journal
of Computer-Mediated Communication, 17(1), 39–59. https://doi.
org/10.1111/j.1083-6101.2011.01561.x
60 S. C. Herring
Katz, J. K., & Crocker, E. L. (2015). Selfies and photo messaging as visual con-
versation: Reports from the United States, United Kingdom and China.
International Journal of Communication, 9, 2387–2396.
Kiesler, S., Seigel, J., & McGuire, T. W. (1984). Social psychological aspects of
computer-mediated communication. American Psychologist, 39, 1123–
1134.
Ko, K.-K. (1996). Structural characteristics of computer-mediated language: A com-
parative analysis of InterChange discourse. Electronic Journal of Communication,
6(3). Retrieved from http://www.cios.org/www/ejc/v6n396.htm
Koteyko, N., Jaspal, R., & Nerlich, B. (2013). Climate change and ‘climategate’
in online reader comments: A mixed methods study. The Geographical Journal,
179, 74–86. https://doi.org/10.1111/j.1475-4959.2012.00479.x
Kramarae, C., & Taylor, H. J. (1993). Women and men on electronic networks:
A conversation or a monologue? In H. J. Taylor, C. Kramarae, & M. Ebben
(Eds.), Women, information technology, and scholarship (pp. 52–61). Urbana,
IL: Center for Advanced Study.
Kress, G., & van Leeuwen, T. (2001). Multimodal discourse – The modes and
Kushin, M. J., & Kitchener, K. (2009). Getting political on social network sites:
Exploring online political discourse on Facebook. First Monday, 14(11). Retrieved
from http://journals.uic.edu/ojs/index.php/fm/article/view/2645/2350. https://
doi.org/10.5210/fm.v14i11.2645.
Lange, P. G. (2007). Commenting on comments: Investigating responses to antago-
nism on YouTube. Paper presented at the Society for Applied Anthropology
Conference, Tampa, FL. Retrieved from https://www.yumpu.com/en/docu-
ment/view/14888058/commenting-on-comments-investigating-responses-
to-antagonism-
Lee, C. K. M. (2007). Affordances and text-making practices in online
instant messaging. Written Communication, 24(3), 223–249. https://doi.
org/10.1177/0741088307303215
Lee, C. K. M. (2016). Multilingualism online. London: Routledge.
Lee, M. K., & Takayama, L. (2011). “Now, I have a body”: Uses and social
norms for mobile remote presence in the workplace. In Proceedings of CHI
2011 (pp. 33–42). New York: ACM.
Lenihan, A. (2011). “Join our community of translators”: Language ideologies
and Facebook. In C. Thurlow & K. R. Mroczek (Eds.), Digital discourse:
Language in the new media (pp. 48–64). New York: Oxford University Press.
Leppännen, S., Pitkänen-Huhta, A., Piirainen-Marsh, A., Nikula, T., &
Peuronen, S. (2011). Young people’s translocal new media uses: A multi-
perspective analysis of language choice and heteroglossia. Journal of

Computer-Mediated Communication, 14(4), 1080–1107. https://doi.
org/10.1111/j.1083-6101.2009.01482.x
Licoppe, C., & Morel, J. (2012). Video-in-interaction: “Talking heads” and the
multimodal organization of mobile and Skype video calls. Research on
Language and Social Interaction, 45(4), 399–429.
Lim, S. S. (2015). On stickers and communicative fluidity in social media.
Social Media + Society, 1(1). http://journals.sagepub.com.proxyiub.uits.iu.
edu/doi/abs/10.1177/2056305115578137
Ling, R. (2005). The sociolinguistics of SMS: An analysis of SMS use by a ran-
dom sample of Norwegians. In R. Ling & P. E. Pedersen (Eds.), Mobile com-
munications: Re-negotiation of the social sphere (pp. 335–349). London:
Springer-Verlag.
Ling, R., & Baron, N. S. (2007). Text messaging and IM: Linguistic comparison
of American college data. Journal of Language and Social Psychology, 26,
291–298.
Locher, M., & Watts, R. J. (2005). Politeness theory and relational work. Journal
of Politeness Research, 1(1), 9–34.
Marcoccia, M. (2004). On-line polylogues: Conversation structure and partici-
pation framework in internet newsgroups. Journal of Pragmatics, 36(1),
115–145. https://doi.org/10.1016/S0378-2166(03)00038-9
Marcoccia, M., Atifi, H., & Gauducheau, N. (2008). Text-centered versus multi-
modal analysis of instant messaging conversation. Language@Internet, 5, article
7. Retrieved from http://www.languageatinternet.org/articles/2008/1621
Marwick, A. (2013). Online identity. In J. Hartley, J. Burgess, & A. Bruns
(Eds.), Companion to new media dynamics (pp. 355–364). Malden, MA:
Blackwell.
Mauntner, G. (2005). Time to get wired: Using web-based corpora in critical
discourse analysis. Discourse & Society, 16(6), 809–828. Retrieved from
http://www.jstor.org.proxyiub.uits.iu.edu/stable/42888717
McDonald, D. (2007). Visual conversation styles in web communities. In
Proceedings of the Fortieth Hawaii International Conference on System Sciences
(HICSS-40). Los Alamitos, CA: IEEE Press. Retrieved from http://www.pen-
sivepuffin.com/dwmcphd/papers/McDonald.HICSS-40.preprint.pdf
McKinlay, A., Procter, R., Masting, O., Woodburn, R., & Arnott, J. (1994). Studies
of turn-taking in computer-mediated communications. Interacting with
Computers, 6(2), 151–171. https://doi.org/10.1016/0953-5438(94)90022-1
62 S. C. Herring
Milani, T. M. (2013). Are ‘queers’ really ‘queer’? Language, identity and same-
sex desire in a South African online community. Discourse & Society, 24(5),
615–633. https://doi.org/10.1177/0957926513486168
Miller, H., Thebault-Spieker, J., Chang, S., Johnson, L., Terveen, L., & Hecht,
B. (2016). Blissfully happy or ready to fight: Varying interpretations of emoji.
AAAI 2016. Retrieved from http://www-users.cs.umn.edu/~bhecht/publica-
tions/ICWSM2016_emoji.pdf
Mitra, A. (1999). Characteristics of the WWW text: Tracing discursive strate-
gies. Journal of Computer-Mediated Communication, 5(1). https://doi.
org/10.1111/j.1083-6101.1999.tb00330.x
Morand, D. J., & Ocker, R. J. (2003). Politeness theory and computer-medi-
ated communication: A sociolinguistic approach to analyzing relational mes-
sages. In Proceedings of the 36th Annual Hawaii International Conference on
System Sciences (HICSS-36). New York: IEEE Press.
Morel, E., Bucher, C., Pekarek-Doehler, S., & Siebenhaar, B. (2012). SMS com-
munication as plurilingual communication: Hybrid language use as a chal-
lenge for classical code-switching categories. Lingvisticae Investigationes,
35(2), 260–288.
Murray, D. E. (1985). Composition as conversation: The computer terminal as
medium of communication. In L. Odell & D. Goswami (Eds.), Writing in
nonacademic settings (pp. 203–227). New York: Guilford.
Murray, D. E. (1988). The context of oral and written language: A framework
for mode and medium switching. Language in Society, 17, 351–373.
Myers, G. (2010). The discourse of blogs and wikis. London: Continuum.
Nakamura, L. (1995). Race in/for cyberspace: Identity tourism and racial pass-
ing on the Internet. Works and Days, 25/26, 13(1 & 2), 181–193.
Nastri, J., Peña, J., & Hancock, J. T. (2006). The construction of away messages:
A speech act analysis. Journal of Computer-Mediated Communication, 11(4).
https://doi.org/10.1111/j.1083-6101.2006.00306.x
Newon, L. (2011). Multimodal creativity and identities of expertise in the digi-
tal ecology of a world of Warcraft guild. In C. Thurlow & K. R. Mroczek
(Eds.), Digital discourse: Language in the new media (pp. 203–231). New York:
Niederhoffer, K. G., & Pennebaker, J. W. (2002). Linguistic style matching in
social interaction. Journal of Language and Social Psychology, 21(4), 337–360.
https://doi.org/10.1177/026192702237953
Nishimura, Y. (2012). Puns in Japanese computer mediated communication:

Observations from misconversion phenomena. AAAI Technical Report FS-12-
02 Artificial Intelligence of Humor. Retrieved from https://pdfs.semantic-
scholar.org/5678/5929485b9c73e260cb930564d66f964ade45.pdf
Nishimura, Y. (2015). A sociolinguistic analysis of emoticon usage in Japanese
blogs: Variation by age, gender, and topic. In Selected Papers of Internet
Research 16. The 16th Annual Meeting of the Association of Internet Researchers.
Phoenix, AZ, USA, October 21–24.
O’Connail, B., Whittaker, S., & Wilbur, S. (1993). Conversation over video
conference: An evaluation of the spoken aspects of video-mediated commu-
nication. Journal of Human–Computer Interaction, 8(4), 389–428. https://
doi.org/10.1207/s15327051hci0804_4
O’Reilly, T. (2005, September 30). What is Web 2.0? Design patterns and busi-
ness models for the next generation of software. O’Reilly Network website.
Retrieved from http://oreilly.com/web2/archive/what-is-web-20.html
Page, R. (2012). The linguistics of self-branding and micro-celebrity in Twitter:
The role of hashtags. Discourse & Communication, 6(2), 181–201.
Paolillo, J. C. (1996). Language choice on soc.culture.punjab. Electronic Journal
of Communication, 6(3). Retrieved from http://www.cios.org/www/ejc/
v6n396.htm
Paolillo, J. C. (1999). The virtual speech community: Social network and lan-
guage variation on IRC. Journal of Computer-Mediated Communication, 4(4).
Retrieved from http://jcmc.indiana.edu/vol4/issue4/paolillo.html
Pastore, M. (2000, August 10). Women surpass men as US web users. ClickZ.
Retrieved from https://www.clickz.com/women-surpass-men-as-us-web-
users/68756/
Peterson, E. E. (2011). How conversational are weblogs? Language@Internet, 8,
article 8. Retrieved from http://www.languageatinternet.org/articles/2011/
Peterson
Phillips, W. (2015). This is why we can’t have nice things: Mapping the relationship
between online trolling and mainstream culture. Cambridge, MA: MIT Press.
Phillips, W., & Milner, R. M. (2017). The ambivalent internet: Mischief, oddity,
and antagonism online. London: Polity Press.
Pihlaja, S. (2011). Cops, popes, and garbage collectors: Metaphor and antagonism
in an atheist/Christian YouTube video thread. Language@Internet, 8, article 1.
Retrieved from http://www.languageatinternet.org/articles/2011/Pihlaja
Puschmann, C. (2010). The corporate blog as an emerging genre of computer-medi-
ated communication: Features, constraints, discourse situation. Göttingen:
Universitätsverlag Göttingen.
64 S. C. Herring
Radford, W., Chisholm, A., Hachey, B., & Han, B. (2016). [Telephone. Person.
Sailboat. Whale. Okhand] or “Call me Ishmael” – How do you translate emoji?
Cornell University Library. arXiv preprint. https://doi.org/arXiv:1611.02027
Rice, R. E., & Love, G. (1987). Electronic emotion: Socioemotional content in
a computer-mediated communication network. Communication Research,
14(1), 85–108.
Rintel, E. S., Mulholland, J., & Pittam, J. (2001). First things first: Internet
Relay Chat openings. Journal of Computer-Mediated Communication, 6(3).
https://doi.org/10.1111/j.1083-6101.2001.tb00125.x
Rintel, E. S., & Pittam, J. (1997). Strangers in a strange land: Interaction man-
agement on Internet Relay Chat. Human Communication Research, 23(4),
507–534.
Riordan, M. A., Markman, K. M., & Stewart, C. O. (2013). Communication
accommodation in instant messaging: An examination of temporal conver-
gence. Journal of Language and Social Psychology, 32(1), 84–95. https://doi.
org/10.1177/0261927X12462695
Rodino, M. (1997). Breaking out of binaries: Reconceptualizing gender and
its relationship to language in computer-mediated communication. Journal
of Computer-Mediated Communication, 3(3). https://doi.org/10.1111/
j.1083-6101.1997.tb00074.x
Rowe, C. (2011). Whatchanade? Rapid language change in a private email sib-
ling code. Language@Internet, 8, article 6. Retrieved from http://www.lan-
guageatinternet.org/articles/2011/Rowe/index_html#d57e2085
Rowe, I. (2015). Civility 2.0: A comparative analysis of incivility in online polit-
ical discussion. Information, Communication & Society, 18(2), 121–138.
https://doi.org/10.1080/1369118X.2014.940365
Santana, A. D. (2014). Virtuous or vitriolic: The effect of anonymity on civility
in online newspaper reader comment boards. Journalism Practice, 8(1),
18–33. http://www-tandfonline-com.proxyiub.uits.iu.edu/doi/full/10.1080
/17512786.2013.813194
Sargeant, P., & Tagg, C. (Eds.). (2014). The language of social media: Identity and
community on the internet. Basingstoke: Palgrave Macmillan.
Schönfeldt, J., & Golato, A. (2003). Repair in chats: A conversation analytic
approach. Research on Language and Social Interaction, 36(3), 241–284.
Sebba, M. (2003). ‘Will the real impersonator please stand up?’ Language and
identity in the Ali G websites. Arbeiten aus Anglistik and Amerikanistik, 28(2),
279–304.
Severinson Eklundh, K. (1986). Dialogue processes in computer-mediated com-

munication: A study of letters in the COM system. Linköping Studies in Arts
and Sciences 6. University of Linköping.
Shaw, D. (1997). Gay men and computer communication: A discourse of sex
and identity in cyberspace. In S. G. Jones (Ed.), Virtual culture: Identity and
communication in cybersociety (pp. 133–145). London: Sage.
Shortis, T. (2007, June). Gr8 txtpectations: The creativity of text spelling.
English Drama Media. Retrieved from http://nl.ijs.si/janes/wp-content/
uploads/2014/09/shortis07.pdf
Siebenhaar, B. (2006). Code choice and code-switching in Swiss-German
Internet Relay Chat rooms. Journal of Sociolinguistics, 10(4), 481–506.
Simpson, J. (2005). Conversational floors in synchronous text-based CMC
discourse. Discourse Studies, 7(3), 337–361.
Sindoni, M. G. (2014). Through the looking glass: A social semiotic and linguis-
tic perspective on the study of video chats. Text & Talk, 34(3), 325–347.
Spilioti, T. (2009). Graphemic representation of text-messaging: Alphabet-
choice and codeswitches in Greek SMS. Pragmatics, 19(3), 393–412.
Spivack, N. (n.d.). Web 3.0: The third generation web is coming. Lifeboat
Foundation. Retrieved from https://lifeboat.com/ex/web.3.0
Squires, L. (2010). Enregistering internet language. Language in Society, 39,
457–492.
Stockton, N. (2015, June 24). Emoji – Trendy slang or a whole new language? Wired
Magazine. Retrieved from http://www.wired.com/2015/06/emojitrendy-slang-
whole-new-language/
Stommel, W. (2008). Conversation analysis and community of practice as
approaches to studying online community. Language@Internet, 5, article 5.
Retrieved from http://www.languageatinternet.org/articles/2008/1537
Stromer-Galley, J., & Martinson, A. (2009). Coherence in political computer-medi-
ated communication: Analyzing topic relevance and drift in chat. Discourse &
Communication, 3(2), 195–216. https://doi.org/10.1177/1750481309102452
Su, H.-Y. (2003). The multilingual and multi-orthographic Taiwan-based Internet:
Creative uses of writing systems on college-affiliated BBSs. Journal of Computer-
Mediated Communication, 9(1). https://doi.org/10.1111/j.1083-6101.2003.
tb00357.x
Subrahmanyam, K., Smahel, D., & Greenfield, P. M. (2006). Connecting devel-
opmental constructions to the internet: Identity presentation and sexual
exploration in online teen chat rooms. Developmental Psychology, 42(3),
395–406. https://doi.org/10.1037/0012-1649.42.3.395
66 S. C. Herring
Tagg, C. (2012). The discourse of text messaging: Analysis of SMS communication.

London: Continuum.
Tagliamonte, S., & Denis, D. (2008). Linguistic ruin? LOL! Instant messaging
and teen language. American Speech, 83, 3–34.
Themistocleous, C. (2013). Digital code-switching between Cypriot and Standard
Greek: Performance and identity play online. International Journal of
Bilingualism, 19(3), 282–297. https://doi.org/10.1177/1367006913512727
new media. Oxford/New York: Oxford University Press.
Tynes, B., Reynolds, L., & Greenfield, P. M. (2004). Adolescence, race, and
ethnicity on the internet: A comparison of discourse in monitored vs.
unmonitored chat rooms. Journal of Applied Developmental Psychology,
25(6), 667–684. http://dx.doi.org.proxyiub.uits.iu.edu/10.1016/j.appdev.
2004.09.003
Vaisman, C. (2013). Beautiful script, cute spelling and glamorous words: Doing
girlhood through language playfulness on Israeli blogs. Language &
Communication, 34, 69–80. http://dx.doi.org.proxyiub.uits.iu.edu/10.1016/j.
langcom.2013.08.006
van Compernolle, R. A. (2008). Morphosyntactic and phonological constraints
on negative particle variation in French-language chat discourse. Language
Variation and Change, 20(2), 317–339. https://doi-org.proxyiub.uits.iu.
edu/10.1017/S0954394508000112
Voida, A., & Mynatt, E. T. (2005). Six themes of the communicative appropria-
tion of photographic images. In Proceedings of the Twenty-Third Annual ACM
Conference on Human Factors in Computing Systems (CHI’05) (pp. 171–180).
New York: ACM.
Walton, S., & Jaffe, A. (2011). “Stuff white people like”: Stance, class, race, and
internet commentary. In C. Thurlow & K. Mroczek (Eds.), Digital discourse:
Language in the new media (pp. 199–219). Oxford/New York: Oxford
University Press.
Warschauer, M., El Said, G. R., & Zohry, A. G. (2002). Language choice online:
Globalization and identity in Egypt. Journal of Computer-Mediated
Communication, 7(4). https://doi.org/10.1111/j.1083-6101.2002.tb00157.x
Weininger, M. J., & Shield, L. (2004). Proximity and distance: A theoretical
model for the description and analysis of online discourse. In 11th CALL
Conference: CALL & Research Methodologies, Addendum to the Proceedings of
the CALL 2004 Conference. University of Anwerp. Retrieved from http://
webh01.ua.ac.be/didascalia/call_2004.htm
Werry, C. (1996). Linguistic and interactional features of Internet Relay Chat.

In S. C. Herring (Ed.), Computer-mediated communication: Linguistic, social
and cross-cultural perspectives (pp. 47–63). Amsterdam/Philadelphia: John
Benjamins.
Wikipedia. (2016, December). World Wide Web. Retrieved December 6, 2016,
from https://en.wikipedia.org/wiki/World_Wide_Web
Yates, S. J. (1993). Gender, computers and communication: The use of com-
puter-mediated communication on an adult distance education course.
International Journal of Computers in Adult Education and Training, 3(2),
21–40.
Yates, S. J. (1996). Oral and written aspects of computer conferencing. In S. C.
Herring (Ed.), Computer-mediated communication: Linguistic, social and cross-
cultural perspectives (pp. 9–46). Amsterdam/Philadelphia: John Benjamins.
Yee, N., Bailenson, J. N., Urbanek, M., Chang, F., & Merget, D. (2007). The
unbearable likeness of being digital: The persistence of nonverbal social
norms in online virtual environments. Cyberpsychology & Behavior, 10(1),
115–121. https://doi-org.proxyiub.uits.iu.edu/10.1089/cpb.2006.9984
Yus, F. (2010). Cyberpragmatics: Internet-mediated communication in context.
Zappavigna, M. (2012). Discourse of Twitter and social media: How we use lan-
guage to create affiliation on the web. London: Continuum.
Zelenkauskaite, A., & Herring, S. C. (2008). Television-mediated conversation:
Coherence in Italian iTV SMS chat. In Proceedings of the Forty-First Hawai’i
IEEE Press.
Part III
Multimodality
3
“Of course I’m married!”
Communicative Strategies
and Transcription-Related Issues
in Video-Mediated Interactions
Maria Grazia Sindoni
3.1 Introduction
In the last twenty-five years, technological advances have made the
unprecedented development of video-mediated communication (VMC)
possible. More recently, VMC has expanded thanks to the advent of inex-
pensive, flat-rate internet and phone plans, increased bandwidth, wide
availability of low-priced devices, such as tablets and smartphones, and
apps for video communication, such as Skype, MSN, Facetime, WhatsApp
and Google Hangouts.
The somewhat unjustified claim that anything on the internet is “new”
does not account for the actual epistemological and socio-semiotic differ-
ences in web-based texts. The distinction between native and non-native
digital genres has been, for example, recognized by the tripartition devised
by Gardner and Alsop (2016), who categorize (1) texts that are born digital
M. G. Sindoni (*)
Department Civiltà Antiche e Moderne, University of Messina, Messina, Italy
e-mail: mgsindoni@unime.it

https://doi.org/10.1007/978-3-319-92663-6_3
72 M. G. Sindoni
(e.g. fandom blogs; see Sindoni, 2016), (2) texts that achieve digitality
(e.g. recontextualized professional genres, such as online medical texts;
see Bloor, 2016; or online university lectures, see Karagevreki, 2016) and
(3) texts that have digitality thrust upon them (e.g. verbal art; see Miller,
2016). The much-heralded claim of the newness of computer-mediated
communication, especially within the domain of the so-called Web 2.0
digital genres, was likewise challenged by Herring at GURT 2011 (see
Herring, 2013) by pointing out the epistemological and heuristic differ-
ences between what she defined as “new”, “emergent” and “reconfigured”
digital texts and genres. Computer-mediated communication scholars
have thus cautioned against simplistic labelling and the automatic identi-
fication of anything digital with something straightforwardly “new”.
However, the use of adjectives such as “unprecedented” and “unparal-
leled” mentioned above is appropriate in the context of VMC. A wide
range of research areas have dealt with developing very delicate systems
for the description, classification and analysis of VMC, but it is true that
face-to-face interactions—happening in real time but with participants
in different places—are a novelty in human communication. Sharing the
same context (see Halliday, 1978; Malinowski, 1923) used to be the pre-
condition of non-mediated face-to-face interaction, but today VMC has
opened up the way to new and unpredictable patterns of communication
that cannot be merely explained in terms of medium affordances.
Questions such as the use of speech, writing and static and moving images
during a communicative event still need to be charted. Furthermore, the
co-deployment of different semiotic resources, their amount and prefer-
ential use, or questions such as embodiment/disembodiment and play-
fulness/performance, to name but a few, are continuously evolving in
such environments, calling for a profound rethinking of traditional cat-
egories for the study of spontaneous interaction.
Synchronous VMC is widespread: from mundane talk between friends
and partners to teleconference corporate business meetings. Academic
environments are no exception. Within them, video interactions can
occur in university/college staff meetings or learning/teaching contexts,
where web-based exchanges are institutionalized sites of learning.
Research into personal interactions can likewise have a variety of applica-
tions, for example, in the context of (extended) families keeping in touch
“Of course I’m married!” Communicative Strategies… 73
across long distances (see Norris, 2004) or English as a Foreign Language

(EFL) learners communicating with their peers (Sindoni, 2014b).
This chapter will explore possible applications and uses of multimodal
transcription and annotation of VMC for educational purposes, with the
specific reference to how students’ transcription of video-mediated con-
versations can reveal their ideologies on language and communication in
general and, in particular, showing that in transcriptions: (1) verbal lan-
guage is prioritized over other non-verbal resources, (2) the chaotic nature
of speech is normalized and, consequently, (3) written norms have a pow-
erful influence on students’ learning experiences. In Sect. 3.2, research
literature is reviewed and research questions are presented; in Sect. 3.3,
the method is illustrated with a view of how video data have been col-
lected. In Sects. 3.4 and 3.5, several examples are discussed regarding a
student’s transcription of a Skype conversation compared to a transcrip-
tion by a multimodal analysist (the author of this study), focusing on the
resource of spoken language and on other semiotic resources (kinesics,
gaze, proxemics), respectively, with additional comments on the student’s
written assignment about his transcription experience. In Sect. 3.5, some
concluding remarks will be outlined with the aim of endorsing the prac-
tice of manual multimodal transcription in contexts of language learn-
ing/teaching as a tool to explore students’ ideas and ideologies on language
representation.
3.2 R
eview of Research Literature
and Research Questions
The study of online social interactions has been mainly grounded on (1)
language-based methodologies and (2) social-ethno-anthropological
methodologies.
The former include linguistics (Crystal, 2006; Georgakopoulou, 2011;
Zappavigna, 2012), sociolinguistics (Androutsopoulos, 2006; Thurlow
& Mroczek, 2011), pragmatics (Herring, Stein, & Virtanen, 2013; Yus,
2011), discourse analysis (Herring, 2004a, b; Myers, 2010), conversation
analysis (Sindoni, 2014a, 2013) and discourse analysis applied to social
74 M. G. Sindoni
interaction (Bou-Franch, Lorenzo-Dus, & Garcés-Conejos Blitvich,

2012), whereas the latter have been informed by interactional sociology
(Goffman, 1981), micro-ethnography (Erickson, 2004) and/or develop
out of research into discursive psychology and health studies (Antaki,
Ardévol, Núñez, & Vayreda, 2005; Flinkfeldt, 2011, 2014; Sneijder & te
Molder, 2004, 2005). Other disciplines take a broader approach, thus
blending the former with the latter, as is the case of studies on naturally
occurring video-based interactions informed by interactional sociolin-
guistics (Gumperz, 1999), linguistic anthropology (Duranti, 1997) and
linguistic ethnography (Creese, 2008).
Studies on VMC have explored mediation within the framework of
mediated discourse analysis (Scollon, 2002), while other studies have
investigated the roles and constraints of technological affordances
(Hutchby, 2001). Other approaches have reflected on digital environ-
ments to grasp how multimodal loci of media convergence and hetero-
glossia can be configured (Androutsopoulos, 2011; Herring, 2011).
Early research on VMC focused on task-oriented interactions in the
workplace (Heath & Luff, 1992) and on educational contexts (Swan
et al., 2008). More recently, however, as mentioned in the previous sec-
tion, more intimate and personal media spaces in the context of one-to-
one and multiparty interactions have been analysed (Sindoni, 2014a). In
the context of family and social VMC, studies have also tended to incor-
porate socio-demographic data, as in studies of teenagers’ use of synchro-
nous video chat (Sindoni, 2011, 2012b). Other researchers investigate
the relationship between VMC and performance to understand how
social behaviour can be interpreted in digital environments (see the
notion of cyberstage, Papagiannouli, 2011), how performance and play-
fulness interact in CMC (Danet, 2001), social presence and the VMC-
distinctive characteristics of focused and unfocused interactions (informed
by Goffman, 1981), and multiple engagements in VMC contexts
(Rosenbaun, Rafaeli, & Kurzon, 2016a, b).
The specific contribution that multimodal studies and socio-semiotic
approaches to communication have brought about in VMC research is
first and foremost epistemological and can be thought of as a twofold
process: (1) de-emphasizing the prominence of verbal language and (2)
addressing the issues of transcription and annotation in video analysis.
Regarding (1), within the agenda of sociolinguistic disciplines men-

tioned above, there seems to have been a tacit, general agreement about
the well-established division between what was labelled as language (i.e.
verbal language) and non-verbal language or paralanguage (e.g. proxemics
and kinesics), which was thought to carry secondary or additional mean-
ings to a core message conveyed by language (Bezemer & Jewitt, 2010).
In this light, a merit of socio-semiotic and multimodal studies was to
contribute to a process where language was starting to be considered as
one resource among others, on a par with other meaning-making
resources, such as proxemics, kinesics, gaze and so on. However, what was
still lacking was a systematic description of all these resources from a
comprehensive standpoint: for example, language was described in lin-
guistic terms, whereas gaze management and movements were explained
by psychology, or psycholinguistics, but mostly in derivational terms,
that is, as deriving from a “superior” discipline, in this case linguistics.
This unifying lens has been provided by multimodal approaches, accord-
ing to which all the resources that come into play in CMC in general, and
VMC in particular, should be described with no aprioristic system of
priorities or with no rigid logocentric taxonomies.
Historically, the socio-semiotic and multimodal approach to commu-
nication is mainly informed by Halliday’s theory of social semiotics,
(Halliday, 1973, 1978) but also draws upon other theoretical frameworks
of visual design of static and moving images, displayed art, music and so
on (see Kress & van Leeuwen, 2001, 2006; O’Toole, 2011; van Leeuwen,
1999, 2005). A social-semiotic and multimodal approach sees communi-
cation as a co-shared and systematic co-deployment of resources, such as
language, image, music, kinesic and proxemics patterns. Such resources
are orchestrated by participants in specific contexts of situation that are
informed by context of culture in meaningful and principled patterns to
produce meanings (Halliday, 1978; Malinowski, 1923). Meaning-
making is thus co-constructed and co-shared and can be understood by
unveiling those patterns that can be likewise called “grammars” on a par
with the grammar of language (Kress & van Leeuwen, 2006).
The theoretical link with transcription and annotation issues is
important, as transcription and annotation are heuristics to help unveil
such patterns, in other words, to describe the grammars that go beyond
76 M. G. Sindoni
language, for example, in the context of VMC, the grammars of kine-

sics, gaze, proxemics patterns and so on. Furthermore, transcription
and annotation in multimodal analyses reflect epistemologies, research
agendas and priorities in VMC. Issues for analysis and interpretation of
video data have been discussed from different theoretical perspectives
(Flewitt, Hampel, Hauck, & Lancaster, 2009; Heath & Luff, 2000;
Norris, 2004; Pink, 2007; Thibault, 2000). Visual ethnographic and
multimodal approaches have been elaborated within academic com-
munities to, among other ends, help analysts in setting research goals.
However, the need and development of “multiliteracy” skills in the
digital age has broadened the previously narrower focus of analysis. The
concept of “multiliteracies” was designed by the group of scholars
known as the New London Group. They set out to devise a pedagogical
toolkit to respond to “the increasing multiplicity and integration of
significant modes of meaning-making, where the textual is also related
to the visual, the audio, the spatial, the behavioral, and so on” (The
New London Group, 1996, p. 64).
Seen under this light, transcription and annotation can potentially
work in two different directions: (1) they can help frame research ques-
tions, priorities and goals in the context of academic communities and
(2) they can orient quality VMC-based study in educational contexts,
with the aim of framing students’ multiliteracies needs and demands.
Thus, the research questions addressed in this chapter are as follows:
• What is students’ perception and understanding of both video-

mediated interactions and their related transcription, considered as
forms of representation of interactions?
• What kinds of bias, if any, are apparent in their transcriptions and
related comments?
To this end, I will discuss a case study of a two-party Skype interaction

between an American student and his Japanese friend. The interaction
was recorded, partially transcribed and commented on by the student
within a research project based on training EFL learners in multimodal
theory and methods, with a specific focus on transcription and annota-
tion. By building upon previous research (Sindoni, 2011, 2012a, b,
2013, 2014a), this chapter will thus compare a student’s transcription of

a two-party Skype interaction to another one, developed by a trained
multimodal analyst (the author of this study). The aim is not to decide
which transcription is better, as the research goal of the project was not
to develop students’ fine-grained skills in multimodal transcription, but
to discern student’s ideologies in making sense of video data and of sev-
eral semiotic resources, such as specific language uses (e.g. mode-switch-
ing, code-switching, turn-taking management and repairs), proxemics,
kinesics and gaze patterns.
3.3 M
ethod and Video Data
The data sets discussed in this chapter have been extracted from a research
project, “MoM. Multimodality on the Move”, carried out at the Italian
national level, with the Universities of Messina (Foreign Languages and
Literatures, second-level degree course), Chieti-Pescara (Foreign Languages
and Literatures, second-level degree course), Rome Tor Vergata (Languages
in the Information Society, first-level degree course) and Florence (Public
and Political Communication Strategies, second-level degree course), in
the 2014–2015 academic year, followed by a pilot project carried out at the
University of Messina in the 2009–2010 academic year (Sindoni, 2014b).
The students, attending one of the universities’ curricular programmes,
were grouped into four cohorts. Students participated in the project by
attending a joint curricular programme in their universities. Students in
the four different universities attended the same classes as the researchers
repeated the same lectures in the four universities: hence, they had the
same classes in their home universities, while the researchers moved to
the four different universities to give the same classes/workshops to dif-
ferent students. The programme included core readings on multimodal
theories and secondary readings on different digital texts. It also included
both regular lectures and workshop sessions with tutorials and practical
learning on how to design a digital text, such as a blog or a website, or
with multimodal training, such as with classes and tutorials on how to
carry out transcription and annotation of video data.
78 M. G. Sindoni
The programme was centred on the critical study of multimodality and

digital texts, and it was designed by the four researchers at the four univer-
sities. However, with the aim of developing multiliteracy skills, students
were not asked to study and comment on ready-made texts, but were
prompted to devise and design a digital text by selecting one from four
digital genres developed by researchers, namely (1) video chats (author,
University of Messina), (2) fanvids (Ilaria Moschini, University of
Florence), (3) blogs (Elisabetta Adami, University of Chieti-Pescara) and
(4) “about us” webpages (Sandra Petroni, University of Rome, Tor Vergata).
At the workshops held by the four researchers, students learned the
same contents and practised with the same learning materials (reading
lists, guidelines, etc.), even though they were in their physical classrooms,
located in different places, each attending their regular course. Researchers
repeated their workshops four times in the four universities, providing
students with instructions (guidelines, tutorials, practical workshops) to
design a digital text (e.g. to design a blog or to record a Skype video con-
versation) and then to critically reflect on their own productions. Critical
reflections had to be shown via a written assignment in the form of a
short academic paper discussing multimodal theory and practice, thus
applied to the specific multimodal text chosen by the student. To develop
receptive and productive skills, together with the enhancement of assess-
ing abilities, students were thus asked to (1) produce their own digital
texts among a choice of the four digital genres mentioned above, (2) write
a short academic paper discussing their linguistic and semiotic choices in
text production and, finally, (3) anonymously evaluate one of their fellow
students from another university on a common peer assessment grid.
Discussing all the findings would go beyond the scope of this chapter, so
we will here focus exclusively on the production of video-mediated inter-
actions, namely video chats.
In my workshop on video chats, students were prompted to (1) screen
record a video conversation with an English-speaking friend, (2) analyse
it by discussing the resources presented at the workshop and (3) select a
short chunk of video interaction and transcribe it by using a time-based
grid with columns designed to report on the use, for each participant
involved in interaction, of the relevant semiotic resources, such as lan-
guage (both spoken and written), mode-switching (switch from speech
to writing and vice versa; see Sindoni, 2013), use of gaze, kinesic action
(i.e. movements) and staged proxemics (i.e. how posture and proximity
to the webcam creates a sense of virtual social distance in interaction).
This chapter will thus consider the materials that were produced during
this project, with the caveat that they are only to some extent spontaneous,
as in the interactions students were partially aware of the research agenda,
that is, they were working as “student collaborators” (SCs henceforth), as
discussed elsewhere (Sindoni, 2012b). However, as will be shown, one
interlocutor per interaction was unaware of the research questions investi-
gated during live video conversations by SCs, thus partly preserving the
spontaneity of conversational exchanges. Furthermore, as will be discussed
in Sects. 3.4 and 3.5, some interactional patterns are very difficult to self-
monitor; hence, SCs provided data as well, for example, in terms of kinesics
and gaze. However, when it comes to other patterns, such as mode-switch-
ing, SCs were not only aware of how it works, but also very likely to initiate
it, in order to have the chance to describe it in their analyses.
The complete data in this study comprises twenty-four different video-
based interactions via Skype, involving forty-eight primary participants
(age range 20–55). Secondary participants are those who joined the con-
versation going on between two primary participants, that is, SC plus
one partner selected beforehand. Secondary participants were mostly
casual appearances onscreen, but contributed to enrich the picture in
terms of overall involvement.
MoM participants signed a consent form, thus allowing a full and fine-
grained study of all moves, turns and interactional patterns. The differ-
ences in these data sets confirm the idea that no single unique interpretative
heuristics can be used when it comes to video-based spontaneous
interactions.
3.4 E
xamples and Discussion of Verbal Data
This section focuses on a case study of one single two-party video interac-
tion, transcribed by SC John and then by the author of this chapter. The
two-party interaction involves SC John, an American student of Italian
ancestry in Pescara, and Nobita, John’s Japanese friend.1 The two tran-
80 M. G. Sindoni
scriptions will be compared to illustrate the student’s priorities in tran-

scription and a further step will illustrate how SC John interprets video
data in his written assignment. In this section, we will focus exclusively
on the transcription of verbal data to see how John, who has been previ-
ously instructed in multimodal theory and in the practice of manual tran-
scription and annotation with the provided grid, perceives, understands
and represents interaction. Some examples will show how students represent
interaction in their transcriptions, because multimodal transcription is a
visual method to represent how interaction develops in time with the aim
of observing it in all its detail (e.g. transcription reports on language, gaze
and movements). In other words, the grid provided requires that students
transcribe the video-recorded interaction by focusing on each resource
involved in interaction, including, but not limited to, language.
The analysis of verbal data and of non-verbal data (e.g. gaze and move-
ment) has been split into Sects. 3.4 and 3.5 to focus separately on the dif-
ferent resources and not with the implied agenda of prioritizing language.
In this section, the transcriptions of the author of this chapter only com-
prise verbal language, although some additional visual information has
been used to contextualize verbal exchanges. More full-length comments
on the use of the other resources will be provided in Sect. 3.5. In Appendix,
transcription conventions for spoken turns and for fillers are reported.
Example 1
1. John: I’m studying in in California and went to San Francisco this

summer…
2. Nobita: ooh ((nods)) nice.. very [nice] ((smiles and nods))*
3. John: [yeah..] very nice place ((smiles)) .. aand and I though I’m goin’..
aah to .. this place
((writes www.middlebury.edu))
4. Nobita: what’s that// ((leans forward to the screen))*
5. John: this one.. uhm

6. Nobita: mmm … I can’t see.. what is it?
7. John: it’s a.. ah – it’s a college
8. Nobita: ah ((nods)) ok
9. John: it’s a college – university / and they they got a good Italian
program – Master’s / and hopefully I’ll be teaching Italian very soon//
((shrugs))
10. Nobita: ((nods)) nice nice nice so – yeah
11. John: what about you / what are you ((smiles)) doing?
12. Nobita: I’m ah – I’m a news reporter ((smiles)) *
13. John: ah-ah

14. Nobita: an covering .. / economy an .. politics
15. John: [really?]
16. Nobita: [so I… ] yeah .. and and in Japanese politics .. ((looks away
and then back toward camera)) we – we have the …b(ig) *
1 7. John: the ele[ctions], right?

18. Nobita: [the ele] yeah the election for a [lower house] ..
82 M. G. Sindoni
1 9. John: [I’m going..] I I like to be informed.. so Shinzo Abe, right?

20. Nobita: yeah yeah.. I know.. yeah ((laughs))
21. John: so what [do you think?]
22. Nobita: [so we] y’know I had ah a lot of .. busy days – these days and
ah I uhm / yeah I have many things to do … so / for example
tomor[row and] the day after tomorrow
23. John: [a-ha]
24. Nobita: I’ve to go work
25. John: REALLY?
26. Nobita: mmm ((nods then half smiles))
27. John: so you have no Christmas – what about Christmas in Japan?
Do you have like holidays / and ..
28. Nobita: no not official holidays ah – we celebrate Christmas but /
we’re not that religious ((laughs))
29. John: oh really ((smiles))
30. Nobita: so we we just imitate the American style .. *
Example 2
1. John: so what about your family / are you – mmm are you married?
Do you have a ..
2. Nobita: oh yeah of course I’m married! ((smiles)) *
3. John: yeah .. I don’t know you’know man ..

4. Nobita: d-d-do you wanna see my wife? D’you wanna see my wife?
[She’s working on ..]
5. John: [oh if she wants ..]

6. Nobita: yeah ah like – but she’s wearing pyjamas now / ((moves the
webcam towards wife)) *
7. John: no no [that’s ok] ..

8. Nobita: [take a look] … ciao ciao ciao – CIAO!
9. Shikuza: CIAO! ((smiles and waves her hand))
10. John: Ciao! ((laughs))
11. Shizuka: Ciao! ((laughs))
12. John: Watashi wa John des!
13. Shizuka: OOH! ((laughs)) Hello! *
1 4. John: HELLO! ((all laughing))

15. Nobita: ((moves camera so as to capture his image)) so / this is my
family *
Examples 1 and 2 have been transcribed by the author applying basic

criteria for transcription in the context of naturally occurring spoken
conversation as reported on Appendix. Students participating in the proj-
ect, however, did not have to follow such conventions as the competence
in conversation analysis transcriptions went beyond students’ background
84 M. G. Sindoni
and the project’s goals. However, these conventions have been here
applied to show more details about the conversation between John and
Nobita, and the occasional participation—in Example 2—of a former
unratified and secondary participant, that is Shizuka, Nobita’s wife.
Even though SC John did not follow transcription conventions, that
were not part of his assignment, he had been instructed to faithfully tran-
scribe all turns and moves of the video conversation, without omission or
editing. However, John repeatedly decided to edit the conversation, both
to summarize and correct it. For example, fillers, false starts, repetitions
and disfluencies are systematically omitted in John’s transcription. John is
normalizing his transcription, as the following excerpt from Example 1,
turn 9, shows:
My transcription
9. John: it’s a college – university / and they they got a good Italian pro-
gram – Master’s / and hopefully I’ll be teaching Italian very soon//
((shrugs))
SC transcription
As mentioned, I am not trying to decide which transcription is better,

as we were not interested in developing students’ fine-grained transcrip-
tion of spoken turns. However, in his transcription, John deletes his self-
correction (“college—university”), changes what he probably perceives as
a non-standard (and, as such, stigmatized) variation of lexical “have”, that
is “got”, into the more acceptable variant, that is “have”, avoids the use of
contraction “’ll” and standardizes the temporal operator “will” by spelling

it out when in fact in speech—that must be even clearer to a native
American speaker—the full pronunciation of “will” in informal contexts
would have been avoided.
In Example 1, Nobita is describing his job, in particular the topics he
usually covers in news reporting:
My transcription:
1 2. Nobita: I’m ah – I’m a news reporter ((smiles))

13. John: ah-ah
14. Nobita: an covering .. / economy an .. politics
15. John: [really?]
16. Nobita: [so I… ] yeah .. and and in Japanese politics .. ((looks away
and then back toward camera)) we – we have the …b(ig)
17. John: the ele[ctions], right?
18. Nobita: [the ele] yeah the election for a [lower house] ..
19. John: [I’m going..] I I like to be informed.. so Shinzo Abe, right?
SC transcription:
In the excerpt above, John has again completely normalized what

Nobita was saying, to the point of summarizing and/or omitting turns.
86 M. G. Sindoni
The several examples of repetition, hesitations and fillers—that show

Nobita’s effort to explain himself to his friend—are omitted in John’s
transcription, and a discourse marker is substituted by an exclamation.
When John does not hear or understand, he decides to guess without
indicating he is doing so. No attempt at reproducing overlapping is made
by John, but he very carefully reports on the time of Nobita’s turn in the
grid. In Example 1, lines 22–24, similar strategies are adopted by John:
My transcription:
22. Nobita: [so we] y’know I had ah a lot of .. busy days – these days and
ah I uhm / yeah I have many things to do … so / for example
tomor[row and] the day after tomorrow
23. John: [a-ha]
24. Nobita: I’ve to go work
SC transcription
John again omits all disfluencies and fillers, but this time he also adds
connectives (“so”) and consistently uses punctuation to impose order
upon spoken turns. He also avoids contracting the auxiliary verb “have”
and omits the transcription of his own turn (line 23 in my transcription)
that overlaps with Nobita’s turn.
In Example 2, John likewise adopts similar strategies, for example,
reducing the number of turns, as the complex and overlapping greetings
exchanged with Nobita’s wife are summarized by attributing a single
greeting to each participant: Nobita suggesting “ciao” as the Italian greet-
ing to his wife Shizuka, John saying “watashi wa John des” (i.e. “My
name is John” in Japanese, with “des” instead of “desu”) and Shizuka first
repeating “ciao” after her husband’s prompt and later “hello”. However,
as can be gathered from lines 8–15, eight turns are reduced to six turns,
and no mention of the relevant linguistic phenomenon of code-switching
is made (Heller, 1988), when in fact four code-switches occur, involving
Italian (John is of Italian origin), Japanese and English. In the video-
recording, furthermore, all names are beeped to protect the participant’s
identities, whereas in both transcription and analysis, all real names are
used, including Nobita’s baby daughter, whose name is also present in all
screenshots in John’s transcription.
My transcription
8. Nobita: [take a look] … ciao ciao ciao – CIAO!

9. Shikuza: CIAO! ((smiles and waves her hand))
10. John: Ciao! ((laughs))
11. Shizuka: Ciao! ((laughs))
12. John: Watashi wa John des!
13. Shizuka: OOH! ((laughs)) Hello!
14. John: HELLO! ((all laughing))
15. Nobita: ((moves camera so at to capture his image)) so / this is my
family
SC transcription
88 M. G. Sindoni
As is evident from the excerpt above, John summarizes and normalizes

also by means of punctuation, that is, in some cases completely irrelevant
(see full stop at the end of turns). Exclamations and overlapped turns are
likewise omitted, even though they constitute the most salient part of this
stretch of conversation.
In his assignment, a short academic paper on the recorded interaction,
John discusses at some length mode-switching, one of the most frequently
covered topics during the workshops on video transcriptions and annota-
tions. Patterns of mode-switching are illustrated by drawing upon models
of conversation analysis (Sacks, 1992; Sacks, Schegloff, & Jefferson,
1974), in particular discussing and giving examples of self-initiated and
other-initiated mode-switching (Sindoni, 2013, 2014a). The former refers
to change from speech to writing and vice versa, and they are performed
by participants on their own initiative, whereas the latter refers to changes
from speech to writing and vice versa that are prompted by another par-
ticipant. Mode-switching can be performed not only for repairing trouble
in conversation (e.g. Schegloff, Jefferson, & Sacks, 1977), but also to add
more precise information, among other reasons (Sindoni, 2011, 2012a).
In Example 1, lines 3–5, one modeswitch from speech to writing is
self-initiated by John, who typed the URL address of the college he has
enrolled in to show Nobita the place. John explains in the following
(unedited) terms his self-initiated modeswitch:
I use text first because I want to be sure that my interlocutor understands

what I am referring to, as I don’t believe he has ever heard the name I men-
tion before. Second, I know that one of the affordances of skype is to rec-
ognize automatically internet links. Doing so I give the chance to my
interlocutor to quickly open the site I am mentioning and have a general
idea of what I am talking about.
He articulates his modeswitch as a way to (1) provide a precise location

and (2) show he is aware of Skype’s affordances, that is, to give the chance
to his interlocutor to semiotically operationalize the suggested option,
thus, in other words, to suggest that Nobita directly click on the link.
However, Nobita not only does not respond to John’s prompt (e.g. Nobita
does not perform any other-initiated modeswitch), but he also shows he
does not understand what he is supposed to do. John explains in the fol-
lowing terms this state of affairs:
When I switched mode, my interlocutor didn’t actually receive any text

notification. The long silence that came right after my typing was an indi-
cation that the interlocutor didn’t have a clue of what I was doing and was
still expecting a follow-up to my deictic statement. [sic]
John realizes that Nobita was using his smartphone and knowing that the
Skype mobile app does not feature any typing indicator, he correctly
understands that Nobita has not read the text message. In fact, Nobita
looks perplexed and even asks “What is that?”, confirming the meaning
of his silence. Interestingly, John argues, “Had I been in an in persona
[sic] conversation with my interlocutor, I doubt I would have given him
this website address”, showing that he is perfectly able to interpret that
the technical possibility of using writing in a video conversation does not
necessarily equate with the semiotic choice of doing so. In the next section,
some further comments on other semiotic resources will be provided.
3.5 E
xamples and Discussion of Non-verbal
Data (Kinesics, Gaze and Proxemics)
This section discusses the use of semiotic resources other than language to
show the student’s understanding of the specific contribution of each
resource to overall meaning-making in conversation. In particular, com-
ments on kinesics and gaze (i.e. use of bodily and eye movements in
online conversations), and proxemics (i.e. use of web-mediated social dis-
tance), will be analysed with a view to understanding how SC John makes
sense of his video data.
Kinesics includes movements that range from fully intentional, such as
deictic gestures, for example, pointing, to unintentional, such as habitual
posture of participants (Norris, 2004). Explorations are usually focused
on head and hand movements, but also on the meaning-making of ges-
tures to control the “behaviour of others, to manipulate the persons in
the environment” (Halliday, 1973, p. 31). The default configuration has
90 M. G. Sindoni
been described as “talking heads”, in which participants “show as much

of their face as possible” (Licoppe & Morel, 2014, p. 4).
Kinesics is also important to provide additional meanings and unspo-
ken thoughts (see Goldin-Meadow, 1999). Technical affordances have
some constraints: the camera typically frames participants in medium
shot, that is, with visible head and torso. Figure 3.1 shows an intentional
movement that has precise on-set and off-set moments that can be tracked
by a linear and time-based model (Sindoni, 2013). Screenshots illustrate
the movements, gaze and proxemics distance that Nobita projects through
the use of the embedded camera in his smartphone. As is evident in the
whole bulk of the data sets used for this study, all recorded interactions
can be seen and analysed only from one perspective, that of SCs’. In other
words, all video-recordings have been screen-captured only by SCs and
not by the other interlocutors. Licoppe and Morel (2014) explain this
state of affairs in terms of the orchestration of “video-in-interaction” and
“talk-in-interaction”, for example distinguishing, among others, between
the “show-er” (in this case, Nobita, as we will discuss below) and the
“viewer” (i.e. John). Paraphrasing Licoppe and Morel, we could also dis-
tinguish between the shown and the showing: the former is the participant
who is called and is thus seen in close shot, whereas the latter is the par-
ticipant who is calling (and recording) the video interaction (the icon is
smaller and can be minimized by the showing).
Conversely, the possibility to screen capture both perspectives, that is,
the ones produced by each participant in a one-to-one video interaction,
Fig. 3.1 Nobita is waving his hands (intentional movement). Nobita is the shown,
and John is the showing
or to screen capture many perspectives, that is, corresponding to the ones

produced by each participant in a multiparty conversation, would be
immensely useful for more fine-grained and delicate analyses. However,
participants in the MoM project who were not students (such as Nobita,
who was involved by SC John) would have had to spend much more time
and effort had they been asked to videorecord the interaction on their
end as well.
Going back to John and Nobita’s interaction, John argues that kine-
sics, distance and use of gaze are rather static and predictable, attributing
this staticity and fixity of posture to Japanese cultural norms. A careful
observation of the video conversation shows that—consistently with
John’s arguments—Nobita’s behaviour is in effect rather rigid, even
though, quite strikingly, he decides at some point to move his smart-
phone so as to capture his wife in pyjamas. The camera movement is
rapid, and he does not listen to John’s tepid agreement or attempt at
minimizing the social imposition (e.g. “if she wants…”) followed by an
embarrassed refusal (“no no no”) so as to capture Shizuka’s personal/inti-
mate space. Nobita is in fact departing from the default “talking heads”
format to be in line with his interactional purpose (see Licoppe & Morel,
2012, 2014). By performing this semiotic action, Nobita is responsible
for this choice—a marked choice in terms of “gazeworthiness” as described
by Licoppe and Morel (2014). By doing so, he is initiating a “showing
sequence”, thus playing the role of show-er (Licoppe & Morel, 2014). As
discussed above, the show-er (Nobita, who moves the camera to show a
gazeworthy participant, his wife, see Fig. 3.2) and the viewer (John) do
not have access to the same set of resources (Licoppe & Morel, 2014,
p. 22), as the viewer, in this case John, has access to the speech turns and
video shots shown by Nobita. Conversely, Nobita has access to John’s
speech turns, but not to video shots, as he is moving the mobile camera
in order to show his wife. The semiotic action of showing by moving the
hand-held device produces the consequence of momentarily preventing
Nobita from viewing John. None of these complex orchestrations of
space and set of semiotic resources is recognized in John’s transcription.
When it comes to social distance in synchronous online interactions, I
have argued elsewhere that it is more appropriate to define it as staged
proxemics, as this notion does not refer to physical distance, but to the
92 M. G. Sindoni
Fig. 3.2 Shizuka’s personal space
distance that is staged through conscious or unconscious webcam posi-

tioning (see Sindoni, 2013). Staged proxemics also refers to configura-
tions of spatial relationships and of timing in face-to-face conversation,
including the study of interpersonal distance (Erickson, 2004; Hall,
1966; Hall & Hall, 1990). In particular, the label “staged” refers to the
fact that participants may consciously or unconsciously change webcam
position to frame different portions of themselves or of the setting/envi-
ronment. Proxemics is scrutinized considering that distance among par-
ticipants is framed via semiotic choices: in this case, Nobita decides to
introduce his wife to John, and he does so by moving his smartphone
(and camera), thus staging proxemics, that is orchestrating (virtual) space
for the so-far-unratified secondary participant Shizuka. He does so by
adjusting camera position to construct a sense of physical space/social
environment that is projected onto the screen for John (see Licoppe &
Morel, 2014). The latter transcribes this sequence as follows in the grid:
This transcription reflects John’s understanding of the interaction. In

VMC, space is delimited by the frame and the social distance between users
is mediated by the webcam, thus suggesting, for example, a sense of detach-

ment if the shot is too long, or, in other words, if the participant is too far
from the screen, she/he may seem as not fully involved in interaction. In
this case, conversely, the participant appears as very close, as Shizuka is
taken in close shot, and this may be due to the familiar and intimate con-
text in which the conversation takes place (i.e. friends talking at home).
As mentioned in previous sections, gaze is a fundamental facilitator in
conversation. Several studies (Goodwin, 1980, 1981; Kendon, 1967,
1990; Everts, 2004) show that turn-taking behaviour in conversation is
mainly regulated through gaze. However, VMC does not allow for simu-
lation of eye contact among participants, so that the lack of it may create
difficulties that participants need to address to achieve successful com-
munication. Gaze management trouble in VMC is mainly due to the
medium’s technical affordances, as found in the difficulty of positioning
camera and monitor/smartphone on the same optical axis.2
In the interaction discussed in this chapter, John notes that Nobita has
maintained direct eye contact: “he has rarely looked away from the screen,
thus seeking direct contact with me”, attributing such behaviour to
Japanese cultural norms: “My interlocutor stayed focused on the screen,
trying to come as close to eye contact as possible. He didn’t use his actions
nor his gaze to manage turn taking, he would rather use silences and
prosody to manage that kind of matters as it is probably of use in the
Japanese culture.”
Being aware that gaze is a turn management device in spontaneous
conversation, John takes pains to show that Nobita strived to keep eye
contact, perhaps to show maximum conversational cooperation.
Additionally, even though Nobita holds his baby daughter on his lap, he
never looks at her, not even when he is introducing her at the beginning
of the exchange. It should be also noticed that a video-mediated conver-
sation that employs a smartphone for connection will be very different in
terms of gaze management and eye contact approximation from a video-
mediated conversation via PC or laptop. Parallax3 between webcam and
user’s position will be significantly different, as it is very likely that a user
that is making a smartphone-based video call will position their smart-
phone (and camera) much closer to their face, thus enhancing the impres-
sion of eye contact due to webcam-eye parallax.
94 M. G. Sindoni
Fig. 3.3 Nobita’s interrogative look
Furthermore, gaze can be interestingly used to regulate the flow of

written turns or to obtain clues about the regular flow of conversation
(Sindoni, 2014b). Visual contact signals that the communication chan-
nel is open, but can also signal that something has gone wrong. As dis-
cussed in the previous section, John admits that he understood that
Nobita had not received text notification of the URL addressed after a
prolonged silence. Before directly asking “What is that?”, however,
Nobita very clearly gives a perplexed gaze that is shown in Fig. 3.3.
Even though John may be unaware of this non-verbal clue, Nobita is
signalling a lack of understanding via gaze and posture well before he
actually utters his question, even though John does not recognize this
state of affairs in his transcription.
Successful communication is the result of the complex working of
intercultural pragmatic competence. The occasion to introduce Nobita’s
wife was prompted by John’s question: “What about your family, are you
married?” The surprise at this question is evident in the use of all the
semiotic resources taken together: Nobita’s amazed look, rising intona-
tion, straightening his shoulders and his rather blunt and piqued reply:
“Of course I’m married!” In Nobita’s culture, having previously intro-
duced his daughter to John, it should have been absolutely clear to John
that Nobita is—of course—married. Conversely, John’s American cul-
tural background does not allow the straightforward equation between
having a daughter and being married.
The orchestration of semiotic resources other than language is so com-

plex that multimodal transcription and analysis can be very challenging
to students and multimodal analysts alike. However, transcription and
annotation can be useful tools to unveil the analyst’s attitudes, bias and
ideologies.
3.6 C
oncluding Remarks
This chapter has discussed video and written data sets that were produced
during a project that involved students in recording video-mediated con-
versations. In particular I singled out one interaction that has been com-
mented on from three different standpoints: (1) direct observation of
video data that have been analysed by applying previous models of analy-
sis (Sindoni, 2013, 2014a) and transcription of spoken turns plus repro-
duction of visuals; (2) symptomatic comparisons between SC’s and my
transcriptions; (3) analysis of the comments on the interaction as emerg-
ing from a short academic paper that was part of SCs’ assignment.
The main goal of my approach is to broaden our understanding of how
VMC is actually used and perceived by both unaware users (in this case,
project participants who were not students) and students in a multimodal
VMC studies programme (primary participants) who are interested in
developing multiliteracy skills. Even though novice learners may be aware
of some of the main theories, transcription methods and research agendas
in VMC, for example in terms of the semiotic resources that they need to
pay attention to, their views on VMC are illuminating from many differ-
ent standpoints. For example, concerns about the correct use of “stan-
dard” variants in language are still powerful (overall, all SCs tended to
standardize their transcriptions).
Furthermore, it could be argued that cultural stigmatizations that still
pervade lay evaluations about the idiosyncrasies and apparent unruliness
of spoken discourse have generally led SCs into normalizing spoken turns
to make it easier to harness data and, in general, make them more accept-
able. Apparently, using written norms to describe spoken language is
much more reassuring, and this comes as no surprise. The power of logo-
centric traditions is still alive, at least in the more formal and institutional
96 M. G. Sindoni
contexts, such as colleges and universities. However, a greater emphasis

on the pragmatic and socio-semiotic use of language variants (e.g. writ-
ten/spoken; standard/non-standard) should be introduced in tertiary
education, especially in the context of foreign-language teacher training
in intercultural environments.
On another level, SCs’ insights are revealing especially when it comes
to tracking and interpreting the directions and ideologies of their efforts
in making sense of their own video-mediated interactions. To my knowledge,
no previous study has explored video interactions by drawing upon this
multilevel approach to data sets yet: that is, taking into account (1) what
happens in VMC, (2) how participants themselves are able to reconstruct
and, more importantly, transcribe what happened and (3) how partici-
pants critically comment on video data in their assignments.
Issues in transcription and annotation of spontaneous interactions are
hotly debated in different fields of social studies, as shown above. Manual
transcription of VMC is particularly complex and time consuming, as it
involves thorny choices for multimodal analysts, for example deciding
how turns can be linearized (Sindoni, 2013). However, in educational
contexts, the time and effort required to transcribe very short chunks of
spontaneous video-based interactions (max. 3–4 minutes each) could
prove immensely useful to enhance students’ skills in understanding,
interpreting and critically meta-commenting naturally occurring video-
mediated conversations.
In other words, our academic understanding of transcription and
annotation should be broadened: instead of seeing these procedures as
highly theoretically developed, but limited to the realms of academia and
substantially separate from educational contexts, they should be incorpo-
rated into university curricula as heuristics for the improvement of stu-
dents’ critical skills and, more importantly, as open windows to gain
precious access into covert (educational) ideologies that still privilege the
written normativity of language.
In the case study analysed in this chapter, SC John consistently normal-
ized speech by referring to the written norms of language, as he reduced
or omitted repetitions, hesitations, fillers and discourse markers and did
not account for overlapping and other characteristics of spoken interac-
tions. This may be also due to the fact that transcribing is time consuming
and standardizing may be a way to get rid of presumed unnecessary,

redundant or incomprehensible information. Furthermore, he repre-
sented verbal language more accurately, whereas other resources were
described in less depth; this may indicate a possible prioritization of lan-
guage over other resources. This interaction is not an isolated case as all
video interactions by all SCs in the project indicate the general tendency
to prioritize language in multimodal transcription and analysis and to
normalize spoken language as to make it appear somewhat more “accept-
able” or “correct”. This is one interpretations among others, as verbal lan-
guage is extremely codified and students are certainly much more
confident in handling written language than any other semiotic resource.
This study may prove useful to chart new directions of research for
teacher training, especially in the context of distance learning and
e-learning, as multimodal transcription and annotation, as has been
shown, can be powerful tools to unveil student’s ideologies in language
representations and can be used to enhance student’s critical skills in
learning and teaching languages.
Acknowledgements I would like to thank my friends and colleagues Elisabetta

Adami, Ilaria Moschini and Sandra Petroni. I am also grateful to Dr Laura
Rosenbaun for her insightful comments on the manuscript. Many thanks go to
all MoM participants who granted their permission to study, reproduce and
publish all the video and written materials developed during the project. Without
their permission and contributions this study would not have been possible.
ppendix: Transcription Conventions

A
for Spoken Turns and for Fillers
Symbol Meaning
// Completion (falling tone)
No end of turn Non-termination (no final intonation)
punctuation
/ Parcelling of talk, breathing time
… Pause of ½ second or more
98 M. G. Sindoni
Symbol Meaning
.. Pause of less than ½ second
? Uncertainty (rising tone)
! Surprised intonation (rising-falling tone 5 in Halliday’s
1994 system)
WORDS IN CAPITAL Emphasis, stress, increased volume
(…) Inaudible/untranscribable utterance
(words within transcriber’s guess
parenthesis)
- False start/restart
(()) Analyst’s description
[] Overlapped turn
* Visual/multimodal insert
Adapted from Eggins & Slade, 1997 and Tannen, 1989
Uhm Doubt
Ah Staller
Mmm Agreement
Eh Query
Oh Reaction
Ooh Surprise
Adapted from Eggins & Slade, 1997
Notes
1. The integral recording of this interaction is to date available at https://
www.youtube.com/watch?v=1px35APDY34&feature=youtu.be
2. However, recent video-teleconferencing research has shown an eye-gaze
scheme correction with stereovision through the application of matching
techniques (Yang & Zhang, 2001) and a 3D teleconferencing system
which allows to transfer the face of a remote participant to an audience
gathered around a 3D screen, to permit gaze reciprocation and realization
of gaze cues akin to what happens in in praesentia communicative events
(Jones et al., 2009).
3. A parallax is the effect whereby the position or direction of an object
appears different if viewed from different positions (see Vertegaal & Ding,
2002).
References
mediated communication. Journal of Sociolinguistics, 10(4), 419–438. https://
doi.org/10.1111/j.1467-9841.2006.00286.x
computer-mediated discourse. In C. Thurlow & K. Mroczek (Eds.), Digital
discourse: Language in the new media (pp. 277–298). Oxford: Oxford
University Press.
Antaki, C., Ardévol, E., Núñez, F., & Vayreda, A. (2005). “For she who knows
who she is:” Managing accountability in online forum messages. Journal of
Computer-Mediated Communication, 11(1), 114–132. https://doi.
org/10.1111/j.1083-6101.2006.tb00306.x
Bezemer, J., & Jewitt, C. (2010). Multimodal analysis: Key issues. In L. Litosseliti
(Ed.), Research methods in linguistics (pp. 180–197). London & New York:
Continuum.
Bloor, M. (2016). The construal of terminal illness in online medical texts:
Social distance and semantic space. In S. Gardner & S. Alsop (Eds.), Systemic
functional linguistics in the digital age (pp. 120–133). London: Equinox.
Bou-Franch, P., Lorenzo-Dus, N., & Garcés-Conejos Blitvich, P. (2012). Social
interaction in YouTube text-based polylogues: A study of coherence. Journal
of Computer Mediated Communication, 17, 501–521.
Creese, A. (2008). Linguistic ethnography. In K. A. King & N. H. Hornberger
(Eds.), Encyclopedia of language and education, Research methods in language
and education (Vol. 10, 2nd ed., pp. 229–241). New York: Springer Science
and Business Media.
Crystal, D. (2006). Language and the internet. Cambridge: Cambridge University
Press.
Danet, B. (2001). Cyberpl@y: Communicating online. Oxford: Berg.
Duranti, A. (1997). Linguistic anthropology. Cambridge: Cambridge University
Press.
Eggins, S., & Slade, D. (1997). Analysing casual conversation. London: Equinox.
Erickson, F. (2004). Talk and social theory: Ecologies of speaking and listening in
everyday life. Cambridge: Polity Press.
Everts, E. (2004). Modalities of turn-taking in blind/sighted interaction: Better
to be seen and not heard? In P. LeVine & R. Scollon (Eds.), Discourse and
technology: Multimodal discourse analysis (pp. 128–145). Washington, DC:
Georgetown University Press.
100 M. G. Sindoni
Flewitt, R., Hampel, R., Hauck, M., & Lancaster, L. (2009). What are multi-
modal data and transcription? In C. Jewitt (Ed.), The handbook of multimodal
analysis (pp. 40–53). London & New York: Routledge.
Flinkfeldt, M. (2011). ‘Filling one’s days’: Managing sick leave legitimacy in an
online forum. Sociology Health Illness, 33(5), 761–776. https://doi.
org/10.1111/j.1467-9566.2011.01330.x
Flinkfeldt, M. (2014). Making equality relevant: Gender, housework, and sick
leave legitimacy in online interaction. Female Psychology, 24(3), 295–313.
https://doi.org/10.1177/0959353513515295
Gardner, S., & Alsop, S. (Eds.). (2016). Systemic functional linguistics in the digi-
tal age. London: Equinox.
Georgakopoulou, A. (2011). Computer-mediated communication. In
J. Ostman & J. Verschueren (Eds.), Pragmatics in practice (pp. 93–110).
Philadelphia & Amsterdam: John Benjamins.
Goffman, E. (1981). Forms of talk. Oxford: Blackwell.
Goldin-Meadow, S. (1999). The development of gesture with and without
speech in hearing and deaf children. In L. S. Messing & R. Campbell (Eds.),
Gesture, speech and sign (pp. 117–132). Oxford: Oxford University Press.
Goodwin, C. (1980). Restart, pauses, and the achievement of mutual gaze at
turn-beginning. Sociological Inquiry, 50(3–4), 272–302.
Goodwin, C. (1981). Conversational organization: Interaction between speakers
and hearers. New York: Academic Press.
Gumperz, J. J. (1999). On interactional sociolinguistic method. In S. Sarangi &
C. Roberts (Eds.), Talk, work and institutional order: Discourse in medical,
mediation and management setting (pp. 453–472). Berlin & New York:
Mouton de Gruyter.
Hall, E. T. (1966). The hidden dimension. New York: Doubleday.
Hall, E. T., & Hall, R. M. (1990). Understanding cultural differences. Yarmouth,
Maine: Intercultural Press.
Halliday, M. A. K. (1973). Explorations in the functions of language. London:
Arnold.
Halliday, M. A. K. (1978). Language as social semiotic: The social interpretation of
language and meaning. London: Arnold.
Heath, C., & Luff, P. (1992). Media space and communicative asymmetries:
Preliminary observations of video-mediated interaction. Journal of Human-
Computer Interaction, 7(3), 315–346.
Heath, C., & Luff, P. (2000). Technology in action. Cambridge: Cambridge
University Press.
Heller, M. (1988). Code-switching: Anthropological and sociolinguistic perspec-

tives. Berlin & New York: Mouton de Gruyter.
Herring, S. C. (2004a). Computer-mediated discourse analysis: An approach to
researching online behavior. In S. H. Barab, R. Kling, & J. H. Gray (Eds.),
Designing for virtual communities in the service of learning (pp. 338–376).
New York: Cambridge University Press.
Herring, S. C. (2004b). Computer-mediated discourse. In D. Tannen,
D. Schiffrin, & H. Hamilton (Eds.), Handbook of discourse analysis
(pp. 612–634). Oxford: Blackwell.
Herring, S. C. (2011). Computer-mediated conversation part II: Introduction
and overview. Language@Internet, 8(2). Retrieved from http://www.lan-
guageatinternet.org/articles/2011/Herring
Herring, S. C. (2013). Discourse in Web 2.0: Familiar, reconfigured, and emer-
gent. In D. Tannen & A. M. Trester (Eds.), Georgetown University Round
Table on languages and linguistics 2011: Discourse 2.0: Language and new
media (pp. 1–25). Washington, DC: Georgetown University Press.
Herring, S., Stein, D., & Virtanen, T. (2013). Pragmatics of computer-mediated
communication. Boston: De Gruyter Mouton.
Hutchby, I. (2001). Conversation and technology: From the telephone to the inter-
net. Cambridge: Polity Press.
Jones, A., Lang, M., Fyffe, G., Yu, X., Busch, J., Bolas, M., et al. (2009).
Achieving eye-contact in a one-to-many 3D video teleconferencing system.
ACM Transactions on Graphics, 28(3), Paper 64.
Karagevreki, M. (2016). Analysis of an online university lecture: Multimodal
perspectives. In S. Gardner & S. Alsop (Eds.), Systemic functional linguistics
in the digital age (pp. 166–183). London: Equinox.
Kendon, A. (1967). Some functions of gaze-direction in social interaction. Acta
Psychologica, 26, 22–63.
Kendon, A. (1990). Conducting interaction. Patterns of behavior in focused
encounters. Cambridge: Cambridge University Press.
Kress, G., & van Leeuwen, T. (2001). Multimodal discourse: The modes and
Kress, G., & van Leeuwen, T. ([1996] 2006). Reading images: The grammar of
visual design (2nd ed.). London: Routledge.
Licoppe, C., & Morel, J. (2012). Video-in-interaction: “Talking heads” and the
multimodal organization of mobile and skype video calls. Research on
Language & Social Interaction, 45(4), 399–429.
Licoppe, C., & Morel, J. (2014). Mundane video directors in interaction:
Showing one’s environment in Skype and mobile video calls. In Studies of
video practices: Video at work (pp. 135–160). London: Routledge.
102 M. G. Sindoni
Malinowski, B. (1923). The problem of meaning in primitive languages. In

C. K. Ogden & I. A. Richards (Eds.), The meaning of meaning (pp. 296–336).
London & New York: Routledge.
Miller, D. (2016). On negotiating the hurdles of corpus-assisted appraisal analy-
sis in verbal art. In S. Gardner & S. Alsop (Eds.), Systemic functional linguis-
tics in the digital age (pp. 211–228). London: Equinox.
Myers, G. (2010). Stance-taking and public discussion in blogs. Critical
Discourse Studies, 7(4), 263–275.
Norris, S. (2004). Analyzing multimodal interaction. A methodological framework.
London & New York: Routledge.
O’Toole, M. ([1994] 2011). The language of displayed art (2nd ed.). London &
New York: Routledge.
Papagiannouli, C. (2011). Cyberformance and the cyberstage. The International
Journal of the Arts in Society, 6(4), 273–282.
Pink, S. (2007). Doing visual ethnography. London: Sage.
Rosenbaun, L., Rafaeli, S., & Kurzon, D. (2016a). Blurring the boundaries
between domestic and digital spheres: Competing engagements in public
Google Hangouts. Pragmatics, 26(2), 291–314.
Rosenbaun, L., Rafaeli, S., & Kurzon, D. (2016b). Participation frameworks in
multiparty video chats cross-modal exchanges in public Google Hangouts.
Journal of Pragmatics, 94(2016), 29–46.
Sacks, H. (1992). Lectures on conversation. Oxford: Blackwell.
Sacks, H., Schegloff, E. A., & Jefferson, G. (1974). A simplest systematics for
the organization of turn-taking for conversation. Language, 50, 696–735.
Schegloff, E. A., Jefferson, G., & Sacks, H. (1977). The preference for self-
correction in the organization of repair in conversation. Language, 53,
361–382.
Scollon, R. (2002). Mediated discourse: The nexus of practice. London &
Sindoni, M. G. (2011). Online conversations: A sociolinguistic investigation
into young adults’ use of video chats. Classroom Discourse, 2, 219–235.
Sindoni, M. G. (2012a). Speech and writing in English: A cultural and linguistic
debate. Firenze: Le Lettere.
Sindoni, M. G. (2012b). Mode-switching: How oral and written modes alter-
nate in video chats. In M. Cambria, C. Arizzi, & F. Coccetta (Eds.), Web
genres and web tools: With contributions from the Living Knowledge Project
(pp. 141–153). Como & Pavia: Ibis.
Sindoni, M. G. (2013). Spoken and written discourse in online interactions: A
multimodal approach. London & New York: Routledge.
Sindoni, M. G. (2014a). Through the looking glass: A social semiotic and lin-
guistic perspective on the study of video chats. Text & Talk, 34, 325–347.
Sindoni, M. G. (2014b). Addressing cultural and linguistic diversity in multi-party
video interaction: A multimodal conversation analysis approach. Paper presented
at Languaging Diversity 2014, Università di Catania, 9–11 October 2014.
Sindoni, M. G. (2016). “There’s power in stories”: A multimodal and corpus-
based functional analysis of fandom blogs. In S. Gardner & S. Alsop (Eds.),
Systemic functional linguistics in the digital age (pp. 1–26). London: Equinox.
Sneijder, P., & te Molder, H. F. M. (2004). ‘Health should not have to be a
problem’: Talking health and accountability in an internet forum on vegan-
ism. Journal of Health and Psychology, 9(4), 599–616. https://doi.
org/10.1177/1359105304044046
Sneijder, P., & te Molder, H. F. M. (2005). Moral logic and logical morality:
Attributions of responsibility and blame in online discourse on veganism.
Discourse Society, 16(5), 675–696. https://doi.org/10.1177/0957926505054941
Swan, P. K., Richardson, J. C., Ice, P., Garrison, D. R., Cleveland-Innes, M., &
Arbaugh, J. B. (2008). Validating a measurement tool of presence in online
communities of inquiry. E-Mentor, 2(24), 1–12.
Tannen, D. (1989). Talking voices: Repetition, dialogue and imagery in conversa-
tional discourse. Cambridge: Cambridge University Press.
The New London Group. (1996). A pedagogy of multiliteracies: Designing
social futures. Harvard Educational Review, 66(10), 60–92.
Thibault, P. J. (2000). Multimodal transcription of a television advertisement:
Theory and practice. In A. Baldry (Ed.), Multimodality and multimediality in
the distance learning age (pp. 311–385). Campobasso: Palladino.
van Leeuwen, T. (1999). Speech, music, sound. London: Macmillan.
van Leeuwen, T. (2005). Introducing social semiotics. London: Routledge.
Vertegaal, R., & Ding, Y. (2002). Explaining effects of eye gaze on mediated
group conversations: Amount or synchronization? In Proceedings of the 2002
ACM Conference on Computer Supported Cooperative Work, 41–48.
Yang, R., & Zhang, Z. (2001). Eye gaze correction with stereovision for video-
teleconferencing. Paper presented at the 7th European Conference on Computer
Vision – Proceedings, 2, May 27–June 2, Copenhagen, Denmark, 479–474.
Amsterdam & Philadelphia: John Benjamins.
4
Multimodality in Memes:
A Cyberpragmatic Approach
Francisco Yus
4.1 Introduction: Image Macros

In 1976, Dawkins proposed the term “meme” to refer to the idea of
disseminated cultural unit. This meant information with the capacity of
infecting people’s minds with a kind of duplicating replicability; imita-
tion, for its part, was said to be the key to the survival of memes. Therefore,
in this initial conceptualization, memes were analogous to the biological
concept of gene, in the sense that they were seen as self-replicating and
being communicated from person to person. More recently, memes have
been defined as units of information, ideas or mental representations,
and cultural instructions that are not only self-replicating but also conta-
gious (Taecharungroj & Nueangjamnong, 2014, p. 152).
From this initial idea of meme, replicating pieces of information that
spread through the net through user-to-user(s) communication have
been labeled similarly: internet memes. This is not surprising, since these
F. Yus (*)
Department of English Studies, University of Alicante, Alicante, Spain
e-mail: francisco.yus@ua.es

https://doi.org/10.1007/978-3-319-92663-6_4
106 F. Yus
memes do share some properties initially ascribed to memes in general.

Among the many definitions of internet meme stand the following: (a)
any artifact that appears on the internet and produces countless deriva-
tives by being imitated, remixed, and rapidly diffused by countless par-
ticipants in technologically mediated communication (Dynel, 2016,
p. 662); (b) a group of digital items sharing common characteristics of
content, form, and/or stance, which were created with awareness of each
other and were circulated, imitated, and/or transformed via the internet
by many users (Shifman, 2014, p. 41); and (c) a relatively complex, mul-
tilayered, and intertextual combination of (moving) picture and text that
is disseminated by the active agency of internet users, becoming popular
among them (Laineste & Voolaid, 2016, p. 27).
Crucially, a distinctive feature of internet memes is that they are invari-
ably ascribed to a “family” and part of their effective processing entails
this initial family ascription (Nissenbaum & Shifman, 2017, p. 484;
Segev, Nissenbaum, Stolero, & Shifman, 2015, p. 418). In this chapter, I
will focus on a particular family of memes: the image macro meme, made
up of a line or two of text on top of the meme, line(s) of text at the bot-
tom, and one picture in the middle. The image macro meme presents a
number of interesting text-picture combinations with interesting prag-
matic and cyberpragmatic implications (Yus, 2011).
4.2 M
ultimodality
Although studies on multimodality have been published for more than
30 years, multimodality is becoming increasingly important nowadays
due to the pervasiveness of discourses on the internet that combine dif-
ferent modes of communication (text, picture, audio, video, etc.). A basic
definition of multimodality is suggested by Stöckl (2004, p. 9), for whom
“multimodal refers to communicative artefacts and processes which com-
bine various sign systems (modes) and whose production and reception
calls upon the communicators to semantically and formally interrelate all
sign repertoires present”. For van Leeuwen (2015, p. 447), the term
“indicates that different semiotic modes (for instance, language and pic-
ture) are combined and integrated in a given instance of discourse or kind
Multimodality in Memes: A Cyberpragmatic Approach 107
of discourse”. In fact, the most interesting picture-text combination for a

pragmatic (i.e. cyberpragmatic) analysis is the one in which the eventual
meaning of the meme cannot be obtained from the partial meanings of
text or picture taken separately, but only from their combined meanings
that yield implications.
Finally, Jewitt (2016) summarizes the key assumptions in studies of
modality: (a) all communication is multimodal; (b) analyses focused
solely or primarily on language cannot adequately account for meaning;
(c) each mode has specific affordances arising from its materiality and
from its social histories which shape its resources to fulfill given commu-
nicative needs; and (d) modes concur together, each with a specialized
role, to meaning-making; hence, relations among modes are key to
understand every instance of communication. In the specific case of
meme communication, co-occurrence of text and picture (fourth assump-
tion in Jewitt, 2016) is especially important and ideal for a relevance-
theoretic analysis, as will be commented upon in the next section.
4.3 T
ext-Picture Combinations and Relevance
In Yus (2016), some space is devoted to the pragmatic implications, in
terms of relevance, of processing combinations of text and picture such as
the ones found in the memes under analysis in this chapter. Yus (2016)
argued that, in the same way as we have explicit interpretations (explica-
tures) and implicit or implicated interpretations (implicatures) of verbal
utterances, visual content also leads to visual explicatures and visual impli-
catures (see Forceville, 2014; Forceville & Clark, 2014; Wharton, 2009
for discussion). That is, when faced with a picture in a meme, the user also
has to make inferential hypotheses concerning the role that it plays in the
overall comprehension of the meme. For instance, the user needs to infer
whether the picture has a purely denotative quality (the user is simply
expected to identify the referent of the picture without any further impli-
cations), in which case its interpretation would be a visual explicature, or
whether it has a wholly inferential connotative quality that can only be
obtained from the combination of the picture and contextual information
(as happens with implicated meanings from utterances). In this case, we
108 F. Yus
would be dealing with a visual implicature. The visual e xplicature is easy

to process, since the user simply identifies the visual information in the
picture and straightforwardly matches it with the most appropriate men-
tal referent. Visual implicatures, on the contrary, are fully inferential and
have to be obtained with the aid of context.
Therefore, interpreting a meme entails a “division of labor” between
the processing of the text, the processing of the picture, and the identifi-
cation of possible connotative meanings for text, picture, and text-picture
combinations. Specifically, upon finding a meme in a Facebook profile, or
within a WhatsApp message, the user will have to engage in the following
inferential strategies, which are not meant to be successive but performed
according to the user’s expectations of, and search for relevance in the
meme:
Strategy 1: To decode and inferentially enrich the verbal content of the
meme (top and bottom lines of text) in order to obtain the explicit inter-
pretation of the text or explicature.
Strategy 2: To derive implicatures from verbal content, if these are nec-
essary to reach a relevant interpretation of the verbal content of the
meme.
Strategy 3: To decode and inferentially enrich the picture to yield a
visual explicature.
Strategy 4: To derive implicatures or implications from the picture in the
meme, if these are necessary to reach an adequate interpretation of the
meme as a whole.
Strategy 5: To infer possible combinations of text and picture to yield
interpretations (typically implicated ones or verbal-visual implicatures)
that are only possible from the combination of these sources of informa-
tion (text and picture) and not from either of them taken separately
(Tsakona, 2009, p. 1172). Very often, the information obtained from the
picture will lead to an inferential backtracking (and reinterpretation) after
the accompanying text has been processed, in the light of the information
provided by visual content. On other occasions, the text will force a new
interpretation of the picture, resulting in either a new visual explicature or
leading to the derivation of a previously unforeseeable visual implicature.
In my opinion, the effects of this combination of text and picture are the
ones that are more interesting for a cyberpragmatic analysis of why memes
turn out effective (e.g. humorous), since the eventual interpretation

demands the user’s active participation in combining sources of informa-
tion for the sake of an eventual satisfactory (i.e. relevant) interpretation.
Combining interpretations from text and picture entails a kind of
iconic literacy in order to process adequately the linear and time-
demanding word-by-word processing of the text and the instantaneous
visual impact of the picture in the meme. El Refaie and Hörschelmann
(2010, p. 200) write about types of literacy concerning the interpretation
of cartoons, which are also applicable to meme interpretation:
“Interpreting cartoons is a matter of drawing on many different types of
literacy, which form the necessary preconditions for readers to be able to
discover relevant connections between the fictional scene of a cartoon
and a political argument. In this particular case, multimodal literacy
included the ability (1) to establish the referents of a cartoon both on the
level of the make-believe world and of the real-life world of current politi-
cal events, (2) to impose a narrative on the cartoon picture, and (3) to
draw on intertextual references.”
Strategy 6: To access as much contextual information as is necessary to
obtain interpretations out of strategies 1–5 above. In the case of some
memes, the reader’s background knowledge on current affairs, newswor-
thy events, political issues, and so on (his/her literacy) is crucial to under-
standing the meme properly, to the extent that, very often, the meme
makes little sense if it is separated from the specific time frame and pieces
of news that justified its publication (El Refaie & Hörschelmann, 2010,
p. 197; Kardaş, 2012, p. 208). As Conradie et al. (2012, p. 41) remark for
cartoons, and equally applicable to memes, we can differentiate two frame-
works of analysis, one related to the reader’s background knowledge
(reader-orientated strategy) and one focused on the semiotic qualities of the
discourse (text-orientated strategy): the first strategy involves looking at
how the reader interprets the discourse (based on his/her subjective con-
ceptual framework), while text-orientated studies involve examining the
discourse (visual and verbal) per se, focusing on its form and style.
Several qualities of the multimodal meme will determine which of the
strategies (1–5) listed above (aided by contextualization, strategy 6), will
be most likely to take place first and in which order the other strategies
will be performed. Among others, the salience of some discursive element
110 F. Yus
in the meme will lead the user to focus his/her attention there before other
parts of the meme are processed. This may have implications for how the
eventual relevance is assessed; for instance, for how the intended implica-
tions from text-picture combinations are derived. In theory, the reading
path for the meme should start at the top (text processing), continue with
the picture in the middle (visual processing), and then finish at the bot-
tom of the meme (text processing). However, the picture may draw the
user’s attention before the lines of text are interpreted for relevance and
this salience will lead to an alteration of the eventual cognitive effects if
the order of processing does not match the initially intended one.
In this sense, van Leeuwen (2015, p. 457) reminds us that texts are not
really linear in their processing. As has been suggested, a top-bottom linear
reading of the meme is expected on many occasions. But reading paths are
mainly created by differences in salience and depend on the textual or visual
element that attracts the reader’s or viewer’s attention over and above other
elements (Kress & van Leeuwen, 2006, p. 218). Differences in salience can
be realized by foregrounding or by differences in size, boldness, tonal con-
trast, or color. In this way, visual compositions can set up particular hierar-
chies between the elements to attract the attention and guide the movement
of the hypothetical internet user’s eyes within and across the different dis-
cursive elements of the meme. Such reading path will begin with the most
salient element, from there it will move on to the next most salient ele-
ment, and so on, in a trajectory that need not be similar to the top-down
order of multimodal discourses such as memes. Besides, salience may vary
enormously from meme to meme even if the memes neatly belong to the
same family. And finally, the actual salience of elements in the meme will
vary for different users; different areas of the meme will draw the attention
of a variety of users in different directions, so there is no guarantee that the
same reading paths will be followed across users.
4.4 Methodology
In this chapter, an analysis is carried out of a corpus of 100 memes ran-
domly selected from a Google search with the text query “image macro
meme”. The main purpose is to determine what category of text-picture
combination is more frequent and why, together with predictions of

interpretive relevance. Along these same lines, several publications have
addressed text-picture interfaces (see for instance Sarapik, 2009; Trifonas,
2015; Tsakona, 2009; van Leeuwen, 2006). However, in this chapter,
McCloud’s (1994) taxonomy of different categories for multimodal com-
binations in comics will be applied to the analysis, since memes share
their textual-visual quality with comics. Actually, only one of McCloud’s
categories cannot be found in the corpus of memes (see below). Along
with McCloud’s taxonomy, other sets of categories for text-picture rela-
tionships of semiotic (Barthes, 1977) and social-semiotic (Chan, 2011;
Gill, 2002; Martinec & Salway, 2005; Salway & Martinec, 2002) orien-
tations will also be applied to the data. The interest lies, of course, in the
text-picture relationships that contribute to specific interpretive out-
comes, and not so much in the purely semiotic interrelation of modes.
For example, Cohn (2013) proposed four ways of connecting text and
pictures: (1) inherent (text and picture are part of each other’s structures),
(2) emergent (text and picture are directly interfaced with each other), (3)
adjoined (text and picture are integrated but not interfaced directly), and
(4) independent (text and picture are fully separate). This taxonomy, while
interesting from a purely semiotic point of view, is not useful for a cyber-
pragmatic analysis of memes, since memes exhibit a rigid placement and
quality of both modes. A more fruitful approach, in my view, is to work
out the inferential implications of the interrelations of both modes and
how their combinations yield relevant interpretive outcomes.
The next step is to analyze the corpus of 100 memes and determine
into which category they may be ascribed, assessing the role that text and
picture play in the eventual interpretation while, at the same time, look-
ing for specific text-picture combinations that may be interesting for a
cyberpragmatic account of their relevance for the user who is processing
the meme. From this analysis, some conclusions are drawn.
Throughout the analysis, special attention is devoted to determining
(a) which contribution each mode (text/picture) makes to the overall
global meaning of the meme; (b) whether there are cases in which the text
makes little or no contribution to the interpretation of the meme, which
is mainly provided by the picture, or it is the opposite relationship: the
picture making little or no contribution to the interpretation of the
112 F. Yus
meme, that is, cases where the picture add little or no meaning to the
meme; (c) in which cases text and picture combine to generate implicated
meanings that may only be obtained from the combination of the infor-
mation from both modes; and (d) whether there is some specificity to
memes that does not neatly correspond to any of the categories suggested
by McCloud (1994) for comics.
4.5 Text-Picture Combinations in Image

Macro Memes
4.5.1 W
ord Specific: Where Pictures Illustrate
but Do Not Significantly Add to a Largely
Complete Text
This category is similar to Barthes’s (1977) concept of illustration, in

which the text is the primary mode and the picture plays no substantial
role in altering the meaning conveyed verbally. It also suits Martinec and
Salway’s (2005) exemplification, in which the text takes prominence over
the picture.
Out of the corpus of 100 memes, 25 instances can be ascribed to this
category. Sometimes, the picture is useful in narrowing the scope of the
meaning of the words (the picture also prevents a potential incongruous
interpretation of the text). Consider (1–2):
(1) Top text: If tomorrow isn’t the due day

Picture: Picture of a university student drinking a beer in a pub.
Bottom text: then today isn’t the do day.
(2) Top text: Why make college the 4 best years of your life
Bottom text: if you can make it 6.
In (1–2), the text is more or less autonomously interpreted, but it is more

adequately processed next to the picture of the person who might possibly
utter these words, that is, a university student. In both memes, the picture
is the same and it exemplifies the kind of person that would typically
produce these utterances. The eventual balance of cognitive effects and

mental effort, as envisaged by relevance theory (Sperber & Wilson, 1995),
is optimized thanks to the contribution of the picture in narrowing down
the potential range of speakers of such utterances; the picture also helps
the user in determining the exact meaning of, for example, do in (1) and
make in (2). Furthermore, the picture helps in the derivation of a number
of implications on a stereotypical university student’s attitude to home-
work and commitments to university duties in general.
The same applies to the adjustment of the concept coded by some
specific word in the text of the meme (more on concept adjustment
below). In (3), for instance, with the same picture as (1–2) above, the
concept coded by attendance is narrowed probably to “attending bars and
cafes in college” but not classes; another possible narrowing of attend
could render “being present in the classroom without paying any atten-
tion to what is being said by the teacher”. Similarly, the concept coded by
the word disturbing in (4) acquires more specific meanings (even if main-
taining its basic denotation) when placed next to a picture of Darth Vader
from Star Wars:
(3) Top text: I have 100% college attendance

Bottom text: but 0% class attendance.
(4) Top text: I find your lack of documentation
Picture: Picture of Darth Vader.
Bottom text: disturbing.
Concept adjustment is one of the inferential strategies that users deploy

when turning the schematic meaning of the words into meaningful and
contextualized interpretations. Another inferential strategy is reference
assignment. In this case, the picture in the meme may also be useful when
identifying who the text is referring to, as in (5):
(5) Top text: Maggie Smith battled cancer while filming the last
Harry Potter movie.
Picture: Photograph of Maggie Smith.
Bottom text: She carried on, not wanting to disappoint the fans.
114 F. Yus
The picture in the meme may also play a part in the eventual success of
an incongruity-resolution pattern (see Yus, 1997, 2016, 2017) in the text
of the meme. Incongruity-resolution is one of the most typical joke struc-
tures, in which the addressees of the joke face an incongruity due to the
manipulation of their inferential strategies, and they have to find a reso-
lution that reconciles the incongruous parts of the joke.
However, in this meme pattern the picture is not essential for the
text to be perceived as humorous (i.e. the text itself follows the
incongruity-resolution pattern without needing the picture for the
derivation of humorous effects). Sometimes, the role that the picture
plays is simply that of an illustration, and therefore it suits this first
category. Take, for instance, the meme in (6). It may well be told or
read without an accompanying picture. But in the meme the picture
helps in associating black with back race, as if a racist (white) joker was
telling the joke, and enhances the subsequent incongruity provided by
the bottom text of the meme.
(6) Top text: What’s black and never works?

Picture: White man smiling.
Bottom text: Decaffeinated coffee, you racist bastard.
On other occasions, though, the incongruity-resolution pattern is con-

structed out of text interpretation only, and the picture could well be
suppressed without significant informative loss, as in (7–8):
(7) Top text: Do you like Big Bang?

Picture: Smiling girl.
Bottom text: ‘Cause I’m sure we’d make fantastic babies.
(8) Top text: Why do people say the sky’s the limit
Picture: Drawing of a dinosaur.
Bottom text: when there are footprints on the moon?
In (7), the meme plays with the initial accessibility to the name of the
famous TV series, but the word Bang acquires a different meaning after
reading the bottom text, which becomes incongruous regarding the ini-
tial interpretation. This is a joke pattern that could well be told orally
without needing an accompanying picture. Something similar occurs
with (8), in which the picture of a dinosaur has no connection whatso-
ever with the text of the meme.
4.5.2 P
icture Specific: Where the Picture Dominates
and Words Do Not Add Significantly
to the Meaning of the Picture
This is similar to the previous category (word specific and pictures illus-
trating), but in the one here under scrutiny it is the picture that is central
to the meme: the words are supplementary and play an exemplification
role. That is, the words exemplify situations in which the gestures pro-
vided by the picture might be produced (Wu, 2014, p. 1417). Only 7
instances of the corpus of 100 memes fit into this category. This indi-
cates that pictures generally play either the role of exemplifying, empha-
sizing or amplifying the meaning of the text, or are combined with the
text in order to generate interpretations which can only be obtained
from this combination. However, instances in which the picture is the
main source of information (and eventual relevance) in the meme are far
less frequent.
One of the cases in which the picture does dominate occurs when the
meme shows a person who is famous for saying something. The picture
itself is worth the user’s attention and the text merely emphasizes the
attributes of the person depicted in the picture. In this case, the user is
expected to be able to retrieve from background knowledge the specific
information which justifies the appearance of the famous person in the
meme (e.g. what he/she typically said or did that leads to the creation of
the meme). An example is (9):
(9) Top text: Either you’re naive or have scruples.

Picture: Picture of actor Hugh Laurie in his role as Dr. House
with an angry look on his face.
Bottom text: I’m not sure which is worse.
116 F. Yus
4.5.3 D
uo Specific: Where Words and Pictures Send
Essentially the Same Message
This category is homologous to the social semiotics category of equiva-

lence (Gill, 2002), since the content is communicated both by the picture
and its accompanying text, resulting in some degree of redundancy in
meaning (Chan, 2011, p. 149). Predictably, almost no instance of the
meme corpus fits into this category, since one of the inherent features of
this kind of meme is the overlap of meaning between text and picture.
Only meme (10) might be close to the attribute of this category. The text
of the meme refers to happiness, and this feeling is paired with the picture
of happiness on a famous actor’s face:
(10) Top text: You find it offensive. I find it funny.

Picture: Picture of actor Leonardo di Caprio smiling and
proposing a toast.
Bottom text: That’s why I’m happier than you.
4.5.4 A
dditive: Where Words Amplify or Elaborate
on a Picture or Vice Versa
By contrast, this category is frequent in the corpus here under analysis.

Up to 37 instances, one way or another, exhibit this text-picture relation-
ship in which there is amplification or elaboration of one mode regarding
the other.
This category reminds us of Barthes’ notion of anchorage, in which
words help the user reach a more fine-grained interpretation of the related
picture, as intended by the creators of the meme, or vice versa. The cate-
gory is also similar to social semiotics notions such as: (1) extension, in
which one mode provides new information on the information from the
other mode; (2) enhancement, where the text adds an informative element
to the picture or vice versa; (3) exposition, where the picture elaborates on
aspects of the text or vice versa; and (4) augmentation, involving a picture
extending or adding new meanings to the text or the text extending the
picture by providing an additional element (Chan, 2011, p. 154).
A frequent text-picture relationship in this category is that in which

the picture amplifies or elaborates on the meaning of the adjacent text.
Consider these memes:
(11) Top text: I think we have to fix this problem

Picture: Actor Marlon Brando in the film The Godfather.
Bottom text: the old-fashioned way.
(12) Top text: I changed all my passwords to “incorrect”
Picture: Famous actor playing a dumb character.
Bottom text: so, whenever I forget, it says, “your password is
incorrect”.
(13) Top text: I want to go to Taco Bell
Picture: Woman in tears.
Bottom text: but I’m on an all-carb diet.
In (11), the user’s background knowledge of Brando and of his role in The
Godfather makes it easier to adjust the concept encoded by the word fix (see
Sect. 4.6 for a description of concept adjustment under relevance theory).
Besides, the user is guided by the picture when trying to narrow and con-
note the meaning of the phrase the old-fashioned way, initially broad and
little specific. In (12), the picture of a famous actor playing a dumb char-
acter enhances the absurd idea that he came up with as described in the text
of the meme. Finally, in (13) the picture of a woman in tears helps enhance
the user’s inference of the woman’s urge to go to this restaurant and in
obtaining a more vivid picture of her addiction to fast food. In all of these
memes, the eventual relevance is optimized because the picture is useful in
obtaining a more accurate interpretation of the accompanying text.
A frequent type of meme which may also be ascribed to this category is
the abundant series of memes which share the same picture but contain
different texts (this text-picture configuration is also compatible with the
meme-specific inferential strategy of ad hoc visual referent adjustment; see
Sect. 4.6). In this case, the main interest lies in a specific kind of gesture
that users find funny (often a gesture made by an actor/actress and repro-
duced as a film frame). The different versions of text are descriptions of
situations in which this gesture might be produced. However, this also
involves some kind of amplification or elaboration of the interpretation of
the picture via adjustment (see Sect. 4.6). An example is provided in (14):
118 F. Yus
(14) Top text: When somebody adds another plate to the sink
Picture: Famous actor with a hateful look on his face.
Bottom text: while I’m washing the dishes.
One further example is reproduced in Fig. 4.1. The meme depicted in

this Figure is one instance in a whole series in which the picture is the
same but the text is changed from meme to meme. In all of the samples
in the series, the strategy is always to show a facial expression of actor
James van der Beek in the famous TV series Dawson’s Creek, one showing
utter pain or desolation, and contrast it with the trivial text accompany-
ing it and justifying it, this time regarding the arrival of the TV Guide.
Finally, another role of the picture is to facilitate the successful out-
come of the incongruity-resolution humorous strategy in the meme. In
the first category above (Sect. 4.5.1), an example was provided where the
picture of the meme played a minor role in the unfolding of an
incongruity-resolution schema. In this category, by contrast, pictures
play more significant roles in the humorous outcome of this schema.
Firstly, the picture may help trigger an incongruity during the construc-
tion of an appropriate scenario for the interpretation of the meme; this
is what Yus (2013a, 2013b) labeled make-sense frame. In a nutshell, infer-
ring the intended interpretation of the meme also involves the extraction
of general information about the world and everyday situations that is
stored as accessible chunks of encyclopedic information (specifically
Fig. 4.1 Adjustment

stored as “I conceptualize X as p” or as a more factual “I believe that p”).

This information is often retrieved almost unconsciously in order to
make sense of the intended “scenario” for the comprehension of the
meme, and may be exploited for humorous purposes. Consider the
meme in (15):
(15) Top text: I feel like nobody loves me

Picture: Woman with tears running down her face.
Bottom text: especially when I eat bananas with my butt.
In this meme, the picture of a woman in tears makes it easier to build up

a make-sense frame of sheer sadness. The picture amplifies the inference
of the negative feelings associated with the fact that nobody loves the
woman. The user is led to believe that the woman is so sad at not being
loved that she finally bursts into tears. However, this make-sense frame
just constructed is radically invalidated by the bottom line of text.
Secondly, the picture may facilitate the successful outcome of a text-
based incongruity-resolution strategy. For example, if this strategy is
based on making one interpretation of an initial part of a joke more
likely to be selected only to invalidate it in a subsequent part of the joke
(thus generating incongruity), the picture may be useful in leading the
user to choose that very sense of the word, as intended. In (16), for
instance, there is an incongruity centered upon the user’s disambiguation
of the verb beat, which encodes several possible senses. The user will
inevitably select the sense that is most accessible, the one demanding
least mental effort, in this case “to strike violently or forcefully and
repeatedly”, and the scary look of the man in (16) facilitates this almost
unconscious, initially relevant, choice of one of these senses of beat,
which is again invalidated by the bottom text of the meme and replaced
with a more unlikely (initially less relevant but eventually correct)
interpretation.
(16) Top text: I love beating women

Picture: Man with a very scary look on his face.
Bottom text: to the door so I can hold it open for them.
120 F. Yus
4.5.5 P
arallel: Where Words and Picture Follow
Different Courses Without Intersecting
This category resembles the social semiotics label of exposition, in which

picture and text are at the same level of generality (van Leeuwen, 2011,
p. 676). Predictably, the quality of this category is absent in the corpus;
no meme seems to contain this text-picture relationship, since the even-
tual interpretation of memes normally relies on the combination of text
and picture, and it makes little sense to put both modes together in the
meme if these are not somehow related.
4.5.6 M
ontage: Where Words Are Treated as Integral
Parts of the Picture
This is a typical category in comics, since artists often skillfully connote

the texts with iconic connotations (dripping words, creative use of bold
letters, or letters with a visually connoted shape). In a way, then, both
modes become semiotically integrated. Consequently, it is similar to the
social semiotics term homospatiality, where different semiotic modes co-
occur in one spatially bonded homogenous entity (Lim Fei, 2004). Chan
(2011, p. 152) provides the example of the words of a poem, Stingray,
which were themselves arranged in the shape of a stingray. It also suits
Cohn’s (2013) aforementioned semiotic term inherent (text and picture
are part of each other’s structures).
This category is absent in the corpus of memes, since text type (font,
size, etc.) is fixed and imposed upon the user by the available software,
and therefore creative iconization of the text is impossible.
4.5.7 Interdependent: Where Picture or Words

Together Convey an Idea That Neither Could
Convey Alone
In contrast, this category is frequent in the corpus with up to 30 instances.

Besides, it is the category that raises more interest for a relevance-centered
cyberpragmatic analysis concerned with how users obtain interpretations
from the available information through interfaces. The fact that interpre-
tations from memes in this category cannot be obtained from the partial
meanings of text or picture taken separately makes the meme utterly con-
text dependent and very significant for the kind of analysis proposed in
this chapter.
This category also suits Barthes’ (1977) term relay, according to which
“text and picture do not ‘say the same thing’ but convey different, com-
plementary content” (van Leeuwen, 2011, p. 657). To some extent, this
category exhibits the qualities of the social semiotics term distribution
(juxtaposed pictures and text jointly constructing information) and diver-
gence (the meanings of text and picture contradict each other and convey
new information out of this contradiction). Besides, the category is
related to what Jewitt (2016) calls multimodal ensemble, where all the
modes combine to convey a message’s meaning. The information is dis-
tributed across modes, and “any one mode in that ensemble is carrying
part of the message only: each mode is therefore partial in relation to the
whole of the meaning” (p. 73).
The most frequent text-picture interaction in this category of memes is
that in which the information from the picture invalidates, to a greater or
lesser extent, the information provided by the text. In (17), for example,
access to the intended ironical-critical remark is facilitated by the picture
of Keanu Reeves with a lunatic look on his face:
(17) Top text: What if Obama needs all of our personal information
Picture: Lunatic-looking picture of actor Keanu Reeves.
Bottom text: because he’s setting up a matchmaking service and
wants us all to find love?
Besides, a typical function of text-picture combination is to trigger the

derivation of implications from text, with the help of a connoted picture.
Consider the following memes (18–20):
(18) Top text: Join the marines they said

Picture: Picture of a marine holding an umbrella over presi-
dent Obama.
Bottom text: You’ll be a hero they said.
122 F. Yus
(19) Top text: Wants a girl who sees him for his personality
Picture: Face of a very ugly man.
Bottom text: if she’s hot.
(20) Top text: Ok class, you have exactly 50 minutes to complete
this exam
Picture: Picture of a beautiful young female teacher in front
of blackboard and with a wide smile on her face.
Bottom text: but before you begin, let’s all take five minutes to cor-
rect the mistakes I made while writing the test.
In (18), the activity depicted in the picture contradicts the presumption

of honor and pride that the soldier should feel for being a marine. In
(19), the picture of an utterly ugly man contradicts the extent of his
demands for a specific type of woman made in the text of the meme.
Finally, in (20) the picture of an apparently happy teacher who seems to
love her profession is at odds with the information of the text concerning
her (in)ability to prepare well-written exams.
Needless to say, the most radical case of text-picture relationship
yielding a different meaning for the text of the meme from the one
obtained from the partial meanings of text and picture occurs when
the picture forces the opposite interpretation to the one obtained from
the text. A good example is the series of memes sharing the same pic-
ture of two girls laughing. This picture triggers a radically different
interpretive course for the texts, invariably related to sex role-connoted
information, to the extent that their interpretation is reversed. An
example is provided in Fig. 4.2, where the girls laughing invalidate
and force a reverse meaning of the accompanying text (“And then I
said I don’t care about your money, I love you for who you are”).
Other instances using this picture of the laughing girls include the
texts in (21–22):
(21) Top text: And then I told him.

Bottom text: Size doesn’t matter.
(22) Top text: And then I said.
Bottom text: I’ll be ready in 5 minutes.
Fig. 4.2 Reversed interpretation
Finally, some pictures also trigger incongruity when the user reads the
text of the meme. The role of the picture is essential here in provoking the
entertainment of two competing interpretations of some ambiguous part
of the text, thus fitting into the category of text-picture combination (the
eventual parallel entertainment of two interpretations is not possible
from text only, but the information from the picture is also required for
this to happen). In relevance-theoretic terms, this simultaneity of inter-
pretations goes against the usual processing strategy, which would nor-
mally be directed at the most accessible—relevant—interpretation of that
bit of text while dismissing, often in an unconscious way, other alterna-
tive interpretations. In fact, when we engage in an inferential choice of an
interpretation, we are generally not aware of other less accessible interpre-
tations once the relevant one has been selected.
By contrast, in some of the memes fitting into this category, the reader
is forced to entertain both senses of some portion of text simultaneously.
The additional mental effort involved in this dual processing has to be
compensated for by extra (or different) cognitive effects: an offset of
humorous effects, or amusement at acknowledging the cunning play
with words, for instance. Meme (23) plays with two simultaneous senses
of credit:
(23) Top text: Gets one of only 12 perfect scores in the world on
macroeconomics.
Picture: Picture of a swot- or nerd-looking student.
Bottom text: Goes to Harvard and gets no credit.
124 F. Yus
In other memes in this category, the text is altered in order to create an

ambiguity that mirrors the accompanying picture and, as a consequence,
the text and the ambiguity arising from it are utterly connected to the
picture, as in (24–25):
(24) Top text: Bitches.

Picture: Portrait of musician Handel.
Bottom text: Can’t Handel my oratorios.
(25) Top text: France adopted a new
Picture: Picture of France’s President Macron.
Bottom text: “Macron” economic policy.
4.6 Meme Specificity: Ad Hoc Visual Referent

Adjustment
For comics, McCloud (1994) proposed a category, montage, which is
absent in memes. Similarly, there is an inferential strategy that seems to
be inherent to memes, which will be labeled ad hoc visual referent adjust-
ment, due to its resemblance with verbal concept adjustment. Through
this inferential strategy, the different texts used for the very same picture
constrain the meanings of the picture, a kind of modal affordance (Jewitt,
2016, p. 72), in which the final interpretation of one of the modes is
constrained by the other mode.
“Concept adjustment”, as proposed by relevance theory (Sperber &
Wilson, 1995; Wilson & Carston, 2006), has already been mentioned
in this chapter in passing; but so far, the term has only been applied to
verbal communication in the relevance-theoretic bibliography. As
summarized in Yus (2010), during the interpretation process, the pro-
totypical concept encoded by a word is adjusted by the hearers so that
it meets their expectations of relevance. The outcome of this adjust-
ment is an ad hoc concept that is similar, but not identical, to the sta-
bilized concept coded by the word. They are ad hoc “because they are
not linguistically given, but are constructed online in response to spe-
cific expectations of relevance raised in specific contexts. There is a
difference then between ad hoc concepts, accessed by a spontaneous
process of pragmatic inference, and lexicalized concepts, which are

context-invariant” (Carston, 2002, p. 322).
In certain contexts, the concept that the speaker intends to communi-
cate is broader (i.e. less exact) than the concept that the word he/she has
chosen literally communicates (or codes), as in (26a–e):
(26) a. There is a rectangle of lawn in the shed.

[not an exact rectangle].
b. We entered a pub, but we left since it was empty.
[there were people in the pub—for example, the waiter—but not
interesting people]
c. I’ve got a thousand things to do this morning.
[many things, but not a thousand].
d. Don’t worry. I’ll be ready in two minutes.
[in a while, surely longer than two minutes].
e. This steak is raw.
[not literally raw, but undercooked].
On other occasions, the concept that the speaker intends to communi-

cate is narrower (i.e. more exact) than the concept that the word he/she
has chosen literally communicates (or codes), as in (27a–e):
(27) a. I’ve got nothing to wear for the party.

[specifically, nothing appropriate, nothing classy, etc.].
b. María has a brain.
[specifically, an outstanding brain, not simply a brain; she is very
intelligent].
c. This boy has a temperature.
[specifically, more temperature than he should].
d. It will take some time to fix this car.
[specifically, longer than you imagine; longer than it would nor-
mally take].
e. Antonio drinks too much.
[specifically, he drinks too much alcohol].
The innovative inferential concept-related strategy found in memes is

that, besides the typical concept adjustment required for verbal content
(involving narrowing or broadening), sometimes the user also has to
126 F. Yus
engage in an inferential concept adjustment but this time applied to the

referent of the picture in the meme. This is what happens in several series
of memes in which the picture is the same across memes and the users
change the accompanying text in humorous ways. These texts trigger
slightly different interpretations of the prototypical referent of the pic-
ture, and therefore the inferential strategy leading to these slightly differ-
ing interpretations may be called ad hoc visual referent adjustment.
Users are aware that there are many different texts for the same picture
in a series of memes and know that sometimes the relationship between
picture and text in the meme is not straightforward. A frequent example
is the memes with a picture of a person making a certain gesture. Upon
finding an instance of this series of memes, the user will infer the gesture
portrayed in the picture, but not simply as the typical (i.e. default) inter-
pretation of the gesture, but with interpretive variations depending on
the text that is above and below the picture, a proper case of adjustment
but this time applied to the referent of the gesture depicted. On some
occasions, the resulting ad hoc visual referent will be narrower than the
prototypical, default one provided by the picture (as happens with verbal
adjustment); on other occasions, the user will infer a broader referent
than the default one in the picture. The outcome will be, in both cases,
an ad hoc visual referent that meets the user’s expectation of relevance
upon interpreting the meme, and which works specifically for this meme
and not for other picture-text relationships in the other memes which
belong to the same series (repeated picture with different texts).
Take, for instance, the memes depicted in Fig. 4.3. The prototypical
gestures depicted in the picture (the little child’s gesture and hand move-
ment) are interpreted, in general, as the child being proud of having
achieved some goal. However, this default, prototypical interpretation
has to be adjusted (narrowed or broadened) so as to infer correctly the
kind of feeling enacted by the actions described in the text that are placed
next to the same picture in those memes. Hence, the ad hoc visual referent
inferred for the gesture will be different when accompanied by the text
Finished homework in class (more like “sheer pride”), or by the text
Divorced parents. Twice the Xmas presents (roughly “intense joy”). In all of
these instances, the user has to adjust the referent of the gesture according
to the accompanying text, even though the gesture remains the same.
Fig. 4.3 Adjusted referents
However, there is a limit to this inferential strategy of visual referent

adjustment. In the same series, there are certainly memes that fit into the
category of ad hoc visual referents from the coded prototypical gesture,
because they resemble the gesture but are not exactly the same as that
prototypical one coded by the gesture. However, it can be stated that
there is a point beyond which the text no longer aids in visual referent
adjustment, but triggers a different referent altogether. An example is the
series of memes sharing the picture of a woman in tears (some instances
of this series have already been quoted in this chapter). In one of the
instances, (28), the user will have no problem in broadening the scope of
the visual referent for the woman in tears so as to include an unusual
source of the tears due to the information communicated by the text:
(28) Top text: I’m so tired of eating

Picture: Tears running down a woman’s face.
Bottom text: at all of the restaurants near work.
In other memes from this series, the picture is the same but the extent of
the pain suffered by the woman has to be adjusted depending on the
accompanying text. In (28), the user can indeed imagine a situation in
which the woman is on the verge of a breakdown due to her being forced
to eat at those locations. However, there is a limit to adjustment by
resemblance between the default visual referent of the coded nonverbal
action and the ad hoc visual referent, beyond which, what the user ends
128 F. Yus
up inferring, is a humorous clash between a serious nonverbal action

(crying) and the text that, in theory, should provide the user with the
reason for supporting its generation, as is the case with memes (29–30):
(29) Top text: I’m so thirsty

Bottom text: but if I drink, I’ll have to get up to go pee.
(30) Top text: I’m hungry
Bottom text: but I already brushed my teeth.
4.7 Concluding Remarks

In this chapter, a corpus of 100 memes of the “image macro family” has
been analyzed in their text-picture combinations and especially concern-
ing the pragmatic implications of these combinations. The default tax-
onomy of categories has followed McCloud’s (1994) proposal for comics
(a medium very close to memes in its verbal-visual semiotic quality),
yielding the following instances of memes: (a) word specific (25 memes),
(b) picture specific (7), (c) duo specific (1), (d) additive (37), (e) parallel (0),
(f ) montage (0), and (g) interdependent (30).
It comes as no surprise that those categories for comics that “separate”
the meanings conveyed by text and picture are not present in the corpus.
Image macro memes such as the ones analyzed here are meant to com-
municate information by combining the partial meanings provided by
text and picture into a slightly or radically different interpretation. In
general, the analysis revealed that texts acquire prominence in the even-
tual interpretation and pictures either illustrate what the text communi-
cates (first category, 25 instances), amplify or elaborate the textual
meaning (fourth category, 37 instances), or aid in altering the textual
meaning, often radically (seventh category, 30 instances).
Furthermore, although most of the interpretive procedures are easy to
explain from a pragmatic point of view, a new inferential strategy has been
found to be inherent to meme communication: ad hoc visual referent
adjustment, similar to the relevance-theoretic proposal of ad hoc concept

adjustment for verbal communication but innovative, at the same time,
considering that this is the first time that such inferential strategy is
applied to visual inputs.
Acknowledgments This chapter is part of the research project PROMETEO

2016/052, titled “Humor gender: Observatory of identity of women and men
through humor”, funded by the Generalitat Valenciana, Conselleria d’Educació,
Investigació, Cultura i Sport.
References
Barthes, R. (1977). Image. Music. Text. London: Fontana.
Carston, R. (2002). Thoughts and utterances. Oxford: Blackwell.
Chan, E. (2011). Integrating visual and verbal meaning in multimodal text
comprehension: Towards a model of intermodal relations. In S. Dreyfus,
S. Hood, & S. Stenglin (Eds.), Semiotic margins: Meaning in multimodalities
(pp. 144–167). London: Continuum.
Cohn, N. (2013). Beyond word balloons and thought bubbles: The integration
of text and image. Semiotica, (197), 35–63. https://doi.org/10.1515/
sem-2013-0079
Conradie, M., Brokensha, S., & Pretorius, M. (2012). No small irony: A dis-
course analysis of Zapiro’s 2010 World Cup cartoons. Language Matters,
43(1), 39–59. https://doi.org/10.1080/10228195.2011.649777
Dawkins, R. (1976). The selfish gene. Oxford: Oxford University Press.
Dynel, M. (2016). I has seen image macros!’ Advice animal memes as visual-
verbal jokes. International Journal of Communication, 10, 660–688.
El Refaie, E., & Hörschelmann, K. (2010). Young people’s readings of a polit-
ical cartoon and the concept of multimodal literacy. Discourse: Studies in
the Cultural Politics of Education, 31(2), 195–207. https://doi.
org/10.1080/01596301003679719
Forceville, C. (2014). Relevance theory as a model for analyzing visual and mul-
timodal communication. In D. Machin (Ed.), Visual communication
(pp. 51–70). Berlin: Mouton de Gruyter.
Forceville, C., & Clark, B. (2014). Can pictures have explicatures? Linguagem
em (Dis)curso (special issue on relevance theory), 14(3), 451–472. https://
doi.org/10.1590/1982-4017-140301-0114
130 F. Yus
Gill, T. (2002). Visual and verbal playmates: An exploration of visual and verbal
modalities in children’s picture books. Unpublished BA (Hons) thesis, University
of Sydney.
Jewitt, C. (2016). Multimodal analysis. In A. Georgakopoulou & T. Spilioti
(Eds.), Handbook of language and digital communication (pp. 69–84).
Abingdon: Routledge.
Kardaş, T. (2012). No laughing matter: Visualizing Turkey’s Ergenekon in polit-
ical cartoons. Middle East Critique, 21(2), 203–223. https://doi.org/10.1080
/19436149.2012.688587
Kress, G., & van Leeuwen, T. (2006). Reading images: The grammar of visual
design (2nd ed.). London: Routledge.
Laineste, L., & Voolaid, P. (2016). Laughing across borders: Intertextuality of
internet memes. European Journal of Humour Research, 4(4), 26–49. https://
doi.org/10.7592/EJHR2016.4.4.laineste
Lim Fei, V. (2004). Developing an integrative multi-semiotic model. In
K. O’Halloran (Ed.), Multimodal discourse analysis: Systemic functional per-
spectives (pp. 220–246). London: Continuum.
Martinec, R., & Salway, A. (2005). A system for image-text relations in new
(and old) media. Visual Communication, 4(3), 337–371. https://doi.
org/10.1177/1470357205055928
McCloud, S. (1994). Understanding comics: The invisible art. New York: Harper
Collins.
Nissenbaum, A., & Shifman, L. (2017). Internet memes as contested cultural
capital: The case of 4chan’s /b/ board. New Media & Society, 19(4), 483–501.
https://doi.org/10.1177/1461444815609313
Salway, A., & Martinec, R. (2002). Some ideas for modelling image-text combina-
tions. Guildford: University of Surrey, Department of Computing.
Sarapik, V. (2009). Picture, text, and imagetext: Textual polylogy. Semiotica,
174(1/4), 277–308. https://doi.org/10.1515/semi.2009.03
Segev, E., Nissenbaum, A., Stolero, N., & Shifman, L. (2015). Families and
networks of Internet memes: The relationship between cohesiveness, unique-
ness, and quiddity concreteness. Journal of Computer-Mediated
Communication, 20, 417–433. https://doi.org/10.1111/jcc4.12120
Shifman, L. (2014). Memes in digital culture. Cambridge, MA: The MIT Press.
Sperber, D., & Wilson, D. (1995). Relevance: Communication and cognition
(2nd ed.). Oxford: Blackwell.
Stöckl, H. (2004). In between modes: Language and image in printed media. In
E. Ventola, C. Charles, & M. Kaltenbacher (Eds.), Perspectives on multimo-
dality (pp. 9–30). Amsterdam: John Benjamins.
Taecharungroj, V., & Nueangjamnong, P. (2014). The effect of humour on viral-

ity: The study of Internet memes on social media. Paper presented at 7th
International Forum on Public Relations and Advertising Media Impacts on
Culture and Social Communication, Bangkok, August.
Trifonas, P. P. (2015). Text and images. In P. P. Trifonas (Ed.), International
handbook of semiotics (pp. 1139–1152). Berlin: Springer. https://doi.
org/10.1007/978-94-017-9404-6_53
Tsakona, V. (2009). Language and image interaction in cartoons: Towards a
multimodal theory of humor. Journal of Pragmatics, 41, 1171–1188. https://
doi.org/10.1016/j.pragma.2008.12.003
van Leeuwen, T. (2006). Word and image. In K. Brown (Ed.), Encyclopedia of
language and linguistics (2nd ed.). Amsterdam: Elsevier.
van Leeuwen, T. (2011). Multimodality. In J. Simpson (Ed.), The Routledge
handbook of applied linguistics (pp. 668–682). Abingdon: Routledge.
van Leeuwen, T. (2015). Multimodality. In D. Tannen, H. E. Hamilton, &
D. Schiffrin (Eds.), The handbook of discourse analysis (2nd ed., pp. 447–465).
Oxford: Wiley Blackwell.
Wharton, T. (2009). Pragmatics and non-verbal communication. Cambridge:
Cambridge University Press.
Wilson, D., & Carston, R. (2006). Metaphor, relevance and the ‘emergent
property’ issue. Mind & Language, 21(3), 404–433. https://doi.
org/10.1111/j.1468-0017.2006.00284.x
Wu, S. (2014). A multimodal analysis of image-text relations in picture books.
Theory and Practice in Language Studies, 4(7), 1415–1420.
Yus, F. (1997). La teoría de la relevancia y la estrategia humorística de la
incongruencia-resolución. Pragmalingüística, 3–4, 497–508.
Yus, F. (2010). Relevance theory. In B. Heine & H. Narrog (Eds.), The Oxford
handbook of linguistic analysis (pp. 679–701). Oxford: Oxford University Press.
Amsterdam: John Benjamins.
Yus, F. (2013a). An inference-centered analysis of jokes: The Intersecting Circles
Model of humorous communication. In L. Ruiz Gurillo & B. Alvarado
(Eds.), Irony and humor: Highlights and genres (pp. 59–82). Amsterdam: John
Benjamins.
Yus, F. (2013b). Analyzing jokes with the Intersecting Circles Model of humor-
ous communication. Lodz Papers in Pragmatics, 9(1), 3–24. https://doi.
org/10.1515/lpp-2013-0002
Yus, F. (2016). Humour and relevance. Amsterdam: John Benjamins.
Yus, F. (2017). Incongruity-resolution cases in jokes. Lingua, 197, 103–122.
https://doi.org/10.1016/j.lingua.2017.02.002
5
Digital and Written Quotations
in a News Text: The Hybrid Genre
of Political Opinion Review
Marjut Johansson
5.1 Introduction
In the recent past, there have been major transformations in news produc-
tion due mostly to digital culture. Since news has gone online, the bound-
aries between broadcast media and printed press have become more and
more blurred, and, today, hybridity is considered one of the central ele-
ments of contemporary media systems (Chadwick, 2013). These changes
affect journalism in many ways, including the manner in which newswrit-
ing takes place (Deuze, 2017). At the same time, social media has become
an important news source and participatory space for journalists (Broersma
& Graham, 2013; Hermida, 2013). This interconnectedness between news
media and social media is reflected in news texts themselves as well as in
how they are composed. The nature of news texts has also changed due to
the characteristics of online texts, whose hypertextuality and multimodality
M. Johansson (*)
Department of French Studies, School of Languages and Translation Studies,
University of Turku, Turku, Finland
e-mail: Marjut.johansson@utu.fi

https://doi.org/10.1007/978-3-319-92663-6_5
134 M. Johansson
connect them to other semiotic m aterials (Chovanec, 2014, p. 1). In addi-

tion, some news texts are composed using the practice of “copy and paste”,
which is a common affordance of contemporary text production (Adami,
2012a, 2012b; Johansson, 2013). Despite these changes, professional news
discourse comprises mediated public discourse, as well as institutional pub-
lic discourse produced by media professionals.
Journalists lead news production: they shift back and forth in their
production of news when looking for material and embedding it into a
new textual context. This process is that of mediation, dependent upon
“the material processes (objects, linkages, infrastructures, platforms)
through which communication, and the construction of meaning, take
place” (Couldry & Hepp, 2016, p. xiii). It can be considered as a process
of entextualization, recontextualization, decontextualization, and retell-
ing (Bauman & Briggs, 1990; Catenaccio et al., 2011; Silverstein &
Urban, 1996). However, in the process of creating actual news texts, a
journalist is constrained by the expectations of the news genre and its way
of representing the meaning of events and objects in the world.
Online news texts are digital texts. They can contain multiple semiotic
modes of presentation, hypertextual links, and algorithm-based techno-
logical affordances for user participation (Johansson, 2014a). A written
news text may not even be the primary news text; this role may now be
assumed by other types of semiotic modes: videos, interactive maps, fig-
ures, polls, slideshows, and so on. News may also include materials from
social media, such as tweets and Facebook posts. This adds to the com-
plexity of the semiotic meaning of digital texts (see Adami, 2012a, 2012b;
Adami & Kress, 2014; Kress, 2010).
This chapter focuses on digital quotations, namely, tweets and a video.
I will study how they are embedded into a news text. Digital quotations
are materials that are copy-pasted from online sources and inserted into a
digital text. In news texts, quotations “convey events, facts, opinions, and
views” from various sources (Johansson, 2013, p. 99). I will study how
quotations are used in one type of news genre. Thus, the analysis will look
at the interplay between digital quotations, ordinary quotations, and the
news story written by the journalist in a news text; this represents a novel
type of digital news genre that I call opinion review (see Johansson, 2013).
Opinion reviews are related to the press review genre. Press review is a
genre through which newspapers participate in opinion formation in the
Digital and Written Quotations in a News Text: The Hybrid… 135
public sphere, even though they do not express their own views (Erbe,
2007, p. 79; Johansson, 2013, p. 99). Press reviews are based on quota-
tions from other newspapers and media, thus linking different public
arenas in the mediated public sphere (Erbe, 2007, p. 79). They are meta-
discourse on current topics in society, forming intertextual and interdis-
cursive links between news stories and other discourses to which they
refer (Claquin, 1993; Erbe, 2007; Johansson, 2013, p. 100). In printed
newspapers, press reviews have traditionally relied on textual quotations,
but on online news sites, the video press review is a new version of the
traditional press review (Johansson, 2013). Both kinds of press review are
based on the journalistic work of selection and interpretation, but the
video press review mixes video clips into a storyline, producing a hybrid
media genre, as it shows the opinions of the quoted persons as a struggle
in the public arena (Johansson, 2013, p. 112).
The objective of this chapter is to examine the opinion review genre
through a case study of a digital news text that combines a written story-
line with quoted material from other media, including social media,
when commenting on a sociopolitical event. I will pay special attention
to how tweets and a video are used as digital quotations and how the
journalist integrates them within the news story. My theoretical approach
is based on digital discourse analysis, which includes an interdisciplinary
view on media studies in the digital context. This will allow an examina-
tion of how discursive and social processes take place in digital media
(Couldry, 2012, p. 8). In addition, I will rely on sociopragmatics in the
analysis of quotations.
This chapter is structured as follows. In Sect. 5.2, I will discuss the
news genres in digital culture. Then, in Sect. 5.3, I will turn the discus-
sion toward digital news texts and quotations. In Sect. 5.4, I will present
the data and methods. I will then analyze one news text in Sects. 5.5 and
5.6. The chapter ends with a few concluding remarks, in Sect. 5.7.
5.2 Digital News Genres and Multimodal Text

How can news genres and news texts be approached? The characteristics of
news discourse and news genres have been the object of linguistic research
since the first studies on written news structure and story production
136 M. Johansson
(Bell, 1991; van Dijk, 1988). News interviews and political interviews
have been studied in sociopragmatics and conversation analysis (Blum-
Kulka, 1983; Clayman, 1992), while the complexity and hybridity of
written or spoken media texts have been widely acknowledged in critical
discourse analysis very early on (Fairclough, 1992, 1995).
Newspapers have now been online for some time. For their part, social
media, such as Facebook and Twitter, have become important news spaces
as well. Social media platforms such as these are changing how news is
published, especially in cases of rapidly unfolding events (Hermida,
2013). Journalists have embraced social media in different ways, fre-
quently incorporating social media as a news source. Journalists may have
active Twitter accounts on which they publish news stories and partici-
pate in discussions, thus creating a hybrid space for public communica-
tion between professionals and individuals from the private sphere
(Giaxoglou, 2017; Hermida, 2013, p. 304). In addition, online newspa-
pers contain multimodal news texts as well as sections devoted to videos.
More recently, news texts and the processes of news production, especially
newswriting, have become the focus of linguistic scrutiny (e.g., Bednarek
& Caple, 2012; Perrin, 2013). In addition, the changing nature of news
texts that combine different types of material within genres has been the
object of study in various domains (Chadwick, 2013; Johansson, 2012,
2013; Mast, Coesemans, & Temmerman, 2016). As my objective is to
study how quotations are used in a specific news genre, I will first discuss
some relevant notions related to digital genres and texts.
Here, the notion of communicative genre—or the parallel notion of
communicative activity type—is considered an organized form in which
participants communicate about social life (Bergmann & Luckmann,
1995; Levinson, 1992; Linell, 1998; Luckmann, 1989, p. 162). A com-
municative genre includes participants and their objectives related to
their social and discursive roles in situated actions and activities; impor-
tantly, it constrains the way interaction and discourse take place (Levinson,
1992; Linell, 1998, p. 240). Digital genres are communicative genres that
are based on mediated interaction and discourse among participants in
digital contexts. They are realized in digital texts that have a sequential
discursive organization, multisemiotic characteristics, and technological
affordances, such as hypertextuality and algorithm-based activities that
enhance user experience. I use the term genre here for brevity, while I use
text to designate how genres are instantiated via different semiotic modes.
Genres can be studied on different levels—macro, meso, and micro.
When genres are considered as multilayered phenomena, research focuses
on different practices and contexts (Fairclough, 1995; Fetzer, 2004).
These distinctions allow analysts to focus on certain processes at a given
time. From the macro-level perspective, news genres are interlinked with
various communicative genres in other areas of social life. For example,
Gruber (2013) analyzed complex genre systems between Austrian parlia-
mentary genres and their coverage by media genres. He showed how par-
liamentary talk was taken up in a series of interconnected political and
media genres (Gruber, 2013). At the meso level, a genre is examined in
its situational context. It can be approached from the perspective of par-
ticipation as well as from how social actors position themselves in situa-
tions through the negotiation of meaning. At the micro level, genres are
analyzed from the perspective of their linguistic context. This study tack-
les the two latter, meso and micro, levels. There, sequential actions and
types of textuality can be studied. In addition, social actors’ stances can
be examined via the ways in which they evaluate objects in the world.
News stories are stories, according to the terminology used in the news-
rooms (van Leeuwen, 2008, p. 349). From a linguistic perspective, a nar-
rative is a textual sequence that is organized chronologically into different
phases. As discourse, a story usually refers to past events and constructs
how these events unfolded (see, for example, Adam, 1992; van Leeuwen,
2008, p. 349). Some news texts, especially news reporting on accidents
and breaking news, may follow a narrative organization. However, not all
news genres are organized as narratives. Instead, news genres are of
numerous types. First, they vary from traditional printed press genres,
such as news reports and editorials, to novel forms of digital news genres,
several of which have been adopted from social media, such as blogs and
discussion forums. Video is a semiotic mode, but video content can be
organized into different genres. News videos may represent traditional
TV news genres, such as reports, announcements, and interviews
(Johansson, 2012, pp. 52–53).
When news genres are examined from the perspective of the type of
journalistic communicative activity on which they are based, they also
138 M. Johansson
vary. Very often, journalistic positioning is referred to as neutralistic,

which is based on the use of quotations (Clayman, 1992; Haapanen,
2016), but this description is not sufficient and needs further examina-
tion. For example, news reports and editorials differ, as the former are
based on reporting activity and the latter on commentary. In reporting,
traditionally, journalists report in an objective manner on events that
have taken place; in commenting, on the other hand, they engage in the
evaluation of events or objects in the world. A breaking news story is an
example of reporting, while editorials and columns are examples of com-
menting. In other words, in commenting, journalistic positioning may
be subjective in various ways. According to Moirand (2007, p. 12), objec-
tivity and subjectivity are not fixed in news genres, but they are dynamic.1
Subjectivity can be marked or unmarked in news texts. For instance, edi-
torials are opinion based, but journalists rarely refer to themselves using
personal pronouns; rather, they express their opinions implicitly (see Le,
2010). In contrast, blogs belong to a genre that allows journalists and
experts to express their views with explicit subjective marking by using
first-person pronouns and other devices that explicitly refer to the author.
Regarding traditional press reviews, which are built entirely on quota-
tions, they represent an objective kind of news genre in which journalistic
positioning is based on the detachment of responsibility (Claquin, 1993,
p. 45). The video press review is not similar, however, as the constructed
storyline reflects the implicit subjective positioning of the journalist
(Johansson, 2013).
How do digital news texts differ from printed news texts? If based on
text only, they may be identical in the printed and online versions.
However, in many cases, digital news texts vary because of their multimo-
dality and hypertextuality. First, digital texts are easy to construct. In her
analysis, Adami (2012b, pp. 28–32) referred to text sharing and remixing
as new forms of the “copy-paste” technique for creating texts. Text shar-
ing involves selection and recontextualization; that is, “editing of the
selected snippets to ‘fit’ the new text” (Adami, 2012b, p. 28). This prac-
tice includes selecting and taking a textual-visual object out of its original
context and inserting it into a new one, resulting in a remix (Adami,
2012b, pp. 28–30). In these kinds of processes, resemiotization takes
place. According to Jones (2009, p. 298), “Digital entextualization is
more immediate, negotiable, communal, and interactive than other forms

of ‘disembodied’ entextualization like writing and analog photography.”
Several approaches to multimodality have explored the visuality and
combination of semiotic modes (Adami, 2016; Kress, 2010; Kress &
Leeuwen, 2008, forthcoming for a review of approaches). News texts
have been analyzed from the perspective of how written texts and moving
images complement each other. Bednarek and Caple (2012) examined
the relationship between the verbal and visual aspects of videos in terms
of how written or visual resources overlap or differ in describing the same
events (Bednarek & Caple, 2012, pp. 121–124). News videos can be
independent news stories with reports, clips of speeches, press confer-
ences, or videos capturing live events integrated within written news
articles (Johansson, 2012, pp. 62–63). A video can be, textually, a mash-
up (or a remix), but videos very often contain only short sound bites that
are used as quotations. In sum, digital news genres are not stable, because
different semiotic modes may have an effect on how a news genre is real-
ized (Johansson, 2014a). The notion of quotation will be discussed in the
next section.
5.3 Q
uotation
5.3.1 Quotation as a Pragmatic and Discursive Object
Here, and from a dialogical perspective, a quotation is considered a prag-

matic object. In my previous work, I refer to quotations as represented
discourse and define this notion in the following way:
[Represented discourse is] a form of mediated action that can be studied

from both a social and individual perspective: a speaker acting as a social
actor who constructs instances of represented discourse is simultaneously
using represented discourse as a cultural and linguistic tool, which then
forms a link between mental action, interaction and action situated in a
social context. The basis of this—cultural tools functioning as mediation—
can be found in Wertsch (1998). (Johansson, 2002, p. 255)
140 M. Johansson
In a more recent definition, quoting is described as a meta-communi-

cative act “of re-contextualizing, re-focusing and reflecting upon a prior
communicative occurrence as constitutive for quoting” (Bublitz, 2015,
p. 4). It involves the process of shifting source text from one context and
then recontextualizing it into ongoing discourse, which may display a new
perspective that includes a possible evaluation (Bublitz, 2015, p. 4). For
example, in a media interview situation, an interviewer can use quotative
clauses that refer to the ongoing situational context (you said a moment
ago), a displaced context (last week, you said on a television talk show), or an
imaginary context (if you say) (Johansson, 2002, 2013). In general, recon-
textualization can be an utterance or a text, but it can be other semiotic
material as well. In this view, a quotation is a multilayered phenomenon.
In general terms, a quotation recontextualizes something from different
contexts. When recontextualized, the quotation is embedded into the new
context; the different levels of quotation are shown in Table 5.1.
Thus, a quotation is a polyphonic linguistic object as it contains the
voice(s) of a social actor or social actors that a writer or speaker embeds
in her or his discourse. In text or speech, a quotation is placed in the
interactional and discursive sequence of actions. Quotations have genre-
specific functions, such as narrative or argumentative functions, depend-
ing on the genre within which they are integrated (Johansson, 2002,
2014b). In the ongoing discourse, a quotation creates relations with
social actors, their texts, and contexts. It also creates relations with other
social actors, and it forms intra- and intertextual as well as interdiscursive
relations with other texts (Johansson, 2002). Furthermore, a quotation
initiates and frames a negotiation of meaning in the new context.
Table 5.1 Quotation as a dialogic object (modified from Johansson, 2002, p. 256)
Quotation is a form of recontextualization in which a social actor constructs
The linguistic object A relation with

• Real or imagined discourse ↔ • Texts and other social actors and their
(speech or thought) texts in other contexts
• (Explicit) voice • Intra- and intertextual and
interdiscursive chaining
• In sequential discursive and ↔ • Negotiation of communicative
interactional activity activities with the co-participant(s)
Each utterance contains a certain type of stance toward the information

expressed that can be analyzed taking linguistic cues into consideration. In
stance-taking, social actors engage in evaluating something, and this con-
cerns the meta-communicative act of quoting as well (Bublitz, 2015;
DuBois, 2007, p. 163). When evaluating, social actors position them-
selves discursively in the situation, showing alignment or disalignment
with other social actors (Davies & Harré, 1990; DuBois, 2007, p. 163;
Weizman, 2008, p. 14). In the next section, the function of quotations in
digital news texts will be discussed.
5.3.2 Quotation and Digital News Texts
The fact that quotations are an integral part of news texts has been well
documented in past research. In informative news texts, quotations are
constructed from a variety of sources, such as interviews, documents, and
so on, where they carry out the function of truthfulness, reliability, and
media interest (Waugh, 1995). According to Charaudeau (1997,
pp. 184–185), quotations are used to introduce basic facts, eyewitness
evidence, expert knowledge or decisions, and opinions into news dis-
course. Quoting is a basic journalistic practice that involves recontextual-
ization. There can be several different sources from which a journalist
decontextualizes a quotation. In an interview situation, a journalist
extracts quotations from interviewees and then inserts them into a writ-
ten article, steering the storyline of the news (Haapanen, 2016). By means
of quoting, a journalist maintains distance from people and the opinions
they convey in news texts, thus respecting the principle of neutrality (see
Haapanen, 2016). However, quoting is different in digital texts, as copy-
pasting makes it easy. In addition, the mere posting of a link can be one
kind of quote that leads to new content (Haapanen & Perrin, 2017).
Recently, Landert (2015) analyzed speech representation in online
news, while quotations have also been studied in other digital genres,
especially in retweeting (Puschmann, 2015) and discussion forums
(Arendholz, 2015). Johansson (2012, 2014b) identified two different
functions for the videos used in online written news texts; these may
function as parallel news stories or as quotations. This latter is the case
with sound bites, announcements, and interview extracts (Johansson,
142 M. Johansson
2012). In their study of several thousand tweets of Dutch and British

newspaper articles, Broersma and Graham (2013) claimed that tweets are
used as quotes in news texts. They found that tweets had a variety of
quoting functions, such as flavoring a story, adding a personal note, or
illustrating or triggering a news story (Broersma & Graham, 2013,
pp. 456–457). In other words, this line of research is starting to identify
how digital quotations are used in news texts and what the functions of
these quotations are. In this study, my aim is to examine the opinion
review news genre. Taking into consideration the study of video press
reviews (Johansson, 2013), I hypothesize that an opinion review based
mainly on quotations is a similar hybrid genre, affected by the practices
of digital text production. The analysis aims to answer the following
research questions, which are based on the description given in Table 5.1:
1. What are the characteristics of the quotations regarding the source

and the quoted voice, and how are they embedded into the sequential
organization of the news text? This will provide an answer as to what
kind of relation the journalist builds with different texts and social
actors in other contexts.
2. What kind of communicative function do quotations carry out in an
opinion review and how does the news story integrate them? This will
provide an answer regarding how the construction of meaning in the
news text takes place.
3. How do quotations affect and construct the news genre of the opinion
review? This will provide an answer regarding how this type of genre is
constructed.
In the next section, I will present the online news text selected for the
analysis, its background, and my analytical approach.
5.4 Data and Methods

This study is primarily based on the ethnographic observation of digital
news texts from August 2010 to August 2013. During this period, a cor-
pus of digital news texts was compiled, and several prior studies con-
ducted based on it (Johansson, 2012, 2013, 2014a, 2014b). The news
text selected for this analysis is called “Les réactions à la réforme des
retraites ne sont pas fait attendre” (“The reactions to retirement reform
did not take long to appear”). It was published by the French Huffington
Post (http://www.huffingtonpost.fr/; 27.8.2013) at a time when reforms
of retirement plans were being discussed in France. Because the article
was not attributed to a specific journalist, it will be referred to as HP in
this chapter. The article length is 621 words and its translation is pro-
vided in the Appendix.
This article comments on a typical situation in French politics: a gov-
ernmental decision and its public reaction. This time the decision was the
very controversial retirement reform being introduced by the socialist
government in the summer of 2013. The decision was opposed by other
political parties as well as by trade unions. One main argument in the
debate was how much workers should pay for their social security with
respect to their retirement. The analysis focuses on how the news text
quotes these reactions.
The analysis was conducted on two levels: micro and meso (see Sect.
5.2). First, the linguistic analysis examined the types of quotations (Sect.
5.5.1). The analysis considered all the quotations found in the news text
in regard to their media source and voices, that is, quoted persons and
their political or institutional background. This was followed by a multi-
modal analysis: that is, how the text’s textual, visual, and linguistic mate-
rial was sequentially organized and how the quotations were integrated
into it (Sect. 5.5.2). The second research question was answered by an
assessment of quotations and their discursive meaning based on what
type of communicative function they carried out through stances and
positioning in quotations and in the news story (Davies & Harré, 1990;
DuBois, 2007; Weizman, 2008; White, 2012). Stance refers to utterance-
level expressions of evaluation, while positioning is determined by the
communicative act the social actors express through their evaluations,
which shows alignment or disalignment with the governmental decision
(Sect. 5.6.1). Lastly, the digital news text was discussed as a news genre in
the media context in order to answer the third research question. Here,
especially, the question of what kind of effect a quotation has on the news
genre was examined (Sect. 5.6.2). The parts of the news text will be
referred to as title, lead, text body (Chovanec, 2014, p. 205). Further, the
text body will be considered as divided into different topical sections.
144 M. Johansson
5.5 Quotations and Multimodal News Text

5.5.1 Quotations and Their Basic Characteristics
The news article is almost entirely composed of quotations, with a total

of 19. Table 5.2 shows who has been quoted and the source of the quota-
tion, if mentioned in the article. The quotations (Q) are numbered from
Q1 through Q19 and are listed in the order in which they were presented
in the lead and three main topical sections of the text body, respectively.
Table 5.2 Basic information concerning the quotations in the Huffington Post article
Q Source of Political party or institution
number Voice quotation Sociopolitical functions
Lead
1 Matignon [Prime Matignon Government
Minister’s office]
2 Majority [a Not mentioned Not mentioned
collective voice]
3 Spokesperson Not mentioned Socialist Party
Topical section 1: Reactions from the Socialist and Green parties
4 David Assouline Twitter Socialist Party
Member of the Senate
5 Frédérique Twitter Socialist Party
Espagnac Member of the Senate
6 Harlem Désir Not mentioned Socialist Party
Member of the European
Parliament
7 Marie-Noëlle Not mentioned Socialist Party
Lienemann Member of the Senate
8 Ecologist group, Press release Green Party
EELV
Topical section 2: Reactions from the right-wing and center parties
9 UMP (collective News article Union for Popular Movement
voice) (hyperlink)
10 Sébastien Huyghe Twitter Union for Popular Movement
(right, center)
Member of Parliament
11 Gérald Darmanin Twitter Union for Popular Movement
(continued)
Q Source of Political party or institution
number Voice quotation Sociopolitical functions
12 Valérie Debord Twitter Union for Popular Movement
13 UMP (collective News article Union for Popular Movement
voice) (hyperlink)
14 Yves Jégo Twitter Radical Party (center, liberal)
15 Jean-François Copé TV news (TF1) Union for Popular Movement
Topical section 3: Reactions from other institutions and from one
left-wing politician
16 Anonymous Not mentioned CFDT—French Democratic
Confederation of Labor
movement
17 Pierre Gattaz Not mentioned Medef—Movement of
Enterprises in France
employer’s organization
18 Eric Aubin Not mentioned CGT—General Confederation
of Labor
19 Jean-Luc Mélechon TV evening news Left Party
(France 2) video Member of the European
Parliament
Table 5.2 reveals that the journalist relies on media sources: Twitter,
the evening TV news, a press release, a news article, and a government
announcement for 12 quotations; while in the other 7 quotations, the
source was not revealed. As regards digital and written quotations, ten of
the quotations were written by the journalist in the text, while there were
seven digital quotations: six tweets and one video. Out of these, two are
written by the journalist containing a hyperlink to another page. In sum,
the quotations link the several media instances that have published com-
ments about the governmental decision.
The quoted voices were those of politicians or other public actors from
different political backgrounds. The first set of quotations (Q1–3) report
about the government’s decision. The next set of quotations (Q4–8) belong
to left-wing and Green Party politicians in favor of retirement reform. The
third set of quotations (Q9–15) represent right-wing or center politicians
who oppose the reform. The last set of quotations (Q16–19) are from
146 M. Johansson
representatives of different organizations, with the last one taken from a

left-wing politician. Their order of occurrence in the text shows that the
basic order used by the journalist has been political and institutional.
As already mentioned, the types of quotations vary. First, there are
traditional third-person quotations with a quotative clause2 in standard
language:
Example 1
Une voix discordante toutefois s’ext exprimée dans la soirée : celle de Marie-
Noëlle Lieneman, animatrice de l’aile gauche du parte: “nous ne pouvons
accepter 43 ans de cotisations”, déclare la sénatrice de Paris.
During the evening, however, one opposing voice was heard: that of Marie-
Noëlle Lienemann, who works for the left wing of the party: “We cannot
accept that we have to pay for 43 years,” declares this senator from Paris (Q7)
In this example, the quoted person and her political role are explained
in the quotative clause by a double identification: a left-wing politician
and a senator. The quoted content reveals that, unexpectedly, she opposes
the government’s decision and is thus in opposition with her own party.
Also, there are mixed cases such as in the following example:
Example 2
Et la critique de la «lâcheté » socialiste (Q13)
And the critique of the socialist party’s “cowardice” (Q13)
This mixed quotation is written by the journalist and contains a hyper-

link to a news article with a full-blown critique given by the right wing.
This quotation does not contain a quotative clause but refers to a speech
event through the word critique and the quoted content through a word
in quotation marks.
Then, one set of quotations comprises tweets:
Example 3 [tweet]
David Assouline
@dassouline
La Gauche démonte qu’une réforme peut être juste, équilibrée et permettre

encore des progrès, pour la droite réforme=matraquage et régression
9:11 PM—27 Aug 2013
[retweeted by 7, favorited by 1]
The Left shows that a reform can be fair, balanced, and can allow us to
make progress, for the Right, reform=hitting and regression (Q4)
This example consists of a tweet, that is, a digital quotation that was
copy-pasted from a social media source into the HP article. The tweet
text template functions as a quotative or reporting frame. It contains a
double identification of the user—the head picture and the Twitter
account name3—as well the means to respond to him @dassouline. The
temporal stamp follows the user’s words; the tweet also displays further
information like the number of times it has been retweeted or favorited.
When the quotation was a tweet, the journalist relied on Twitter’s own
identification affordances, previously selected by the user. The journalist
did not explain who these politicians were but left it up to the readers to
do so. In sum, this digital quotation was a literal and first-person quota-
tion that was not modified by the journalist.
The third type of quotation in the HP article was a video quotation: an
extract from a TV news interview in standard spoken French. This digital
quotation was presented by a quotative:
Example 4
Le Front de gauche viendra également grossir les rangs des cortèges. Jean-
Luc Mélechon a confirmé sur France 2 que son mouvement appelait à
manifester (Q19)
The Left Front is going to come to the protest. Jean-Luc Mélechon con-
firmed on France 2 [television channel] that his movement has asked [peo-
ple to come to] protest.
Here, the journalist included the name of the politician, a speech verb,
and the place where the interview took place—a TV studio. The quoted
content expresses the negative stance toward the reform. In the next sec-
tion, I will study how quotations are integrated in the news texts.
148 M. Johansson
5.5.2 Integration of Digital Quotations in the News Text
In this section, the sequential organization and how digital quotations are
integrated in news text is described. The main textual components of a
basic news story consist of the headline, the lead, and the main text body
(Chovanec, 2014, p. 205). After the headline, the HP news article is
organized into the lead, while the main news text contains three different
topical sections.
In the lead, the story opens with a picture of crowds in the street pro-
testing the French government’s decision. The picture shows the reaction
of people to the proposed governmental plan: they are on the streets pro-
testing. It is not only visual but it also contains written/printed text.
Several banners indicating the names of the trade organizations partici-
pating in the demonstration or expressing their opposition can be seen.
After the picture, the lead describes the public reactions to the govern-
ment’s announcement.
The different parts of this news text are organized as indicated in
Table 5.3. I use letters (a, b, c, etc.) to point out different parts of the texts
(see Appendix, the news text).
As described in the Table 5.3, the tweets are inserted within the first
and second topical sections (b, e, f ) and they are framed by text para-
graphs. The tweets are presented as screen captures. They contain both
written material and the pictures of their writers. The third and last topi-
Table 5.3 Organization of Main title

news text Picture
First topical section
a Lead: two paragraphs
b Two tweets
c Three short paragraphs
Second topical section
Title
d One short paragraph
e Two tweets and one short phrase
f Two tweets and one short text paragraph
Third topical section
Title
g Two paragraphs
h Video
cal section of the article contains a text paragraph and a video at the end
of the article. It is a still image from TV news (h), which is a video in
which an interviewer and a left-wing politician, Jean-Luc Mélenchon, are
engaged in dialogue. In sum, the news text contains a written news story,
a photo, tweets, and a video. The news article ends with a video that con-
tains both visual and spoken modes.
This section, Sect. 5.5, examined how the digital quotations were
embedded in the multimodal new texts. The next section, Sect. 5.6, will
focus on the functions of quotations and how they affect the news genre.
5.6 N
ews Genre of Opinion Review
5.6.1 Quotations and Their Discursive Meaning
The object of the stances in all of the quotations was the retirement
reform, which each of the quoted individuals/institutions evaluated dif-
ferently. The quotations contain positive, negative, or mixed evaluations.
Table 5.4 categorizes the type of quotation: digital or those written or
entextualized by the journalist. Table 5.4 also summarizes the types of
stances and positionings of the quotations.
Table 5.4 explains what the object of evaluation is in each quotation;
that is, what part of the reform is being evaluated and whether the quoted
person favors or opposes it.
In the following text, I will concentrate on tweets and the video. The
tweet that was already quoted above, in Example 1, will be examined
again, but from another angle:
Example 5
David Assouline
@dassouline
La Gauche démontre qu’une réforme peut être juste, équilibrée et permettre
encore des progrès, pour la droite réforme=matraquage et régression
The Left shows that a reform can be fair, balanced and can allow us to make
progress, for the Right, reform=hitting and regression (Q4)
150 M. Johansson
Table 5.4 Object of stances by the quoted persons

Q
number Digital quotation Quotation written by the journalist
1 • Information report: the decision
has been made
2 • Information report: reactions are
positive in the Socialist Party
3 • Information report: reactions are
positive in the Socialist Party
4 • Praise: right kind of reform
• Anticipates criticism:
political right opposes the
reform
5 • Praise: the reform is good
6 • Thanking: the reform will have
good consequences
7 • Denial: we cannot accept the
costs the reform will entail
8 • Regret: it’s not a global reform
• Acceptance: the reform does not
go far enough
9 • Complaint: the Socialist Party
fails in the reform
10 • Complaint: the reform
brings increase in taxes,
unemployment
11 • Complaint: the reform
shows inequality between
public/private sectors and an
increase in taxes
12 • Complaint: the reform increases
taxes
13 • Complaint: the government is a
coward
14 • Complaint: companies will
suffer and pensions will be
smaller because of the
reform
15 • Complaint: the prime minister
showed his “true” face
• Complaint: tax increase
16 • Praise: the reform is equitable
17 • Warning: dangerous reform
(continued)
Q
number Digital quotation Quotation written by the journalist
18 • Complaint: our proposals were
not heard
19 • Complaint: the previous
reform of the right was
accepted
• Complaint: salaries will be
lower
• Demand: we cannot accept
it, we want to raise salaries
• Complaint: of the type of
politics
• Invitation: to participate in
the planned protest
In Example 5, the tweet contains two evaluations that present polarized

views. First, the politician engages in self-praise. When he considers that
“a reform can be fair, balanced and can allow us to make progress”, he uses
positive axiological adjectives and substantives that refer to democratic
values to point out the outcome to the larger public. He presents progress
as a consequence of the reform, thus anticipating a better future. In the
last part of the tweet, he presents an account of how the reform will be
viewed by his political opponent: “hitting and regression”. His positioning
is that of aligning with the decision of the Socialist Party, of which he is a
member. In his view, the political decision will enable a better future
(make progress). In sum, his positioning is that of praise: he compliments
his own party, which is behind the retirement reform.
A similar kind of tweet is presented by his fellow party member:
Example 6
Frédérique Espagnac
@FEspagnac
:#retraites JM Ayrault propose une réforme structurelle, juste et équilibrée,
permettant de financer durablement la retraite par répartition (Q5)
152 M. Johansson
:#Retirement JMAyrault [Prime Minister] proposes a structural, just

and fair reform, allowing the finance of pensions in a sustainable and dis-
tributed way
In Example 6, the politician uses similar evaluative adjectives (just, bal-

anced) to those used by her colleague, but she also mentions the Prime
Minister by name, positioning him as the decision-maker. In her view,
the reform will have a positive economic consequence (distribution of
revenues). Her positioning is also that of praise.
However, several right-wing politicians oppose the reform, as we can
observe in the following tweets.
Example 7
Sébastien Huyghe
@SebastienHuyghe
#retraites nouvelles hausse d’impôts et renchérissement coût du travail
conséquences = hausse du chômage et inefficacité sur financement (Q9)
#retirement new tax increases and rise of cost of work consequences =
unemployment increases and inefficient financing (Q9)
In Example 7, the tweet does not contain a verb; instead, the author’s
stance is expressed by pointing out the outcomes of the political decision.
In the previous examples, Examples 5 and 6, the consequences of the
reform were implicitly conveyed; in this example, however, they were
spelled out through the use of the noun consequence. The negative stance
is conveyed through the use of nouns referring to economic and labor
affairs (tax increases, cost of work). The consequence is seen as a worsening
of the economic situation (unemployment, financing). This politician does
not align with the reform and expresses his critical positioning, which is
that of complaint, as with several other quotations in the article. The
same positioning is expressed in the tweets in Q11–13.
In the last digital quotation (Q19), which is a five-minute clip extracted
from a TV interview, the politician explains his view in a more elaborate
manner. He conveys the same positioning as the others. In addition, he
announces a demonstration that his party is going to organize.
5.6.2 News Genre of Political Opinion Review
In this section, I will examine news texts as a genre and the role of the
journalist in it. In the lead, the journalist introduces the news topic:
Example 8
Politique. Très attendue, la réforme des retraites n’a donc pas manqué de
faire réagir. Aussitôt les annonces faites à Matignon, chacun y est allé de
son commentaire. Mais si le timing des décisions a pu surprendre quelques
minutes seulement après la fin des consultations avec les partenaires
sociaux, les réactions étaient convenues.
Politics. As expected, the retirement reform did not take long to gather
comments. As soon as the announcement was made in Matignon [the
Prime Minister’s office] everybody gave their commentaries. But if the tim-
ing of the decision was surprising, that is, only a few minutes after the
consultation of the social partners, the reactions were usual.
Basically, the lead introduced the topic of the news article and the
public reactions to the decision concerning the retirement plan. In the
other sections of the text, the journalist’s role was to present only the
quotations, that is, she or he introduced them through quotative clauses,
tweet templates, or a still image from the TV news. In addition, the jour-
nalist gave short evaluations or summaries of the opinions expressed. The
following extract is from the third topical section of the text, in which the
views of the political right and center were grouped:
Example 9
La droite est pour sa part beaucoup plus critique. Ses responsables ont
repris le refrain entamé après l’intervention du Premier ministre dimance
soir sur France 2. Deux angles d’attaque sont choisis […]
As for the right, it is much more critical. Its responsible persons started to
repeat the refrain after the announcement of the Prime Minister on
Sunday night on France 2 [television channel]. Two perspectives have
been chosen […]
154 M. Johansson
In the third topical section of the news article, the journalist summa-
rizes the comments. As has been shown through this analysis, this news
text is a multimodal compilation or a mix of quotations that are inte-
grated within a news story in a way that ties the quotations together. The
news text unfolds based on how the quoted persons comment on a given
news event.
How can this news genre of opinion review be classified further? First,
it is a remix of opinions, which is similar to the video press review
(Johansson, 2013) even though it differs as all the sources of quotations
are not mentioned in the news text as they were in the video press review.
Second, the construction of meaning is done through the journalist’s
selections of quotations. Despite the fact that the journalist’s positioning
has remained unmarked or seemingly objective, the quotations were
organized in a certain order in the text: first, those in favor of the reform,
followed by those who opposed it. In this task, the journalist emerged as
a mediator on the one hand and as a gatekeeper on the other. In the pro-
cess of recontextualization, the journalist thus became a mediator between
social media and readers. Digital quotations establish relations with other
media sites and social media, adding to media circularity and the reme-
diation of material (Bolter & Grusin, 1999; Chadwick, 2013). Quotations
in the printed press may do the same, but they do not allow the reader
instant access to original sources in the way that tweets and videos do.
Finally, the news text constructs an imaginary dialogue, as the quoted
politicians do not necessarily speak to each other as they would do during
a conversation. This genre could be called a political opinion review. This
dialogue takes place at a certain time across media; here, the comments
are shown together. Based on the communicative activity of c ommenting,
this news genre creates, as in the case of the video press review, opinions
not as being shared but as polarized sites of struggle.
5.7 Conclusion
A digital quotation is a tweet, video, or other digital element that is
recontextualized from a digital context by being copy-pasted into digital
news texts. Digital quotations are literal quotations, and they disseminate
information in the traditional sense of news information. However, they

can have several functions and can consequently contribute to several
discursive processes when they are embedded in the news text.
In regard to textual organization when quotations are abundant (as in
the news text examined in this study), quotations fragment a news text.
A news text thus becomes a compilation or remix of quotations, orga-
nized by a journalist, that have been recontextualized from various
sources. This type of text can present several voices. In this process, the
journalist is both a gatekeeper and a mediator: it is he or she who chooses
the quotes and arranges them in a certain order.
A news text that comprises a remix of quotations can be characterized
as a commenting type of news text. It is an opinion-based, hybrid media
genre: a review of opinions. It constructs a dialogue that has not actually
taken place between the quoted persons; it presents a cross-media strug-
gle at a certain moment in time (Jones, 2009).
In contemporary news making, multimodality (e.g., videos, interactive
graphics, and picture slideshows) has gained more ground compared to the
print press. Digital quotations represent a shift in news making, and their
use resembles popular culture practices and a type of vernacularization.
These new ways of composing news texts have an effect on reception prac-
tices as well: readers are asked to participate in deconstructing the layered
meaning. To do so, they have to know how to “read” the playful digital
culture (see Johansson, 2013). As they are presented with a problem and
are asked to take sides, active participation from the reader is demanded.
Appendix
Translation of article:
Reactions to the retirement reform did not take long
(a) Politics. As expected, the retirement reform did not take long to gather
comments. As soon as the announcement was made in Matignon [the
Prime Minister’s office] everybody gave their commentaries. But if the
timing of the decision was surprising, that is, only a few minutes after
the consultation of the social partners, the reactions were usual.
156 M. Johansson
In the ranks of the majority, there is a general satisfaction that is

addressed to Jean-Marc Ayrault [Prime Minister]. There is even a
competition in praises if one reads the commentaries of the spokes-
persons of the Socialist Party.
(b) David Assouline
@dassouline
The Left shows that a reform can be fair, balanced and can allow us
to make progress, for the Right, reform=hitting and regression
Frédérique Espagnac
@FEspagnac
:#Retirement JMAyrault [Prime Minister] proposes a structural, just
and fair reform, allowing the finance of pensions in a sustainable
and distributed way
(c) This has what it takes: “The distribution [of costs] will save the system”,
congratulates the First Secretary [of the Socialist Party] Harlem Désir.
During the evening, one opposing voice was heard: that of Marie-
Noëlle Lienemann, who works for the left wing of the party: “We
cannot accept that we have to pay for 43 years,” declares this senator
from Paris.
As regards the ecological allies, they are very restrained. In a press
release, the Ecologist Group in the Parliament “regrets” the absence
of a global reform. But the ecological representatives are also pleased
that “two red lines were not crossed”, notably relying on CSG [social
charges (deductible from income tax)].
(d) Opposition exposes pressure on taxes
As for the right, it is much more critical. Its responsible persons started
to repeat the refrain after the announcement of the Prime Minister on
Sunday night on France 2. Two perspectives have been chosen…
(e) Sébastien Huyghe
@SebastienHuyghe
#retirement new tax increases and rise of cost of work consequences=
unemployment increases and inefficient financing
Gérard Darmanin
@GDarmanin
Inequality between different #retirement systems public/private and
financing by +taxes. Still +taxes #PS
And the critique of the socialist party’s “cowardice”.
(f ) Valérie Debord
@DebordValerie
#Retirement the only solution of the government is to add to pres-
sure on taxes through cowardice and popularity seeking
Yves Jégo
@yvesjego
New costs for firms. No convergence between public/private.
Decrease in pensions in the future. #Reform without courage
Resumed by Jean-François Copé interviewed on TF1 [television
channel]: “The Prime Minister threw a masque and showed clearly
that this was unfortunately not a reform of retirement, but simply
a rise in taxes, a rise in social security charges that will take place
instead of retirement reform”.
(g) Protest September 10th is on

Social partners are also quite divided. While the CFDT [trade union]
is content with the “achieved fair measures”, the president of Medef
[employer’s organization], Pierre Gattaz, says in the Figaro [newspa-
per] that it is “a dangerous anti-reform”. This is also true for the CGT
[labor union], but for totally opposing reasons. It [la centrale de
Montreuil] says that it was not heard. “The Prime Minister did not
care at all about our propositions” complains the Secretary, Eric
Aubin. In consequence, the protest that it has planned for September
10th with Force Ouvrier [labor union] is on.
The Left Front is going to come to the protest. Jean-Luc Mélechon
confirmed on France 2 [television channel] that his movement has
asked [people to come to] protest.
(h) Video: Evening news with Jean-Luc Mélenchon
158 M. Johansson
Notes
1. Moirand (2007, p. 12) defines news genres as “objectivisés” and
“subjectivisés”—“objectifying and subjectifying” news genres. That is, she
considers this a dynamic feature of news genres.
2. The quotative clause is underlined, and the quoted content is in quotation
marks.
3. Not shown here for copyright reasons.
Source
Les réactions à la réforme des retraites ne se sont pas fait attendre. Article pub-
lished 27.8.2013, updated 5.10.2016. Huffington Post (fr). Retrieved from
April 20, 2018, from http://www.huffingtonpost.fr/2013/08/27/reactions-
reforme-retraites_n_3824466.html
References
Adam, J. M. (1992). Les textes: Types et protypes. Paris: Nathan.
Adami, E. (2012a). The rhetoric of the implicit and the politics of representa-
tion in the age of copy-and-paste. Learning, Media and Technology, 37(2),
131–144.
Adami, E. (2012b). Mashing genres up, breaking them down: Habitus and lit-
eracy in the age of copy-and-paste. Linguagem em Foco/Language in Focus, 1,
25–42.
Adami, E. (2016). Multimodality. In O. Garcia, N. Flores, & M. Spotti (Eds.),
Oxford handbook of language and society (pp. 451–472). Oxford: Oxford
University Press.
Adami, E., & Kress, G. (2014). Introduction: Multimodality, meaning making,
and the issue of “text”. Text and Talk, 34(3), 231–237.
Arendholz, J. (2015). Quoting in online message boards: An interpersonal per-
spective. In J. Arendholz, W. Bublitz, & M. Kirner-Ludwig (Eds.), The prag-
matics of quoting now and then (pp. 53–69). Berlin: De Gruyter Mouton.
Bauman, R., & Briggs, C. L. (1990). Poetics and performance as critical per-
spectives on language and social life. Annual Review of Anthropology, 19,
59–88.
Bednarek, M., & Caple, H. (2012). News discourse. London and New York: A
and C Black.
Bell, A. (1991). The language of news media. Oxford: Blackwell.
Bergmann, J. R., & Luckmann, T. (1995). Reconstructive genres of everyday
communication. In U. M. Quasthoff (Ed.), Aspects of oral communication
(pp. 289–304). Berlin/New York: Walter de Gruyter.
Blum-Kulka, S. (1983). The dynamics of political interviews. Text and Talk,
3(2), 131–154.
Bolter, J. D., & Grusin, R. (1999). Remediation: Understanding new media.
Cambridge, MA: MIT Press.
Broersma, M., & Graham, T. (2013). Twitter as a news source: How Dutch and
British newspapers used tweets in their news coverage, 2007–2011. Journalism
Practice, 7(4), 446–464.
Bublitz, W. (2015). Introducing quoting as a ubiquitous meta-communicative
act. In J. Arendholz, W. Bublitz, & M. Kirner-Ludwig (Eds.), The pragmatics
of quoting now and then (pp. 1–26). Berlin: De Gruyter Mouton.
Catenaccio, P., Colleen, C. M., De Smedt, G., Garzone, G., Jacobs, F.,
Macgilchrist, L., et al. (2011). Towards a linguistics of news production.
Journal of Pragmatics, 43(7), 1843–1852.
Chadwick, A. (2013). The hybrid media system: Politics and power. Oxford:
Charaudeau, P. (1997). Le discours d’information médiatique. La construction du
miroir social. Paris: Nathan.
Chovanec, J. (2014). Pragmatics of tense and time in news, From canonical head-
lines to online news texts. Amsterdam/Philadelphia: John Benjamins.
Claquin, F. (1993). La revue de presse: Un art de montage. Langage et Société,
64, 43–71.
Clayman, S. E. (1992). Footing in the achievement of neutrality: The case of
news-interview discourse. In P. Drew & J. Heritage (Eds.), Talk at work:
Interaction in institutional settings (pp. 163–198). Cambridge: Cambridge
University Press.
Couldry, N. (2012). Media, society, world: Social theory and digital media prac-
tice. Cambridge and Malden: Polity.
Couldry, N., & Hepp, A. (2016). The mediated construction of reality. Chicester:
Polity Press.
Davies, B., & Harré, R. (1990). Positioning: The discursive production of selves.
Journal for the Theory of Social Behaviour, 20(1), 43–63.
Deuze, M. (2017). What is journalism? Journalism, 6(4), 442–464.
van Dijk, T. A. (1988). News as discourse. New Jersey: Erlbaum, Hillsdale.
160 M. Johansson
DuBois, J. (2007). The stance triangle. In R. Engelbretson (Ed.), Stancetaking in

discourse: Subjectivity, evaluation and interaction (pp. 139–182). Amsterdam/
Erbe, J. (2007). “What do the papers say?” How press reviews link national
media arenas in Europe. The Public, 12(2), 75–92.
Fairclough, N. (1992). Discourse and social change. London: Polity Press.
Fairclough, N. (1995). Media discourse. London: Edward Arnold.
Fetzer, A. (2004). Recontextualizing context. Amsterdam/Philadelphia: John
Benjamins.
Giaxoglou, K. (2017). Storying leaks for sharing: The case of leaking the
“Moscovici draft” on Twitter. Discourse, Context and Media, 19, 22–30.
Gruber, H. (2013). Genres in political discourse: The case of the “inaugural
speech” of Austrian chancellors. In P. Cap & U. Okulska (Eds.), Analyzing
genres in political communication: Theory and practice (pp. 29–71). Amsterdam/
Haapanen, L. (2016). Monologisation as a quoting practice: Obscuring the
journalist’s involvement in written journalism. Journalism Practice, 11, 1–20.
Haapanen, L., & Perrin, D. (2017). Media and quoting: Understanding the
purposes, roles, and processes of quoting in mass and social media. In
C. Cotter & D. Perrin (Eds.), The Routledge handbook of language and media
(pp. 424–441). London and New York: Routledge.
Hermida, A. (2013). #JOURNALISM. Digital Journalism, 1(3), 295–313.
Johansson, M. (2002). Sequential positioning of represented discourse in insti-
tutional media interaction. In A. Fetzer & C. Meierkord (Eds.), Rethinking
sequentiality: Linguistics meets conversational interaction (pp. 249–272).
Johansson, M. (2006). Constructing objects of discourse in the broadcast politi-
cal interview. In A. Fetzer & E. Weizman (Eds.), Pragmatic aspects of politi-
cal discourse in the media. Journal of Pragmatics, 38(2), 216–229.
Johansson, M. (2012). Political videos in digital news discourse. In A. Fetzer &
L. Berlin (Eds.), Dialogue in politics (pp. 43–67). Amsterdam/Philadelphia:
John Benjamins.
Johansson, M. (2013). Remixing opinions: Video quotations and stances in
video press reviews. Recherches en Communication, 39, 95–115.
Johansson, M. (2014a). Reading digital news: Participation roles, activities, and
positionings. Journal of Pragmatics, 72, 31–45.
Johansson, M. (2014b). Faktoista mielipiteisiin. Referointi verkkolehtien medi-
aesityksissä [From facts to emotions. Quotations in news articles and news
videos]. In M.-L. Helasvuo, M. Johansson, & S.-K. Tanskanen (Eds.), Kieli
verkossa. Näkökulmia digitaaliseen vuorovaikutukseen [Language in the Net.

Perspectives to Digital Interaction] (pp. 149–176). Helsinki: Suomalaisen
Kirjallisuuden Seura.
Jones, R. H. (2009). Dancing, skating and sex: Action and text in the digital age.
Journal of Applied Linguistics, 6(3), 283–302.
Kress, G. (2010). Multimodality: A social semiotic approach to contemporary com-
munication. London: Routledge.
Kress, G., & Leeuwen, T. V. (2008). Reading images: The grammar of visual
design. London: Routledge (original work published in 1996).
Landert, D. (2015). Reportable facts and a personal touch: The functions of
direct quotes in online news. In J. Arendholz, W. Bublitz, & M. Kirner-
Ludwig (Eds.), The pragmatics of quoting now and then (pp. 29–52). Berlin:
Walter de Gruyter.
Le, E. (2010). Editorials and the power of media. Amsterdam/Philadelphia: John
Benjamins.
van Leeuwen, T. (2008). News genres. In R. Wodak & V. Koller (Eds.), Handbook
of communication in the public sphere (pp. 345–362). Berlin/New York: Walter
de Gruyter.
Levinson, S. (1992). Activity type and language. In P. Drew & J. Heritage (Eds.),
Talk at work: Interaction in institutional settings (pp. 66–100). Cambridge:
Linell, P. (1998). Approaching dialogue: Talk, interaction and contexts in dialogical
perspective. Amsterdam/Philadelphia: John Benjamins.
Luckmann, T. (1989). Prolegomena to a social theory of communicative genres.
Slovene Studies, 11(1–2), 159–166.
Mast, J., Coesemans, R., & Temmerman, M. (2016). Hybridity and the news:
Blending genres and interaction patterns in new forms of journalism.
Journalism, 18, 1–8.
Moirand, S. (2007). Les discours de la presse quotidienne. Observer, analyser, com-
prendre. Paris: Presses Universitaires de France.
Perrin, D. (2013). Linguistics of newswriting. Amsterdam/Philadelphia: John
Benjamins.
Puschmann, C. (2015). The form and function of quoting in digital media.
Discourse, Context and Media, 7, 28–36.
Silverstein, M., & Urban, G. (1996). Natural histories of discourse. Chicago:
University of Chicago Press.
Waugh, L. R. (1995). Reported speech in journalistic discourse: The relation of
function and text. Text and Talk, 15(1), 129–173.
162 M. Johansson
Weizman, E. (2008). Positioning in media dialogue. Amsterdam/Philadelphia:

John Benjamins.
Wertsch, J. V. (1998). Mind as action. Oxford & New York: Oxford University
Press.
White, P. R. R. (2012). Exploring the axiological workings of “reporter voice”
news stories: Attribution and attitudinal positioning. Discourse, Context and
Media, 1, 57–67.
6
Emoticons in Relational Writing
Practices on WhatsApp: Some
Reflections on Gender
Carmen Pérez-Sabater
In 2015, Oxford Dictionaries chose face with tears of joy as its Word
of the Year.
http://blog.oxforddictionaries.com/press-releases/announcing-the-
oxford-dictionaries-word-of-the-year-2015/
C. Pérez-Sabater (*)
Department of Applied Linguistics, Universitat Politècnica de València,
Valencia, Spain
e-mail: cperezs@idm.upv.es

https://doi.org/10.1007/978-3-319-92663-6_6
164 C. Pérez-Sabater
6.1 Introduction
Whereas early computer-mediated communication (henceforth, CMC)
studies sought to find universal norms, today’s CMC research looks at
community-level, specific practice, as McKeown and Zhang (2015) posit.
The study presented in this chapter investigates several online communi-
ties as entities of interest in their own right.
In the past, many online communities were formed because participants
required the affordances the medium offered: anonymity and invisibility.
Anonymity allows participants to feel less vulnerable and to open up more
easily because what they say in the forum stays in the forum; it cannot be
linked to the rest of their lives. Meanwhile, invisibility further disinhibits
participant behavior since posters do not have to worry about how they
look or sound, especially in online support groups that address physical or
speech problems, as Barak, Boniel-Nissim, and Suler (2008) indicate.
However, although online communities which favor anonymity still exist,
today the internet is seen as a tool for self-promotion. This more recent
phenomenon was observed by Jones and Hafner (2012), who also note
that people now gather on social networking sites which encourage partici-
pants to reveal their “true” identities. Among these popular social network-
ing services and instant messaging (henceforth, IM) applications, online
communities interacting via WhatsApp1 are particularly “real” because a
user must provide their mobile phone number to be able to interact with
others. Users must share this very private contact information with other
participants in their WhatsApp chats; therefore, community participants
frequently tend to be connected by strong ties. As Ling (2005) demon-
strates, the circle of people with whom we are in regular contact by phone,
by means of either calls or texts, is quite small. Other online communities
on social networking platforms are generally larger. For example, Facebook
users tend to have many “friends” because it is easy to find them by typing
their names on the search window. However, as WhatsApp users share
their private telephone numbers with each other, they are often linked by
close social relationships and a shared relational purpose: to maintain those
friendships. Other WhatsApp communities are formed by colleagues and
members of clubs and political parties, among many others.
Emoticons in Relational Writing Practices on WhatsApp: Some… 165
This study examines WhatsApp communities formed by close friends

to see if the affordances of the medium, such as the enormous gallery of
emoticons provided, determine their choices in writing style; particular
attention will be given to the use of iconic elements in written practice.
Initially, the study focused on the difference between teenagers’ and
adults’ use of emoticons. However, after processing an initial questionnaire
on the habits of these WhatsApp users, the age variable appeared not to be
significant. Other more interesting results were found regarding the gen-
der variable, thus, the study was therefore redirected toward examining
women’s and men’s participation in chat groups and their contrasting use
of emoticons. Thurlow and Brown (2003) suggest it is hardly surprising that
gender differences emerge in young people’s interactions. In this chapter,
we see how these gender differences manifest in adults’ use of emoticons in
WhatsApp interactions that take place in a Spanish context.
Few linguistic studies have examined emoticon use on new mobile
applications in non-Anglophone contexts, and even less attention has
been paid to gender differences in adults’ use of emoticons. Two very
recent exceptions are the doctoral dissertations by Al Rashdi (2015), con-
cerning emoticons in WhatsApp messages written by Omani men and
women in Arabic, and Sampietro (2016a), who provides an overview of
how emoticons feature in messages written in Peninsular Spanish.
Bieswanger (2013) claims that authors publishing in English need to
indicate whether they are describing computer-mediated English, lan-
guage use in CMC, or modes of CMC in a language or particular lan-
guages other than English, or whether they are attempting to make
universal claims as to patterns of language use in CMC in all languages.
Far from seeking to generalize and make universal claims, this study
attempts to shed light on the linguistic conventions of use of several com-
munities communicating in Peninsular Spanish, and thus fill the gap in
the literature regarding these conventions as displayed by users of new
messaging applications for smartphones.
To contextualize the present study in the wider field of CMC research,
the main findings of other relevant studies of texting and emoticons will
be briefly reviewed in the next section.
6.2 Literature Review

6.2.1 Texting
The study of what Markman (2013) calls “text-based conversations”

attracted the attention of academia at the beginning of the new millen-
nium. However, despite this initial research, Thurlow and Poff (2013)
contend that texting has been relatively underexamined by linguists in
comparison to other modes or genres of CMC. The recent proliferation
of messaging applications that run on smartphones and their relevance to
our everyday lives seem to have caused resurgence in scholarly interest in
this particular field (see, e.g. Ling & Baron, 2013).
In general, most studies have been conducted in the West and into
Western languages. Scholars have focused on age and gender differences
in English as well as in several other languages. Regarding the age vari-
able, for example, Plester, Wood, and Joshi (2009) found that young
texters were better at spelling than those young people who do not regu-
larly use textisms.
As for the gender variable and texting, the study of gender similarities
and differences in IM by Fox, Bukatko, Hallahan, and Crawford (2007)
showed that women sent messages that were usually more expressive than
those sent by men. Baron’s study (2004) of IM yielded similar conclu-
sions: women wrote longer texts and longer goodbyes than men, and they
also used more emoticons. Ling (2005) studied SMS interactions in
Norwegian and found that young girls wrote more complex SMSs than
young boys. In general, these studies posit that female users of IM are
more focused on building and maintaining relations with others through
messaging applications than male users (Fox et al., 2007). On the whole,
these studies conclude that the interlocutor’s gender plays an important
role in the speech style chosen.
Furthermore, many studies have revealed the relational character of
SMS. For instance, texting is a tool to maintain and reinforce relation-
ships (Ishii, 2006) or to initiate new ones (Ling, Julsrud, & Yttri, 2008)
or is simply an arena in which to negotiate roles and relationships (Spilioti,
2011). These studies clearly demonstrate the social function of the tech-
nology (Thurlow & Poff, 2013).
In the case of Spanish, although extensive research has been done into
chat interactions (e.g. Sanmartín, 2007), little work on written interac-
tions via mobile phones is available, despite recent interest in this CMC
mode. An exception is the ethnographic study conducted by Sabaté i
Dalmau (2014) concerning transnational SMS practices among migrant
people. Her study challenges the assumption that a certain level of com-
petence is needed for successful communication in an information and
communication technology (ICT) multilingual environment as texters in
her sample are able to communicate effectively by creating their own
intergroup lingua franca. Another interesting recent study is that by
Vázquez-Cano, Mengual-Andrés, and Roig-Vila (2015) into the linguistic
characteristics of teenagers’ messages. The research concluded that the
corpus presents orthotypographic and audiovisual characteristics condi-
tioned by the size of device display, hours of conversation, and the rela-
tionship between speakers.
Notwithstanding these studies, Thurlow and Poff (2013) argue that a
great deal of research in this area is still needed, especially into the inter-
actions between adult texters.
6.2.2 Emoticons and Emoji
According to its definition, emoji is a loanword from Japanese defined as

“a small digital image or icon used to express an idea or emotion in elec-
tronic communication”. It is derived from the Japanese words “e” (pic-
ture) and “moji” (letter, character), whereas emoticon is derived from the
English words “emotion” and “icon”. Emoji are well established on many
social media platforms. The affirmation that emoticons are used less fre-
quently in online writings than tends to be believed is a long-established
conclusion of CMC research (Baron, 2008; Herring, 2012; Pérez-Sabater,
Turney, & Montero-Fleta, 2008; Thurlow & Poff, 2013). For example, in
Baron’s (2008) data, emoticons used in texting and IM were <1% of the
words studied. This situation has dramatically changed over the last few
years with the increasing use of graphical means of communication such
as emoji, stickers, and GIFs to supplement images on online writing. The
increase in their use stems mainly from the incorporation of large emoji
galleries into messaging applications such as WhatsApp and Facebook

messenger. They are continuously evolving, and, although Stark and
Crawford (2015) noted that there are “many white emoji faces, hands,
and body parts” (p. 7), they no longer represent only white and hetero-
sexual cultures. For example, WhatsApp has just included pictures of
same-sex families and users can choose the skin color of many of the
images. Stickers, on the other hand, known as “next-level emoji” (Stark
& Crawford, 2015), are proprietary to the platform that sells them. For
the purpose of this research, emoticons or emoji will be referred to indis-
tinctly throughout the text, since the distinction between these graphical
icons is not relevant here. Moreover, most research to date employs the
term “emoticons” predominantly to refer collectively to new emoticons,
traditional ASCII emoticons, and emoji (see, e.g. Yus, 2014).
The use of emoticons has traditionally been related to sociolinguistic
factors such as gender, age, and CMC mode (Bieswanger, 2013). Many
studies consistently highlighted the relationship between women and
emoticon use. For instance, Wolf (2000) discovered that women used
emoticons more than men in online newsgroups. Similarly, Baron (2004,
2008) found that women included emoticons more often than men when
communicating via IM. Likewise, in her study of blogs, Nishimura (2015)
concluded that gender is the most salient factor in explaining the differ-
ence of usage in Japanese: younger women employ emoticons far more
frequently than older women and men, and old men far less frequently
than all other users. Nishimura also emphasized that topic is a determin-
ing factor for emoticon use: they are more appropriate for expressing the
“brighter aspects of life” (Nishimura, 2015). In Kapidzic and Herring
(2011), girls used more emoticons than boys in teenagers’ chat rooms,
especially those representing smiles and laughter. Exceptions to these
studies are found in Huffaker and Calvert (2005), whose analysis showed
that male teenagers used flirty emoticons more than girls in blogs, and in
Maíz Arévalo (2014), whose male participants on Facebook messenger
were more emotional than their female counterparts. Her research revealed
that males make frequent use of the sad face emoticon because they are
affectively protected by the screen, a fact that allows them to express their
emotions more freely than they would in a face-to-face conversation.
As for the mode variable, whereas in other modes of CMC, such as
email, the general absence of emoticons is of academic interest (see, e.g.
Pérez-Sabater et al., 2008 for lack of emoticons in a large corpus of

academic emails, or Lee, 2007), emoticons are now increasingly common
in the corpora of messages exchanged through new mobile IM applica-
tions. For example, Al Rashdi’s (2015) corpus of WhatsApp messages
from Omani participants in two chat groups contains so many emoji
that, in her view, they can be considered as a defining feature of Omani
WhatsApp interactions, although, as she indicates, her two corpora may
not be truly comparable. It is also interesting to note that the Omani
women studied use far more iconic elements than men in men-only
groups. Al Rashdi also found that some emoji are only used by men, for
example, the thumbs-down representation or the policeman, while women
predominantly employ the kissing face. Similarly, Sampietro’s (2016a)
empirical research revealed that 50% of her corpus included an emoti-
con. The conclusions of her doctoral dissertation state that men make less
frequent use of the face throwing a kiss emoticon, an emoji which is almost
a conventionalized farewell, because men rarely kiss each other to open or
close face-to-face encounters in Spain2 (except in some social contexts,
such as among families and gay male groups). Despite these conclusions,
Sampietro’s main concern is that emoticons should be studied as part of
a more comprehensive approach to multimodal communication, and
that the inclusion of emoticons should be considered a meaningful
choice. Finally, Sampietro also notes that some linguistic studies evaluate
sociolinguistic differences in emoticon usage relying on nonrepresenta-
tive corpora, or corpora which are not truly comparable.
In light of the above, this research addresses these concerns by analyz-
ing gender differences in a comparable corpus of men-only and women-
only chat groups formed by close friends. Two social categories highlighted
by Bieswanger (2013) are the basis of this study: gender similitudes and
differences in emoticon use, and the role of mode choice, specifically
WhatsApp, one of the new mobile applications for smartphones.
The research questions this study aims to answer are as follows:
1. Is the relationship between emoticon use and gender still relevant in

WhatsApp?
2. To what purpose are these emoticons used in closed single-gendered
communities?
6.3 Methodology, Participants, Parameters

Studied, and Delimitation of Study
6.3.1 Methodology
The methodology for the study follows that suggested by Orgad (2006)
and Spilioti (2011) for the analysis of online data on breast cancer com-
munities and SMS interactions, respectively. Basically, these studies
involved the use of online and offline data. In the present analysis, three
types of texts were taken into consideration: an online questionnaire
about emoticons on WhatsApp, case studies of online texts, and face-to-
face interviews with a selection of the participants in the study. The dis-
course analysis of online interactions was contextualized by the offline
data taken from the interviews, while the questionnaire worked as an
anonymous source of information and an initial point of departure.
Different sources were used to collect and generate good-quality data to
answer the research questions. In the methodology employed by Orgad,
no hierarchy was imposed on the different texts; consequently, online and
offline data were treated as interwoven rather than separate entities.
However, in this study, the questionnaire was used to delimit the scope of
the study since, as will be described below, the unexpected questionnaire
results were decisive in reorienting the research.
Orgad (2006) suggests that researchers must ask themselves questions
concerning the adequacy and usefulness of these combined methods. In
other words, academics should ask themselves whether obtaining offline data
could reveal important information about the context under study, which
would otherwise be impossible to obtain through other means. In this case,
the results obtained offline were used to triangulate those obtained online
and vice versa. Interviews with participants were very revealing and useful in
helping contextualize their text messages, while the study of text messages
was decisive in gaining a fuller understanding of what participants said in
their interviews and in the questionnaire, as discussed in the results section.
The questionnaire was subjected to statistical analysis. A mean com-
parison using ANOVA was undertaken. The variability of emoticon use
according to gender and age was studied for the ANOVA study.
6.3.2 Participants
The questionnaire was distributed among 400 participants (n ≈ 400):

200 teenage daily users of WhatsApp groups (100 females and 100 males)
and 200 adult daily users of WhatsApp groups (100 females and 100
males). It consisted of a five-point Likert scale about the frequency of
emoticon use in WhatsApp messages.
A reduced number of individuals (23 people) from among the ques-
tionnaire respondents volunteered to participate in the research. They
were interviewed, and also provided examples of their real written
WhatsApp exchanges to be examined by the researcher. In total, eight chat
threads were selected from the corpus provided: four from men-only
groups and four from women-only groups made up of close friends, whose
constant message exchanges constitute only a snapshot of a long interac-
tion over many years, as in Spilioti’s study (2011). In detail, this chapter is
based on a total dataset of 2087 messages/utterances gathered in 2015
from 47 participants, volunteers, and their friends (7 groups formed by 6
people and 1 group formed by 5), aged 35 to 49. This age range for adults
was chosen following the methodology of the WhatsApp, Switzerland
research project (www.whatsup-switzerland.ch). The total number of
words in the corpus is 8556, which means there is an average of 4.1 words
per message. In this regard, it is necessary to explain that, unlike Al
Rashdi’s (2015) data analysis, in this corpus, emoticons were counted as
words following Baron (2008), who considers them to be lexical issues.
The text messages collected were overwhelmingly relational in their
orientation, ranging from friendly salutations to social arrangements,
or substantial friendship maintenance. They were basically one-to-
many texts sent and received in a closed online community formed by
friends. Moreover, it is important to clarify that all the examples pre-
sented and discussed are from naturally occurring private electronic
discourse exchanged between friends. They were provided by partici-
pants in the study who gave permission for their use.3 None of the
group members knew in advance that their WhatsApp conversations
would be used in the study; therefore, these conversations are unelicited
and actually occurring.
6.3.3 Delimitation of the Study
Initially, in accordance with the literature on texting, the study sought to

examine whether there were any significant differences between teenag-
ers’ and adults’ online interactions in the use of emoticons and emoji. In
other words, the object of research was whether the age of the interactants
influenced the use of these iconic elements in their online texts.
After the results of the questionnaires were analyzed statistically, prac-
tically no differences between teenagers’ and adults’ use of emoticons
were found but there was significant difference in the way men and
women included emoticons in their chat groups. Consequently, the
research was redirected toward the observation of the distinct participa-
tion of both genders in their online communities. In particular, the study
focused on men-only and women-only online communities.
The threads are analyzed using the systematic taxonomy of the prag-
matic functions of iconic elements developed by Yus (2014). He classified
the pragmatic functions of emoticons into eight categories that range
from signaling the propositional attitude underlying the utterance, which
would be difficult to identify without the aid of the emoticon, to adding
a feeling or emotion toward the communicative act. On the whole, these
categories correspond “to the different ways in which emoticons satisfy
the user’s search for relevance” (Yus, 2014, p. 511). These emoticonical
expressions of attitudes can reveal the general underlying stance of the
user. In the next section, this taxonomy will often be referred to explain
the examples discussed.
In this corpus, Yus’s taxonomy is applied to what has been named coor-
dination activities (Ling, 2005) or micro coordination (Ling & Baron,
2013). The most common themes used in SMS are coordination, infor-
mation, answers, grooming, and others such as questions and requests
(Ling, 2005). The analysis of the eight cases selected and mentioned above
is restricted to those excerpts devoted to coordination activities such as
making arrangements with other close friends. CMC studies on the func-
tion of emoticons pointed out that emoticons are known to be used more
frequently in socioemotional contexts than in task-oriented contexts; for
example, we use these iconic elements habitually in environments where
there is camaraderie or friendship between participants, and in tight-knit

groups than in more neutral scenarios such as the workplace (Derks, Bos,
& von Grumbkow, 2007). Yus (2011) explains that IM users are more
likely to express feelings and emotions by means of emoticons when com-
municating with intimate friends, whereas emoticons are absent from
exchanges among mere acquaintances.
Attempting to analyze WhatsApp exchanges without this type of clear
focus would be fraught with many methodological problems and make a
comparison between genders impossible.
In the next section, interactional sociolinguistics is used to interpret
the findings. Further, in light of previous research on emoticons, some
explanations for their use are proposed. In conclusion, the broader impli-
cations of the growing use of emoticons in social media for relational
maintenance and in-group identification are considered.
6.4 R
esults and Their Interpretation
The questionnaire asking whether respondents include emoticons in
messages (1 for never, 2 for seldom, 3 for sometimes, 4 for very often, and
5 for always) revealed that, although no great differences exist in the rou-
tines of teenagers and adults when participating in their online commu-
nities, there is indeed a clear dissimilarity in the way men and women
interact in these environments. This is represented in Table 6.1.
Table 6.1 clearly demonstrates that the difference in the use of emoti-
cons by subjects’ gender is statistically significant. Thus, as mentioned
above, the study was subsequently reoriented to a discourse analysis of
parts of these chats to examine gender differences in as much detail as
possible, focusing specifically on threads that deal with coordination
activities in men-only and women-only groups of adult users aged 35–49.
Generally, the discourse analysis of these excerpts from WhatsApp chat
threads dedicated to coordination tasks corroborates the statements of
questionnaire participants and shows the following: women include
emoticons profusely in their messages, while men make sparse use of the
wide gallery provided by the messaging company.4
Table 6.1 Mean table with the results of the questionnairea

Gender Age Use of emoticons
Men Teenagers Mean 2.67
Std. dev. 1.397
Adults Mean 2.93
Std. dev. 0.730
Total Mean 2.79
Std. dev. 1.114
Women Teenagers Mean 4.47
Std. dev. 0.743
Adults Mean 4.25
Std. dev. 0.775
Total Mean 4.35
Std. dev. 0.755
Total Teenagers Mean 3.57
Std. dev. 1.431
Adults Mean 3.63
Std. dev. 0.999
Total Mean 3.60
Std. dev. 1.224
The mean comparison using ANOVA shows that the p-value was lower than
a
0.001, which indicates that there is significant difference between the results.
This attests the validity of the analysis carried out
To illustrate this finding, let us now observe in detail some extracts

dedicated to organizing a meal out.
In this thread (Table 6.2), there is no introduction or farewell; men go
straight to the point and simply agree or disagree with the proposal.
Their style could be defined as brisk and short, as seen in Ling, Baron,
Lenhart, and Campbell’s (2014) study of male teenagers’ texts. This
example also shows that the participants in the group are close friends,
since no surnames are added to the names of the members, and one
friend, Manolín, is identified with a diminutive, suggesting that the
owner of the telephone may have known him since childhood. The time
sequence is worth observing as well: they start to organize the event in
the early evening for the same day, although they may have had a previ-
ous face-to-face meeting where they agreed on the day of the get together.
What is significant is that they start this online coordination activity on
Sunday at 18:19 and finish the same day at 21:17, with their capacity
diminished by alcohol intake (utterance 11: “we are all blotto”). The only
Table 6.2 Example 1. A group of close male friends organizing a meal out, with
English translationa
Utterance Day Time Message Translation
1 Sun 18:19 Julio: Siete y algo vamos Julio: we’ll go there
pallaaaaa around seven
2 Sun 18:50 Hugo: Oki Hugo: okay
3 Sun 19:31 Julio: Hoy pizaaas Julio: pizzas today
4 Sun 19:31 Julio: Vesubio y napolitana? Julio: Vesuvius and
Napolitana?
5 Sun 19:32 Lino: Si, se lo has dicho a Lino: Yes, did you tell
Manolo? Manolo?
6 Sun 19:59 Manolín: Voy Manolín: I’m coming,
coming
7 Sun 19:59 Manolín: Y quesos Manolín: and cheese
8 Sun 20:03 Julio: A menos cuarto aqui Julio: be here at quarter
to
9 Sun 20:03 Julio: Y me llevas Julio: and give me a lift
10 Sun 20:05 Manolín: Manolín: okay emoticon
11 Sun 21:17 Julio: Ya estamoooos Julio: we are all blotto
torpedooos
12 Sun 21:17 Hugo: Voy cagando leches Hugo: I’ll peddle to the
metal
13 Sun 21:17 Hugo: Que los nanos me Hugo: the kids have
tienen machacao knackered me out
a
These excerpts have been chosen because they are clear representative
examples of the corpus gathered
graphical element displayed is the thumbs-up sign, a recurrent indicator

of approval commonly used by men in the corpus studied (see, e.g.
Sampietro, 2016b, for a study dedicated to this graphical element). This
is usually a stand-alone emoji that commonly closes an utterance with a
strong informal character (Sampietro, 2016b).
Example 1 reflects what sociologists call “lad culture” or “laddish
behavior”, a term usually applied to British heterosexual groups of mid-
dle class, young boys, young adults, and often middle-aged men who
reinforce their masculinity by alcohol consumption, objectifying women,
and liking and playing sport (Francis, 1999; Wheaton, 2004). In this
sociological context, there does not seem to be a place for smiley faces
and kisses; group identity is performed through a complete absence of
smiling or winking little faces.
Let us now turn our attention to another representative example of

coordination activities, but this time by a group of close female friends
organizing a meal out for the weekend (Tables 6.3 and 6.4):
Table 6.3 Example 2. Part one of an exchange among women organizing a meal
out with the English translation on the right
1 Tue 18:54 Rochi: Hola, propngo Rochi: Hello, I suggest
grupos para la comida y groups for lunch and
q cada una elija, vale? each of us can choose,
Somos 11 adultos ok? We are 11 adults
G1: vasos, platos, G1: glasses, plates, cutlery,
cubiertos, aperitivos starters
G2: bebidas G2: drinks
G3: postre G3: desert
G4, 5 y 6, comida, jeje G4, 5, and 6 food (pasta
(ensaladas de pasta, salads, Spanish omelets,
tortilla patata, pies … whatever you
empanadas … lo q se os come up with)
ocurra) Start choosing
Empieza la eleccion! Kisses (diminutive)
Besitis
2 Tue 19:14 Blanca: Si queréis yo m Blanca: If you want, I’ll go
pongo en el 3 y seguro q into group 3 and I’m sure
alguien se alegra, jeje someone will be happy,
haha
Table 6.4 Example 3. Part two of an exchange among women organizing a meal
out with the English translation on the right
1 Tue 23:13 Xtina: Bueno xicas, os Xtina: Right girls, I’m off to
dejo me voy a la bed, we’ll continue with
camita, mañana the menu stuff
seguimos con el tema tomorrow!!! I’m still not
menu!!! Sigo con sure about all the
dudas!!!! Bona nit details!!!! Goodnight
2 Wed 08:24 Laura: Es verdad … x q (The next morning) Laura:

no encargamos una You’re right … why don’t
paella???? we order a paella???
In this thread, the women start to coordinate the organization of the

event on a Tuesday at 18:45 and continue their conversation during the
evening and into the following day. However, things take a turn in the
morning when they suggest canceling the meal plans they made so far
and ordering takeaway paella instead. Their style can be described as
lengthy and full of unnecessary elements at a purely transactional or
informational level: there are many emoticons and often more than one
per utterance, as seen in Al Rashdi’s (2015) corpus. It can also be observed
that these emoticons usually occupy a final sentence position. According
to Yus’s (2014) taxonomy, the emoticons in these excerpts are simply used
to communicate the intensity of a feeling or an emotion that has already
been encoded verbally, for example, in the case of “kisses” followed by a
face throwing a kiss emoticon (Example 2, utterance 1), an emoticon
included frequently in women’s chats, as seen in Al Rashdi’s (2015) study,
or “haha” followed by several winking smiley faces (Example 2, utterance 2).
This excerpt also presents many of the traditional textual features histori-
cally associated with CMC such as reduplication of punctuation marks,
phonetic orthography, and other strategies of oralized written texts (Yus,
2011) or what Thurlow (2007) calls “textese” language. In parallel with
informality, the text is well organized and follows some patterns of formal
writing with paragraph divisions, use of accents and punctuation marks,
commas, and brackets. Furthermore, this short text contains one case of
code switching: the use of Bona nit (the Catalan phrase for goodnight).
Scholars have suggested that code switching is more common in face-
to-face oral communication than in writing (Li, 2002). Chat messaging
applications seem to be designed to simulate the immediacy and interac-
tivity of face-to-face conversation, consequently this “motivates multilin-
gual users to code-switch as they would do in a conversation” (Lee, 2011,
p. 11). In these examples, code switching, together with other strategies
such as code-alternation, switching to both formal and informal varieties,
is a clear strategy to reinforce affective ties, construct participant alliances,
and support playfulness (Georgakopoulou, 2011). Finally, as in Example
1, the absence of a surname in the identification of the members attests
to the close friendships between the chat members, as does the abbrevia-
tion of the name of one of the friends, with Cristina spelt Xtina by the
owner of the mobile phone.
Table 6.5 Example 4. Male coordination chat to watch a film

Line Day Time Message Translation
1 Thu 19:59 Pepe: Yeeeeee diez y cuarto? Pepe: Wasssuuuup 10:30?
2 Thu 20:14 Miguel: Luis pilla la peli Miguel: Luís, get the film
Horns Horns
3 Thu 20:15 Luís: Ok Pepe Luís: Ok, Pepe
4 Thu 20:15 Hector: Yo no Hector: I can’t
5 Thu 20:16 Manolo: Yo no puedo :( Manolo: I can’t :(
Turning to another conversation between men (Table 6.5), in which a

group of middle-aged men are arranging to meet to watch a film, we
observe considerable differences when compared to the previous female
coordination examples.
Similarly to Example 1, this men-only chat reflects the briskness of
male interactions, in the sense that there are usually no greetings or fare-
wells, and many imperative forms, although, in this case, we can see some
sort of greeting by means of a very colloquial “Yeeeee”, an equivalent of
the English “Wassuuup”. This short text also exemplifies another salient
point that has surfaced in the study and which requires further research
with a broader corpus: traditional text-deformation emoticons (Yus, 2014)
are used only by men. Here, the sad face emoticon adds a feeling or emo-
tion toward the propositional content of the utterance, the example exhib-
its a negative emotion (sadness) that the user feels toward the information
provided by the propositional content of the utterance. The emoticon
occupies the final position in the thread, one of the most habitual posi-
tions of emoticons according to Sampietro (2016a). This group usually
adopts ASCII emoticons as a sign of group identity, as a unique social
language shared by the group members to create shared and secret unique-
ness (Kelly & Watts, 2015). In the interviews, some of the group members
said they work as engineers and computer scientists, and this may be the
reason why they chose traditional emoticons to construct group identity.
Regarding interview findings, participants underlined the conversa-
tional character of these written interactions. The men interviewed
emphasized the fact that these chats between close friends are continu-
ously active, what Spilioti (2011) calls the frame of “perpetual contact” in
mobile technologies. Consequently, there is no need to use what the
interviewees call “unnecessary elements”, such as introductions, farewells,

and emoticons to maintain affective ties in this maximum speed, mini-
mum effort communication medium. The structure of these never-
ending conversations among men appears to coincide with findings of
other studies which highlight the conversational frame of texting and its
adherence to conversational rather than prescriptive forms for writing
(e.g. Thurlow & Poff, 2013).
In the interviews with women, however, the “unnecessary” elements
were reported to be indispensable. WhatsApp messages without saluta-
tions or closings may sometimes be considered inadequate but those
without emoji are definitely believed to look brusque and even rude, as
also reported by Al Rashdi (2015). Another interesting opinion expressed
in these semi-structured interviews was that online practices reflect simi-
lar face-to-face interactions. As Yus (2014) states, these online exchanges
mirror face-to-face interactions: women kiss each other in face-to-face
meetings and in their WhatsApp exchanges, whereas men only kiss other
men when they are members of their family or in groups of gay male
friends.5 This stands in stark contrast to Al Rashdi’s (2015) results, which
show that men and women include the face throwing a kiss emoticon
repeatedly in their messages because, in Oman, men and women kiss and
hug each other when they greet their same-sex friends in person. This
brings to the fore what Derks et al. (2007) note: emoji use is heavily
influenced by linguistic and social contexts, and by both cultural and
personal conventions. The results show the need to add gender and per-
sonal or group conventions to this statement, as crucial elements in deter-
mining emoticon preference.
Overall, the distinction between men’s report talk and women’s rapport
talk (Tannen, 1991) is not applicable to the cases examined, as both gen-
ders use these communities mainly for relational purposes. The detailed
discourse analysis of the written messages shows that although the rela-
tional function is at the heart of all of them, it is expressed differently in
men’s and women’s chats. Whereas men usually go straight to the point,
women habitually introduce their texts with a salutation or addressivity,
and also interrupt their written conversations with graphic emoticons,
which function as community conventions which are used with the inten-
tion of building intimacy. It seems that texters in women-only groups
need emoticons to build relationships, while their male counterparts feel

no need to emphasize or build an already-existing relationship by using
what they consider to be unnecessary elements. Moreover, the analysis
demonstrates that friends and peer groups establish their own local stylis-
tic norms. As in the study of Thurlow and Poff (2013), the groups of
friends under scrutiny here are able to creatively transform multimodal
digital literacy resources to construct group identities. Yus (2014) observes
that in the chatrooms he analyzed, emoticons and emoji simply generate
a colorful visual arrangement of the text “so that the message typed ‘stands
out from the textual crowd’ and arouses an interest in the other users”
(p. 513). In most female chats studied here, emoticons are used simply to
communicate the intensity of a feeling or emotion that has already been
coded verbally, for example, “kisses” followed by a face throwing a kiss
emoticon (Example 2, utterance 1), or to add a feeling or emotion toward
the communicative act itself, as in Example 3, utterance 2 (Yus, 2014). In
other words, female participants use emoticons to emphasize belonging to
the group regardless of content. The excerpts examined suggest that men
and women desire to build friendships and enhance their relationships
through the technology. In this context, emoticons, as Maíz Arévalo
(2014) argues, work as group solidarity and rapport boosters, mirroring
face-to-face interactions (Yus, 2014).
A recent article by Kelly and Watts (2015) underlines how emoji are
appropriated in mediated discourse. Following Dix (2007), they define
“appropriation” as “usage that lies beyond a designer’s original intent”
(Kelly & Watts, 2015). Beyond the role of emoticons for conveying emo-
tional states, they discuss three categories of appropriation: to maintain a
conversational connection, to permit play, and to create shared and secret
uniqueness. For example, emoticons are used to control a thread or to
encourage playful behavior so as to maintain social bonds among partici-
pants. Specifically, in close personal relationships such as those among
interactants in this corpus, Kelly and Watts (2015) suggest that emoji are
used to maintain connections. They give the example of emoji sent by the
recipient to acknowledge a message, which prevents the sender from feel-
ing ignored due to a lack of response. They can also serve to promote
feelings of intimacy within the context of a relationship.
This is also in line with Jones and Hafner’s (2012) claim that users of
mobile communication, especially young people, do not use the technol-
ogy to exchange information but to exchange friendship. Similarly, in
their study of texting between teenagers, Berg, Taylor, and Harper (2005)
compare text messages to the practice of gift giving.
In this relational function of the technology, academic studies have
highlighted that women often employ mediated communication for rela-
tionship maintenance (Colley, Todd, White, & Turner-Moore, 2010).
Women use texting for social purposes, while men’s texting is more
instrumental; in other words, women see the smartphone as a social
channel, whereas men view it as a mere tool (Ling et al., 2014). However,
this does not appear to be the case for the messages studied in this chapter
since male-only communities also use their mobile phones for relation-
ship maintenance, although these relationships are maintained differently
and their exchanges are stylistically different from those of the female-
only communities. Building on Stark and Crawford (2015), it can be
affirmed that emoji and emoticons have become decisive elements in the
affective mix of relationship maintenance, sustenance, and continuation,
especially in women’s chats.
Finally, although the affordances of mobile technologies can enhance
the use of stylistic elements such as emoticons, a simplistic deterministic
approach should be avoided. In spite of the immense gallery of graphic
elements available, men include them infrequently in their chats, while
women, in contrast, make abundant use of them. One plausible reason
for this imbalance is that women tend to use standard language more
often than their male counterparts, along the lines argued by Squires
(2012) regarding IM and gender variation. Therefore, bearing Squires’s
(2012) ideas in mind, it could be claimed that the use of standard lan-
guage on WhatsApp involves the usage of all the affordances provided by
this application, namely a large, continuously updated gallery of emoti-
cons, emoji, and stickers. Indeed, in the eight cases examined, women
appear to be adopting standard language practices that require an abun-
dant use of emoticons. It seems that “there is something feminine about
conforming to standard written expectations in this medium, and/or
something masculine about not conforming” (Squires, 2012, p. 312).
6.5 Conclusions
The findings show that gender-based differences persist in communica-
tions via internet-based IM applications, specifically WhatsApp. The
study reveals that stereotypes regarding gendered emotional expression
are also present in exchanges taking place via online messaging applica-
tions. On the other hand, within the debate of the sensitiveness of CMC
to technological and social constraints (Lorenzo-Dus & Bou-Franch,
2013), the examples provided demonstrate that the affordances of
WhatsApp do not determine the actions of users: we have seen that
women employ galleries of emoticons profusely to maintain their already
close friendships, whereas men do not; men maintain their relationships
by omitting what they consider to be superfluous elements.
With regard to emoticon interpretation, it must be noted that there is
no difficulty in interpreting the emoticons in these threads, as may be the
case in other studies, since graphic representations in this corpus are mainly
kisses, winks, and flowers. Here, unlike in other studies on social network-
ing and IM (e.g. Maíz Arévalo, 2014), the participants form a homoge-
neous audience who share the same degree of (in)formality and consider
the use of emoticons in their community appropriate. The study of other
communities may yield results that could reveal problems associated with
emoticon contextualization and interpretation of their function.
The study is, of course, limited. A large multinational comparative
project such as the one currently underway in Switzerland (www.what-
sup-switzerland.ch) would be required to provide results with a wider
scope, which could account for the greater variability between texters and
the messages they send.
How men and women interact in men-only and women-only groups has
been analyzed in this chapter. Future studies could analyze how participants
in mixed groups accommodate their writing practices to those of the other
gender. Research into online accommodation has indicated that a common
feature of online groups, which is similar to face-to-face interaction, is that
online members accommodate to each other (Pérez-Sabater, 2017).
However, despite its limitations, this study has provided a first glimpse
into some communities that use the WhatsApp messaging application and
can be seeing as a starting point for further investigations in this promising
CMC field, where technological developments advance rapidly. WhatsApp

is constantly growing and adopting new channels and new facilities, which
may make these results obsolete almost overnight, as explained by Spooren
and van Charldorp (2014). We must be aware that the factors that influ-
ence the adoption and usage of new technologies are complex and some-
times unpredictable. However, as the primary objective of WhatsApp is to
enable communication, the need to interact will trigger the adoption of
another popular messaging service with similar affordances, making the
research published in this article scholarly relevant again.
As a final remark, it is necessary to clarify that these WhatsApp groups
seem to be formed, to a great extent, by heterosexual participants. The
study of other chat groups formed by lesbian and gay male participants,
or those with different gender identities, may yield different results which
may help us understand gender similarities and differences in language
use in messaging applications. Another issue to note is that this sample is
not systematic, and, since it was not selected at random from a larger
pool of chats, the findings cannot be extrapolated and applied to the
general populace of users. The analysis can only show how some online
communities interact. Future studies, which may involve several research-
ers and/or a larger corpus, could yield more general outcomes.
Lastly, future research can involve the study of emoticons in public dis-
course. As a striking example of how emoticons are increasingly occupy-
ing public spaces these days, the obituary shown below published in El
Periódico de Catalunya on 28 May 2016 features the winking face emoti-
con in the space traditionally occupied by the Christian cross in Spanish
obituaries.6 An obituary featuring an emoticon would have been incon-
ceivable a few years ago. In the threads analyzed, women are shown to
incorporate emoji profusely in their interactions, and, in line with the
findings, this obituary is to announce the death of a woman who requested
that the emoticon be included in her death notice. In previous research
about mourning sites on Facebook, such as the one for the death of Steve
Jobs (Holiman, 2013), emoticons were often incorporated into the text
to express emotion, usually a sad face; in other corpora, however, such as
the one examined by Giaxoglou (2014), there is a clear lack of pictorial
elements. In Spain, where tradition has excluded humor from mourning
ceremonies, this emoticon appearing in the header of an obituary is an
example of the ongoing evolution of newspaper obituaries toward less
formal styles, as documented by Ollanquindia (1998), and is intimately

related to a broader tendency in the general evolution of literacy and
public discourse toward informality (Montero-Fleta, Montesinos-López,
Pérez-Sabater, & Turney, 2009):
Acknowledgments Thanks to the editors, Patricia and Pilar, and the anony-
mous reviewers for their valuable suggestions that have greatly improved the
original manuscript. Special gratitude to my colleague and friend Begoña
Montero-Fleta for her help in processing the questionnaire and the corpus.
Notes
1. According to the United Nations, mobile broadband is the most dynamic
market segment; globally, mobile broadband penetration reached 47% of
the world population in 2015 (http://www.itu.int/en/ITU-D/Statistics/
Pages/facts/default.aspx accessed 12 September 2015). In September
2015, Jan Koum, the founder of WhatsApp, announced on Facebook
that it had 900 million monthly active users.
2. In Spain, conventional face-to-face greetings and farewells among friends
involve kissing the person you are meeting twice, once on each cheek.
3. All examples reproduced in this article are from the corpus gathered.
Participants’ telephone numbers have been removed to protect their
privacy.
4. Men include an emoticon in 17% of their utterances, while women
include them in 82%. However, since only the parts of this corpus devoted
to organizing events have been examined in detail, no more detailed sta-

tistics will be given with regard to occurrences and type of emoticon per
utterance. This quantitative analysis has been left for another time.
5. The case studies analyzed are chats between heterosexual participants, as
participants declared in the interviews. The analysis of gay male groups
may yield different results related to the use of the face throwing a kiss
emoticon.
6. Used with kind permission of El Periódico de Catalunya.
References
Al Rashdi, F. (2015). Forms and functions of emojis in WhatsApp interaction
among Omanis (Doctoral dissertation). Retrieved from ProQuest Dissertations
& Theses Global (1703437877).
Barak, A., Boniel-Nissim, M., & Suler, J. (2008). Fostering empowerment in
online support groups. Computers in Human Behavior, 24(5), 1867–1883.
Baron, N. S. (2004). ʽSee you online’: Gender issues in college student use
of Instant Messaging. Journal of Language and Social Psychology, 23(4),
397–423.
Baron, N. S. (2008). Always on: Language in an online and mobile world. Oxford:
Berg, S., Taylor, A. S., & Harper, R. (2005). Gift of the gab. In R. Harper,
L. Palen, & A. Taylor (Eds.), The inside text (pp. 271–285). Dordrecht:
Springer.
Bieswanger, M. (2013). Micro-linguistic structural features of computer-
mediated communication. In S. C. Herring, D. Stein, & T. Virtanen (Eds.),
Pragmatics of computer-mediated communication (pp. 463–487). Berlin/
Boston: Walter de Gruyter.
Colley, A., Todd, Z., White, A., & Turner-Moore, T. (2010). Communication
using camera phones among young men and women: Who sends what to
whom? Sex Roles, 63(5–6), 348–360.
Derks, D., Bos, A. E. R., & von Grumbkow, J. (2007). Emoticons and social
interaction on the internet: The importance of social context. Computers in
Human Behavior, 23, 842–849.
Dix, A. (2007). Designing for appropriation. In D. Ramduny-Ellis &
D. Rachovides (Eds.), Proceedings of the 21st BCS HCI Group Conference
(pp. 27–30). Lancaster: The British Computer Society.
Fox, A. B., Bukatko, D., Hallahan, M., & Crawford, M. (2007). The medium
makes a difference: Gender similarities and differences in instant messaging.
Journal of Language and Social Psychology, 26(4), 389–397.
Francis, B. (1999). Lads, lasses and (new) labour: 14–16-year-old students’
responses to the ‘laddish behaviour and boys’ underachievement’ debate.
British Journal of Sociology of Education, 20(3), 355–371.
Georgakopoulou, A. (2011). On for drinkies?: E-mail cues of participant align-
ments. Language@Internet, 8(4). Retrieved from http://www.languageatint-
ernet.org/articles/2011/Georgakopoulou
Giaxoglou, K. (2014). “RIP man … u are missed and loved by many”:
Entextualising moments of mourning on a Facebook Rest in Peace group
site. Thanatos, 3(1), 10–28.
Herring, S. C. (2012). Grammar and electronic communication. In C. Chapelle
(Ed.), The encyclopedia of applied linguistics (pp. 1–11). Hoboken, NJ: Wiley-
Blackwell. Retrieved from http://ella.slis.indiana.edu/~herring/e-grammar.
pdf
Holiman, J. M. (2013). iGrieve: Social media, parasocial mourning and the death
of Steve Jobs. Master’s thesis. Retrieved from https://www.suu.edu/hss/comm/
masters/capstone/thesis/igrieve-holiman-j.pdf
Huffaker, D. A., & Calvert, S. L. (2005). Gender, identity, and language use in
teenage blogs. Journal of Computer-Mediated Communication, 10(2).
Ishii, K. (2006). Implications of mobility: The uses of personal communication
media in everyday life. Journal of Communication, 56(2), 346–365.
Jones, R. H., & Hafner, C. A. (2012). Understanding digital literacies. London/
Kapidzic, S., & Herring, S. C. (2011). Gender, communication, and self-
presentation in teen chatrooms revisited: Have patterns changed? Journal of
Computer-Mediated Communication, 17(1), 39–59.
Kelly, R. M., & Watts, L. A. (2015). Characterising the inventive appropriation
of emoji as relationally meaningful in mediated close personal relationships.
In Proceedings of experiences of technology appropriation: Unanticipated users,
usage, circumstances, and design (workshop held at ECSCW 2015). Retrieved
from https://www.google.es/#safe=strict&q=Characterising+the+inventive+a
ppropriation+of+emoji+as+relationally+meaningful+in+mediated+close+per
sonal+relationships
Lee, C. (2011). Micro-blogging and status updates on Facebook: Text and prac-
tices. In C. Thurlow & K. Mroczek (Eds.), Digital discourse: Language in the
new media (pp. 110–128). Oxford: Oxford University Press.
Lee, C. K. (2007). Linguistic features of email and ICQ instant messaging in
Hong Kong. In B. Danet & S. C. Herring (Eds.), The multilingual Internet:
Language, culture, and communication online (pp. 184–208). Oxford: Oxford
University Press.
Li, D. C. S. (2002). Cantonese-English code-switching research in Hong Kong:
A survey of recent research. In K. Bolton (Ed.), Hong Kong English: Autonomy
and creativity (pp. 79–99). Hong Kong: Hong Kong University Press.
Ling, R. (2005). The sociolinguistics of SMS: An analysis of SMS use by a ran-
dom sample of Norwegians. In R. Ling & P. Pedersen (Eds.), Mobile com-
munications (pp. 335–349). London: Springer.
Ling, R., & Baron, N. S. (2013). Mobile phone communication. In S. C.
Herring, D. Stein, & T. Virtanen (Eds.), Pragmatics of computer-mediated
communication (pp. 191–216). Berlin/Boston: Walter de Gruyter.
Ling, R., Baron, N. S., Lenhart, A., & Campbell, S. W. (2014). “Girls text really
weird”: Gender, texting and identity among teens. Journal of Children and
Media, 8(4), 423–439.
Ling, R., Julsrud, T., & Yttri, B. (2008). New tech, new ties: How mobile com-
munication is reshaping social cohesion. Cambridge, MA: The MIT Press.
Lorenzo-Dus, N., & Bou-Franch, P. (2013). A cross-cultural investigation of
email communication in Peninsular Spanish and British English: The role of
(in)formality and (in)directness. Pragmatics and Society, 4(1), 1–25.
Maíz Arévalo, C. (2014). A pragmatic and multimodal analysis of emoticons
and gender in social networks. In A. Sánchez Macarro & A. Cabrejas Peñuelas
(Eds.), New insights into genedered discoursive practices: Language, gender and
identity construction (pp. 175–197). Valencia: Universitat de València.
Markman, K. M. (2013). Conversational coherence in small group chat. In
S. C. Herring, D. Stein, & T. Virtanen (Eds.), Pragmatics of computer-
mediated communication (pp. 539–564). Berlin/Boston: Walter de Gruyter.
McKeown, J., & Zhang, Q. (2015). Socio-pragmatic influence on opening salu-
tation and closing valediction of British workplace email. Journal of
Pragmatics, 85, 92–107.
Montero-Fleta, B., Montesinos-López, A., Pérez-Sabater, C., & Turney, E.
(2009). Computer mediated communication and informalization of dis-
course: The influence of culture and subject matter. Journal of Pragmatics,
41(4), 770–779.
Nishimura, Y. (2015). A sociolinguistic analysis of emoticon usage in Japanese blogs:
Variation by age, gender, and topic. The 16th Annual Meeting of the Association
of Internet Researchers, Phoenix, AZ.
Ollanquindia, R. (1998). Las esquelas y los cambios de mentalidad. Cuadernos

de etnología y etnografía de Navarra, 30(71), 39–52.
Orgad, S. (2006). The cultural dimensions of online communication: A study of
breast cancer patients’ internet spaces. New Media & Society, 8(6), 877–899.
Oxford Dictionary, Language Matters. Retrieved February 2, 2016, from http://
blog.oxforddictionaries.com/press-releases/announcing-the-oxford-diction-
aries-word-of-the-year-2015/
Pérez-Sabater, C. (2017). Linguistic accommodation in online communication:
The role of language and gender. Revista Signos, 50(94), 265–286.
Pérez-Sabater, C., Turney, E., & Montero-Fleta, B. (2008). Orality and literacy,
formality and informality in email communication. Ibérica: Revista de la
Asociación Europea de Lenguas para Fines Específicos (AELFE), 15, 71–88.
Plester, B., Wood, C., & Joshi, P. (2009). Exploring the relationship between
children’s knowledge of text message abbreviations and school literacy out-
comes. British Journal of Developmental Psychology, 27(1), 145–161.
Sabaté i Dalmau, M. (2014). A sociolinguistic analysis of transnational SMS
practices. In L. A. Cougnon & C. Fairon (Eds.), SMS communication: A lin-
guistic approach (pp. 169–190). Amsterdam/Philadelphia: John Benjamins.
Sampietro, A. (2016a). Emoticonos y emojis: análisis de su historia, difusión y uso
en la comunicación digital actual. Retrieved from http://roderic.uv.es/han-
dle/10550/53873
Sampietro, A. (2016b). Emoticonos y multimodalidad. El uso del pulgar hacia
arriba en WhatsApp. Aposta: Revista de Ciencias Sociales, 69, 271–295.
Sanmartín, J. (2007). El Chat. La conversación tecnológica. Madrid: Arco Libros.
Cuadernos de Lengua Española.
Spilioti, T. (2011). Beyond genre: Closings and relational work in text-messaging.
In C. Thurlow & K. Mroczek (Eds.), Digital discourse: Language in the new
media (pp. 67–84). Oxford: Oxford University Press.
Spooren, W., & van Charldorp, T. (2014). Challenges and experiences in col-
lecting a chat corpus. Journal for Language Technology and Computational
Linguistics, 29(2), 83–96.
Squires, L. (2012). Whos punctuating what? Sociolinguistic variation in instant
messaging. In A. Jaffe, J. Androutsopoulos, & M. Sebba (Eds.), Orthography
as social action: Scripts, spelling, identity and power (pp. 289–323). Munich:
De Gruyter Mouton.
Stark, L., & Crawford, K. (2015). The conservatism of emoji: Work, affect, and
communication. Social Media + Society, 1(2). Retrieved from http://sms.sage-
pub.com/content/1/2/2056305115604853.full.pdf+html
Tannen, D. (1991). You just don’t understand: Women and men in conversation.
New York: William Morrow.
Thurlow, C. (2007). Fabricating youth: New-media discourse and the technolo-
gization of young people. In S. Johnson & A. Ensslin (Eds.), Language in the
media: Representations, identities, ideologies (pp. 213–233). London:
Continuum.
Thurlow, C., & Brown, A. (2003). Generation Txt? The sociolinguistics of
young people’s text-messaging. Discourse Analysis Online, 1(1), 30–57.
Retrieved form http://extra.shu.ac.uk/daol/articles/v1/n1/a3/thur-
low2002003-paper.html
Thurlow, C., & Poff, M. (2013). Text messaging. In S. C. Herring, D. Stein, &
T. Virtanen (Eds.), Pragmatics of computer-mediated communication
(pp. 162–189). Berlin/Boston: Walter de Gruyter.
Vázquez-Cano, E., Mengual-Andrés, S., & Roig-Vila, R. (2015). Análisis lexico-
métrico de la especificidad de la escritura digital del adolescente en WhatsApp.
RLA. Revista de Lingüística Teórica y Aplicada, 53(1), 83–105.
Wheaton, B. (2004). New lads’? Competing masculinities in the windsurfing
culture. In B. Wheaton (Ed.), Understanding lifestyle sports: Consumption,
identity, and difference (pp. 131–153). London/New York: Routledge.
Wolf, A. (2000). Emotional expression online: Gender differences in emoticon
use. Cyberpsychology & Behavior, 3(5), 827–833.
Yus, F. (2014). Not all emoticons are created equal. Linguagem em (Dis)curso,
14(3), 511–529.
Part IV
Face and Identity
7
From “My Manly Husband…” to
“… Sitting Down to Take a Pee”:
The Construction and Deconstruction
of Gender in Amazon Reviews
Camilla Vásquez and Addie Sayers China
7.1 Introduction
Over the past decade, online reviews have emerged as a popular form of
vernacular literacy. Research has demonstrated that online reviews fea-
tured on websites such as Amazon have an impact on consumer decision-
making and consumer spending (e.g., Chevalier & Mayzlin, 2006; Ghose
& Ipeirotis, 2011). As prospective consumers read these reviews, they pay
attention not only to the product information that is presented therein,
but they are also sensitive and responsive to implicit cues as well as to the
more explicitly communicated information about the social identities of
the authors who have created online reviews (e.g., Sen & Lerman, 2007;
Vermeulen & Seegers, 2009). Gender and family roles and relationships
have been identified as among the most common demographic categories
C. Vásquez (*) • A. Sayers China

Department of World Languages, University of South Florida,
Tampa, FL, USA
e-mail: cvasquez@usf.edu; sayersal@mail.usf.edu

https://doi.org/10.1007/978-3-319-92663-6_7
194 C. Vásquez and A. Sayers China
that authors of online reviewers inscribe into their review texts (Vásquez,
2014). Because discourse is one of the primary means through which
identity work is accomplished (Benwell & Stokoe, 2006; Bucholtz &
Hall, 2005; Gee, 2011; Page, 2012), this chapter focuses on how Amazon
review writers use discourse to construct and orient to gendered identities,
as well as on how they invoke gender in specific product descriptions.
The analysis in this chapter draws on two datasets of Amazon reviews:
100 bona fide or “legitimate” reviews1 and 100 parody reviews (Ray,
2016; Skalicky & Crossley, 2015). Our analytic focus on references to
relational identities and explicitly marked gender roles in legitimate
reviews written about two different categories of products (i.e., high-
speed blenders and diaper bags) reveals the constitutive role of discourse
in the gendering of certain product features and highlights the ways in
which reviewers reproduce normative gender stereotypes in these texts.
We then contrast these bona fide product reviews with a corpus of 100
Amazon parody reviews written about two popularly parodied products.
We illustrate how these parody reviews critique different realizations of
hegemonic discourses. In one case, authors of parody reviews challenge
the corporate gendering and gendered marketing of a product (Bic for
Her Pen); in the second case, authors interject site-external political dis-
courses into discussions of a semiotically more “neutral” product (Avery
Durable View Binder). By contrasting these two different text types (i.e.,
legitimate reviews and parodies of reviews), it becomes apparent that
the commercial online review space can also serve as a site for contesta-
tion and resistance of gender stereotypes, particularly when viewed as
post-structuralist instances of deconstructive language play (Gannon &
Davies, 2012). We propose, therefore, a plurality of gender construc-
tions in Amazon review discourse. That is, some product reviewers
reproduce and reify essentialist gender binaries, whereas authors of par-
ody reviews posted on the same site use discourse to destabilize, denatu-
ralize, and deconstruct gender ideologies. As a result, the Amazon review
space simultaneously affords both modernist and postmodernist con-
ceptualizations of gender identities. We further argue that the review
space on Amazon can thus be considered a virtual agora, a site where
commercial activities and discourses coexist alongside political and
moral discourses.
From “My Manly Husband…” to “… Sitting Down to Take a Pee”… 195
7.2 G
ender Identities in Online Contexts
Multiple, and even contradictory, conceptualizations of gender identities
can be found in research on gender and digital discourse practices. From
the early, foundational explorations of Usenet discussion boards and
email listservs (Balka, 1993; Herring, 1993) to more recent studies of
social media platforms such as Tumblr (Kanai, 2015) and Pinterest
(Ottoni et al., 2013), researchers have observed that both the larger social
context and the technological affordances associated with specific plat-
forms impact gender performances and constructions online.
Early computer-mediated communication (CMC) research empha-
sized the democratizing potential of the internet, highlighting the pri-
macy of words, rather than bodies, in text-based communication (Graddol
& Swann, 1989; Turkle, 1995). Researchers such as Danet (1998), for
instance, argued that the anonymity afforded by Internet Relay Chats
(IRC) allowed for “carnivalesque” language play and experimentation
with gender identities. Similarly, Rodino’s (1997) IRC participants broke
out of traditional gender roles despite the presence of a binary gender
system operating in the larger IRC environment.
More recently, however, researchers of digital communication have
challenged the notion that anonymity or pseudonymity can erase or miti-
gate gender asymmetries (Herring & Stoerger, 2013), contending both
that language constrains social practice online (Wynn & Katz, 1997) and
that gender is intrinsic to language (Herring, 2004). Herring and Stoerger
(2013) found a correlation of gender with interaction style—for exam-
ple, in the use of emoticons, gendered lexical choices, and intensifiers—
across several text-based digital domains. Similar findings regarding
gender and style have also been found in studies of communication on
social networking sites (SNS) Twitter (Kivran-Swaine, Brody, & Naaman,
2013) and MySpace (Fullwood, Morris, & Evans, 2011). In one study,
women and men were found to use different rhetorical strategies on
Twitter, with women using more personal involvement strategies and
men more directly persuasive approaches (Cunha, Magno, Almeida,
Gonçalves, & Genevuto, 2012). Similarly, women and men were found
to differ in their interaction type on the image-based SNS Pinterest:
women participated in more “lightweight” interactions and made more
efforts to reciprocate in social links as more active and more generalist

posters, while men, on the other hand, tended to act as specialists by
expressing themselves in more assertive terms and by curating content
more of personal, rather than social, interest (Ottoni et al., 2013).
Finally, while investigations of language and digital discourse practices
highlight the varying performances and constructions of gender across
different sites, a smaller set of studies with a slightly different focus have
examined the various affordances associated with particular digital plat-
forms with respect to women’s agency and objectification. Despite the
fact that certain aspects of digital culture, such as the mashup and creativ-
ity of meme humor, have been found to perpetuate sexist tropes against
women (Marwick, 2014), both the internet and SNS also provide poten-
tial for feminist and activist practices. Harris (2008) and Marwick (2014)
discuss how women use blogs and SNS to take up alternative subject
positions in order to counter-hegemonic discourses. For instance, in
Marwick’s study, blogger Julia Alison, in “presenting herself as an object,
suggests an agented subjectivity that threatens the male dominated social
hierarchy” (Marwick, 2014, p. 69). Women and girls also use SNS and
blogs as part of a do-it-yourself (DIY) cultural framework in which they
seek control to construct their public selves, build spaces for public peer
communities, and develop new modes of activism and political subjec-
tivities (Harris, 2008, p. 492). Some platforms, such as the blogging SNS
Tumblr, appear as critical sites for female and feminine subjectivities.
Kanai (2015), for example, locates practices of feminine authenticity at
the intersection of SNS and remix culture afforded by Tumblr, while
Shorey (2015) also underscores the importance of Tumblr’s affordances of
common norms, anonymity, and audience specificity for providing spaces
for girls for self-expression, particularly with respect to expression of cul-
turally devalued emotions for women, such as sadness and anger. Finally,
Connelly (2015) illustrates through textual and thematic analysis how
the affordance of anonymity in Tumblr contrasts with anonymity in other
feminist spaces online by creating a space for feminist “world” building,
through streamed collective consciousness.
The various affordances, digital practices, and linguistic resources
deployed across digital platforms provide users with multiple possibilities
in the performance and construction of gender identities, as well as in the
contestation of those identities. While research generally focuses on only

one aspect of this domain of gender and digital practice, we will consider
both here: gender performance and construction in bona fide reviews, as
well as gender deconstruction in parody reviews. We explore these issues
on the site of e-retailer, Amazon.
7.3 O
nline Consumer Reviews and Their
Parodies
As an exemplar of participatory culture (Jenkins, 2006), the user-
generated online consumer review has evolved over the last decade, into
a global, widespread genre of digital communication. Research interest in
the topic of online consumer reviews has been growing as well, with an
increasing number of publications found across a wide variety of disci-
plines, including marketing, economics, computer science—and, more
recently, discourse studies. As discourse analytic research about online
review language has shown, this digital genre is a distinctive one, owing
to the ways in which authors of online reviews claim particular identities
(Mackiewicz, 2010a, 2010b) in the typical forms of evaluation they use
(Skalicky, 2013; Taboada, 2011; Tian, 2013), in how they create and
represent audience involvement (Vásquez, 2012, 2015), as well as in their
intertextual and narrative practices (Jurafsky, Chahuneau, Routledge, &
Smith, 2014; Vásquez, 2014).
In earlier work (Vásquez, 2014), we identified several features shared
by 1000 online reviews sampled from various websites, including
Amazon. Among these features were the discursive constructions of
specific, context-relevant reviewer identities. Research on reviews has
shown that many review writers claim situationally relevant identities
in order to establish their credibility, and to provide review readers
with additional context for interpreting the evaluative claims made
within a review. Both gender and family roles and relationships have
been identified as some of the most common demographic or mem-
bership categories that authors of online reviewers inscribe into their
review texts.
As online reviews have become established as a genre of CMC, and

have continued to grow both in their number and their influence over the
last decade, parodies of the genre have increasingly appeared on popular
review sites (such as Amazon), alongside legitimate reviews. Addressing
the question of differences between parodic and nonparodic Amazon
reviews, a recent corpus-based analysis (Skalicky & Crossley, 2015) com-
pared a sample of 375 parody reviews with a similar sample of legitimate
reviews. The analysis of lexical, grammatical, and semantic features found
that the parodic texts were characterized by significantly higher frequen-
cies of past tense words, as well as significantly higher levels of word
concreteness. Just as authors of legitimate reviews inscribe their social
identities into their texts, authors of review parodies also provide various
types of discursively constructed “personal” information. This performed
self-disclosure serves two functions in parody texts: it mimics a discourse
convention that appears in legitimate Amazon reviews, and it also func-
tions as an anchoring device for the unusual or improbable events which
are presented within the narrative texts.
Another recent study (Ray, 2016) has also taken a corpus approach to
examining the language of parody reviews. Ray’s analysis centered on the
notion of “stylization” and the gender stereotypes found in Amazon reviews
of one popularly parodied product: Bic for Her Pens. While scholars from
other disciplines have begun to notice that “there is, in fact, a thriving and
popular genre of humorous Amazon reviewing” (Kozinets, 2016, p. 836),
Ray’s study represents a unique contribution because it moves beyond the
humorous aspect of parody and recognizes its potential for social critique.
Specifically, Ray considered the ways in which authors of parody reviews
have “reframed patriarchal discourses” about a “sexist product” (2016,
p. 42). With respect to identifying review texts as parodies, Ray noted that
the parodies in his corpus typically mentioned “an implausible or impossi-
ble event,” “unreasonable explanations […or] expectations,” or expressions
of “emotional states that seemed unlikely” (2016, p. 63). Drawing on the
sociolinguistic notion of “genderlect” (Tannen, 1991), Ray argues that par-
ody reviewers utilize the highly salient, and often stereotypical, linguistic
features associated with gendered speech styles—specifically, exaggerated
and hyperbolic lexical items and empty adjectives—as part of their innova-
tive, technofeminist rhetorical practices (2016, p. 43).
While academic scholarship on parodies of consumer reviews is still

nascent, numerous forms of digital media—such as online newspapers,
blogs, and content aggregator sites—have been showcasing these paro-
dies (sometimes referred to as “ironic,” “spoof,” “sarcastic,” or “funny”
reviews) since they first began appearing on Amazon around 2005 (e.g.,
Doward & Craig, 2012; Pogue, 2010; Zeller, 2006). Even though paro-
dies of reviews do not reflect the primary goals of Amazon, the fact that
they currently number in the tens of thousands suggests that they are a
textual phenomenon which merits further investigation. Building on
existing corpus-based research about the language of Amazon reviews and
their parodies (Ray, 2016; Skalicky & Crossley, 2015), we take a closer,
qualitative, discourse analytic approach to examining how gendered
identities and related ideologies are performed and constructed in 100
legitimate reviews and 100 parody reviews.
7.4 M
ethods
Our approach to the study of social identities in general—and gender
identities more specifically—is informed by scholars who view discourse as
the primary means through which identity work is accomplished (Benwell
& Stokoe, 2006; Bucholtz & Hall, 2005; Gee, 2011). In particular, Judith
Butler’s (1990, 2004) insights about the culturally constructed and perfor-
mative nature of gender identities are especially relevant to our discussion.
To this end, we will consider how gender is constructed and performed,
both to engender reviewers and the products and practices of their reviews
and to challenge and destabilize essentialized engenderment.
In order to consider what kinds of gender cues are both given and “given
off” (Goffman, 1959) by authors of Amazon reviews, our analysis is divided
into two parts. In the first, we focus on a set of legitimate reviews and con-
sider what that those cues tell us about normative gender ideologies. In
other words, we focus attention on the kinds of gender ideologies that are
being circulated in this genre of online discourse. In the second half of our
analysis, we shift our attention to how issues of gender are addressed in
parodies of review also found on Amazon. In particular, we consider the
potential of such parodies to resist normative gender ideologies—as well as
their potential to produce counter-hegemonic discourses.
Both the legitimate and the parody Amazon review datasets examined
here were collected as part of other, larger projects. The legitimate reviews
analyzed here consist of a subcorpus of 100 reviews in total: 50 reviews of
diaper bags of various brands and 50 reviews of high-speed blenders of
various brands. (See Vásquez, 2014, for more information about the semi-
random data sampling procedures followed.) Gender was not a central
focus of the original study for which these data were collected. However,
questions about gender emerged from the initial analysis when it was
found that gender and/or relational identities were commonly made
explicit in several review texts. The second dataset, of parody reviews, was
created for a different project. This dataset also consists of 100 reviews in
total: 50 reviews of Bic for Her Pen and 50 reviews of Avery Durable View
Binder. In contrast to the legitimate review data, these products were pur-
posefully selected because the parody reviews for both products do focus
specifically on gender issues. We sorted by “Top” reviews for each of these
products and saved the first 50 that appeared. “Top” reviews on Amazon
are determined via an algorithm that weights various factors, including
the number of a review’s “helpful” votes as well as its recency.
In the analysis below, we consider both the distribution of gender cues
and references, as well as their functions, in order to learn how gender is
constructed in these texts, as well as what purposes it serves in both sets
of texts.
7.5 Findings
7.5.1 B
ona Fide Reviews: Diaper Bags
and High-Speed Blenders
As far as the overall distribution of gender cues in the bona fide review
dataset is concerned, 20% of the texts include one or more instances of
these types of cues (i.e., 19 out of a total of 100 reviews), as can be seen
in Table 7.1. In other words, one in every five reviews contains some ref-
erence (whether direct or indirect) to a reviewer’s gender, or to the gender
of their partner. Most of these references are communicated in relational
terms (my wife/my husband or I am a mother). More specifically, female
Table 7.1 In-text references to bona fide review authors’ gender and/or rela-
tional identities
Product type Male Female Total
Diaper bags 0 13 13
e.g., my husband
I am a mother
Blender 1 5 6
e.g., my wife e.g., my husband
Total 1 18 19
reviewers tend to define themselves in relational terms more frequently

than male reviewers. In fact, only one reference to my wife is made in all
100 reviews. It is important to note that these data were collected prior
to the passing of marriage equality laws in several US states, so the default
assumption here is that a mention of husband likely entails the conven-
tional semantic pairing of wife. However, we recognize that this assump-
tion is problematic, and we are very well aware that this may not be the
case in all instances; it is possible that some references to wife and hus-
band may come from female and male users, respectively.
We now take a closer look at how these gendered or relational identi-
ties are used as resource for reviewers describing their experiences with a
product. We begin with the diaper bag reviews, which featured the most
instances of these types of references. Typically, these references have to
do with what type of diaper bag a reviewer’s husband might be willing to
carry, as in Example 1, where the reviewer begins by verbally underscor-
ing her husband’s masculinity (my manly husband).
Example 1
My manly husband doesn’t think twice about carrying this bag. It’s com-
fortable for both of us to wear. [Diaper Bag, 41]
The word comfortable here likely refers not so much to physical comfort
as it does to the reviewer’s husband’s social comfort with a normatively
feminine practice traditionally associated with mothering, not fathering.
The use of the idiom doesn’t think twice juxtaposes the manly husband
with the bag; the pragmatic presupposition is that diaper bags are inher-
ently feminine in their default and unmarked forms, and unless marked
otherwise (as in Example 1) carrying a diaper bag is a normative practice

of mothering that may threaten masculinity. According to the reviewer,
this particularly bag is not overtly marked for gender, so neither feminin-
ity nor masculinity is threatened: it’s comfortable for both.
The second example includes a reference to the reviewer’s husband, fol-
lowed by an ideologically loaded finite clause, which offers an account for
why the reviewer decided to purchase this particular style of bag: that is, so
that he wouldn’t feel silly carrying around a diaper bag. The reviewer’s lexical
choice in the counterfactual construction used to express an affective state
she did not wish for her husband (i.e., […not] feel silly) again implies that
it is nonnormative for a male person to carry a bag designed to accom-
modate the products associated with an infant. That diaper bags are obvi-
ously a type of female gear is, once again, the implicit message here, and it
is further reinforced in the final clause of the example. In spite of the
reviewer’s efforts to select a (presumably) gender-neutral product type, she
indicates that she is nevertheless the one who is left—quite literally—
holding the bag. The normative nature of such an arrangement is under-
scored by the reviewer’s use of the stance adverbial, of course, which
modifies the statement I am the one using it everyday. Although the reviewer
is the one using it everyday, her choice of product was nonetheless inspired
by protecting her husband from feeling silly, reflecting the underlying
male hegemony of such normatively guided consumer decision-making.
Example 2
I got this diaper bag mainly for my husband so he wouldn’t feel silly carry-
ing around a diaper bag, but of course I am the one using it everyday!
[DiaperBag, 44]
The reviewer in the third example also includes reference to her hus-
band. And just as in Example 2, this reviewer indicates that she specifi-
cally had her husband’s needs in mind (I opted for the chocolate brown for
my husband’s sake) in the product selection process. This time, the prod-
uct feature of color choice is associated with masculinity. Presumably,
brown is understood as a more normatively masculine color than the
light or pastel-colored options that diaper bags often come in.
Example 3
Since I was taught to be positive I’ll start with two things I love … the
many color choices are fabulous—I opted for the chocolate brown for my
husband’s sake. [DiaperBag, 3]
Similarly, in Example 4, the review’s husband is also mentioned in the

context of the product description (it’s not overly feminine). The adverbial
marker as well implies the reviewer’s expectation that both parents will be
able to use the product. However, the husband’s future use of the product
is framed tentatively, through the use of a modal of possibility/ability
(i.e., he can carry it, rather than he will carry it). As in the reviews above,
this review also implies negative social consequences for husbands carry-
ing a “feminine” product.
Example 4
It’s not overly feminine so my hubby can carry it around as well.
[DiaperBag, 21]
Example 5 presents a husband’s reaction to the reviewer’s purchase of a

diaper bag, conveyed via direct reported speech (“oh no … another diaper
bag”). Presumably the husband’s negative reaction is linked to a belief that
the purchase was nonessential or unnecessary, due to the premodification
of diaper bag by another. While it is unclear whether this husband actually
participated in the carrying of the diaper bag, he is portrayed as ultimately
experiencing of change of heart about his assessment of it, as the excerpt
ends with his positive assessment of the product. Once again the hus-
band’s perspective is communicated by the reviewer via direct reported
speech (he said “that was a really good buy; it worked out great for our trip”).
Example 5
When I first bought this my husband said, ‘oh no … another diaper bag’
but after this trip he said ‘that was a really good buy; it worked out great for
our trip’… [DiaperBag, 2]
The final example includes another instance of a diaper bag being pur-
chased for a husband. In this instance, the product’s name even specifies
that it is intended to be used by a man (i.e., DaddyGear), demonstrating

how product marketing draws on and reproduces existing gender ideolo-
gies and stereotypes; the reviewer even calls it a backpack rather than a
diaper bag and contrasts the superior functionality of the DaddyGear
backpack (roomier, more convenient) with a different bag, which is suitable
only for quick day trips. While the reviewer frames the family’s future
activities and use of this product from a plural perspective (We will use),
this is followed by her indication that she purchased the product specifi-
cally for her husband. Nevertheless, it still appeals to more feminine
expectations of a diaper bag being stylish.
Example 6
I’ll use this bag for quick trips around town but for day trips, forget about
it. We will use the DaddyGear backpack I got for my husband—roomier,
more convenient with an outside wipe dispenser, and still stylish.
[DiaperBag, 4]
These excerpts from Amazon reviews of diaper bags illustrate the con-
struction, reproduction, and circulation of a number of normative gen-
der ideologies. The first of these is that diaper bags are a woman’s domain,
and that when men carry a diaper bag, that action represents a deviation
from the normal state of affairs. Second, some products in this category
are now being made specifically for a man to use. These include diaper
bags made in dark or subdued, neutral colors (e.g., the chocolate brown
ones), or the ones explicitly made for and marketed to fathers, such as the
DaddyGear, for example.2 However, even when this type of product is
manufactured with male consumers in mind, the excerpts above suggest
that women are the ones who are doing most of the consumer decision-
making about what kind of diaper bag to purchase. Furthermore, the
excerpts indicate that even when a woman does end up purchasing a
specific type of product that has been designed for a man, she may end
up being the one using that product.
References to gender or relational identities in reviews of high-speed
blenders (arguably a more “gender-neutral” product type than a diaper
bag) offer a useful point of comparison. As indicated above in Table 7.1,
while there were far fewer of these references in blender reviews, some
reviewers did mention their husbands in reviews of these products as

well. Predictably, most of these references were ideologically far more
neutral than those which appeared in the diaper bag reviews, as can be
seen in Examples 7–9.
Example 7
I am a vegan who tends to eat mostly raw blended meals, as do my son and
husband. [Blender, 16]
Example 8
My husband scoffed at the idea of adding spinach or lettuce to smoothies,
until he tried one from the Vitamix. [Blender, 44]
Example 9
My husband will not TOUCH anything dark green. As a result of this and
just as importantly is his serious illness from a DIET SODA HABIT (five
to six cans per day) I had to figure out a way to sneak spinach into his ice
cream smoothies. [Blender, 34]
While reviewers of blenders do make reference to their gendered, rela-

tional identities, these references do not emphasize gender dualities or
draw on gender stereotypes in the same way that the diaper bag reviewers
do. (However, the author of Example 9 does present herself as a wife who
is a nurturer, responsible for her husband’s well-being: I had to figure out
a way….) In contrast, relational identities are used here primarily to con-
struct shared (Example 7) or nonshared (Examples 8 and 9) dietary pref-
erences—a topic which is relevant given this particular type of product.
Or, as in the example below (the only mention of wife in the 100 reviews),
relational identities are used in the framing of consumer decision-making
as a joint activity.
Example 10
My wife and I are both 47 years old and have owned our fair share of
blenders. After the motor died on our Oster last year, we splurged for
Christmas…. [Blender, 21]
7.5.2 P
arody Reviews: Pastel Pens and Durable
Binders
Over the last ten years, parody reviews for a specific set of products have
appeared on Amazon and have also been circulated via other forms of
digital media—online newspaper articles, blogs, on Facebook, and so on.
These parodies of reviews combine humor and critique to mock a given
product, for various reasons. We focus here on parody reviews of two
products that center exclusively on themes of gender: one product that
was gendered in its corporate marketing within Amazon and elsewhere
(Bic for Her Pen) and one that was gendered as a result of site-external
political discourse (Avery Durable View Binder). As Table 7.2 illustrates,
references to review authors’ gender and/or relational identities are far
more frequent in parody reviews than in the legitimate reviews
discussed.
One of the best-known popularly parodied products is the Bic Crystal
Pen for Her. While this product is no longer available for purchase on
Amazon, at the time we collected our data, there were over 2000 reviews
posted for this product, the majority of which were parodies. Ray (2016),
who manually screened a set of 700 reviews of this product, found that
only 12 of them were not parodies. As evidence of the popularity of paro-
dies of this product, the highest-ranked “review” for it (which is a parody)
has nearly 40,000 helpfulness votes on Amazon.
Parodies of this product most typically mock marketer’s attempts to
“gender” a gender-neutral product: in this case, a pen. The following
excerpts from some of these parodies illustrate that in some cases counter-
Table 7.2 In-text references to parody review authors’ gender and/or relational
identities
Product type Male Female Total
Bic for Her Pen 11 29 40
e.g., as a man e.g., us ladies
my wife my lady parts
Avery Binder 6 16 22
e.g., as a man e.g., As a single mother
I … with a male As a businesswoman
perspective…
Total 17 45 62
hegemonic discourses can also appear in the review space on Amazon.

Most authors of Bic for Her Pen parodies adopt a mock first-person narra-
tive style and construct a fictional persona to create humor, as can be seen
in the examples below. The dominant themes invoked in Bic for Her Pen
parodies are related to common gender stereotypes, as well as to traditional
gender roles. Parody reviews of the Bic for Her Pen include references to
sewing, baking, shopping, knitting, child-rearing, or being financially
dependent on men. They also make use of specific discursive strategies that
have been known to deny women’s agency (Hardman, 1993; Penelope,
1990; Russ, 1983; Taylor, Hardman, & Wright, 2013). The female narra-
tor, in Example 11, illustrates this tendency. References to herself occur in
three semantic roles that are stereotypically feminine: the role of an expe-
riencer (I have never been so happy), the role of beneficiary and recipient
of action of a male agent (My husband bought them home for me), and the
role of an agent co-occurring with verbs denoting traditionally “femi-
nine” activities (I cook, clean, and take care of). While she also takes the role
of an agent with a more active verb (push), she does so in a construction
where that action is negated: I don’t have to push so hard.
Example 11
I have never been so happy to have a Lady’s pen in my life. My husband
brought them home for me. They are perfect for writing down recipes
while I cook, clean and take care of our seven children. My favorite aspect
is that I don’t have to push so hard on them, considering my frail, weak
lady-like hands…. [Pen, 50]
The product’s description on Amazon includes bullet points which read:

“Elegant design—just for her!” and “Thin barrel to fit a woman’s hand.”
The authors of Example 11 and Example 12 play intertextually with
this product description, in their mentions of weak lady-like hands and
tiny womanly hands. Semantic roles and discursive strategies interact in
Example 12, as well. My tiny womanly hands as a synecdoche for the
woman herself appears as another agent associated with a negated verb,
expressing inability (my tiny womanly hands were unable to open). The
female author also appears as a beneficiary, again, but this time with her
cats in the role of agent (one of my 257 cats will chew open the package
for me).
Example 12
Sadly, my tiny womanly hands were unable to open the package. Thanks a
lot Bic. (I’m hoping one of my 257 cats will chew open the package for me.)
[Pen, 18]
In parodies of this product, humor is often created through exaggeration,

and by pushing stereotypes to absurd extremes, as can be seen in the
hyperbolic reference to having 257 cats, in the example above.
The author of Example 13 uses a slightly different discourse strategy to
ridicule the association of a particular gender identity, or gender identifi-
cation, with this class of product.
Example 13
First of all I’m a male. I picked a pink one up by mistake to write a quick
note… Next thing I know I’m sitting down to take a pee. Be careful.
[Pen, 12]
Instead of writing from a female perspective, this author adopts the per-
sona of a man who accidentally uses this pink “feminine” product and,
consequently, engages in a behavior normally associated with women
(i.e., sitting down to take a pee). This alternative perspective makes use of
the semantic role typically associated with masculinity—that of a male
agent (I picked […] one up). This author follows his concise narrative
with a warning to others. The humor here—and in other parodies simi-
lar to this one—is based on comically implying a causal relationship:
using a “for her” product resulting in noticeable changes in one’s behav-
ior and/or biology. As can be seen in Table 7.2, 80% (i.e., 40/50) of Bic
for Her Pen parody reviews include some mention of the authors’ gender
identity. This contrasts with legitimate reviews, where such references
are less frequent.
In addition to some of the more implicit critiques of the product
(where the humorous dimension of parody may be more salient than its
critical dimension), some authors of parodies engage with gender politics
more explicitly, as we see in the next two examples, where the authors
draw attention to gender inequality, especially in terms of the issue of fair

pay. To this end, instead of primarily utilizing feminine lexical items as
the basis for their critique, they include references to women’s political
issues, such as pay cuts, the pay gap, and women’s silence, as seen in
Examples 14 and 15.
Example 14
I gave these to all of the men in my office and they all received pay cuts a
few weeks later! Thanks Bic for helping me to bridge the pay gap—at least,
in my office. [Pen, 22]
Example 15
These are so great. Not only can I write recipes and outfit ideas with abso-
lute ease, they even let me know when I’m getting too ambitious. […]
Thanks bic, no other pen could silence my silly opinions the way yours
does. Not to mention, gender inequality stings so much less in pastel pink.
[Pen, 33]
Appearing in the review space alongside these parodies are also review
texts like the next two examples, which are not parodies. However,
because they do not necessarily report on a direct experience with the
product, they cannot be considered legitimate reviews either. Instead,
they are simply much more direct critiques of the product. For instance,
in Example 16, the author alerts consumers to the fact that these pens
cost twice as much as their gender-neutral counterparts and encourages
other consumers not to support the gender inequality being perpetuated
by this company.
Example 16
A ten-pack of the same pen that’s not marketed ‘for her’ is about half the
price.
Stop buying this kind of crap!
The author of Example 17 addresses the company directly and takes them
to task for making and marketing this type of product in the first place.
Example 17
Really Bic? You actually manufacture this pen? You want to insult every woman
on the planet? Well, you did it. I would never buy one of these pens no matter
how good it is. But as long as you pull this kind of b.s. on your customers, you
won’t get another dime from me on ANY of your pens.
Another product that became the target of parody reviews, which also
highlight gender politics, is the Avery Durable View Binder. Unlike the Bic
for Her Pen parodies, in this case, it was not the product itself that stimu-
lated these kinds of parodic discourses, but rather it was a “site-external”
political event. While there is nothing inherently problematic from a gen-
der perspective about this product, it became the object of many parody
reviews for political reasons. In mid-October of 2012, during a political
debate, US presidential candidate, Republican Mitt Romney, made a
comment about “binders full of women” in response to a question about
inequalities in the workplace during the time that he served as governor
of the state of Massachusetts. Romney’s gaffe went viral across various
forms of social media at the time, and—perhaps unpredictably—one of
the sites where his comment received further attention was in the review
space for Avery Durable View Binder on Amazon.
In Example 18, the author creates a fictional persona of a helpless
female incapable of making decisions, and does so within a negative
semantic frame (I’m not adept at making decisions). This author makes an
indirect reference to Romney, which links his comment to this type of
product through a random association, while simultaneously performa-
tively underscoring Romney’s male agency and contrasting it with the
lack of her own: the presidential candidate that KNOWS.
Example 18
As a woman, I’m not adept at making decisions that concern me. So when
I need the right choice, I turn to the presidential candidate that KNOWS….
[Binder, 1]
The author of Example 19 draws on some of the similar gender stereo-

types that appeared in the Bic for Her Pen reviews—for instance, situating
herself in the domain of the kitchen, and being financially dependent on
her husband. She also uses similar discursive strategies in which she rep-
resents herself as a semantic patient, being acted upon (it keeps me in my
place and allows me to get my dinner ready on time) alongside lexical choices
that underscore the agency of her husband (head of household and owner
of the binder).
Example 19
As a wife and mother, I LOVE this binder. It keeps me in my place, allows me
to get dinner ready on time. Some people might think it’s sexist, but sheesh,
I’m not binding my feet, just my brain. Extra bonus, if you sit on it just right,
it can act as an effective method of birth control! Full disclosure: I submitted
this under my husband’s account, with his full permission. He is the head of
our household, and the owner of the binder. [Binder, 5]
Interestingly, the author of Example 19 begins her text by identifying

herself in relational terms, a strategy typically associated with women, as
was seen in the legitimate reviews of diaper bags. In this way, although
the author of Example 19 adopts some of the same discourse practices
associated with legitimate reviews in the construction of a parody review,
the orientation to the absurd (i.e., I’m not binding my feet, just my brain;
if you sit on it just right, it can act as an effective method of birth control)
serves to invert the social meanings associated with such discursive
constructions of gender.
7.6 Discussion and Conclusions

Our analysis has shed light on how normative discourses about gender,
gender roles and the gendering of products occur in reviews on Amazon.
As was shown in the diaper bag examples, consumers often participate in
these types of discourses by circulating and reproducing normative gen-
der ideologies, as well as by engendering certain products through the
discourses that they use. Not only are diaper bags viewed as feminine, but
carrying the diaper bag is habituated through discourse as a woman’s (or
mother’s) activity. Nevertheless, the (apparently, predominantly female)
reviewers still frame their consumer choices around the needs of their
husbands—often considering their husbands’ masculine needs as their

primary motivation for choosing a particular product. As multiple
reviewers suggest, the act of husbands/fathers carrying bags deemed “too
feminine” can threaten their masculinity. Normativity is realized at mul-
tiple levels of such discourses, echoing assumptions that women are not
only emotionally responsible for the needs of others, but also that it is
their heteronormative duty as women to maintain and develop relation-
ships (Cameron, 2000; Coates, 1996; Holmes, 1995; Lutz, 1990;
Tannen, 1991); this emotional duty of protecting their husbands from
emasculation outweighs women’s choices on a product projected as part
of their mothering duty.3 Predictably, this turns up more in reviews of
certain type of products—especially those more often associated with
gender or gendered behavior—than in other types of products, which
may be considered by consumers to be more gender neutral. However,
even in discussion of arguably more “gender-neutral” products, such as
blenders, some reviewers still subtly invoke discourses of women’s gender
socialization in their reviews: for instance, as wives responsible for their
husband’s nutritional well-being, or by evoking the relational identities
that they are responsible for maintaining.
Nevertheless, counterdiscourses to these normative gender ideologies
also appear on Amazon. Often, these take the forms of parodies and other
nonlegitimate types of reviews.
In some of these cases, the corporate “gendering” of products is con-
tested by users (e.g., Bic for Her Pens). Moreover, the review site can also
become an arena for engaging with issues of gender politics in ways that are
not even dependent on a particular product, as in the conflation of
Romney’s comments with Avery Binders in mock reviews. In both instances,
reviewers draw upon specific discursive strategies, including lexical items,
semantic roles and structural properties, hyperbole, and exaggeration, to
tease apart and draw attention to gender stereotyping and gender inequali-
ties. Whereas some legitimate reviews construct, reify, and engender prod-
ucts through their discourses, the majority of parody reviews appear to
deconstruct, undo, and critique the engenderment of such products. While
authors make use of stereotypical ideologies and normativities in both
parodic and nonparodic text types, product engenderment is approached
from different angles in legitimate versus “nonlegitimate” reviews.
The references to gender—in both legitimate reviews and in parodies—

rely on a shared set of indexical links to normative discourses. In this way,
parody reviews represent a form of “authorized transgression” (Hutcheon,
2000). Whereas the unquestioned, discursive performances of gender in
legitimate product reviews, along with the assumptions and presupposi-
tions inherent in their indexicality, help to habituate gender and gender
roles as normal, the exaggeration and implausibility in parody reviews
disrupt such performances. Drawing from Gannon and Davies (2012)
and Butler’s (1990) notion of performativity, it can be argued that these
reviews make use of deconstructive language play. This language play
destabilizes fixed and essentialist notions of gender. That is, particularly in
conjunction with products like pens and binders, which share no cultur-
ally indexical links to gender ideologies, gender performance is visible as
performance. The exaggeration and hyperbole, along with incorporation
of both stereotypically feminine discursive features in discussions of seem-
ingly neutral products draws attention to, and thus disrupts, the process
of engenderment, because there are no prior established stereotypical
indexicalities from which to draw and on which to rely. In legitimate
product reviews, however, performance of gender appears natural and is
not directly visible as performance that constructs gender difference.
Ultimately, Amazon reviewers offer two different conceptualizations of
gender identities simultaneously within the same platform: modernist
gender identities, with women and men in presumed essentialist binary
opposition, and postmodernist deconstruction of gender, with gender
unmoored from essence and instead constructed through discourse.
Gender is not a function of reviews as a digital genre, but, rather, a range
of possible gender identities result from how users make use of both tech-
nological and linguistic affordances within this digital genre. In other
words, gender identities emerge from the specific ways in which users
take up the technological affordances of the platform to accomplish their
particular goals. In this way, we might think of Amazon as a twenty-first
century’s virtual agora—a marketplace and forum—where multiple
voices overlap, and where various points of view are represented: a place
where individuals exploit the site’s available resources to construct—and
occasionally, deconstruct—ideologies as much as they do to shop, con-
sume, and discuss consumer products.
Notes
1. Throughout this chapter we contrast parody reviews with what we refer to
variously as “bona fide” or “legitimate” reviews. This is our way of distin-
guishing between parodic and nonparodic reviews. There are, of course,
other issues related to “real” versus “fake” reviews, especially given mass
media exposés of fraudulent reviews. Both parody reviews and fraudulent
reviews may be considered “fake” reviews; however, Hutcheon (2000)
provides the following useful distinction between the two: the former is
“to imitate with critical irony,” whereas the latter is “to imitate with intent
to deceive” (2000, p. 40).
2. Our internet research about “masculine” diaper bags turned up a number
of blog posts, which provided insights into the cultural constructions of
gender in product design. Some blog posts entitled “Diaper Bags My
Husband Would Carry” (or some variation thereof ) feature predomi-
nantly over-the-shoulder messenger-style bags or backpacks, mostly in a
restricted range of colors: black, gray, or brown. “Masculine” diaper bags
often look as though they have been designed for some other, work-
related, purpose, such as carrying a laptop: one model even looked like a
toolbox. In the rare instances that “masculine” diaper bags featured prints
or graphic elements, these included camouflage patterns, Star Wars images,
and a scull-and-crossbones motif—images which index a number of non-
infant domains of experience. These designs contrast sharply with the
two-strap, large purse or tote-style diaper bags produced in light, pastel
colors or cheerful prints that are associated with traditionally “feminine”
diaper bags.
“Diaper Bags My Husband Would Carry.”
http://www.chockababy.com/2013/10/5-manly-diaper-bags/
http://lifeasmama.com/7-manly-diaper-bags-any-dad-will-carry/
http://forums.thebump.com/discussion/5503207/diaper-bags-and-
your-husband-partner
http://www.whattoexpect.com/forums/march-2015-babies/topic/do-
you-and-so-each-have-your-own-diaper-bag.html?page=4
3. Incidentally, our perusal of reviews of the “Daddy Gear” or the “Diaper
Dude” diaper bags did not turn up any outraged reviewer comments, paro-
dies, or counterdiscourses. This supports our observation that most consum-
ers unquestioningly accept that diaper bags are a woman’s domain unless
such products are explicitly marked (and marketed) as being masculine.
References
Balka, E. (1993). Women’s access to on-line discussions about feminism.
Electronic Journal of Communication, 3(1). Retrieved from http://www.cios.
org/EJCPUBLIC/003/1/00311.HTML
Benwell, B., & Stokoe, E. (2006). Discourse and identity. Edinburgh: Edinburgh
University Press.
Bucholtz, M., & Hall, K. (2005). Identity and interaction: A sociocultural lin-
guistic approach. Discourse Studies, 7(4–5), 585–614.
Butler, J. (1990). Gender trouble: Feminism and the subversion of identity.
Abingdon: Routledge.
Butler, J. (2004). Undoing gender. New York: Routledge.
Cameron, D. (2000). Styling the worker: Gender and the commodification of
language in the globalized service economy. Journal of Sociolinguistics, 4(3),
323–347.
Chevalier, J., & Mayzlin, D. (2006). The effect of word of mouth on sales:
Online book reviews. Journal of Marketing Research, 43, 345–354.
Coates, J. (1996). Women talk: Conversations between women friends. Oxford:
Blackwell.
Connelly, S. M. (2015). “Welcome to the FEMINIST CULT”: Building a feminist
community of practice on Tumblr. Student Publications, Paper 328. Retrieved
from http://cupola.gettysburg.edu/student_scholarship/328
Cunha, E., Magno, G., Almeida, V., Gonçalves, M. A., & Benevenuto, F.
(2012). A gender based study of tagging behavior in Twitter. Proceedings of the
23rd ACM Conference on Hypertext and Social Media, 323–324. https://
doi.org/10.1145/2309996.2310055
Danet, B. (1998). Text as mask: Gender, play, and performance on the internet.
In S. Jones (Ed.), Cybersociety 2.0: Revisiting computer-mediated communica-
tion and community (pp. 129–158). Thousand Oaks, CA: Sage Publications.
Doward, J., & Craig, E. (2012, May 5). Amazon spoof reviews bring art of satire
to website. The Guardian. Retrieved from http://www.theguardian.com/tech-
nology/2012/may/05/amazon-spoof-reviews-satire
Fullwood, C., Morris, N., & Evans, L. (2011). Linguistic androgyny on
MySpace. Journal of Language and Social Psychology, 30(1), 114–124.
Gannon, S., & Davies, B. (2012). Postmodern, post-structural, and critical the-
ories. In S. Nagy Hesse-Biber (Ed.), Handbook of feminist research: Theory and
praxis (pp. 65–91). Los Angeles, CA: Sage.
Gee, J. P. (2011). An introduction to discourse analysis: Theory and method (3rd
ed.). London: Routledge.
Ghose, A., & Ipeirotis, P. (2011). Estimating the helpfulness and economic
impact of product reviews: Mining text and reviewer characteristics. IEEE
Transactions on Knowledge and Data Engineering, 23(10), 1498–1512.
Goffman, E. (1959). The presentation of self in everyday life. New York: Doubleday
Anchor.
Graddol, D., & Swann, J. (1989). Gender voices. Oxford: Blackwell.
Hardman, M. J. (1993). Gender through the levels. Women and Language,
16(2), 42–50.
Harris, A. (2008). Young women, late modern politics, and the participatory
possibilities of online cultures. Journal of Youth Studies, 11(5), 481–495.
Herring, S. C. (1993). Gender and democracy in computer-mediated commu-
nication. Electronic Journal of Communication, 3(2). https://doi.org/10.1016/
b978-0-12-415040-9.50121-4
Herring, S. C. (2004). Computer-mediated communication and woman’s place.
In M. Bucholtz (Ed.), Language and a woman’s place: Text and commentaries
(pp. 216–222). Oxford: Oxford University Press.
Herring, S. C., & Stoerger, S. (2013). Gender and (a) nonymity in computer-
mediated communication. In M. Ehrlich, M. Meyerhoff, & J. Holmes (Eds.),
Handbook of language, gender, and sexuality (pp. 567–586). Hoboken, NJ:
John Wiley & Sons. https://doi.org/10.1002/9781118584248.ch29
Holmes, J. (1995). Women, men and politeness. London: Longman.
Hutcheon, L. (2000). A theory of parody: The teachings of twentieth-century art
forms (Vol. 874). Champaign, IL: University of Illinois Press.
Jenkins, H. (2006). Convergence culture: Where old and new media collide.
New York: New York University Press.
Jurafsky, D., Chahuneau, V., Routledge, B., & Smith, N. (2014). Narrative
framing of consumer sentiment in online restaurant reviews. First Monday,
19. Retrieved from http://firstmonday.org/ojs/index.php/fm/article/view/
4944/3863
Kanai, A. (2015). What should we call me? Self-branding, individuality and
belonging in youthful femininities on tumblr. M/C Journal, 18(1). Retrieved
from http://journal.media-culture.org.au/index.php/mcjournal/article/view-
article/936
Kivran-Swaine, F., Brody, S., & Naaman, M. (2013). Effects of gender and tie
strength on Twitter interactions. First Monday, 18(9). https://doi.org/10.5210/
fm.v18i9.4633
Kozinets, R. (2016). Amazonian forests and trees: Multiplicity and objectivity in
studies of online consumer-generated ratings and reviews. Journal of Consumer
Research, 42, 834–839.
Lutz, C. A. (1990). Engendered emotion: Gender, power and the rhetoric of
emotional control in American discourse. In C. A. Lutz & L. Abu-Lughod
(Eds.), Language and the politics of emotion (pp. 69–91). Cambridge:
Mackiewicz, J. (2010a). Assertions of expertise in online product reviews.
Journal of Business and Technical Communication, 24, 3–28.
Mackiewicz, J. (2010b). The co-construction of credibility in online product
reviews. Technical Communication Quarterly, 19, 403–426.
Marwick, A. E. (2014). Gender, sexuality and social media. In J. Hunsinger &
T. Senft (Eds.), The social media handbook (pp. 59–75). New York, NY:
Routledge.
Ottoni, R., Pesce, J. P., Las Casas, D. B., Franciscani, G., Jr., Meira, W., Jr.,
Kumaraguru, P., et al. (2013). Ladies first: Analyzing gender roles and behav-
iors in Pinterest. ICWSM.
Page, R. (2012). Stories and social media. London: Routledge.
Penelope, J. (1990). Speaking freely: Unlearning the lies of the fathers’ tongues.
New York, NY: Teachers College Press.
Pogue, D. (2010, December 16). Amazon provides a dose of humor. The New York
Times. Retrieved from http://pogue.blogs.nytimes.com/2010/12/16/amazon-
provides-a-dose-of-humor/?_r=0
Ray, B. (2016). Stylizing genderlect online for social action: A corpus analysis of
‘Bic Crystal for Her’ reviews. Written Communication, 33(1), 42–67.
Rodino, M. (1997). Breaking out of binaries: Reconceptualizing gender
and its relationship to language in computer-mediated communica-
tion. Journal of Computer-Mediated Communication, 3(3). https://doi.
org/10.1111/j.1083-6101.1997.tb00074.x
Russ, J. (1983). How to suppress women’s writing. Austin, TX: University of Texas
Press.
Sen, S., & Lerman, D. (2007). Why are you telling me this? An examination
into negative consumer reviews on the web. Journal of Interactive Marketing,
21, 76–94.
Shorey, S. (2015). Fragmentary girls: Selective expression on the Tumblr platform
(Unpublished MA). Amherst, MA: University of Massachusetts.
Skalicky, S. (2013). Was this analysis helpful? A genre analysis of the Amazon.
com discourse community and its “most helpful” product reviews. Discourse,
Context & Media, 2(2), 84–93. https://doi.org/10.1016/j.dcm.2013.04.001
Skalicky, S., & Crossley, S. (2015). A statistical analysis of satirical Amazon.com
product reviews. European Journal of Humor Research, 2, 66–85.
Taboada, M. (2011). Stages in an online review genre. Text & Talk, 31, 247–269.
Tannen, D. (1991). You just don’t understand: Women and men in conversation
(1st ed.). New York, NY: Ballantine.
Taylor, A., Hardman, M., & Wright, C. (2013). Making the invisible visible:
Gender in language. Bloomington, IN: iUniverse.
Tian, Y. (2013). Engagement in online hotel reviews: A comparative study.
Discourse, Context & Media, 2, 184–191.
Turkle, S. (1995). Life on the screen: Identity in the age of the internet. New York,
NY: Simon and Schuster.
Vásquez, C. (2012). Narrativity and involvement in online consumer reviews:
The case of trip advisor. Narrative Inquiry, 22(1), 105–121.
Vásquez, C. (2014). The discourse of online consumer reviews. London:
Bloomsbury.
Vásquez, C. (2015). “Don’t even get me started…”: Interactive metadiscourse in
online consumer reviews. In E. Darics (Ed.), Digital business discourse
(pp. 19–39). London: Palgrave Macmillan.
Vermeulen, I., & Seegers, D. (2009). Tried and tested: The impact of online
hotel reviews and consumer considerations. Tourism Management, 30,
123–127.
Wynn, E., & Katz, J. E. (1997). Hyperbole over cyberspace: Self-presentation
and social boundaries in internet home pages and discourse. The Information
Society, 13(4), 297–327. https://doi.org/10.1080/019722497129043
Zeller, T. (2006, August 9). On Amazon, all of a sudden everyone is a milk critic.
The New York Times. Retrieved from http://www.nytimes.com/2006/08/09/
technology/09milk.html?_r=0
8
Linguistic Expert Creation in Online
Health Practices
Marie-Thérèse Rudolf von Rohr, Franziska Thurnherr,
and Miriam A. Locher
8.1 Introduction
In this chapter, we focus on the linguistic creation of expert identities in
online health practices. Previous research by, for example, Armstrong,
Koteyko, and Powell (2011), Kouper (2010), Gross (2015), Harrison
and Barlow (2009), Locher (2006), Rudolf von Rohr (2015), Sillence
(2010), Thurnherr, Rudolf von Rohr, and Locher (2016), and Veen, te
Molder, Gremmen, and van Woerkum (2010) identified a range of dis-
course strategies that seem to be commonly used to establish expertise in
online health practices, such as referring to other informational sources,
referring to one’s professional status, listing numerical facts, displaying
empathy, using humor, and mobilizing personal narratives. The four data
sets examined in the present chapter are an online advice column of an
American higher educational institution, email counseling exchanges
from a British university, peer forums to quit smoking, and professional
M.-T. Rudolf von Rohr (*) • F. Thurnherr • M. A. Locher

Department Language and Literatures, University of Basel, Basel, Switzerland
e-mail: m.vonrohr@unibas.ch; f.thurnherr@unibas.ch; Miriam.locher@unibas.ch

https://doi.org/10.1007/978-3-319-92663-6_8
220 M.-T. Rudolf von Rohr et al.
websites for smoking cessation in the UK. In order for their users to take
up information or advice, accept support, or remain on site, participants
need to position themselves as trustworthy and credible experts in profes-
sional as well as peer-to-peer interaction (Harvey & Koteyko, 2013;
Locher, 2013; Richardson, 2003; Sillence & Briggs, 2015). In the case of
the advice column, the advice-giver is a fictional persona named Lucy.
The team behind the advisor persona has to use language in such a way
that the advice will be taken seriously and will ideally be followed.
Regarding email counseling, the trustworthiness of the service is a require-
ment for clients. Further, expertise continues to be negotiated in interac-
tion, as counselors want clients to engage with the support provided. In
the case of peer forums to quit smoking, contributors also face several
‘rhetorical challenges’ that are connected to issues of trust (Harvey &
Koteyko, 2013, p. 165). In the role of help-givers, contributors have to
establish their expertise and credibility to position themselves as trust-
worthy advice-givers (Eichhorn, 2008; Harvey & Koteyko, 2013;
Richardson, 2003; Rudolf von Rohr, 2015; Thurnherr et al., 2016, etc.).
Additionally, clients of email counseling and help-seekers on forums can
also position themselves as experts in interaction. Professional websites
communicating on smoking cessation face the challenge of being per-
ceived as a trustworthy source, giving expert advice. In other words,
studying how expert identities are created in our different e-health con-
texts is of interest for this chapter. Expertise is especially relevant since it
facilitates important activities such as advising and informing in online
health practices. We approach expertise from a constructionist point of
view. Thus, we see expertise as being multifaceted and discursively con-
structed. Further, we argue that expertise does not only consist of scien-
tific or experiential knowledge.1 Similarly, specific roles in interaction do
not necessarily equal access or rights to only one type of expertise (such
as laypersons having access to only experiential knowledge). Rather,
expertise can refer to scientific, experiential, or personal knowledge, or
knowledge of how to interact with a participant to help them manage
their health. Moreover, in peer settings, interactants can even validate the
knowledge of the entire community when they praise previous help
received on the forums and recommend further interaction. Thereby,
they simultaneously position the entire group as having expertise and
Linguistic Expert Creation in Online Health Practices 221
strengthen their ties to the group. In other words, a gamut of strategies

can be utilized in interaction to display various notions of expertise. Our
aim is not only to describe the strategies used in the specific practices
researched and to compare them to previously reported results, but also
to explore their link to contextual factors (social and technological) and
how the strategies interplay with each other. In the next section, we will
motivate our choice to study the creation of expertise further by review-
ing some of the literature.
8.2 T
heoretical Background to the Study
of Expertise and Identity Construction
in Health, Online Contexts
Expertise in medical contexts has been researched in previous studies and
yielded interesting results with regard to a wide range of aspects such as
power asymmetry, identity construction, and other contextual factors.
Health practitioners need to project the identity of credible experts if they
want patients to align with the advice and information provided (see
Segal, 2005, who argues that expertise equals ethos in health contexts).
Being perceived as a credible expert is connected to establishing trust.
Sillence, Briggs, Harris, and Fishwick (2006, p. 697) have even suggested
that ‘trust is key’ in online health advice (see also Sillence & Briggs, 2015).
In face-to-face doctor-patient interaction, the power asymmetry between
patients and medical experts used to be clearly established due to the
institutional nature of the interaction (Bigi, 2011). The internet has con-
tributed to a change in the positions available to patients in medical con-
sultations. It has been argued that the information provided on the
internet facilitates more symmetrical decision-making between health-
care providers and patients/clients. In this context, the role of accessibility
to information has been emphasized as it empowers patients to make
responsible health decisions (Heaton, 2011). Lay consumers surfing the
Web can potentially access the same scientific knowledge and informa-
tion as experts on the internet (Sarangi & Clarke, 2002). Ziebland and
Wyke (2012) even argue that ‘[o]nline resources are now established as a
primary route to health information and support’ (p. 220). Thus, lay
consumers can independently become more knowledgeable without the
guidance of health professionals. Moreover, lay consumers can produce
health information themselves (Seale, 2003). In fact, lay expertise in the
form of personal experience has gained in authority over the last decade
and can be mobilized to counter evidence-based information. Experiential
advice and lay expertise seem to be highly valued by consumers surfing
the Web (Furedi, 2008; Seale, 2002; Sillence, 2010). If the expertise of
health practitioners is potentially in contest with lay sources online, this
raises the question of how practitioners construct their expert identities in
order to be perceived as trustworthy advice-givers. Moreover, the ques-
tion of how lay expertise is constructed online also arises; especially con-
sidering that laypeople usually cannot rely on institutional backing. If
laypeople want to be taken seriously as trustworthy advisors, they need to
establish expertise. All of the authors mentioned point toward a vital need
to understand how expertise is constructed in an online health context, in
which the participants’ main goal is to improve someone’s well-being.
Three studies are singled out in this review as they are instrumental in
illustrating the theoretical background on health communication for the
present chapter (Locher, 2006; Richardson, 2003; Sillence & Briggs,
2015). Sillence and Briggs (2015) explored how people evaluate the
trustworthiness of e-health sites from a psychological perspective. They
list four factors that shape how internet users develop trust, suggesting
these factors be seen as stages of how trust evolves: (1) visual appeal, (2)
branding of the site, (3) evaluation of site content, and (4) ‘the impor-
tance of personalization’ and ‘relationship management’ (Sillence &
Briggs, 2015, p. 472). The first factor refers to the initial decision of
whether internet users interact with websites at all, whereas the second is
concerned with whether there are trust markers, such as the logos on, or
branding of, the site, with which internet users are familiar. These first
two factors do not lend themselves for linguistic analysis in the strict
sense. Nevertheless, they are essential if we want to consider a holistic
picture of how expertise is built up. Also, taking these two factors into
account allows us to address the influence of medium and social factors
on the practices (Herring, 2007) in our analyses. In a third stage, internet
users assess whether they find the information provided to be of good
quality, believable, and altruistic. Finally, internet users are more likely to
develop ‘longer-term trust’ if they engage in continued interaction where
they can self-disclose or if content is personalized, for instance, through
‘patient-authored material’ (Sillence & Briggs, 2015, p. 481).
Several studies on written online health discourse have analyzed war-
ranting strategies, that is, discourse strategies used to create expertise and
to display credibility2 (Richardson, 2003), in order to generate trustwor-
thiness in different practices (e.g. Eichhorn, 2008, on an eating disorder
forum; Morrow, 2006, on a forum dealing with depression; and Kouper,
2010, on a LiveJournal community blog about motherhood). Richardson’s
study of an online newsgroup discussing the risk of cell phones has set the
ground for several later studies on the discursive negotiation of expertise
and credibility, including ours. In her study of discussions about the
health risks of mobile phones in internet newsgroups, she found that par-
ticipants positioned themselves as credible expert contributors by (1)
referring to other sources, (2) referring to their own experience, (3) refer-
ring to the expertise of friends, (4) self-categorizing as experts, or (5) using
a technical register. Further studies corroborated the strategies Richardson
found to enhance credibility and trustworthiness in online health support
groups dealing with a range of conditions but also added different ones.
For example, presenting key health statistics (a case of using technical
register) or requesting further information (a new strategy), not unlike in
traditional medical discourse, have been reported in Fage-Butler and
Jensen (2013) on informational and relational aspects in a thyroid forum,
or in Sillence (2010) on advice in a prostate cancer forum. In particular,
the role of how participants share their own experience to position them-
selves as authoritative sources while simultaneously showing empathy
when giving advice has been found an important characteristic of peer-to-
peer online health groups (e.g. Fage-Butler & Jensen, 2013; Kouper,
2010; Sillence & Briggs, 2015; Wright, 2015; Veen et al., 2010, etc.).
Locher’s (2006) work on an online advice column discussed how the
expert identity of the advisor Lucy emerged in the online practice in
question. In this professional-lay context, Locher (2006) found that
seven factors contributed to making up the expert persona Lucy. First,
the advisor was mostly referred to by her first name, which boosted her
identity as expert but also reminded readers of her fictionality. Second,
Lucy directed readers to other sources, invoked her expert status as health
educator, and appeared to ‘quote facts in numbers and percentages’
(Locher, 2013, p. 348). She expanded on the original question of the
advice-seeker and gave ‘detailed background information’ (ibid.). Further,
Locher (2006, pp. 193–194) suggests that Lucy ‘makes readers think and
gives options’, pointing to the preference for non-directiveness typical in
counseling contexts. These strategies can be straightforwardly linked to
boosting Lucy’s expert status. Moreover, Lucy’s response letters were
characterized by a register which avoided medical jargon in order to cater
to the target audience consisting of students. Further, the advisor Lucy
clearly marked her stance on the questioner’s actions, which helped her to
sanction behavior and attitudes in line with good health information.
Additionally, Lucy engaged in relational work, such as the display of
empathy as well as bonding and hedging through humor. Thereby, she
emerged as caring advice-giver who was ‘clever and witty’ (Locher, 2006,
p. 201). Finally, Locher (2013) also points to the importance of the over-
all visual design, explaining that it clearly marks the advice column as an
expert professional source.
These and other studies have informed our theoretical background
with regards to the creation of expertise in health discourses. Further,
discursive identity construction lends itself especially well to studying
how expertise is negotiated in discourse. Thereby, we adopt a construc-
tionist approach to identity and follow Bucholtz and Hall (2005), who
argue that identity should be seen as a social, discursive, emergent, and
relational phenomenon. In other words, it cannot be captured simply by
taking account of variables such as age, gender, education, and so on. We
make use of the concept of positioning as developed in discursive psy-
chology (Bucholtz & Hall, 2005; Davies & Harré, 1990; Hall &
Bucholtz, 2013). This means that we are interested in how interactants
position themselves vis-à-vis each other and how they take each other’s
face concerns into account during interaction.
Our brief review of the previous literature has, on the one hand, high-
lighted that there is already an extensive body of research on the discursive
creation of expertise and credibility and its link to trust in online health
discourse. On the other hand, it also positioned our study within a con-
structivist approach to the creation of expertise and identity. In view of
the above finding that the creation of expertise is key in e-health contexts,
this chapter will try to answer the following research questions for our
specific corpus: (1) How do writers create expert identities in health con-
texts where credibility is important? (2) To what extent does the computer-
mediated context shape/facilitate/empower the practices observed?
8.3 M
ethodology and Data
Our study aims to add to existing research on expertise by adopting a
comprehensive and holistic understanding of its creation and potential
resulting trustworthiness from a discourse-analytic perspective. We do so
by combining an analysis of the warranting strategies that are used to cre-
ate expertise, as reported by Richardson (2003) and Locher (2006), with
Sillence and Briggs’ (2015) model of how trust is established. The exten-
sion of Locher’s (2006) initial work on an online advice column allows us
to reveal not only the types of discursive strategies used to construct
expertise, but also how these strategies interact with each other and how
certain characteristics of practices, such as interactivity, can influence the
use and interplay of strategies. To answer our first research question of
how expert identities are created, we thus focus on what kind of, and
how, discourse strategies are used to display and negotiate expertise and
trustworthiness. In a qualitative process, we identify linguistic strategies
by engaging in close readings of the texts. We draw on our prior knowl-
edge of the individual practices (see Locher, 2010; Rudolf von Rohr,
2015, 2017; Thurnherr et al., 2016) to identify particular strategies used
to display expertise, which can then be put into relation with the con-
struction of an expert identity. Moreover, our second research question,
regarding how the computer-mediated communication (CMC) context
shapes the practices observed, leads us to touch upon the influence of the
differing social and medium factors (Herring, 2007) of the practices
examined. Throughout, we highlight how Sillence and Briggs’ model
(2015) is reflected in the participants’ discursive work.
This chapter belongs to a project entitled ‘Language and Health
Online’, which is both a continuation and an expansion of Locher’s
(2006) study of the advice health column ‘Lucy Answers’. We work with
the same theoretical foundations, delving into the discursive setup of

online health advice practices and interpersonal aspects such as credibility
and trustworthiness that are associated with it. One of our project’s goals
consists in being able to expand the initial study to include a range of
asynchronous online health advice settings that use different modes (such
as email, forums, and less interactive websites) and vary in topic and in
participation structure (Herring, 2007). We are working with four data
sets that, despite differences in medium and social factors, all have in
common that language is at the forefront to communicate on health and
instantiate change in interactants. The data sets are introduced consider-
ing whether authoritative health professionals are part of the practices
and the level of interactivity. An overview is presented in Table 8.1. We
work with subsamples for three of the four sets.
Table 8.1 Overview of our four data sets, ordered according to decreasing pres-
ence of health professionals and increasing level of interactivity
Data source Location/sites Interactants Source size
Anti-smoking UK nonprofit and Professionals, 7 websites
websites commercial charities, broad analyzed
websites audience (corresponds to
23% of larger
corpus)
Online advice American Uni, Health team ‘Lucy’ 280 texts
column health program Anonymous analyzed
questioners (corresponds to
10% of larger
corpus)
Email British Uni, 1 counselor (BACP 5 threads (this
counseling counseling accredited, trained corresponds to
exchanges service in online the entire
counseling), 5 counseling
clients (students of corpus)
British university) 10–12 emails per
thread
Online forums UK non-commerc. Peers, that is, 27 threads
providing peer peer support smoking quitters analyzed
support for groups: Between 3 and 14 (corresponds to
smoking (a) interactants per 34% of larger
quitters Smokingisbad; thread corpus)
(b) Between 11 and
Nosmokingday. 19 posts
co.uk
The subsample of anti-smoking websites for the present study con-

sists of seven websites (collected in 2012) and is the least interactive of
our four sets.3 The sources include different institutional and organiza-
tional backgrounds (governmental, commercial, and nonprofit) and are
all set in the UK. The sample is characterized by its extensive coverage
of information on smoking cessation, ranging from how smoking
affects one’s beauty to smoking during pregnancy. Further, the websites
of this subsample highlight the importance of smoking cessation ser-
vices while also offering opportunities for readers to get in contact.
Expertise is created in various ways on these multimodal platforms of
information (see Herring, 2013), as it is established in different sec-
tions, such as Tips sections, FAQs or testimonials. In these subsites, the
hierarchy of knowledge between website providers and readers may be
more or less emphasized.
The second data set consists of 280 problem and response letters from
an online health advice column, which represents 10 percent of the entire
corpus (collected in 2002 and 2004). The column is run by professional
health experts at an American university, who create the fictional agony
aunt Lucy. It covers seven topic categories from emotional health to
information on drugs and fitness. There is only one exchange between the
advice-seeker and the advice-giver. The advice-seekers are entirely anony-
mous. Previous work on this practice has been published in Locher
(2006, 2010) and Locher and Hoffmann (2006).
The third set consists of five threads of email between one counselor
and five different clients, obtained from a counseling service at a British
university in 2013 and 2014. Client and counselor only interact online,
that is, they do not meet face-to-face. Each thread consists of 10–12
emails, with themes ranging from relationship issues to depression, anxi-
ety, or stress. The exchanges were carried out in two ways: the counseling
content is either in the body of the email (exchanges obtained in 2013)
or in a password-protected word document that is sent back and forth
between the interactants (exchanges obtained in 2014). In the exchanges,
the client and counselor discuss the client’s problems, devise coping
strategies, and evaluate the client’s progress. We have gained retrospective
informed consent. In other words, we have asked the interactants for
access to their exchanges only after the counseling was concluded. Finally,
the counselor anonymized the exchanges to protect the clients’ identity.
The fourth set consists of data from two forums that are open access,
noncommercial independent, or charity peer support groups on the
topic of smoking cessation, in which participants provide each other
with support, advice, and information: (a) SmokingisBad and (b)
nosmokingday.co.uk.4 The threads from the corpus were posted between
March and April 2012 in subforums where participants who had just
started a fresh attempt at quitting smoking mainly initiate the interac-
tion.5 For this study, we compiled a subcorpus of 27 threads, dealing
either with explicit requests for help, support, or advice by initiating
participants (15 threads) or with relapse/starting over announcements
by initiating participants (12 threads). In requests for help threads, the
advice-giving frame, including the asymmetry in knowledge, is brought
about by initiators themselves. In contrast, in relapse threads initiating
posters are experienced in quitting and often in being a community
member and are primarily looking for support. Both contexts raise inter-
esting questions regarding the interpersonal or interactional purpose for
which expertise and credibility need to be discursively constructed. In
terms of ethics, we did not ask for informed consent, basing our decision
on the heuristic principles outlined by the Association of Internet
Researchers (Ess & the AoIR Ethics Working Committee, 2002;
Markham, Buchanan, & the AoIR Ethics Working Committee, 2012).
This is because, with respect to the public-private continuum, we observe
that interactants place group interaction in the public sphere since they
refer to the private messaging function if they want to continue talking
about more delicate issues.
In Table 8.2, we compare the individual data sets with respect to three
important medium and social factors: interactivity, the number of parties
involved, and whether the main participants were health professionals, lay-
people, or both. Our notion of interactivity is closely linked to turn-taking,
designating the fact that users can interact with each other (see Warnick,
2007, for a discussion of different understandings of interactivity). Data
sets were comparable due to their similarity regarding online health con-
cerns and a focus on problem-solving, advising, and information-giving.
Table 8.2 Three important medium and situational factors of the data sample
Type of
Data source Interactivity interactivity Participants
Anti-smoking websites Not interactive No turn-taking Professional
Online advice column Limited Problem letter— Professional—
interactivity Response letter lay
Email counseling Very Two-party Professional—
exchanges interactive interaction lay
Several turns
Online forums providing Highly Multiparty Lay interaction
peer support for interactive interaction
smoking quitters Several turns
8.4 Results and Discussion

We established four dimensions of how expertise surfaces in our online
practices:
1 . Embeddedness of the practice in online health context

2. Interplay between strategies
3. Interactivity of medium
4. Clients/laypeople and professionals equally construct expertise
The first three dimensions are intertwined. This demonstrates how strate-
gies for creating expertise are embedded, act together, and are shaped by
the interactivity afforded by the medium. The fourth dimension high-
lights that it is not just the health professionals who engage in acts of
constructing expertise but also laypeople. In what follows, we discuss
each of these four dimensions in turn. Due to space constraints, we can-
not present examples and detailed analysis for all four data sets in each
section. Instead, we will choose one practice per dimension to illustrate
our insights and will then summarize how this dimension panned out in
the other practices. The sequence of discussion will vary, depending on
which practice was chosen for illustration.
8.4.1 Embeddedness in Online Health Context
It is important to point out that the linguistic strategies we found (to be

illustrated in Sect. 8.4.2) are embedded within the framework of the
entire online interaction. In our data sets, the visual designs and the use
of trust markers are available as pointers of institutional or contextual
expertise (Sillence & Briggs, 2015). To illustrate this, we use the profes-
sional anti-smoking websites. The branding and visual design is highly
dependent on the institutional background of these sources. Governmental
websites can invoke their institutional standing by using their well-known
logos. Screenshot shows how the UK national anti-smoking website is
professionally made with a layout that facilitates information seeking.
Moreover, the logos index the institutional background of the NHS
(frames top left and top right). This prominent placement at the top con-
veys trustworthiness to website visitors, possibly encouraging them to
stay on the website (Sillence & Briggs, 2015). The institutional setting is
further reinforced by the possibility to call a helpline (frame top center) or
to order a smokefree kit (frame bottom right), which demonstrates the
professionalism and scope of the service. In combination with these trust
markers, advice and information offered on a range of aspects related to
smoking cessation position this practice as an informationally founded
and noncommercial source.
Screenshot: smokefree.nhs.uk, bold framed squares added

Continuing with professional health sites, we observe that the letter

exchange of the advice column Lucy Answers is framed by information
about the health service provided by the educational institution to which
it belongs. Thereby, it profits from the latter’s reputation and standing.
Additionally, the hyperlinks to further resources of a medical nature and
the logos of the health department and the university give the advice
column a professional appearance.
In email counseling, the initial construction of expertise for the coun-
selor precedes the first email exchange. Clients approach the counselor
via the university health website. The website is verified as a university
service, displayed by the specific URL containing the university’s web
address, the university’s official logo, and the general style of the site with
regards to font, colors, and other design features that are in line with the
rest of the university’s website. Additionally, the overall counseling service
is described as accredited, professional, and confidential on the introduc-
tory page and each counselor has their professional credentials listed
separately on a subpage. Finally, the client is asked to fill in a question-
naire regarding their well-being, indicating that clients and their issues
are taken seriously and are assessed in a professional manner from the
start of the interaction.
In nonprofessional interaction between lay participants in smoking
cessation forums, we can also observe embeddedness in a larger context.
First, the thematic direction of the forum embeds the interaction in a
health context. While there is no institutional backing of individual
interactants, there is system generated information about how often a
contributor is active in the forum (i.e. the number of posts made per
contributor overall are listed next to the date of the post). In addition,
participants get titles depending on their number of posts (‘senior, junior
member’, indicated as labels under the user name). This can indirectly be
seen as information that gives them expert status. Further, initial posters
to early subforums Day1 or Day2 automatically position themselves as
being at the beginning of their quitting journey, thus presenting them-
selves as in need of help or as being less experienced in quitting.
These various design features and strategies highlight what Sillence and
Briggs (2015) termed ‘visual appeal’ and ‘the branding of the site’ (p. 472).
While the logos referring to the institutional backing of some of our
practices are more evident in the construction of expertise, we also see

system-given information in the lay forums that contribute to the cre-
ation of expertise. The design features and strategies thus serve to create
identities of experts who are credible, and may serve as the first step in
establishing a trusting relationship between help-seekers and helpers.
8.4.2 Expert Identity Strategies and Their Interplay
Secondly, we focus on the linguistic strategies used for the creation of

expertise. Table 8.3 lists the recurring strategies that we identified as con-
tributing to identity construction in our various practices. Each of our data
sets has a particular combination of strategies that is unique and can also
change over time or in location (e.g., on the website and within the thread).
The crosses indicate presence of strategies. They are organized into six
groups to show that some of these strategies are related to a specific aspect
of problem-solving. Expertise is brought to the fore in some groups more
clearly than in others, especially the ‘information-related’ group might
seem logically connected to constructing expertise. However, the other
strategies also contribute to an expert identity. We decided to classify strat-
egies into groups according to what we judged to be their primary function
in the interactions. The table is necessarily a simplification of what occurs
in interaction, as some strategies could be attributed to more than one
group of strategies. Together the strategies function to establish a particular
impression that may—of course—differ in various readers’ minds.
We will briefly illustrate some of the strategies employed by the health
team to create the agony aunt Lucy. Lucy makes use of all the strategies
marked with [x] in Table 8.3, column 3. And we argue that it is in fact
the interplay of these strategies that creates a credible and trustworthy
expert identity appropriate for this particular context (i.e. an online
health column aimed at college students). First, it is noteworthy that
Lucy refers to herself in the third person (Examples 1 and 2):
(1) ucy assumes that this is what you mean by contamination. (LA,
L
drugs)
(2) Lucy noticed that you signed your letter ‘argh’ and wondered about
that. (LA, sexuality)
Table 8.3 List of recurring expert identity strategies
Smoking Lucy in Email Smoking

cessation Lucy counselinga cessation
websites Answers Counselor Client forums
Status related:
Name, self-reference, x x x x
address terms
Information related:
Competent and x x x x x
knowledgeable
source of accurate
information
Referring to further x x x x
help
Presence/absence of x x x x
jargon
Predicting future x x x x
developments
Advice related:
Providing strategies x x x x
step-by-step
Making the addressee x x x x
think and giving
options
Assessment related:
Having an opinion x x x x x
(praise and criticism)
Personal experience x x x x
Interpersonal:
Showing empathy x x x x
Displaying a sense of x x
humor
Showing unconditional x x x
positive regard (UPR)
Showing active x
listening
Being welcoming x x x
Expressing confidence x x x x x
in someone’s abilities
(continued)
Smoking Lucy in Email Smoking

cessation Lucy counselinga cessation
websites Answers Counselor Client forums
Interactive:
Requesting x x x
information that only
the advice-seeker has
access to
Agreement x x x x
Evaluation of a x x
suggested coping
strategy
Recommend continued x x x
interaction
Adapted and expanded from Locher (2006)
a
We have separated the use of strategies according to interactant in the email
corpus, as the counselor and client both construct expertise in different ways
and to different ends
Referring to herself in the third person is a strongly marked feature (see

Locher & Hoffmann, 2006). We argue that this strategy simultaneously
serves two functions. First, it is a way to point to the team of health edu-
cators that stand behind the pseudonym Lucy. This reinforces her identity
as an authoritative expert. At the same time, it is also an honest way of
reminding the reader that Lucy is not real. Secondly, the frequent men-
tion of her name writes Lucy into being. This happens only over time,
when readers access the site repeatedly.
Lucy’s main task is to pass on accurate health information (strategies in
group 2) and provide advice (strategies in group 3). The major part of the
response texts is thus dedicated to giving information (Example 3), advice
(Example 4), and providing further suggestions (Example 4). Consequences
of following advice are also often detailed (‘predicting future developments’).
(3) anic attacks are periods of heightened anxiety often coupled with an
P
extreme fear of being in crowded or closed places. At first, these
attacks are sudden and unexpected, but, if they continue, are often
triggered by environment, like going through tunnels, traveling
across bridges, or being in crowded elevators. Accompanying symp-
toms include [etc.] (LA, emotional health)
(4) W
ith counseling for yourself as well, you may be better able to help
your partner. If you are at AEI, call Counseling and Psychological
Services (CPS) at <phone number>. (LA, emotional health)
While Lucy is clearly positioned as being created by health profession-

als and belonging to the health services of a university (which legitimizes
the practice), she nevertheless avoids medical jargon and chooses an easily
accessible, informal, and nonoffensive range of vocabulary items that is
suitable for her target audience.
The ways in which advice is rendered (e.g. that Lucy praises and criti-
cizes) and the fact that Lucy invests much into interpersonal strategies
such as showing empathy, being welcoming, and displaying humor (not
shown in examples) makes her an approachable and human health
expert—all this despite the fact that Lucy actually does not exist, but is
created by a team of health experts.
Turning to website practices, these use a gamut of strategies for expert
identity construction to give advice and information (groups 1–4, as
shown in Table 8.3, column 2). While strategies from all groups are
employed, Table 8.3 shows that there is fewer use of strategies in the
interpersonal and interactive group when compared to the other data
sets. Since the websites do not allow users to contribute content, the
explanation for the use of fewer strategies is obvious in the interactive
group. Sometimes websites compensate for the absence of user interac-
tion in presenting testimonials, which are supposedly personal experi-
ences by real readers. However, we can only speculate as to the reasons for
fewer strategies in the interpersonal group. It could be that professional
information websites abstain from using interpersonal strategies such as
‘show unconditional positive regard’ or ‘display a sense of humor’ to
avoid misunderstandings and an impression of currying favor.
The strategies used in the email counseling exchanges are manifold,
especially on the counselor’s side (Table 8.3, column 4). The counselor uses
all strategies but one: sharing ‘personal experience’ is avoided due to rules
of professional conduct (Green, 2010). The counselor, therefore, has many
different strategies available to construct her expertise, and she uses specific
strategies at different times during the exchanges according to changes in
her role as advice-giver (see Sect. 8.4.3 for a detailed illustration). Finally,
clients (Table 8.3, column 5) use discursive strategies to share knowledge of

their own lives on which they are the experts rather than the counselor and
position themselves as accurate sources of information (strategy in group
2). Additionally, clients can illustrate their expertise concerning specific
coping mechanisms through such strategies as ‘having an opinion’ (group
4), thereby reporting on progress (for a brief discussion, see Sect. 8.4.4).
A wide range of recurring strategies are employed for expert identity
construction in the request and relapsing forum threads, as shown in the
list of strategies in Table 8.3, column 6. Mostly, these strategies are
employed by responding participants, who, as advice-givers, need to
establish their expertise to have their recommendations accepted (strate-
gies from groups 1 to 4). Some strategies such as ‘personal experience’
and ‘express confidence in one’s abilities’ are also used by initiating par-
ticipants when they report back on their progress in ongoing interaction.
Two strategies, ‘be welcoming’ and ‘recommend continued interaction’,
do not only contribute to the construction of an individual as expert but
reinforce the knowledge and resources of the entire community. The
listed strategies need not occur in every individual thread but characterize
the practices as a whole.
We cannot illustrate all strategies in detail in this section due to space
(see Sects. 8.4.3 and 8.4.4 for an illustration and explanation of several
strategies). While our overview of strategies demonstrates that (1) many of
those found in our corpora overlap with strategies reported in previous
literature, (2) we want to highlight that specific interplays of strategies are
utilized to form particular identities in the interactions and are responsible
for the successful creation of expertise. Not every interactant in our data
used all of them, and it is one of our findings that it is the particular com-
bination of strategies that differentiates our participants in subtle ways.
8.4.3 Interactivity of Medium
As mentioned in our data description (Sect. 8.3), all four practices differ
according to their interactivity. Forums and email counseling are highly
interactive and threads extend over several exchanges. Lucy Answers is spe-
cifically designed to contain one problem letter and an advisory response,
hence is limited to two turns. Finally, the websites are not interactive in the
sense that visitors cannot contribute content. Due to these different degrees
of interactivity, we can also witness various ways of constructing expert
identities in the four practices. On the one hand, there are strategies that
are more prone to occur in interactive practices (e.g., ‘requesting informa-
tion that only the advice-seeker has access to’, ‘agreement’, ‘evaluation of a
suggested coping strategy’, and ‘recommend continued interaction’). On
the other, the continuous interactivity allows for the combination of strate-
gies to change over time in a practice. In other words, while some strategies
are favored at the beginning of the interaction, others might be more suit-
able to create expert identities toward the end of the interaction. In what
follows, we will show examples from the email counseling corpus to illus-
trate interactivity in more detail, before looking into how interactivity
influences the construction of expertise in the other three data sets. We will
use the concept of (self and other) positioning, derived from discursive
psychology (Davies & Harré, 1990), to point out the stances that interac-
tants take up toward each other.
As a highly interactive medium, email allows client and counselor to
exchange multiple messages. The ‘interactive’ strategies mentioned above
are closely connected to the interactivity the medium affords participants.
All four practices show use of these strategies, but highly interactive ones
utilize them particularly often and with the possibility of further interac-
tion taking place. While the counselor makes use of all four ‘interactive’
strategies, we will illustrate two with Example (5): ‘request information
that only the client has access to’ and ‘agreement’. The client has intro-
duced anxiety as a problematic topic in the first email by stating that she
feels anxious and suffers from panic attacks. The counselor confirms that
what the client describes are in fact panic attacks and therefore takes up
the issue of anxiety:
(5) Th
at certainly sounds like panic attacks Anna, and I’m aware of how
frightening they can be, especially when you are away from home and
comfort and support of your family, boyfriend, and friends. Help me
to understand whet [sic] you have done in these situations to help get
through the panic? 6 Maybe there are some things you have done in the
past to cope with anxiety (…). (Counselor: email 2)
By agreeing with the client’s self-diagnosis of panic attacks, the coun-

selor positions herself as a knowledgeable source of information about
psychological concepts and reclaims the task of diagnosing a mental
issue—a task clearly assigned to the expert in the therapeutic relation-
ship. After this confirmation, the counselor shows empathy toward the
client by acknowledging the terrifying nature of panic attacks, further
highlighting her understanding of the subjective perception of a psycho-
logical problem and the client’s emotional reaction. Further, the coun-
selor proceeds to request further information of how the client has so far
dealt with these panic attacks. The counselor aims to establish what kind
of coping mechanisms the client has already in place. These activities—
diagnosing and assessing a problem, showing understanding of psycho-
logical concepts, and asking questions that increase understanding—are
tasks closely linked to an expert role. The counselor carries out several
such acts of positioning to create her expert advice-giver identity.
Interactivity therefore allows the counselor to execute such specific tasks
and simultaneously create her expertise with the help of strategies that are
closely linked to the interactive nature of the medium.
While the counselor might initially use strategies that are closely
related to establish her role as advice- and information-giver, we find that
the counselor focuses on (aspects of ) strategies that position her as an
encourager of progress in later stages of the interaction. On the one hand,
information-related strategies can, for example, change with regards to
content depending on the client’s progress. In other words, if the client
has improved, the counselor might provide the client with background
information on why the application of coping strategies was successful
rather than information about the underlying psychological concept that
caused the client’s problem. On the other, recognizing and assessing
progress are foregrounded when the client improves and instructing the
client about psychological concepts becomes less important. Example (6)
shows an instance in which the counselor praises the client’s progress by
giving an evaluation of the client’s handling of her anxiety:
(6) I t sounds like you are making positive progress, and challenging your
anxiety, even though it’s really tough. (Counselor: email 8)
The counselor does not need to highlight her own status as a profes-
sional counselor anymore, but rather highlights the client’s progress. As
progress demonstrates the successful work the counselor and client have
carried out, the counselor’s praise can also be seen as a positive assessment
of the suggested coping mechanisms. Recognizing progress and praising
become essential tasks of the expert to help clients find and continue to
use coping techniques. The client reinforces the counselor’s accurate
interpretation of her progress by responding You’re right, I am making
progress in email 9. This development finally leads the counselor to fur-
ther praise Anna and to encourage her to continue on the same track:
(7) Th
is is great progress Anna, and I can hear how you have found the
courage to challenge your anxiety, and this has clearly been part of
what has helped you move forward. (…) Keep on doing what you
doing, and you will continue to progress :-) (Counselor: email 12)
After the client has established herself as successfully applying sug-

gested coping techniques, the counselor confirms the client’s improve-
ment by praising the client’s progress. Hence, the counselor positions
herself as an encourager. As she was responsible for suggesting the specific
coping techniques, she simultaneously reaffirms her expertise as success-
ful advice-giver. Further, she expresses her confidence in the client’s abili-
ties to continue to progress. Thus, the counselor has moved from using a
combination of strategies that focused more on establishing her position
as advice- and information-giver to a combination that centers on inter-
active and interpersonal aspects to foreground the clients’ progress. This
move is not random but rather reflects the clients’ needs of specific sup-
port at different stages of the counseling process.
In sum, the interactivity of the medium can influence the construction
of expertise in two ways: first, due to the fact that interactants can respond
in following turns, the use of specific strategies that rely on continued
interaction, such as to ‘request information only the advice-seeker has
access to’, is facilitated. Second, the possibility of several exchanges
enables continuous interactions over space and time, in which the use or
combination of expert strategies can change. Both aspects can be sub-
sumed under Sillence and Briggs’ (2015) fourth point concerning the
evolvement of trust: ‘relationship management’ (p. 472).
Turning to the other practices in sequence of increasing interactivity:

we find smoking cessation websites offer no options for users to directly
interact with other users. Nonetheless, website providers encourage their
users to get in contact with them via other communicative means (email,
chat), that is, they ‘recommend continued interaction’ (see Table 8.3).
Thereby, they try to move communication into a one-to-one setting.
Moreover, while the interplay of expertise strategies changes over time in
highly interactive mediums (see email counseling and forums), website
providers use different patterns of strategies depending on the section of
the websites. Our analysis showed that subsites such as FAQs make exten-
sive use of information- and status-related strategies but relatively little of
interpersonal ones. In contrast, sections that are explicitly related to
advice-giving (e.g. Stop Smoking tips) prefer to employ advice-related and
interpersonal strategies.
In the case of the less interactive practice Lucy Answers, interactivity is
limited to a problem and response letter format in Lucy Answers.
Strikingly, Lucy still uses one of the interactive strategies to create exper-
tise, namely to request further information from the advice-seeker. While
the advice-seeker is not able to respond due to the format of the interac-
tion, Lucy still asks questions aimed at making the reader think. In more
interactive mediums, such as the counseling data, these questions could
readily be answered, and might also be expected to be answered by cli-
ents. In Lucy Answers, the questions are used rhetorically and remind the
readers of a dialogue format despite the system’s limited interactivity.
Finally, forums are highly interactive, and posters use interactive
strategies to construct their own or each other’s expertise. For instance,
they employ ‘agreement’ to align with the posters to either boost their
own claim or to bond with them, creating group expertise. Respondents
can also ‘request more information’ to properly assess initiators’ situa-
tion, thus adopting an identity of conscientious and considerate advice-
givers. Moreover, the topic of threads seems to influence the extent to
which interactive strategies are used. In passages requesting help, inter-
active strategies are more prominent, suggesting that giving tailored and
targeted advice is part of constructing the identity of an expert advice-
giver. In contrast, in relapsing, there is a clear preference for interper-
sonal over interactive strategies and for assessment- to advice-related
ones. This highlights the increased relational needs of initiators in the

difficult situation, which advice-givers acknowledge and for which they
show their understanding.
8.4.4 C
lients/Laypeople and Professionals Equally
Engage in the Construction of Expertise
Not only professionals but also clients in counseling and lay interactants
in forums construct themselves as experts and authorities both when ask-
ing for and when giving advice. We will illustrate this by drawing on the
forum data. In the absence of institutional backing, advice-givers (respon-
dents) index their individual expertise with the 15 strategies mentioned
in Table 8.3, column 6. They thus write their expertise into being with
the aim that advice-seekers consider them trustworthy and comply with
their recommendations. Respondents also signal their expertise as com-
munity members when they reply to newbies. Moreover, the interactive
nature of forum interactions allows clients and initial posters to gradually
build up expertise. Hereafter, we use excerpts from a sample thread from
the requesting help corpus to illuminate how lay participants interac-
tively construct their own and each other’s identities as experts, which is
common in the entire forum corpus.
In the sample thread, an initiator asks whether her experience of quit-
ting is normal (Example 8):
(8) Initiator—first post:

I have just past 3 days and 10 hours since my last smoke but it seems
more like 3 months. Is it usual for the time to ‘drag’ on like this…
(Requesting help, t20, post 1)
By sharing her stopping smoking statistics (past 3 days…), implying

that quitting is a struggle (more like 3 months), and positioning herself as
a new member (smiley), she takes on the identity of a credible participant
in the sense that she has a legitimate right to contribute to the forum.
Further, she adopts the position of an advice-seeker by her request for
normalization (is it usual). At the same time, potential respondents are
constructed as experts, since the initiator wants an evaluation based on

their experience. Finally, her question opens up an adjacency pair which
other members are expected to complete.
Her question results in several responses before she posts again. The
technical nature of the forums thus allows an immediate reaction by
respondent 1. He establishes similarity between himself and the advice-
seeker, a move that works to enhance his credibility and, possibly, his
chance to influence the initiator (see also Wright, 2015, on how similar-
ity between interactants is beneficial to the persuasiveness of a sender).
(9) Respondent 1:
I’m also on day three [name].
Yes it does seem slow but try to busy yourself and dont [sic] get bored
whatever you do.
[…].
You will have lots of support on here if you can log on regularly but
stay strong and positive!
You can do this! (Requesting help, t20, post 2)
Respondent 1 fills the second part of the adjacency pair by agreeing

with the initiator, which normalizes her concern (Yes it does seem slow).
Thereby, respondent 1 accepts the identity of expert quitter even though
his initial statement hedges his expertise to a certain extent, since he is no
further along in his quitting journey. After having pointed to his position
of expertise and the similarity between him and the initiator, the respon-
dent can ‘give options’ on coping as credible advice-giver (try to…, don’t,
etc.). Moreover, respondent 1 ‘recommends continued interaction’ with
the forum. His use of the will-future indicates that he has already received
lots of support, which stresses his experienced quitter identity, highlights
group expertise, and reinforces the authority of the community.
The interactive nature of the forum allows respondents to back up
other contributors across posts through ‘agreement’. Thus, respondents
engage in co-constructing each other’s expertise as advice-givers in inter-
action. In Example (10), the respondent refers to previous posts to intro-
duce her normalizing statement (As others have said):
(10) Respondent 2:
Well done on 3 days and a huge welcome to the forum. As others
have said its perfectly normal to feel like times crawling. I noticed it
quite a lot when I first quit and found it dificult [sic] but stick with
it and occupy yourself and you’ll soon feel like times flying by nicely.
There’s loads of info on the forum to help with your quit and post
often and let us know how you’re doing. We’re all here doing the
same thing and support you. (Requesting help, t20, post 6)
By agreeing with earlier respondents, respondent 2 simultaneously legiti-

mizes preceding and her own assessments. Further, she uses her ‘personal
experience’ to position herself as an expert quitter who felt like the initia-
tor in terms of time being slow (noticed it quite a lot). The use of the past
tense implies that the respondent has in fact overcome this initial stage
without having to explicitly say so. This strengthens her ensuing advice
and especially attests to her ‘prediction of future developments’ (you’ll
soon feel like). Finally, respondent 2 draws the initiator’s attention to the
forum as an archive of information (loads of info on the forum), which is
only possible due to the persistence of transcript (see Herring, 2007).
Moreover, it constructs the online group as having a history of helping
contributors, institutionalizing its expertise and authority. To conclude
her post, the respondent ‘recommend[s] continued interaction’ with the
community (let us know). She adopts a voice for the entire group, which
is indexed by the use of the first-person plural, clearly marking her as a
representative member. The use of the plural makes the respondent’s
bonding attempt more powerful and possibly more effective in tying the
initiator to the group (We’re all here doing the same thing). This type of
involvement and proffer of support is inextricably linked to the user-to-
user interactivity on forums.
The initiator returns to the conversation after five posts. She thanks
contributors, which indicates that she accepts their stance as experts but
also positions herself as an authentic advice-seeker who wants to take up
advice-givers’ suggestions (thank you all so much). Further, she reacts to
respondents’ recommendation of continued interaction with the forum.
In fact, she reifies the group and gives it additional authority and exper-
tise by referring to the community as a space (place to be). In the second
paragraph, we can observe how the interactivity of forums allows partici-

pants to continuously modify their identity. While the initiator positions
herself as struggling in her initial post, she returns and constructs herself
as persevering and proud of her success so far.
(11) Initiator—second post:

Oh WOW … thank you all so much […]! I’m sure this is the place
to be to find support when I think I’m going to fall off the wagon!!
I quit on Friday at approx 3 pm, and it is now Tuesday 11.30—so
am I still on day 3 or have I moved to day 4 yet? I am quite proud
of myself (Requesting help, t20; post 7)
To sum up, this sequence of examples is illustrative of how lay partici-

pants position themselves and each other as experts and peers in matters
of quitting smoking. It is important to maintain that there is no fixed
division in status between advice-givers and advice-seekers: these roles are
continuously reassigned and renegotiated in each new thread. Moreover,
advice-seekers can adopt a more experienced identity within the interac-
tion of the thread itself. Further, in the context of online health support
groups, the expertise of the entire community can be mobilized rhetori-
cally in interaction, especially to involve new members. Contributors can
tap into the experience and authority of the entire group by indexing
their community membership.
We now turn to the other practices in sequence of decreasing interac-
tivity. In the counseling exchanges, clients also point toward their exper-
tise. On the one hand, they are experts on their own lives and experiences
and share details on this knowledge with the counselor. The counselor
co-constructs this identity when requesting information that only cli-
ents have access to. Additionally, clients also construct themselves as
experts on applications of specific coping techniques. After the client
agrees with the progress recognized by the counselor in Example (6), she
illustrates this progress by narrating how she dealt with missing a flight
by writing [I] managed to calm myself down relatively quickly by thinking
logically. She therefore positions herself as an expert in applying the cop-
ing technique she developed with the counselor’s support (for further
information on narratives and their use for identity construction, see
Thurnherr et al., 2016).
In Lucy Answers, problem letter writers rarely construct themselves as

experts. However, they use strategies to establish their credibility as
advice-seekers, such as sharing their background, describing symptoms,
and requesting help. Due to the format of the advice-column, which only
allows for limited interactivity, problem letter writers cannot report on
their progress regarding their initial problems. Even so, the way Lucy
gives nondirective optional advice assigns problem letter writers an iden-
tity as active self-helpers, who know what is best for them.
In the smoking cessation websites, laypeople are shown as experts in
quitting smoking in the case of testimonials. The success stories of ordi-
nary smokers becoming smoke-free are used to serve as inspiration and
example to readers who are positioned as being in the process of quitting
smoking. The fact that there is a specifically designated space where the
transformation of other lay smokers into nonsmokers is illustrated has a
rhetorical function (see Thurnherr et al., 2016). On the one hand, it pro-
vides a basis for identification for readers, who may consider testimonials
as giving them a motivational spur. On the other, testimonials help to
construct the expertise of the smoking cessation websites since successful
quitters vouch for the services provided.
8.5 C
onclusions
In this chapter, we explored four different online health practices and
asked the following two research questions:
1. How do writers create expert identities in health contexts where cred-

ibility is important?
2. To what extent does the computer-mediated context shape/facilitate/
empower the practices observed?
To answer question 1, we found that we could not stop at simply list-

ing a set of strategies (as shown in Table 8.3), but need to include both
the wider context in which they are employed and the possibility for
interactivity and the ensuing dynamics. Our results thus show that four
aspects are vital to the use of these discursive strategies: the embeddedness
of the strategies in the online health context (e.g. as part of an NHS-

supported smoking cessation website), the interplay between strategies
(i.e. each practice has its particular combination of strategies), the inter-
activity of the medium (e.g. there are specific patterns of strategy use
according to the stage of counseling), and the fact that not only profes-
sionals, but also clients and laypeople construct their expertise (such as
peers in an online support group by referring to their own experience).
Results revealed that previously found strategies to create expertise also
occur in our corpus. Additionally, the findings showed that the interplay
of several strategies is in fact needed to explain the creation of credible
and trustworthy expert identities for all participants involved. This inter-
play of strategies varies according to the practice in which expertise is
made relevant. Moreover, expert identities are always created in relation
to interactants, as evident in the forums where interactants position
themselves as experts by answering to advice-seekers’ posts. Further,
interactants also create expertise for the other interactants, such as clients
positioning the counselor as having expertise on how to deal with a spe-
cific troubling situation. The discursive approach to expertise especially
highlights these interpersonal dimensions of creating expertise. Finally,
the interplay of strategies can change (a) over time in highly interactive
practices and (b) depending on where they occur in less interactive
practices.
The discourse-analytic approach of close qualitative analyses to four
online practices revealed a complex pattern of highly interdependent
strategies for creating linguistic expertise for both lay and professional
interactants. The fact that we are dealing with online practices became
especially salient when taking the different levels of interactivity into
account as well as the persistence of the written data.
In choosing health contexts for our analysis, we may have tapped into
a field in which the creation and negotiation of expertise can be expected.
This is especially the case since we are dealing with advisory situations. In
future research, the question of how authority, credibility, and the poten-
tially resulting trust are intertwined bears further scrutiny. In addition,
other, nonhealth-related contexts should be explored for comparison.
Acknowledgments We wish to thank the Swiss National Science Foundation

for funding the project ‘Language and Health Online’ (100016_143286/1) and
especially the counselor and clients who consented to be part of this study. We
also thank Patricia Bou-Franch and Pilar Garcés-Conejos Blitvich, who gave us
the opportunity to present an early version of this work in Valencia at the first
ADDA conference, and the reviewers for their constructive feedback.
Notes
1. In contrast to Prior (2003), which focuses on doctor-patient interaction
and stresses on professional expertise, we want to highlight that expertise
is a multifaceted concept.
2. It is worthwhile to point out that the notions of expertise and credibility
can sometimes become muddled. We follow Sillence (2010), who argues
that showing expertise is one way of establishing credibility and trust.
3. We work with a bigger corpus of 30 sources in the larger project.
4. To safeguard the confidentiality of participants, we refer to the first online
support group using a pseudonym, SmokingisBad. For copyright reasons,
we refer to the second group’s original address. However, names and loca-
tion markers have been changed for both groups. In the meantime, both
groups have ceased to exist in the form studied.
5. Nonetheless, more experienced quitters also post in subforums dealing
with the early stages of quitting.
6. Italics in the original: the counselor mentions in the introductory state-
ment in her first response to the client that any text passages that she
writes in italics should be understood as questions that she would like the
client to answer. To stay as true to the data as possible, we have decided to
keep the italics in the examples presented in this chapter.
References
Armstrong, N., Koteyko, N., & Powell, J. (2011). ‘Oh dear, should I really be say-
ing that on here?’: Issues of identity and authority in an online diabetes com-
munity. Health, 16(4), 1–19. https://doi.org/10.1177/1363459311425514
Bigi, S. (2011). The persuasive role of ethos in doctor-patient interactions.
Communication & Medicine, 8(1), 67–76.
guistic approach. Discourse Studies, 7(4–5), 585–614.
Eichhorn, K. C. (2008). Soliciting and providing social support over the
Internet: An investigation of online eating disorder support groups. Journal
of Computer-Mediated Communication, 14(1), 67–78. https://doi.
org/10.1111/j.1083-6101.2008.01431.x
Ess, C., & the AoIR Ethics Working Committee. (2002). Ethical decision-
making and Internet research: Recommendations from the AoIR Ethics Working
Committee. Approved by AoIR, November 27, 2002. Retrieved from http://
www.aoir.org/reports/ethics.pdf
Fage-Butler, A.-M., & Jensen, M. N. (2013). The interpersonal dimension of
online patient forums: How patients manage informational and relational
aspects in response to posted questions. Hermes, 51, 21–38.
Furedi, F. (2008). Medicalisation in therapy culture. In D. Wainwright (Ed.), A
sociology of health (pp. 97–114). Los Angeles: Sage.
Green, J. (2010). Creating the therapeutic relationship in counselling and psycho-
therapy. Exeter: Learning Matters.
Gross, A. (2015). Asymmetrie und (Patienten-)Expertise in der HIV-
Sprechstunde. In A. Busch & T. Spranz-Fogasy (Eds.), Handbuch Sprache in
der Medizin (pp. 282–299). Berlin and Boston: de Gruyter.
Hall, K., & Bucholtz, M. (2013). Epilogue: Facing identity. Journal of Politeness
Research, 9(1), 123–132. https://doi.org/10.1515/pr-2013-0006
Harrison, S., & Barlow, J. (2009). Politeness strategies and advice-giving in an
online arthiritis workshop. Journal of Politeness Research, 5(1), 93–111.
https://doi.org/10.1515/JPLR.2009.006
Harvey, K., & Koteyko, N. (2013). Exploring health communication: Language
in action. London: Routledge.
Heaton, L. (2011). Internet and health communication. In M. Consalvo &
C. Ess (Eds.), The handbook of internet studies (pp. 212–231). Malden, MA:
Wiley-Blackwell.
discourse. Language@Internet, 4. Retrieved from http://www.languageatint-
ernet.org/articles/2007/761
gent. In D. Tannen & A. M. Trester (Eds.), Georgetown University Round
Table on languages and linguistics 2011: Discourse 2.0: Language and new
Kouper, I. (2010). The pragmatics of peer advice in a LiveJournal community.

Language@Internet, 7(article 1). Retrieved from http://www.languageatinter-
net.org/articles/2010/2464
Locher, M. A. (2006). Advice online: Advice-giving in an American Internet health
column. Amsterdam: Benjamins.
Locher, M. A. (2010). Health Internet sites: A linguistic perspective on health
advice columns. Social Semiotics, 20(1), 43–59. http://www.tandfonline.
com/doi/pdf/10.1080/10350330903438402
Locher, M. A. (2013). Internet advice. In S. Herring, D. Stein, & T. Virtanen
(Eds.), Pragmatics of computer-mediated communication (pp. 339–362).
Berlin: Mouton de Gruyter.
Locher, M. A., & Hoffmann, S. (2006). The emergence of the identity of a fic-
tional expert advice-giver in an American Internet advice column. Text and
Talk, 26(1), 69–106. https://doi.org/10.1515/TEXT.2006.004
Markham, A., Buchanan, E., & the AoIR Ethics Working Committee. (2012).
Ethical decision-making and Internet research recommendations from the AoIR
Ethics Working Committee (Version 2.0). Retrieved from http://www.aoir.org/
reports/ethics2.pdf
Morrow, P. R. (2006). Telling about problems and giving advice in an Internet
discussion forum: Some discourse features. Discourse Studies, 8(4), 531–548.
NHS SmokeFree. Retrieved May 1, 2012, from http://www.smokefree.nhs.uk
No Smoking Day Stop Smoking Forum. Retrieved May 1, 2012, from http://
www.nosmokingday.co.uk
Prior, L. (2003). Belief, knowledge and expertise: The emergence of the lay
expert in medical sociology. Sociology of Health and Illness, 25, 41–57.
Richardson, K. P. (2003). Health risks on the internet: Establishing credibility
on line. Health, Risk and Society, 5(2), 171–184.
Rudolf von Rohr, M.-T. (2015). “You will be glad you hung onto this quit”:
Sharing information and giving support when stopping smoking online. In
C. A. Smith & A. Keselman (Eds.), Meeting health information needs outside
of healthcare: Opportunities and challenges (pp. 263–290). Waltham, MA:
Chandos/Elsevier.
Rudolf von Rohr, M.-T. (2017). Persuasion in smoking cessation online: An inter-
personal pragmatic perspective. Unpublished doctoral dissertation, University
of Basel, Basel, Switzerland.
Sarangi, S., & Clarke, A. (2002). Zones of expertise and the management of
uncertainty in genetics risk communication. Research on Language and Social
Interaction, 35(2), 139–171.
Seale, C. (2002). Media and health. London: Sage Publications.
Seale, C. (2003). Health and media: An overview. Sociology of Health & Illness,
25(6), 513–531. https://doi.org/10.1111/1467-9566.t01-1-00356
Segal, J. Z. (2005). Health and the rhetoric of medicine. Carbondale: Southern
Illinois University Press.
Sillence, E. (2010). Seeking out very like minded others: Exploring trust and
advice issues in an onlinc health support group. International Journal of Web
Based Communities, 6(4), 376–394.
Sillence, E., & Briggs, P. (2015). Trust and engagement in online health: A time-
line approach. In S. Shyam Sundar (Ed.), The handbook of the psychology of
communication technology (pp. 469–487). Chichester: Wiley Blackwell.
Sillence, E., Briggs, P., Harris, P., & Fishwick, L. (2006). A framework for
understanding trust factors in web-based health advice. International Journal
of Human-Computer Studies, 64(8), 697–713. https://doi.org/10.1016/j.
ijhcs.2006.02.007
Thurnherr, F., Rudolf von Rohr, M.-T., & Locher, M. A. (2016). The functions
of narrative passages in three written online health contexts. Open Linguistics,
2(2), 450–470. https://doi.org/10.1515/opli-2016-0024
Veen, M., te Molder, H., Gremmen, B., & van Woerkum, C. (2010). Quitting
is not an option: An analysis of online diet talk between celiac disease patients.
Health, 14(1), 23–40. https://doi.org/10.1177/1363459309347478
Warnick, B. (2007). Rhetoric online: Persuasion and politics on the world wide
web. New York; Bern etc.: Peter Lang.
Wright, K. B. (2015). Computer-mediated support for health outcomes:
Psychological influences on support processes. In S. S. Sundar (Ed.), The
handbook of the psychology of communication technology (pp. 488–506).
Chichester: Wiley Blackwell.
Ziebland, S., & Wyke, S. (2012). Health and illness in a connected world: How
might sharing experiences on the Internet affect people’s health? The Milbank
Quarterly, 90(2), 219–249.
9
How Social Media Shape Identities
and Discourses in Professional Digital
Settings: Self-Communication
or Self-Branding?
Sandra Petroni
9.1 Introduction
The ubiquitous nature of Web 2.0, its collaborative and participatory
culture (Jenkins, 2006), and the pursuit of shareability have led users to
remediate1 their processes of meaning-making as well as their methods of
self-presentation (Goffman, 1959), whether they refer to personal or to
corporate profiles (Chouliaraki & Morsing, 2009).
Social networking practices which give rise to new models of identity
construction are so deeply embedded in our daily routines that we are no
longer able to separate our Self performed in a private or in a public
semiosphere, whether our Self is offline or ‘always on’, as Naomi Baron
claimed in 2008. Over the last decade, digital spaces are the arenas where
traditional and new social practices have been reshaped.
S. Petroni (*)
School of Humanities, University of Rome “Tor Vergata”, Rome, Italy
e-mail: sandra.petroni@uniroma2.it

https://doi.org/10.1007/978-3-319-92663-6_9
252 S. Petroni
Most of these processes are instantiated by the presence of the

‘consumer culture’ (Featherstone, 2007; Singh, 2011) which pervades
and cuts across all digital domains. This occurs for two reasons. First, the
main property of the Web is that of being a hyperdomain where the
Greek prefix hyper- stands for its capability to conflate different domains
by blurring their borders, going beyond borders. Being hyper is the inevi-
table condition of the digital world. The Web setting is a hyperdomain
because it merges different practices from different domains within each
single digital artifact, such as a site or a profile in a social network. This
means that the digital domain should be seen as a ‘blended’ domain
stemming from the functions which are fulfilled simultaneously during a
meaning-making process, and one of these functions is mainly “promo-
tional” (Petroni, 2011, p. 73). In social media networks, but also in com-
munity blogs, due to their collaborative nature and their shareability,
human interactions are shaped following patterns of community practice
which entail alignment, affiliation (Zappavigna, 2012) and sometimes
loyalty. Generally, these processes require ‘appreciation’ (Martin & White,
2005), for example, liking or rating systems from the community. To be
positively appreciated requires the use of persuasive and promotional dis-
cursive strategies (Cook, 2012) but this is not necessarily a conscious
process in social media interactions.
The second reason resides in information itself. Since we are wit-
nesses to a process called ‘informalization’ of public discourse, as
Fairclough (1992, 1995) argues, and since this implies a transformation
in style and register within communicative practices, it is possible to
claim that this striking and pervasive phenomenon is today further sup-
ported by the advent of new digital media. In fact, new communication
technologies have been directly involved in the process of informaliza-
tion and globalization: on the one hand, private styles have crossed bor-
ders into public, official and business situations; on the other, the public
domain seems to invade private domain practices. The way digital media
have been developed and consumed by the global audience reflects a
precise purpose that is of ‘commodifying’ the information conveyed
through them. In 1992, Fairclough, in his discussion of institutional
and social domains, claims,
How Social Media Shape Identities and Discourses… 253
[…] the process whereby social domains and institutions, whose concern is
not producing commodities in the narrower economic sense of goods for
sale, come nevertheless to be organized and conceptualized in terms of
commodity production, distribution and consumption. […] In terms of
orders of discourse, we can conceive of commodification as the coloniza-
tion of institutional orders of discourse, and more broadly of the societal
order of discourse, by discourse types associated with commodity produc-
tion. (1992, p. 207)
Consequently, with information being a commodity, the representa-

tion of technology, by which information is produced, distributed and
consumed, merges two basic functions that are informative and persuasive
and tends to be informal in order to popularize its consumption and to be
perceived as user-friendly. The commodification of information points to
its consequential ‘marketization’. Thus, persuasive, evaluative and descrip-
tive rhetorical actions, characterizing promotional genres and settings, are
often present and integrated in the meaning-making process on the Web,
be it instantiated in a site or an interface (Petroni, ibid., p. 74).
These processes also affect social network environments where self-pro-
filing is the predominant aim and the most far-reaching consequence of
this mechanism is the construction of ‘commodified identities’ thanks to
the processes of self-promotion and self-branding (Herring & Kapidzic,
2015; Page, 2012). In Fairclough’s terms, ‘self-promotion is becoming part-
and-parcel of self-identity […] in contemporary society’ (1995, p. 140).
Moreover, digital technologies play a crucial role since, thanks to their
affordances (Gibson, 1977), they can affect users’ habits and values and
push them to consume digital products or endorse digital actions accord-
ing to well-established algorithmic frameworks which envisage precise
behaviors. There is, in fact, a branch of computing science called
Persuasive Technology (Fogg, 2003, 2009) which allows any interactive
product to be designed with the aim to change attitudes or behaviors by
making desired outcomes easier to achieve. Thus, we find persuasion
attempts in many clicks in the online world. Potentially, most digital
artifacts have a persuasive purpose and their creators intend to affect user
attitudes or behaviors in some way: sign up for service, tell a friend about
this video and enter your email address, and all these actions occur thanks
to a simple click (Petroni, 2016).
254 S. Petroni
More importantly, this ‘algorithmic ideology’ (Amoore, 2011; Cheney-

Lippold, 2011) is adopted by Google, Facebook, LinkedIn and other
social networks which in turn are governed by the world’s most hidden,
commercial and political interests. From this viewpoint, social media
platforms decide the value of information, construct users’ worlds and
the identities they perform in this process. In fact, in LinkedIn, for exam-
ple, reputation building and networking, two of the main techniques of
digital self-branding (Isaksson & Jørgensen, 2010; Wenger, 1999), rely
more on the interactivity technologies designed and embedded in the
platform architecture rather than on what professionals write in their
short resumes or CVs.
This premise is necessary to place and contextualize the processes and
practices that will be analyzed and discussed in the following sections.
The current study in fact focuses on how users discursively present and
promote themselves in a professional digital setting, namely, LinkedIn,
and how social media technology affects user’s agency and, as a conse-
quence, their identity construction. Of course, both facets entail social
and psychological implications which arise from the transformation of
selves, as isolated personae, into ‘networked selves’ (Papacharissi, 2011),
or rather, as identities performed exclusively in combination with other
social connections.
9.2 Theoretical Rationale and Literature

Review
9.2.1 LinkedIn
LinkedIn’s founding mission was to provide recruitment and advertising

services to corporations and agencies, although initially its aim was to
help professionals to connect to each other. LinkedIn is the largest profes-
sional networking site in the world. The brand name itself accounts for
the mission of this platform since the two words (the past participle
Linked and the preposition In) represent the crucial principle of social
networking: users must be present on the platform but it is necessary for
them to be present by providing and consuming updated information
(the fact that they are linked in describes users as being integrated into
the platform). Therefore, LinkedIn functions as a front-stage environ-
ment (using Goffman’s metaphor) where the members’ professional iden-
tity is performed.
It was launched in 2003, and today, with over 400 million members
worldwide, precisely located in 200 countries, and with a rate of 2 new
profiles per second (Nishar, 2014; Sordello, 2014). More than 80% of its
users are aged between 36 and 45 years, although the average is decreas-
ing over time (Archambault & Grudin, 2012) while the recruiting prac-
tices employing LinkedIn have been increased constantly.2 In fact, the
traditional recruiting process based on the classic recruitment agencies
has changed significantly over the last few years. Traditionally, companies
provided their potential applicants with a position description via recruit-
ers (Joos, 2008; Lee, 2005). In turn, applicants had to respond with an
extremely careful edited resume, adapt it to that precise position, and
include only relevant information. Candidate’s identity and credibility
could hence be built only within the boundaries established by the orga-
nization/recruiter through the position description. There was only a pri-
vate company-to-applicant communication and vice versa, and for this
reason the resume could be changed or refined according to the diverse
position applications.
In LinkedIn, candidates have their profile posted and edited within the
boundaries set by the platform, and it is composed of different facets but
only one resembles, in terms of discursive style, the traditional resume,
namely, the summary section. The other components of a traditional CV
such as Education, Experience, Interests, Additional Info, Volunteer
Experiences, Honors and Awards, Publications, Projects and so on are
parts of the profile and have dedicated sections to be completed with con-
cise and brief narratives. On the one hand, a profile is a continuously
updated Web presence (Schawbel, 2013), but on the other is less flexible
to various position descriptions at the same time. Professionals in fact have
to possess a new set of skills in order to recreate and manage their personal
brand in broader boundaries (Labrecque, Markos, & Milne, 2011).
Furthermore, they are required to receive recommendations, to join
groups (related to their professional fields), to follow influencers and to be
updated and informed on top conversations from their connections—
256 S. Petroni
including people they are following—via the new ‘Trending Storyline’

link. Since this amount of information becomes visible and transparent,
prior positions, background, network and endorsements are well exhib-
ited and ready to be analyzed by recruiters, influential people and unin-
tended audiences (Stutzman & Hartzog, 2012). For this reason, it has
been also claimed that information displayed on LinkedIn profile pages is
more reliable and accurate (Brouer, Stefanone, Badawy, & Egnoto, 2015)
and less misleading (Guillory & Hancock, 2012) compared with that pro-
vided by classic CVs.
Therefore, effectiveness and credibility rely on how users promote
themselves, on how they communicate to the broader audience and if
they are able to create a network and increase their connections (Kampf,
Broillet, & Emad, 2014). Peters claims that personal branding requires
professionals to go beyond the classic resume and to become able to ‘mar-
ketize’ instead their skills, and build what Peters (1997) calls ‘the brand
You’. This is precisely the social role LinkedIn should play: to make all
public profile information visible to the global networks of society.
Additionally, LinkedIn profiles also become new information sources
that can be retrieved or linked to by external websites, and these proce-
dures are supported by the Google general search engine where LinkedIn
has a great Web authority (Kelly & Delasalle, 2012). The fact that
LinkedIn profiles are accessible via URLs makes the conversion of these
personal data into units suitable for link analysis possible. As Orduna-
Malea, Font and Ontalba-Ruipérez (2017, p. 128) state, ‘LinkedIn
appears as a cybermetric feasible websource’ that can analyze professional-
company interactions.
Self-profiling in online public spaces has also been the object of study
of socio-cognitivists (Fiske & Tylor, 2013; Tomlinson, 2016), and what
has emerged is that LinkedIn allows users to make a distinction between
the ‘ought self ’ and the ‘ideal self ’, respectively, who one is expected to be
according to the conceptualizations of appropriate versus inappropriate
behaviors, and ‘who one wants to be’ (Fiske & Tylor, 2013, p. 130).
As seen so far, a large part of the literature devoted to LinkedIn arises
from marketing studies, business and corporate communication, as well as
from psychology, sociology, social psychology and Social Cognition Theory.
Building identities and reputation in digital settings are however social
practices that also have their discursive and linguistic dimensions (Ellison,
Heino, & Gibbs, 2006; Ellison, Steinfeld, & Lampe, 2007; Hargittai, 2008;
Herring, 2013; Hoffmann, 2012; Marwick, 2013; Thurlow & Mroczek,
2011). Unfortunately, with the exception of a few studies focused on the
promotional discursive resources (Díez Prados & Cabrejas-Peñuelas, 2015)
employed within this specific ‘virtual marketplace’ (Page, 2012) by the pro-
fessionals and by those who recommend them, there is little evidence of
linguistic analyses aimed at identifying the main promotional language
markers used to write the summary section only.
9.2.2 S
elf-Presentation, Identity Construction
and Reputation Building in Social Media
Social media are a fertile ground for personal and professional identity
construction. The functionality of the Web 2.0 platform, that is, its tech-
nical affordances, is not a matter of a technical updating if compared to
the Web 1.0 system but rather a new way of experiencing this system in
terms of agency. It encourages users to be progressively engaged with
broader and more powerful forms of global communication where iden-
tities need to be renegotiated constantly. As Jenkins states, ‘all human
identities are by definition social identities. Identifying ourselves or oth-
ers is a matter of meaning, and meaning always involves interaction:
agreement and disagreement, convention and innovation, communica-
tion and negotiation’ (2014, p. 18).
Creating profiles and accounts, opening blogs, taking part in social
networks like Facebook, Twitter or LinkedIn, or sharing media files on
YouTube, Flickr or Instagram means, on the one hand, to post comments,
share photos and connect with friends and professionals; on the other, it
means that these actions definitely recontextualize users’ identities in
these new contexts and users ‘perform’ their identity, in Goffman’s terms.
Goffman’s original framework (1959) is of great utility as an analytical
framework for investigating identity through interaction and self-pre-
sentation in the digital world. By using metaphors taken from drama-
turgy, Goffman theorized a new conceptualization of identity
construction in the study of human interaction. In his influential work,
258 S. Petroni
The Presentation of Self in Everyday Life, he studied interpersonal

interactions and how participants perform their actions aimed at pro-
jecting a desirable image that will be their public self-image, their face
(Goffman, 1967). In interactions, individuals are seen as actors that act
on the front stage of their lives and are perfectly aware of being observed
by an audience. For this reason, they perform by following social rules
and conventions, hoping to improve their reputation. On the backstage,
they live their private stories without any performance.
Although computer-mediated communication spaces and social net-
works are nonphysical online environments for social interaction,
Goffman’s framework is still applicable to digital interactions. Social
media interactions go beyond the virtual dimension of the Web and regain
their ‘trueness’, the real face-to face dimension, since multimedia features
such as photos, videos, gadgets, music, friends’ lists and links to others’
social networking profiles are identity markers which surrogate the physi-
cal interplay. According to Jenkins, these features are ‘perhaps, among the
most elaborate examples of impression management that one can imag-
ine’ (2010, p. 264).
Along with self-presentation and impression management (Darics &
Koller, 2017; Gardner & Martinko, 1988; Goffman, 1959; Leary &
Kowalski, 1990), reputation is another concept extremely close to self-
presentation and at the same time is a highly fluid and fortuitous attri-
bute deriving from the perception, attention and approval by others.
Building a reputation entails a constant process of image-making and
impression management and, as such, can be continuously renegotiated
and adapted to the situation. Historically, one’s reputation had to dem-
onstrate and display clearly the inherent quality of a person’s work or
achievement, a variable that is extremely dependent on contextual, cul-
tural, institutional and economic factors. In social networks, the acquisi-
tion of reputation has very little to do with any specific competence or
talent since it is mostly related to users’ ability of attention-catching
online, including expressing opinions, values, establishing connections
and so on (Rodden, 2006). Reputation is a cultural product, and as such
is affected by its mode of production. This is thus totally immersed in the
traditional ‘promotional culture’ (Wernick, 1991) produced by the wide-
spread use of the processes of promotion and marketing. Goods, services,
information and, most centrally, people are all involved in a promotional

culture. Wernick describes this phenomenon as ‘a subject that promotes
itself, constructs itself for others in line with the competitive imaging
needs of the market. Just like any other artificially imaged commodity,
then, the resultant construct is a persona produced for public consump-
tion’ (Wernick, 1991, p. 192).
When users construct their reputations in a global open-access envi-
ronment, for example, in a social network such as LinkedIn, they show
themselves to a networked public (boyd, 2011; Zappavigna, 2012) and
the transformation of selves, as isolated personae, into ‘networked selves’
(Papacharissi, 2011) inevitably carries implications in terms of agency. As
Herring and Kapidzic state, ‘[s]elf-presentation is generally considered to
be motivated by a desire to make a favorable impression on others, or an
impression that corresponds to one’s ideals. As such, self-presentation is
centrally involved in impression management and the projection of an
online identity […]’ (2015, p. 146).
Recent studies (Krämer & Winter, 2008) show that among the most
cogent reasons to host an online personal profile are self-expression and
impression management. Identity construction has been speculated as a
public process that implies both ‘identity announcement’ made by the
individual claiming that identity and ‘identity placement’ made by others
who endorse and empower the claimed identity through digital anchored
relationships (Rosenberg, 1986; Zhao, Grasmuck, & Martin, 2008).
Social networks can be defined as popular stages for self-expression, com-
munication and self-promotion. For example, Facebook is particularly ori-
ented to facilitate personal self-presentation, whereas LinkedIn meets the
need for professional self-promotion. Originally, networks such as
Facebook or MySpace were generally considered as spaces for personal self-
expression and for establishing connections between friends. Depending
on the platform, and how this supports self-expression, and on the target
audience, users have progressively developed techniques of online self-
presentation and are becoming increasingly aware of the relevance of social
networks as tools for professional self-promotion and self-branding.
Additionally, online social networking discursive practices have given
rise to methods of self-presentation and reputation building which in turn
may offer a stage for representing forms of narcissism and self-esteem, as
260 S. Petroni
many social and psychological studies demonstrate. Narcissism refers to a

personality trait reflecting an extravagant and sometimes augmented self-
concept (Buffardi & Campbell, 2008), but this is not necessarily negative.
In fact, there are different forms of narcissism and this phenomenon must
be seen as if it moves along a continuum where the (−) pole represents the
natural and often spontaneous personality trait that refers to an individu-
al’s positive self-perception, whereas the opposite (+) pole reflects an over-
estimated positive and exaggerated self-view of agentive traits like superior
intelligence, strong power and physical attractiveness as well as a pervasive
sense of uniqueness (Emmons, 1984) and entitlement (Campbell, Bonacci,
& Shelton, 2004), which can be transformed into pathological traits of
extreme narcissism and mental disorder, but this study does not focus on
them obviously. Basically, narcissism is correlated with a high degree of
extraversion/agency (Brown & Zeigler-Hill, 2004; Campbell, Rudich, &
Sedikides, 2002; Gabriel, Critelli, & Ee, 1994; John & Robins, 1994) and
a low level of agreeableness or communion (e.g. Miller & Campbell, 2008;
Paulhus & Williams, 2002). As Buffardi and Campbell claim:
these on line communities may be an especially fertile ground for narcis-

sists to self-regulate via social connections for two reasons. First, narcissists
function well in the context of shallow (as opposed to emotionally deep
and committed) relationships. […] Second, social networking Web pages
are highly controlled environments (Vazire & Gosling, 2004). Owners
have complete power over self-presentation on Web pages, unlike most
other social contexts. In particular, one can use personal Web pages to
select attractive photographs of oneself or write self-descriptions that are
self-promoting. (2008, p. 1304)
Needless to say, narcissism is not the sole reason that motivates partici-
pation in these social media sites.
Self-esteem is described as a person’s overall self-evaluation of their
value. Implicit and explicit self-esteem are subtypes of self-esteem.
Implicit self-esteem is an automatic, unconscious self-evaluation; explicit
self-esteem is a more conscious, reflective self-evaluation. According to
Mehdizadeh, ‘[r]egardless of the type of self-esteem, one of the most per-
vasive facts about this construct is that all humans have a vital need to
maintain and/or raise it in both online and offline social settings’ (2010,
p. 358). ‘Friending’, ‘liking’, ‘commenting’, ‘connecting’, ‘endorsing’ and
so on are only some of the specific agentive practices carried out on social
networks in this respect.
Narcissism and self-esteem fuel self-promotion and self-branding, and
these have also become normalized and well-accepted phenomena in
ordinary people’s lives (Buffardi & Campbell, 2008). Most online identi-
ties are thus often shaped by exploiting the same strategies as those used
by celebrities (Marwick, 2013; Page, 2012; Rein, Kotler, & Shields,
2006), corporations and professionals (Ward & Yates, 2013) to promote
themselves. In doing so, users hope to gain popularity and hopefully
reach an appropriate level of recognition, reputation and connectedness
(Ward & Yates, 2013; Zappavigna, 2012).
Today, online platforms and their affordances have allowed personal
branding to be transformed into an important marketing task for ordi-
nary users. Potentially, they are able to manage their own brand and they
can even be their own marketers (Peters, 1997). This is at the basis of
personal branding. Furthermore, if individuals do not manage their own
face, that is, the individual’s public self-image, someone else will do it for
them, giving the power to others and ‘chances are that their brand descrip-
tion won’t be what you have in mind’ (Kaputa, 2005, p. 8). Literature has
also discussed the marketization of identity in terms of personal branding
(e.g. Labrecque et al., 2011; Lair, Sullivan, & Cheney, 2005). In the age
of Web 2.0, self-branding strategies imply building reputation, creating
and maintaining social networking profiles but also using search engine
optimization techniques to boost access to one’s information (Labrecque
et al., 2011, p. 39).
9.2.3 Self-Branding and Technology
Today, social media offer a new ‘protocol’ for social relations which is
inscribed in the technologies and affordances provided by the system.
They allow individuals’ personal connections to become more perma-
nent, presentational, ever broader and, most importantly, public. As
Hearn claims:
262 S. Petroni
[n]ot only can we see the numbers of relationships a person has, but we can
assess their quality and conduct as well; […]. With the emergence of pub-
licly available information about a person’s affective bonds, we get a sense
of their total social impact, an amalgam of their digital activity, which can
then be measured, rationalized, and represented as their ‘digital reputation’.
(2010, p. 429)
This means that what can be used as private or intimate information is

now becoming a public measure that is exploited to brand and evaluate
the overall social value of a person or organization, both by the users and
by those people who make use of these data.
What is difficult to monitor is the extent to which users are conscious
that they are not totally responsible for their identity construction, repu-
tation building and self-branding. Social media platforms have their own
‘logic’ (van Dijck & Poell, 2013)—that is to say their norms, strategies,
technologies and economies which support its dynamics—and this logic
is not clearly identifiable since there are crucial connections which link
together these platforms with users, institutions, economies and tech-
nologies: social media are used by individuals, driven by technologies,
scaffolded by economic patterns and embedded in institutional bodies
(government agencies, news, corporate, education, etc.). Although social
media are able to move their logic beyond their platforms, yet their spe-
cific technological, discursive, economic and organizational tactics strive
to be implicit or appear ‘natural’.
As van Dijck and Poell claim, there are four technical mechanisms
which underpin social networks and which will be analyzed in this study
(Sect. 9.4.2). These are: programmability, popularity, connectivity and
datafication. By programmability the scholars mean ‘the ability of a social
media platform to trigger and steer users’ creative or communicative con-
tributions, while users, through their interaction with […] coded environ-
ments, may in turn influence the flow of communication and information
activated by such a platform’ (2013, p. 5). The first part of this statement
looks at technology and refers to computer code, data, algorithms, proto-
cols, interfaces and the platform organizations that are entangled in pro-
gramming. According to Critical Internet Studies (Beer, 2009; boyd &
Crawford, 2012; Mager, 2012; Milioni, 2015; van Dijck, 2009, 2013),
algorithms are no longer sets of coded instructions but they are able to
steer user experiences, content and user relations through relational activi-
ties, such as liking, favoriting, recommending, sharing, endorsing and so
on, and these technological mechanisms are not visible. The second part
of the statement, instead, refers to human agency: users have crucial
agency in the process of running and directing programmability thanks
both to their own contributions, when they upload contents, and to the
possibility, when known, to refuse coded instructions or protocols, for
instance by maneuvering the privacy settings button in LinkedIn. Content
is no longer programmed just by a central agency, though this still has
remarkable control, and users can manipulate coded interactions. This
happens, for example, when users heavily retweet posts on Twitter, creat-
ing in this way a ‘trend’ topic, or click on the Like button to establish
affiliation (Zappavigna, 2012), and these actions are unpredictable.
Popularity makes use of the same mechanisms as programmability. It
depends on both algorithmic and socioeconomic components. Each plat-
form has its distinct algorithm for boosting the popularity of people,
things or ideas, which is mostly quantitative rather than qualitative. The
Like button aims to brand a social experience or event but the underlying
technology immediately adds it to an automated ‘like-economy’. The
logic of online popularity resides in links for ‘Most viewed’ profile on
LinkedIn, friend stats on Facebook, or follower counts on Twitter.
‘Platform metrics are increasingly accepted as legitimate standards to
measure and rank people and ideas; these rankings are then amplified
through mass media and in turn reinforced by users through social but-
tons such as following and liking’ (van Dijck & Poell, 2013, p. 7).
Connectivity differs slightly from connectedness since it conflates both
the meaning of participation (connectedness) and the technological
valence. Connectivity, which belongs to the semantic field of ‘hardware’,
implies the socio-technical affordance of networked platforms to connect
content to user activities (agency). Connectivity, on the one hand, allows
users to exert influence over the content uploaded by both creating strate-
gic alliances or communities and shaping target audiences through tactics
of automated group formation or personalized recommendations (e.g.
‘People you may know’ on LinkedIn). On the other, connectivity always
mediates users’ agency and establishes how to construct connections.3
264 S. Petroni
Datafication stands for ‘the ability of networked platforms to render

into data many aspects of the world that have never been quantified
before: not just demographic or profiling data yielded by customers in
(online) surveys, but automatically derived metadata from smart phones
such as time stamps and GPS-inferred locations’ (ibid., p. 9). When con-
tents such as music, books, videos and even relationships (friends, likes,
trends and endorsements) are mediated by social media platforms, they
become data. All three principles so far illustrated—programmability,
popularity and connectivity—are interwoven in datafication. Mainstream
thinking conceives of data as raw resources simply being transported
through online channels. Yet, datafication is a crucial facet in social media
logic since it adds a real-time data dimension to social media’s notion of
liveness. The fact that the platform can detect instantaneous activities of
user behavior, that it can translate them into data and process them,
translate again the data into valuable information about individuals,
groups or society in real time, reflects much of social media data’s value.
Datafication processes remain invisible, and this poses questions about
the real link between data and users and about how monitoring and steer-
ing can be manipulated. The overall logic of social media affects social
agency and the shaping of social relations accordingly.
A claim from Kampf et al. (2014, p. 3) says that when professionals
build a personal profile in LinkedIn, they are ‘actively engaging in build-
ing their professional ethos, or credibility, especially in the summary sec-
tion’. This assertion has triggered the analysis of this chapter. In fact, the
aim of this study is to verify whether de facto the summary section in
LinkedIn, thanks only to the concise verbal resources utilized, builds
reputation and credibility or whether the reputation building process
relies instead on the overall platform architecture and in particular on its
digital interactive affordances. In order to verify this problem statement,
two research questions have been posed:
(RQ1) How can online self-presentations be commodified and mar-

ketized as brands in terms of promotional discursive strategies?
(RQ2) To what extent can these marketized self-profiles be constructed
exclusively through verbal resources, or instead, be generated by inter-
active affordances and functionalities embedded in social network
technology?
9.3 D
ata and Procedure
For this study, an empirical research has been conducted on a sample of
LinkedIn profiles. More precisely, the sample is composed of 80 profiles,
40 women and 40 men, equally distributed into 4 professional catego-
ries—journalists, photographers, web managers and project managers—
all of them holding a higher education degree. In this phase (RQ1), only
their summary sections have been extracted from their profiles and pro-
cessed using a freeware corpus analysis toolkit for concordancing and text
analysis, Antconc. The procedure adopted follows the corpus linguistic
approach (Biber, 2007; Biber & Conrad, 2009; Sinclair, 1991; Tognini-
Bonelli, 2001) which enables researchers to detect, for example, the per-
centage of word frequency in a corpus of texts. The quantitative analysis
has addressed those markers which can mirror impression management,
traits of narcissism and self-esteem and contribute to the reputation-
building process. These markers have been codified and then divided into
the three grammar classes suitable for marketizing one’s reputation,
namely, nouns, adjectives and verbs, and within each category only the
first top ten occurrences have been taken into consideration.4 Within
these three categories, those items which convey referential and denota-
tive meaning (e.g. project, business and professional) have been consid-
ered nonnarcissistic and markers of implicit self-esteem, while those with
a promotional meaning (e.g. skills, strategic and excellent) have been
considered narcissistic and markers of explicit self-esteem.
The second phase (RQ2) aims at identifying the most relevant interac-
tive affordances embedded in the social network technologies, namely,
programmability, popularity, connectivity and datafication (cf. par. 2.3).
These affordances reside in other sections on the LinkedIn platform and,
in our opinion, the most strategic are: Skills and Recommendations,
Question and Answer, People You May Know and Who’s Viewed Your
Profile. In contrast with the first phase, here the analysis will not be lan-
guage oriented but rather process oriented. Said differently, the focus will
be on the meaning potential of these allegedly technical affordances which
can be envisaged as playing a crucial role in promoting and branding
identities. A more critical approach which takes inspiration from Social
Construction of Technology Theory (Bijker, Hughes, & Pinch, 1987;
266 S. Petroni
Bijker & Law, 1992; Pinch & Bijker, 1997) and Critical Internet Studies
(Beer, 2009; boyd & Crawford, 2012; Mager, 2012; Milioni, 2015; van
Dijck, 2009, 2013) will be adopted to answer the second RQ.
9.4 Results and Discussion

9.4.1 R
Q1 How Can Online Self-Presentations
Be Commodified and Marketized as Brands
in Terms of Promotional Discursive Strategies?
After extracting data from the summary sections of the sample, Antconc
software has processed them, and the first results are displayed in the follow-
ing figures. Figure 9.1 shows the percentage related to the most frequent
nouns utilized in the sample. The first top-ten nouns are project, business,
experience, strategy, skills, communication, product, leadership, success
and solution, all belonging to the business and communication domain.
For its part, Fig. 9.2 shows the percentage of occurrence of the most
frequent verbs, namely, be, have, work, help, write, build, report, plan,
develop and consult. It is important to underline that ‘be’ and ‘have’ have
been evaluated only when used as stative verbs and not as auxiliaries.
Fig. 9.1 The top-ten nouns

Fig. 9.2 The top-ten verbs
Fig. 9.3 The top-ten adjectives
In Fig. 9.3, the top-ten adjectives are shown: professional, exceptional,

strategic, creative, innovative, excellent, responsible, effective, expert and
dynamic.
As for nouns, regardless of the professional peculiarities, the analysis
demonstrates that in all four professional categories words such as ‘proj-
ect’ and ‘business’ are the most frequent but at same time the most gen-
eral in meaning. They can be in fact classified as ‘semi-technical terms’
since they represent the ‘general vocabulary that has a higher frequency in
a specific field’ (Dudley-Evans & St John, 1998, p. 83). They need hence
268 S. Petroni
to be collocated and contextualized to convey meaning which expresses

reputation building.
Words such as ‘skills’, ‘leadership’ and ‘success’ which can be more
immediately matched with the semantic fields of reputation and credibil-
ity have, instead, a lower frequency. These results suggest that profession-
als do not seem to brand themselves by using more strategic words which
would more clearly manage their ‘identity announcement’, express their
narcissistic traits (intelligence, uniqueness, power) or build their explicit
self-esteem by using promotional markers. Verb category shows that,
even though eight out of ten verbs are dynamic, the two stative verbs, ‘be’
and ‘have’, prevail over the others in terms of occurrences. While the
dynamic verbs denote the profession such as write, plan, build, develop
and so on, the stative verbs express state of being and conditions.
‘Have’ mainly collocates with nouns (knowledge, skills, experience,
etc.), and ‘be’ is followed by the user’s profession (I’m a photographer,
I’m a journalist, etc.), but without transforming the identity into a
brand. As for the adjectives, self-branding seems to work more effec-
tively: ‘exceptional’, ‘strategic’, ‘excellent’, ‘innovative’ and so on reflect
identity announcement, narcissistic traits and explicit self-esteem, and in
this way impression management and reputation building are more
clearly constructed.
However, the most frequent term is ‘professional’, which is the most
general and neutral adjective. These results5 mirror what has already
emerged in other studies. For example, van Dijck claimed that ‘LinkedIn’s
profiles look clean and factual, with only one formal picture as eye-catcher
and text arranged mostly in the form of lists’ (2013, p. 208). In this
analysis the focus is only on the summary section simply because this is
the unique space where professionals can describe themselves through the
elaboration of a very concise resume, as the platform itself recommends.
Nevertheless, there are other sections such as Education, Experience,
Skills, Interests, Additional Info, Volunteer Experiences, Honors and
Awards, Publications, Projects and so on that combined with the sum-
mary section constitute the digital CV. But these spaces are predefined by
the platform and require only brief and factual information. Self-branding
verbal potential resides then in the summary section where results have
shown a scarce use of self-branding techniques. ‘Online presentations
resemble formatted CVs containing only the most relevant facts on edu-
cation, current and past positions, as well as former experience’ (ibid.).
Paradoxically, from a linguistic point of view, the platform seems to ‘dis-
suade’ users from inserting valuable forms of self-expression only for
brevity and clarity’s sake.
What are then the resources exploited in order to brand and marketize
professional profiles? In LinkedIn, the interface or rather the narrative6 of
the layout and the connectivity potential are the real resources which
prompt and trigger the slow transformation of LinkedIn profiles into
commodified and branded identities. The analysis, thus, runs the risk of
being incomplete if we do not investigate, at the same time, how this
platform really works, by looking not only at the language used but also
at the allegedly technical resources and affordances (RQ2).
9.4.2 R
Q2 to What Extent Can These Marketized
Self-Profiles Be Constructed Exclusively
Through Verbal Resources, or Instead,
Be Generated by Interactive Affordances
and Functionalities Embedded in Social
Network Technology?
LinkedIn’s interface features mirror the architectural principles of pro-

grammability, popularity, connectivity and datafication to redirect users’
self-expressions. Thanks to these affordances, the platform maneuvers
and merges users’ needs for self-communication and self-branding with
those for LinkedIn’s datafication. Thus, most profiles are designed on a
‘programmed’ self of one’s professional identity and not on his/her life
story and career. Despite its obvious preference for a clear-cut presenta-
tion of the professional self, LinkedIn exploits the same strategies and
techniques of social media logic. In fact, profiles on LinkedIn are more
than CVs uploaded with the aim to be recruited. It is possible to claim
without being overstated that LinkedIn profiles function as ‘inscriptions
of normative professional behavior’ (van Dijck, 2013, p. 208). Each pro-
file mirrors a well-defined template of one’s professional identity, and this
is thanks to the programmed format following the strategic narrative
which is shown to peers and anonymous evaluators.
270 S. Petroni
There are some LinkedIn sections7 in particular that contribute to the

creation of this narrative professional profile and that fully exploit the four
principles (descriptors) of social media logic. These sections are: Skills and
Recommendations, Question and Answer, People You May Know and
Who’s Viewed Your Profile. Skills and Recommendations is the most rel-
evant space where professionals can merge their ‘identity announcement’
(Skills) with their ‘identity placement’ (Recommendations). LinkedIn
does not require users’ life story but induces them to highlight specific
skills in order to brand their strength (see the Profile Strength at the right-
top side of the personal homepage). Members are also triggered to add at
least two recommendations which can display members’ relations with
authoritative people who advocate their knowledge and skills (popularity
and connectivity). If users do not complete this feature, it will be as if
nobody was willing to endorse them and the Profile Strength area will
signal this gap. Vice versa, the Profile Strength will be at its best when all
sections are compiled and the number of contacts (groups, people who
recommend and validate skills) increases.
As for the Question and Answer space, members’ professional identity
may be boosted by contributing to this area if they answer questions posed
by people in their professional field. Thus, this becomes part of users’ pro-
file, unless they opt out of this feature, but to do this a conscious change
of privacy settings is necessary (connectivity and programmability).
The third LinkedIn section, People You May Know (PYMK), perfectly
reflects the principles of connectivity and popularity, through a tab
prominently featured on each personal page. In fact, LinkedIn manipu-
lates this feature to automatically connect users to all possible ties, that is,
contacts of contacts, and these can be people you may desire to know or
need to know. The PYMK algorithm automatically recommends social
relations on the basis of inferred data in order also to recruit new mem-
bers for its services: LinkedIn continuously sends out invitations to non-
members, pushing them to sign up for the service and connect with the
people already part of ‘your extended network’.
Professional reputation is built on the base of Who’s Viewed Your
Profile space where users are allowed to measure their own professional
value, their ‘identity placement’, by looking at their profile stats, indi-
cating the names, titles and companies of people who look at the user’s
profile. This area shows the ‘state of one’s professional brand’ and
hence users can understand how impressive they are by analyzing the
popularity of their profile. Datafication here is clearly represented and
made visible through profile stats which are aimed at increasing users’
personal scores. Ongoing performances of data keep users’ profiles
always updated demonstrating that they are well connected, sociable
and skilled.
In conclusion, the evidence of self-branding in LinkedIn profiles
mainly emerges not from the information provided in the summary sec-
tion, the unique area where users can describe their life verbally as in a
CV, but from the information users select or add in the other sections
and above all from the application of the four principles. As stated above,
at this level, what provides evidence of the self-branding process is not a
question of quantitative data, for example, the skills selected, the num-
ber of connections for each professional profile, which are, of course,
strategic elements for impression management. The analysis, in fact, is
not data driven but rather process driven. Therefore, what needs to be
highlighted is the presence of these algorithmic principles and how they
process users’ data in order to transform impression management into
self-branding. These principles endorse virtually, though practically,
every strategy of the professional network, including people who are
already members of LinkedIn but also those who can be potentially
recruited for the service.

The two levels of conscious and unconscious self-performance, respec-
tively self-expression and self-promotion, are today newly relevant when
applied to online platforms, and to LinkedIn in particular. Users con-
struct their own profile purposely, and in doing so they provide purpose-
oriented information. In this study, the focus has been on how
professionals build their profiles and hence their reputation and brand.
The two RQs posed aim at identifying the resources through which users
carry out this process. Writing a sort of brief resume and compiling the
rest of the digital CV by incorporating brief and factual information, on
272 S. Petroni
the one hand, and making use of the interactive affordances embedded in
the platform, on the other, are the resources taken into consideration.
Data collection and analysis draw attention only to the summary section
as it is the part written exclusively by the profile’s owner. Thus, threads,
forums, recommendations, endorsements, Q&A section, trending story-
line and so on, which are narratives composed by the owner together
with other users or directly by the other connected users, have not been
included in the analysis of verbal texts.
The first part of the analysis has investigated how four categories of
professionals, journalists, photographers, Web managers and project
managers, exploit verbal resources to brand themselves. Results have
demonstrated that they do not use those markers that have been consid-
ered as self-branding descriptors which most clearly highlight impression
management, self-esteem and narcissistic traits. In fact, as regards impres-
sion management, findings suggest that ‘identity announcement’ is pro-
cessed softly and does not seem to trigger promoting and branding. As far
as the self-esteem trait is concerned, these profiles move along a contin-
uum between implicit and explicit self-esteem where the implicit seems
to be more relevant than the explicit. This claim is also supported by the
factual top-ten nouns and the agentive value conveyed through the exten-
sive use of the stative verbs ‘be’ and ‘have’.
It is interesting to emphasize that what affects the ‘profile strength’ is
not what the users write but only if this section is actually compiled.
What increases the value of the measurement system is the number of
sections of the platform the professional fills in and the number of con-
nections established, along with the number of recommendations posted
by users’ connections and matched with each single skill.
The answer to the second RQ, in fact, is in the analysis of the digital
technical resources of LinkedIn which are not just simple technical
devices but, rather, they channel meaning potential. The way the plat-
form triggers connecting and networking procedures by pushing users to
establish connections with a wider audience so that their profile strength
grows represents the core of the self-branding process. The descriptors
used for RQ2, programmability, popularity, connectivity and datafica-
tion, show how the platform itself shapes professionals’ profiles.
These technological potentialities, therefore, make the conventional

discourse analysis approach, even when it is critical, ineffective in some
ways since the evidence and findings which can emerge from this analysis
do not mirror the real meaning-making processes generated within and by
the platform. As seen in the case study of LinkedIn profiles, the summary
and the digital CVs’ sections resemble the printed resume and CV text
types in the attempt to fulfill their function. There are no noteworthy
changes in terms of rhetorical strategies and persuasive discourse markers
deriving from the digital environment, or rather these strategies are para-
doxically less exploited. The effective and strategic rhetoric of LinkedIn
profiles no longer resides in the data uploaded by users but also in how the
algorithms which fuel the platforms manipulate these data.
Unfortunately, users are not aware of creating behavioral data that are in
turn gathered and gauged by social media owners along with their clients.
Users are thus profiled by social media platforms and this profiling affects
them in the processes of data creation and identity construction. Data
deriving from personal behavior, attitude, values expressed and so on, once
a simple product of a mere online sociality, have now become a kind of
‘social asset’ to exploit (Mager, 2012; Milioni, 2015; van Dijck, 2013).
Thus, it is possible to conclude that practices of self-profiling, self-
presenting and self-branding are only partially carried out by users them-
selves since the platforms’ technologies are aimed at the construction of
identities which are shaped by an underlying algorithmic ideology.
Notes
1. Traditionally, remediation (McLuhan, 1964) means a shift from an old
medium, an old technology (e.g. writing), to a newer one (e.g. printing).
Bolter and Grusin (1999) claim that today new visual media carry out
their cultural significance precisely by borrowing from, by paying homage
to, rivaling, and refashioning earlier media: photography remediates
painting, film remediates stage production and photography, and televi-
sion remediates film, vaudeville and radio. A website, a portal or a social
network, thanks to their technological affordances, remediate television,
radio, TV, news, journals, letters (email) and face-to-face conversation
(chat) simultaneously.
274 S. Petroni
2. https://www.omnicoreagency.com/linkedin-statistics/
3. The human connectedness efficacy of social media recalls early network
sociology. Network sociologists (Rainie & Wellman, 2012) claim that
new media technologies have given rise to ‘networked individualism’ due
to the shift in sociality from massively linked groups to loosely bounded
social networks of relations. From this viewpoint, social networks lead
users to build their own customized social networks and communities.
Consequently, connectivity can be tackled as a breakthrough affordance
for connecting users to content, users to users, platforms to users, users to
advertisers and platforms to platforms algorithmically, and blurs the
boundaries between private and public and between commerce and state.
4. This procedure takes inspiration from diverse studies carried out on CVs,
resumes and motivational letters (Basthomi, 2012; Bhatia, 1993; DeKay,
2006). However, it is important to underline that most literature concern-
ing these topics is mainly based on both a prescriptive approach (Enelow
& Kursmark, 2012; Johnson, 2016) and the Genre Analysis framework
(Bhatia, 1996; Furka, 2008; Swales, 1990).
5. Some gender correlations have been created in order to focus on the use
of these markers from another perspective. The most interesting data to
split into the two binary gender categories are those related to verbs
and adjectives, since men and women tend to use the same nouns to
express their identities and hence the results would resemble those
shown in Fig. 9.1. If we look at the adjective category, women mainly
make use of adjectives such as professional, responsible and effective,
focusing more on reliability rather than on narcissism and explicit self-
esteem, whereas men describe their qualities and/or activities mostly as
strategic, creative and excellent. As for verbs, men tend to use ‘have’
more than ‘be’, whereas women seem to be eager to state their agentive
value. With ‘have’ being a stative verb in these contexts, its meaning
implies possession, and this result combined with the adjectives used
demonstrates that men are more prone to brand themselves through
identity announcement, explicit self-esteem and narcissistic traits than
women. As regards the other verbs, women seem to use more dynamic
verbs than men, and, as said above, these verbs are extremely profession
oriented and objective. Of course, a more detailed gender-related inves-
tigation would need to be contextualized in the wide area of gender
studies in professional settings but this is not the rationale of this
contribution.
6. By narrative the author means the shift from a hypertextual database

structure toward a linear narrative structure which steers data input into a
more uniform format (Manovich, 2001).
7. These sections are present at the time of data collection.
References
Amoore, L. (2011). Data derivatives. Theory, Culture & Society, 28(6), 24–43.
Archambault, A., & Grudin, J. (2012). A longitudinal study of Facebook,
LinkedIn, & Twitter use. CHI, 2012, 2741–2750.
Baron, N. S. (2008). Always on: Language in an online and mobile world.
New York: Oxford University Press.
Basthomi, Y. (2012). Curriculum vitae: A discourse of celebration with narcis-
sistic allusions. TEFLIN Journal, 23(1), 1–24.
Beer, D. (2009). Power through the algorithm? Participatory web cultures and
the technological unconscious. New Media & Society, 11(6), 985–1002.
Bhatia, V. (1993). Analysing genre: Language use in professional settings. New York:
Longman.
Bhatia, V. (1996). Methodological issues in genre analysis. Journal of Language
and Communication Studies, 16, 39–60.
Biber, D. (2007). Discourse on the move: Using corpus analysis to describe discourse
structure. Amsterdam: John Benjamins Publishing.
Biber, D., & Conrad, S. (2009). Register, genre, and style. New York, NY:
Bijker, W., Hughes, T., & Pinch, T. (1987). The social construction of technological
systems: New directions in the sociology and history of technology. Cambridge
MA/London: MIT Press.
Bijker, W., & Law, J. (1992). General introduction. In W. Bijker & J. Law
(Eds.), Shaping technology/building society: Studies in sociotechnical change
(pp. 1–16). Cambridge, MA: MIT Press.
Bolter, J. D., & Grusin, R. (1999). Remediation: Understanding new media.
boyd, D., & Crawford, K. (2012). Critical questions for big data: Provocations
for a cultural, technological, and scholarly phenomenon information.
Communication & Society, 15(5), 662–679.
boyd, D. M. (2011). Social network sites as networked publics: Affordances,
dynamics, and implications. In Z. Papacharissi (Ed.), A networked self:
276 S. Petroni
Identity, community, and culture on social network sites (pp. 39–58). London:

Routledge.
Brouer, R. L., Stefanone, M. A., Badawy, R. L., & Egnoto, M. J. (2015). Losing
control of company information in the recruitment process: The impact of
LinkedIn on organizational attraction. In Proceedings from the 4th Hawaii
International Conference on System Science (pp. 1879–1888). Hawaii, HI:
IEEE.
Brown, R. P., & Zeigler-Hill, V. (2004). Narcissism and the nonequivalence of
self-esteem measures: A matter of dominance? Journal of Research in
Personality, 38, 585–592.
Buffardi, L. E., & Campbell, W. K. (2008). Narcissism and social networking
web sites. Personality and Social Psychology Bulletin, 34, 1303–1314.
Campbell, W. K., Bonacci, A. M., & Shelton, J. (2004). Psychological entitle-
ment: Interpersonal consequences and validation of a selfreport measure.
Journal of Personality Assessment, 83, 29–45.
Campbell, W. K., Rudich, E., & Sedikides, C. (2002). Narcissism, selfesteem,
and the positivity of self-views: Two portraits of self-love. Personality and
Social Psychology Bulletin, 28, 358–368.
Cheney-Lippold, J. (2011). A new algorithmic identity: Soft biopolitics and the
modulation of control. Theory Culture & Society, 28(6), 164–181.
Chouliaraki, L., & Morsing, M. (2009). Media, organizations and identity.
Basingstoke: Palgrave Macmillan.
Cook, G. (2012). Persuasion in English. In D. Allington & B. Mayor (Eds.),
Communicating in English: Text, talk, technology (pp. 225–266). London &
Darics, E., & Koller, V. (2017). Language in business, language at work.
Basingstoke: Palgrave.
DeKay, S. H. (2006). Expressing emotion in electronic job cover letters. Business
Communication Quarterly, 69, 435–439.
Díez Prados, M., & Cabrejas-Peñuelas, A. B. (2015). Persuasion at work: Digital
media for self-promotion. Paper presented at ADDA (Approches to Digital
Discourse Analysis), 1st International Conference, 18–20 November 2015,
Valencia.
Dudley-Evans, T., & St John, M. J. (1998). Development in English for specific
purposes: A multi-disciplinary approach. Cambridge: Cambridge University
Press.
Ellison, N., Heino, R., & Gibbs, J. (2006). Managing impressions online: Self-
presentation processes in the online dating environment. Journal of Computer-
Mediated Communication, 11(2), 415–441.
Ellison, N. B., Steinfeld, C., & Lampe, C. (2007). The benefits of Facebook
‘friends’: Social capital and college students’ use of online social network sites.
Journal of Computer-Mediated Communication, 13(1), 210–230.
Emmons, R. A. (1984). Factor analysis and construct validity of the narcissistic
Personality inventory. Journal of Personality Assessment, 48, 291–300.
Enelow, W., & Kursmark, L. (2012). Expert resumes for managers and executives
(3rd ed.). St. Paul, MN: JIST Works.
Fairclough, N. (1992). Discourse and social change. Cambridge, MA: Polity
Press.
Fairclough, N. (1995). Critical discourse analysis. London: Longman.
Featherstone, M. (2007). Consumer culture and postmodernism (2nd ed.).
London: Sage.
Fiske, S. T., & Tylor, S. E. (2013). Social cognition from brain to culture. London:
Sage publications.
Fogg, B. J. (2003). Persuasive technology. San Francisco: Morgan Kaufmann.
Fogg, B. J. (2009). A behaviour model for persuasive design. Persuasive 2009.
Proceeding of the 4th International Conference on Persuasive Technology
April 2009, Claremont, CA, USA. Retrieved from http://bjfogg.com/fbm_
files/page4_1.pdf
Furka, I. (2008). The curriculum vitae and the motivational letter: A rhetorical
and cultural analysis. WoPaLP, 2, 18–37.
Gabriel, M. T., Critelli, J. W., & Ee, J. S. (1994). Narcissistic illusions in self-
evaluations of intelligence and attractiveness. Journal of Personality, 62, 143–155.
Gardner, W. L., & Martinko, M. J. (1988). Impression management: An obser-
vational study linking audience characteristics with verbal presentations.
Academy of Management Journal, 31, 42–65.
Gibson, J. J. (1977). The theory of affordances. In R. Shaw & J. Bransford
(Eds.), Perceiving, acting, and knowing: Toward an ecological psychology
(pp. 67–82). Hillsdale, NJ: Erlbaum.
Goffman, E. (1959). The presentation of the self in everyday life. New York:
Anchor Books.
Goffman, E. (1967). Interaction ritual: Essays on face-to-face behaviour. New York:
Doubleday.
Guillory, J., & Hancock, J. (2012). The effect of LinkedIn on deception in
resumes. Cyberpsychology, Behavior, and Social Networking, 15(3), 135–140.
Hargittai, E. (2008). Whose space? Differences among users and non-users of
social network sites. Journal of Computer-Mediated Communication, 13,
276–297.
278 S. Petroni
Hearn, A. (2010). Structuring feeling: Web 2.0, online ranking and rating, and
the digital ‘reputation’ economy. Ephemera, 10(3/4), 421–438.
Herring, S., & Kapidzic, S. (2015). Teens, gender, and self-presentation in social
media. In J. D. Wright (Ed.), International encyclopedia of social and behav-
ioral sciences (2nd ed., pp. 146–152). Oxford: Elsevier.
gent. In D. Tannen & A. M. Trester (Eds.), Discourse 2.0: Language and new
Hoffmann, C. R. (2012). Cohesive profiling: Meaning and interaction in personal
weblogs. Amsterdam: John Benjamins publishing.
Isaksson, M., & Jørgensen, P. E. F. (2010). Communicating corporate ethos on the
web: The self-presentation of PR agencies. Journal of Business Communication,
47(2), 119–140.
Jenkins, H. (2006). Convergence culture: Where old and new media collide.
New York: New York University Press.
Jenkins, R. (2010). The 21st century interaction order. In M. Hviid Jacobsen
(Ed.), The contemporary Goffman (pp. 257–274). London: Routledge.
Jenkins, R. (2014). Social identity. London: Routledge.
John, O. P., & Robins, R. W. (1994). Accuracy and bias in self-perception:
Individual differences in self-enhancement and the role of narcissism. Journal
of Personality and Social Psychology, 66, 206–219.
Johnson, N. L. (2016). Improving the accounting student resume language:
Accounting faculty best practice tools. The Accounting Educators’ Journal, 26,
18–33.
Joos, G. J. (2008). Social media: New frontiers in hiring and recruiting.
Employment Relations Today, 35(1), 51–59.
Kampf, C., Broillet, A., & Emad, S. (2014). What and how do we learn from
LinkedIn forums? An exploratory investigation. Retrieved from https://hesso.
tind.io/record/479/files/Emad_Linkedinforums_2014.pdf
Kaputa, C. (2005). UR a brand! How smart people brand themselves for business
success. Mountain View, CA: Davies-Black Publishing.
Kelly, B., & Delasalle, J. (2012). Can LinkedIn and Academia.edu enhance
access to open repositories? In OR2012: The 7th International Conference on
Open Repositories. Bath: University of Bath.
Krämer, C. N., & Winter, S. (2008). Impression management 2.0: The relation-
ship of self-esteem, extraversion, self-efficacy, and self-presentation within
social networking sites. Journal of Media Psychology, 20, 106–116.
Labrecque, L. I., Markos, E., & Milne, G. R. (2011). Online personal branding:
Processes, challenges, and implications. Journal of Interactive Marketing, 25,
37–50.
Lair, D. J., Sullivan, K., & Cheney, G. (2005). Marketization and the recasting
of the professional self: The rhetoric and ethics of personal branding.
Management Communication Quarterly, 18, 307–343.
Leary, M. R., & Kowalski, R. M. (1990). Impression management: A literature
review and two-component model. Psychological Bulletin, 107, 34–47.
Lee, I. (2005). The evolution of e-recruiting: A content analysis of fortune 100
career web sites. Journal of Electronic Commerce in Organizations, 3(3), 57–68.
Mager, A. (2012). Algorithmic ideology: How capitalist society shapes search
engines. Information, Communication & Society, 15(5), 769–787.
Manovich, L. (2001). The language of new media. Cambridge, MA: MIT Press.
Martin, J. R., & White, P. R. R. (2005). The language of evaluation. New York:
Palgrave Macmillan.
Marwick, A. E. (2013). Status update: Celebrity, publicity, and branding in the
social media age. New Haven, CT: Yale University Press.
McLuhan, M. (1964). Understanding media. The extension of man. London &
Mehdizadeh, S. (2010). Self-presentation 2.0: Narcissism and self-esteem on
Facebook. Cyberpsychology, Behavior, and Social Networking, 14(4), 357–364.
Milioni, D. L. (2015). Opening the ‘black box’ of user agency: A critical cultural
studies approach to Web 2.0. Retrieved from https://www.researchgate.net/
publication/300246504_Opening_the_‘Black_Box’_of_User_Agency_A_
Critical_Cultural_Studies_Approach_to_Web_20
Miller, J. D., & Campbell, W. K. (2008). Comparing clinical and social-personal-
ity conceptualizations of narcissism. Journal of Personality, 76, 449–476.
Nishar, D. (2014). The next three billion [INFOGRAPHIC]. Retrieved from
LinkedIn Corporation: http://blog.linkedin.com/2014/04/18/the-next-
three-billion/
Orduna-Malea, E., Font, C. I., & Ontalba-Ruipérez, J. A. (2017). From univer-
sty to private company: A measureable route of LinkedIn users. In M. Cabrera
& N. Lloret (Eds.), Digital tools for academic brending and self-promotion
Page, R. (2012). The linguistics of self branding and micro-celebrity in Twitter:
Papacharissi, Z. (Ed.). (2011). A networked self: Identity, community, and culture
on social network sites. London: Routledge.
280 S. Petroni
Paulhus, D. L., & Williams, K. (2002). The dark triad of personality: Narcissism,
machiavellianism, and psychopathy. Journal of Research in Personality, 36,
556–568.
Peters, T. (1997). The brand called you. Retrieved from https://www.fastcom-
pany.com/28905/brand-called-you
Petroni, S. (2011). Language in the multimodal web domain. Rome/Ottawa:
Aracne/Legas.
Petroni, S. (2016). Digitality and persuasive technologies: New social actions and
practices in digital settings. In S. Gardner & S. Alsop (Eds.), Systemic functional
linguistics in the digital age (pp. 29–44). United Kingdom: Equinox Publishing.
Pinch, J., & Bijker, W. (1997). The social construction of facts and artifacts: Or
how the sociology of science and the sociology of technology might benefit
each other. In W. Bijker, T. Hughes, & T. Pinch (Eds.), The social construction
of technological systems: New directions in the sociology and history of technology
(pp. 17–50). Cambridge, MA: MIT Press.
Rainie, L., & Wellman, B. (2012). Networked: The new social operating system.
Rein, I. J., Kotler, P., & Shields, B. (2006). The elusive sports fan, reinventing
sports in a crowded marketplace. New York: McGraw-Hill.
Rodden, J. (2006). Reputation and its vicissitudes. Society, 43(3), 75–80.
Rosenberg, M. (1986). Conceiving the self. Malabar, FL: Robert E. Krieger
Publishing Company.
Schawbel, D. (2013). Reasons why your online presence will replace your
resume in 10 years. Forbes. Retrieved from http://www.forbes.com/sites/
danschawbel/2011/02/21/5-reasons-why-youronline-presence-will-replace-
your-resume-in-10-years/
Sinclair, J. (1991). Corpus, concordance, collocation. Oxford: Oxford University
Press.
Singh, P. R. (2011). Consumer culture and postmodernism. Postmodern
Openings, Year 2, 5(5), 55–88.
Sordello, S. (2014). LinkedIn’s Q1 2014 earnings. Retrieved from LinkedIn
Corporation: http://blog.linkedin.com/2014/05/01/linkedins-q1-2014-earn-
ings/
Stutzman, F., & Hartzog, W. (2012). Boundary regulation in social media. In
Proceedings from the ACM 2012 Conference on Computer Supported Cooperative
Work (pp. 769–778).
Swales, J. (1990). Genre analysis: English in academic and research setting.
Cambridge: Cambridge University Press.
Tognini-Bonelli, E. (2001). Corpus linguistics at work. Amsterdam: John
Benjamins Publishing Company.
Tomlinson, E. (2016). First encounters in professional cyberspace: Writer’s
exploration of LinkedIn. In P. Thomas & P. Takayoshi (Eds.), Literacy in
practice: Writing in private, public, and working lives (pp. 163–176). London
and New York: Routledge.
van Dijck, J. (2009). Users like you? Theorizing agency in user-generated con-
tent. Media, Culture & Society, 31(1), 41–58.
van Dijck, J. (2013). ‘You have one identity’: Performing the self on Facebook
and LinkedIn. Media, Culture & Society, 35(2), 199–215.
van Dijck, J., & Poell, T. (2013). Understanding social media logic. Media and
Vazire, S., & Gosling, S. D. (2004). E-perceptions: Personality impressions
based on personal websites. Journal of Personality and Social Psychology, 87,
123–132.
Ward, C., & Yates, D. (2013). Personal branding and e-professionalism. Journal
of Service Science, 6(1), 101–104.
Wenger, E. (1999). Communities of practice: Learning, meaning, and identity.
Wernick, A. (1991). Promotional culture: Advertising, ideology and symbolic
expression. London and Thousand Oaks: SAGE Publications.
Zhao, S., Grasmuck, S., & Martin, J. (2008). Identity construction on Facebook:
Digital empowerment in anchored relationships. Computers in Human
Behavior, 24, 1816–1836.
10
Losing Face on Facebook: Linguistic
Strategies to Repair Face in a Spanish
Common Interest Group
Carmen Maíz-Arévalo
10.1 Introduction
Traditionally, the study of face-repairing strategies or corrective facework
has concentrated on face-to-face interactions in different contexts and
within different disciplines (e.g. social predicaments in Hodgins,
Liebeskind, & Schwartz, 1996). In the last few years, however, the
increasing importance of computer-mediated communication (CMC
henceforth) has switched the focus to this way of interaction
(Androutsopoulos, 2006; Herring, Stein, & Virtanen, 2013; Tannen &
Trester, 2013, among others). More specifically, the main interest now
lies in the ethnography (or netnography) of CMC (Androutsopoulos,
2006; Georgakopoulou, 2006; Kozinets, 2010).
Within the field of CMC, Facebook interaction is one of the areas that
has led to burgeoning research dealing with a myriad of different features
such as self-presentation and identity construction and management
C. Maíz-Arévalo (*)
Department of English Studies: Linguistics and Literature, University
Complutense of Madrid, Madrid, Spain
e-mail: cmaizare@ucm.es

https://doi.org/10.1007/978-3-319-92663-6_10
284 C. Maíz-Arévalo
(e.g. Barash, Ducheneaut, Isaacs, & Bellotti, 2010; Dalsgaard, 2008;

Lim, Vadrevu, Chan, & Basnyat, 2012; McKay, 2010), education
(Davies, 2012; Dyson, 2012; Sauter, 2013; Selwyn, 2009); pragmatics
(Carr, Schrok, & Dauterman, 2012; Ilyas & Khushi, 2012; among oth-
ers), persuasion in political discourse (Bronstein, 2013; Keat, 2012) and
a vast amount of studies which are beyond the scope of this paper. Most
research on Facebook, however, has focused upon English, while other
languages—like Spanish—have generally received less attention.
Furthermore, research has concentrated on the study of interaction
among Facebook users who also have an offline relationship. In this line,
the study by West and Trester (2013, p. 133) is especially noteworthy. In
their paper, they follow an ethnographic approach in an attempt to “cap-
ture the way participation takes place on particular sites […] by consid-
ering the expectations and norms surrounding facework on the social
networking site Facebook”. Their research reveals that Facebook is mostly
targeted to boosting positive face among its users. As a result, face-
threatening acts (FTAs) (Brown & Levinson, 1987) tend to be avoided
and facework is mostly accomplished by intertextuality. However inter-
esting, their analysis does not consider corrective facework in detail,
which is the main focus of the present study.
Thus, the aim of this chapter is twofold: on the one hand, it intends to
redress the imbalance in favor of Spanish, focusing on the interactional
strategies and relational work of a Peninsular Spanish Facebook commu-
nity. This community consists of more than 3000 users and it is adminis-
trated by three Facebook users, who grant friendship to all those users
who are fans of a Spanish male singer and ask to join his fan club. As such,
the group is formed by an ever increasing and open community of users
joined by a common interest, and everybody asking for friendship is
admitted with the only condition that the posts and updates are always
related to this singer.1 On the other hand, this paper intends to analyze a
sub-genre within Facebook since it focuses on the interactions taking
place in a semi-public common interest group, whose members do not
have an offline relationship and have just met online precisely because of
their common interest (a particular Spanish singer). More specifically, in
this chapter I shall focus on the study of restoration or face-repairing strat-
egies (corrective facework) used by this specific group of Facebook users.
Losing Face on Facebook: Linguistic Strategies to Repair Face… 285
10.2 Theoretical Background

The concept of facework has been widely researched as a core element in
human interaction (Arundale, 2006; Cupach & Metts, 1994; Goffman,
1955; Ting-Toomey, 1994, inter alia). Defined as “the carrying out of
communicative acts that influence face” (Turnbull & Saxton, 1997,
p. 152), facework is closely related to politeness but they are not inter-
changeable terms (Penman, 1990; Tracy, 1990; Wood & Kroger, 1994).
Prior to a definition of facework, however, it is important to define the
notion of face itself. In Goffman’s (1967) words, face is “an image of self
delineated in terms of approved social attributes” (p. 5); that is, our public
self-image in social interaction. Wood and Kroger (1994) usefully distin-
guish between face as a state and facework as a process. Hence, interac-
tants do facework “in order to injure, prevent bad effects on, repair bad
effects to, or benefit face” (Turnbull & Saxton, 1997, p. 152). Thus, the
effects of facework on face are those of maintenance, enhancement, resto-
ration or damage (ibid.). This is in line with Goffman’s premises that:
(1) One values one’s own and the other’s face; (2) one usually maintains
or saves one’s own and the other’s face; (3) when one loses face, one tries
to restore one’s own face; and (4) when one loses face, the other party
present in the situation tries to help the first party restore the lost face.
Corrective facework, or face-repairing strategies, can thus be carried
out either by the speaker whose face has been damaged (self-repairing
strategies) or by other(s), in a solidarity attempt to restore the social equi-
librium (other-repair strategies). However, repairing our own damaged
face may involve a further threat to others or to oneself, which makes
corrective facework a terrain where interlocutors should tread carefully.
In fact, social psychologists like Hodgins et al. (1996, p. 301) claim that
striking the right balance revolves around “the degree of threat to one’s
own face and motivation to maintain the relationship”. However, this
equilibrium becomes even more difficult to manage when a (large) group
of interactants is involved.
According to Jeon and Mauney (2014), in face-damaging or face-
threatening situations, users are commonly more concerned about repair-
ing the social relationship with those other users who are well known to
them rather than with those users who are either unknown or whose
relationship is weaker. Thus, in relationships in which the interpersonal link

among participants is weak, such as the one under scrutiny in this chapter,
it can be argued that avoiding taking sides or ignoring the whole situation
might be the general trend while dealing with others’ face-damage.
Corrective facework “may be performed by the person whose face was
threatened, or by others who are assisting in the protection or repair of
the person’s face” (Guerrero, Andersen, & Afifi, 2014, p. 48). According
to social psychologists Guerrero, Andersen, and Afifi (2014), there are six
general corrective strategies for repairing a damaged face (see also Cupach
& Metts, 1994), which can appear in isolation or combined:
1. Avoidance/Ignorance: it occurs when “individuals pretend that the act

never happened or ignore its occurrence” (Guerrero, Andersen, & Afifi,
2014, p. 49). For instance, by attempting to change the topic rather
than focusing on face-threat. This is illustrated by extract (1),2 where
the user tries to change the topic (discussing the reasons why another
user has chosen not to participate actively in the group any longer) by
resorting to an “adapted” version of one of the Singer’s most popular
songs where he asks his girlfriend to reconsider their break up. In the
new lyrics, rather than using the original version “from my place to
your place”, the user opts from “from my Facebook to your Facebook”
to convince the offended user not to “break up” with the group:
(1)
y si se lo decimos con una cancion de [Cantante] … alla voy yo: “Por
todo el camino de mi facebook a tu facebook, cómo convencerte venía
pensando…”
[and … what about telling her with a song by [Singer]… here I go:
“all the way from my Facebook to your Facebook, I was thinking
about how I could convince you…”]
2. Humor: “Sometimes it is best to laugh at yourself so others will laugh

with you, not at you” (Guerrero, Andersen, & Afifi, 2014, p. 49). This
is precisely what happens in extract (2), where the user has realized
that he has spelled the verb “cantar” (to sing) incorrectly, as he has
written “catar”. However, in Spanish, the verb “catar” means “to
taste”. This why the user refers to the fact that, if they really meet in a
bar to sing the singers’ songs, they will do both things (sing and taste
a few drinks together). In this way, he laughs at his own mistake, as he

also makes clear by the onomatopoeic repetition of laughter at the end
of the comment:
(2)
he puesto “catamos”, cantamos, joperrrrr, cantamos, catar podemos catar
también, pero me refería a cantar, jajajajajajajajaja [I’ve written
“taste”, sing, darnnnnn, sing, taste we can taste too, but I meant to
sing, hahahahahahahaha]
3. Apologies: the speaker expresses regret for having committed a previous

offense, as a social ritual to restore their relationship with the
interlocutor(s) (Davies, Merrison, & Goddard, 2007; Fraser, 1981;
Grainger & Harris, 2007; Holmes, 1990; Olshtain, 1989). In the data-
set at hand, however, apologies are commonly used not to express regret
for having committed a previous offense but in order to mitigate other
FTAs like disagreeing with the interlocutor(s) (see Analysis section).
4. Accounts: involve attempting to explain the FTA, mostly by means of
excuses or justifications. “Excuses are explanations that minimize per-
sonal responsibility of the actor for the actions” (Guerrero, Andersen,
& Afifi, 2014, p. 49). On the other hand, “with justifications, actors
do not try to distance themselves from the act, but instead ‘reframe an
event by downplaying its negative implications’” (Cupach & Metts,
1994, p. 10). In a similar line, Hodgins, Liebeskind, and Schwartz
(1996) distinguish a cline from concessions (where the perpetrator
takes full responsibility and hence threatens their own face), excuses
(which mitigate the speaker’s own threat by reducing responsibility),
justifications (where full responsibility is accepted but justified as a
legitimate behavior) to, finally, refusals (where all personal responsibil-
ity is denied) (p. 300). Interestingly enough, previous studies (Hodgins,
Liebeskind, & Schwartz, 1996) have shown that gender has an effect
on account length, with women providing longer and linguistically
more complex accounts.
5. Remediation: “involves attempts to repair physical damage (e.g. clean-
ing up after you have spilt some drink on your partner)” (Guerrero,
Andersen, & Afifi, 2014, p. 49). In the case of CMC, however, there is
not a real possibility to carry out remediation except linguistically, for
example, by promising not to do it again or offering compensation.
6. Aggression: on some occasions, “individuals feel the need to repair their

damaged face by using physical aggression […] People may also become
aggressive when they are embarrassed or violate a norm” (Guerrero,
Andersen, & Afifi, 2014, p. 49). It may be argued that aggression per se
does not restore face-loss but it has been demonstrated to be a frequent
aftermath (Cupach & Metts, 1994) as an attempt “to reassert the indi-
vidual’s capability or strength after the person feels threatened” (Ting-
Toomey, 2005, p. 79). Aggression can be physical in face-to-face
encounters, but it becomes purely linguistic in the case of CMC—for
example, resorting to insults, using sarcasm, and so on.
Taking the above into account, the present chapter aims to answer the
following research questions:
1. What face-repairing strategies are used by the participants in a

Facebook common interest group?
2. Does self-repair occur more frequently than other-repair?
In the group under scrutiny, users only share an online relationship,

with their common interest in a Spanish singer as their only link. It is
therefore hypothesized that face-repair strategies will mostly be produced
by those users whose face gets damaged—that is, self-repair rather than
other-repair in line with Jeon and Mauney’s (2014) argument that inter-
locutors are more worried about repairing the social relationship with
those interlocutors who are well known to them than with those with
whom they have a weaker relational link. Other users, however, are
expected to ignore or avoid conflict by remaining silent, hence avoiding
damaging their own face by “taking sides” and therefore helping to keep
the group’s conviviality and harmony.
10.3 Methodology
This section focuses on the methodology employed in the present paper
by paying attention to the data gathering method, corpus description and
the participants involved. A brief section on the ethics followed closes up
the section.
10.3.1 Data Gathering Process and Corpus Description
The data was compiled following a netnographic approach, defined as “a

participant-observational research based in online fieldwork [that] uses
computer-mediated communications as a source of data to arrive at the
ethnographic understanding and representation of a cultural or commu-
nal phenomenon” (Kozinets, 2010, p. 60). Thus, the author of this study
was a participant observer who has been part of the community for
almost a year, even if not as a pro-active member. The data was randomly
collected during January and February 2015, so as to gauge the way par-
ticipants tend to interact and to gather completely natural data, unbiased
by the researcher’s objectives. Once the data was gathered, the only limit-
ing criterion was to eliminate all exchanges which did not include textual
comments. Thus, exchanges which only included an update (e.g. a video
link to one of the singer’s songs) followed only by ‘Likes’ were not consid-
ered for the present analysis. The resulting corpus involves 12,327 words
and 412 conversational turns in Spanish.
10.3.2 Participants
At the time of data compilation, the group under study consisted of 2849
members (on April 2018 it already includes 3875). However, most mem-
bers are—like me—merely lurkers with only 117 (4.1%) participating
with comments and posts rather than simply clicking on Like. To calcu-
late this figure, all the participants at the time of the data compilation
were manually listed and only those who had ever posted were considered
while those who only chose to like others’ posts or updates were not con-
sidered (see Sect. 10.3.1). Out of these, 65 are female, 43 are male and
there were 9 participants whose gender could not be identified by their
username, despite Facebook’s policy in this respect.3 The rest of the par-
ticipants simply lurk or click on the ‘like’ button. Following Goffman
(1955), this differentiation has been described as users who are on the
front stage or remain on the backstage (Keehn, 2013). Active members
may, hence, be argued to also have a more active behavior when repairing
self and other’s face-damage while less active members are expected to
ignore face-loss or avoid entering conflict.
The group is based on a common interest in a Spanish singer; where all

the members only share an online relationship despite a few failed
attempts to meet offline. Given the local character of the singer (most
likely to be unknown outside the Spanish-speaking world), all the partici-
pants are assumed to share the same mother tongue—that is, Spanish–
even if they speak different regional varieties, whose analysis is beyond
the scope of the present paper.
Regarding the participants’ age, given that all of them share interest in
this ‘generational’ singer (who was famous in the 80s), they are all assumed
to be within the same age span (late 30s to mid-40s).
10.3.3 Ethics
It has often been pointed out that CMC offers more advantages regard-
ing data gathering than face-to-face interaction since painstaking tran-
scription is not necessary and “interaction results in the immediate
production of a text file” (Mann & Stewart, 2000, p. 22). However,
texts need to be slightly edited to respect the participants’ anonymity.
In this sense, I have followed the guidelines provided by previous
researchers. Therefore, messages posted on the internet are considered
as public acts and, even if researchers must act cautiously, there is no
need to take more than “normal precautions” (Paccagnella (1997),
quoted in Mann & Stewart, 2000, p. 46). In this respect, Kozinets
(2010) argues that “analyzing online community or culture communi-
cations or their archives is not human subjects’ research if the researcher
does not record the identity of the communicators” (emphasis in the origi-
nal) (p. 142). Hence, in order to preserve anonymity and ensure confi-
dentiality, the present study only quotes textual data as examples to
illustrate the aspects under analysis. Thus, all photographs have been
eliminated. As for the participants’ names, they have also been elimi-
nated (see note 2). Whenever there is more than one user, they are
referred to as U1, U2—that is, user 1, user 2, in the order they are
intervening in the exchange. Even the name of the singer has been sub-
stituted by [Singer] whenever it appears. Finally, vulnerability poses no
problems since the study deals with research that “involves no risk [and
focuses on] activities typical of normal daily behavior and where the
research does not involve the collection of identity in association with
response data” (Kozinets, 2010, p. 143).
10.4 Data Analysis

In general terms, users may need to repair their own public image (face)
in two situations: when their face has been previously damaged by others
or when they have damaged their own face—for example, by making a
blunder. In the dataset at hand, there are seven situations where face-loss
has occurred. The analysis focuses upon these seven situations, which are
summed up in Table 10.1.
The data was analyzed using Guerrero, Andersen, and Afifi’s (2014)
categories (see Sect. 10.2). However, it was observed that their categories
mostly apply to self-repairing strategies but do not account for other-
repair strategies emergent in the data such as expressing support for the
other or appealing to group’s unity. Figure 10.1 summarizes the different
strategies, providing an example of each of them. The two latter strategies
have been shadowed to indicate they have been added and did not origi-
nally appear in Guerrero, Andersen, and Afifi’s (2014) repertoire.
As the analysis shows, most strategies, such as apologies, can be used as
self-repair or other-repair. However, some of these strategies are exclusively
employed when repairing other members’ damaged face—that is, express-
ing support. For the sake of methodological clarity, the following sections
will deal with the different strategies in isolation. However, the combina-
tion of different strategies can also take place. Table 10.2 below summarizes
the frequency of occurrence of these different strategies in a summative way
(i.e. taking into account all the different situations described in Table 10.1).
10.4.1 Expressing Support for the Other
As can be observed, the most frequent strategy is expressing support for

the member whose face has been damaged (29%). Following Maíz-
Arévalo and Sánchez-Moya’s (2017) study of support strategies, express-
ing support can take the following forms (Fig. 10.2):
Table 10.1 Situations of face-loss in the corpus

Situationa Description
The meeting One of the members has suggested meeting in a friend’s bar to
arrange an informal concert session. His suggestion has been
criticized as an attempt to benefit his friend’s business.
The Embedded within “The meeting”. One of the members has
misspelling misspelt a word. This leads to a humorous sub-topic on
singing and drinking. It is the only example of ‘self-inflicted’
face-loss.
A personal One of the most active members has been criticized by another
offense member for posting too much and trying to steal the groups’
attention from the real protagonist of the group (the Singer).
She feels offended and leaves the group after an emotional
goodbye to the group.
The interview One of the most active members has uploaded an interview
where the private life of the singer and his drug problems are
discussed. Because of the critics, the person who uploaded
the interview deleted it and abandoned the group. This
generated a division in the group between those who are
against invading the singer’s privacy and those who think
everything can be commented on, provided it is respectfully
done. The latter want this member to rejoin the group.
Disagreement Embedded within “The Interview”, two of the members (who
represent the two different sides of the issue) start a
discussion on the topic.
The warning One member posts a comment where he compares the Singer
to another popular singer. Another member warns him he
might be criticized for this comparison since there are
members of the group who do not like the other singer (and
they have been explicit about it) and hence should be careful.
The shoes After posting a photograph, one of the members starts a
sub-topic on the name of the singer’s shoes but her comment
is misunderstood. This leads to more comments on this
sub-topic.
For the sake of clarity, the situations have been given a name to facilitate
a
further reference to them
Support can be shown in two major ways: by showing approval of the

other and/or their action(s) or by showing kindness toward them. The
following exchange took part in the situation labeled “the personal
offense” after the offended user, who had abandoned the group, eventu-
ally decided to rejoin it. In this exchange, it can be observed how users
E.g. "Y ahora qué pasa? No me entero de

Expressing ignorance User(s) pretend not to know nada..."
(Avoidance) what the face loss is about [What's going on now? I don't understand
anything]
E.g. "No busco q estes de acuerdo,solo

User commits to action in
Remediation order to remediate face loss
trato de aclarar las cosas" [I'm not trying
to make you agree, just trying to clear
things up]
User employs humor to E.g. Hay que cantar? Jodooo, habrá que ir
haciendo gárgaras!
Humor
Face repair strategies
diminish face loss as [Do we have to sing? Jee, I'll have to

unimportant gargle!]
E.g. "me fui porque me sentía en una

página que no había libertad para hablar
lo que se quiera sin ofender a alguien"
Accounts User justifies action as licit [I left because I felt there wasn't any
freedom to say whatever one wans to say
without offending anybody"]
E.g. "Siento si he te he ofendido"

Apologies User apologises for action [I'm sorry if I have offended you]
User attacks other user(s) to E.g. "pero tú de qué vas?"

Aggression reestablish his/her strength [What's your story, man?]
User expresses support E.g. "Te entiendo perfectamente"

Expressing support for other user(s) [I totally understand you]
E.g. "Estamos todos aqui por un

Appealing to group's User appeals to group unity to sentimiento común"
unity avoid conflict [We are all here thanks to a common
feeling]
Fig. 10.1 Face-repairing strategies
Table 10.2 Frequency of face-repairing strategies in the corpus

Face-repairing strategy N = 216 Ratio
Expressing support 63 29%
Giving account 36 17%
Expressing ignorance 34 16%
Using humor 32 15%
Aggression 27 12%
Remediation 13 6%
Appealing to group’s unity 11 5%
interactively repair the face of the offended participant (U4), showing her
their full support by welcoming her back (U1 and U2), or by compli-
menting her (U3). U4 also responds to their support either by thanking
them or returning the compliment in turns 4 and 54:
Fig. 10.2 Strategies to express support (Maíz-Arévalo & Sánchez-Moya, 2017,

p. 215)
(3)
1. U1 (m): bienvenida de nuevo
[welcome back]
2. U2 (f ): Me alegro de tenerte de vuelta [U4], y me alegra que hayas
sido tan clara, xq por ese mismo motivo tb se ha ido mas gente
[I’m glad to have you back [U4], and I’m glad that you have been
so clear, because some others also left for the same reasons]
3. U3 (f ): ainsssss…a la mierda!!! [U4] eres un sol enormeeeee…no te
vayes!!!
[awwww…. Let’s them go to hell!! You are a huge sunshineeeeeee…
don’t ever go!!!]
4. U4 (f ): Gracias amigos !!!!
[Thanks friends!!!]
5. U4 (f ): [U1] me encanta la canción. Mil besosss. Eres genial.
[U1 I love the song. A thousand kisses. You are great.]
Showing support for a member whose face has been lost or damaged
does not support the initial hypothesis that others would not risk their
own face to help someone else, especially in a group with a low relational
link. By showing support of others’ actions (which have been criticized by
other group members), these users are risking their own face, since they
might be criticized by those members who initially criticized the member
they are now supporting. There are two situations (“The personal offense”
and “The Interview”) where this strategy is most often employed.
Interestingly enough, both contexts share two common features. On

the one hand, the member whose face has been damaged threatens to
abandon the group. On the other, both situations divide the group into
two “factions”. Thus, in the case of the personal offense, the leaving mem-
ber is a very active participant who has both detractors and supporters,
and both groups engage in further discussion once she is gone. In the case
of the second leaving member, her uploading of a very controversial
interview has given rise to the discussion about whether this group should
exclusively focus on the singer’s music or on everything related to him. In
fact, this is the only requirement or policy in the group, where it is stated
that the only condition to join is to contribute with photos or songs of
the singer, but not with polemics about him. Given that the group’s only
link is sharing information about the singer, it is important for its mem-
bers to determine the kind of information to be shared but also to ensure
that disagreements like the present one do not pose a real threat to the
group’s continuity. This could explain why other members are so eager to
restore harmony either by not intervening in the discussion (as in “The
Disagreement”) or by expressing their support for the member who
intends to leave, so that she feels part of the group and does not disrupt
it by abandoning it.
10.4.2 Giving an Account
Giving an account is the second most frequent strategy (17%), used both
as self- and other-repair strategy. Explaining why face should be restored
acts as a strategy to restore it. As can be seen in extract (4), the user employs
other-repair by excusing or minimizing the importance of the action (“I
can’t see anything wrong”, “I can’t see anything odd or out of place”):
(4) U1 (m): No veo nada malo es una entrevista con preguntas coherentes a
un periodista que ha escrito un libro sobre [Singer] yo y creo la mayoria
de nosotros hemos leido “[Título de libro]” asi que tampoco veo algo raro
o fuera de lugar en las preguntas.
[I can’t see anything wrong it is an interview with coherent ques-
tions to a journalist who’s written a book about [Singer], I think I
myself and most of us have read [Book’s title] so I can’t see anything
odd or out of place in the questions]
When used as a self-repair strategy, interlocutors tend to opt for justifying

their action, as in extract (5), where the user explains why she left the
group after being heavily criticized for posting some comments (“I left
because…”) and, as she says, she felt her freedom of speech was being
disrespected (“there was no freedom to talk”). This is her first update after
re-joining the group again:
(5)
me fui porque me sentía en una página que no había libertad para hablar
lo que se quiera sin ofender a alguien. Y para mi la libertad de expresión
es fundamental, siempre desde el respeto.
[I left because I felt I was in a group where there was no freedom
to talk about what you like without offending anybody. And for me
freedom of expression is crucial, always from respect.]
A close inspection of the data reveals that, when some member’s face is
damaged by others, the need to repair becomes more pressing than when it
is self-damaged (e.g. by making a personal blunder), given the speaker’s need
to feel part of the group. In extract (6), the user, who has been criticized for
trying to organize an offline encounter in a friend’s bar, tries to restore his
face by justifying and fully explaining the reason behind his actions:
(6)
mi idea al organizar la quedada es que nos pusieramos cara … nos cono-
cieramos … rindieramos homenaje a Cantante … para nada pretendo
lucrarme, hacer negocio … me encantaria que todos los que quisieran y
se atrevieran subieran al escenario a cantar … no pretendo que sea un
concierto de uno o dos grupos … la sala donde se va a hacer es grande, ir
avisando e invitando a vuestros amigos del grupo … gracias
[my idea when I organized the meeting was to get to know one
another, to honor [Singer]… I don’t intend to make any profits at all
or to do any business… I’d love it if everybody who wanted and dared
could come onstage and sing… I don’t intend it to be a one or two-
group concert … the place where it’s taking place is big, tell and
invite your friends in the group … thanks]
As can be observed, the user opts for giving a long account, justify-
ing his proposal as a positive thing for the group and the singer (“my
idea when I organized the meeting was to get to know one another, to
honor [Singer]”) and hence redeeming any suspicion of a hidden

agenda (“I don’t intend to make any profits at all or to do any busi-
ness”). Indeed, giving an account seems to be favored when self-repair-
ing face (damaged by others).
10.4.3 Expressing Ignorance
Expressing ignorance of why there is a disruption in the group is the third

most common strategy (16%). As opposed to Guerrero, Andersen, and
Afifi’s (2014) findings, where it was often used as a self-repair strategy by
ignoring face-loss (e.g. pretending to ignore an ironic remark uttered by
an interlocutor or pretending no harm was made after slipping down on
the street in front of other passers-by). In the corpus under analysis,
expressing ignorance is employed differently. When faced with these con-
flictive situations, other users resort to showing their puzzlement and
ignorance of the matter. It could be argued that this is not a face-repair
strategy if users are actually unaware of what is happening and simply ask
to be informed, as shown in extracts (7) and (8):
(7) Por que [sic] te vas [Name]? [Why are you leaving, [Name]?]
(8) Yo sí que estoy perdida … no sé qué ha pasado [I’m really lost, I don’t
know what happened]
However, expressing ignorance may also serve as a highly convenient

face-repairing strategy since participants show their interest in the
group while preserving their own face by not taking sides. The strategic
use of expressing ignorance becomes more obvious on those occasions
when a member’s ignorance seems to be feigned as shown by their con-
tradictory comments. For example, in extract (9), the user states that he
does not know what another user has published (“I don’t know what
you have published”) and the reasons for the whole disruption (“I don’t
understand they are upset because of an interview”). However, he ends
up his post by adding that the publication is an interview, showing
hence that he does indeed know what the attacked user has published,
which contradicts his previous statement:
(9)
[Name] ni se te ocurra irte!!!!!! No se [sic] lo que has publicado lo que no
entiendo es que se molesten por una entrevista. [[Name] don’t even
think of leaving!!!!! I don’t know what you have published but I don’t
understand they are upset because of an interview]
Furthermore, some users may also ‘take advantage’ of others’ ignorance

to reinforce their own face and raise up their status as more powerful
members within the group, since they have the piece of information
other users lack. This is the case of a very active member who has aban-
doned the group twice and generated different controversies herself. In
extract (10), and given the different expressions of ignorance by some
members, she states she knows the reason why another user has left but
will not reveal that information as a sign of respect for her (“I know what
it is but if [Name] doesn’t want to, I’ll respect her”). However, she imme-
diately asks another member to upload the controversial interview so as
to clear the “mystery” up for those other users whose information is less
complete than hers (“[Name], if you have it, upload it. It is an exclusive
interview of Singer”):
(10)
Chicos, perdonad, pero a mi no me corresponde aclararlo. Se lo que es
pero si [Name] no quiere … yo la respeto. [Name], si la tienes pomla
[sic]. Es una entrevista exclusiva a [Singer].
[Guys, sorry, it is not me who should clear it up. I know what it
is but if [Name] doesn’t want to, I’ll respect her. [Name], if you have
it, upload it. It is an exclusive interview of Singer]
10.4.4 Using Humor
As pointed out by Guerrero, Andersen, and Afifi (2014), humor is often

resorted to by users who have committed a blunder and who prefer laugh-
ing at themselves before being laughed at by others. In this dataset, it is
employed in 15% of the cases. The data shows that, in such contexts,
self-inflicted humor usually leads to a stream of humorous comments by
other members, as a way to boost the group’s conviviality and rapport.
This is illustrated in the situation I have termed “The misspelling”, which
was already quoted as extract (2) but is repeated here for the sake of
clarity as extract (11):
(11)
1. U1 (m): pos claro, catamos tós [sic] [some minutes later, he adds
the following comment]
2. U1 (m): he puesto “catamos”, cantamos, joperrrrr, cantamos, catar
podemos catar también, pero me refería a cantar, jajajajajajajajaja.
As can be observed by his exaggerated used of onomatopoeia at the

end of the second comment, the user prefers resorting to humor and
laughing at himself for his ‘innocent’ blunder. This gives rise to an imme-
diate long set of humorous comments and onomatopoeic representations
of laughter by the other interlocutors, who expand the joke while boost-
ing the group’s rapport and camaraderie and leading the topic away from
the user’s initial mistake. As can be seen in extract (12), other users also
resort to onomatopoeic representations of laughter (as U2 and U3), and
repetition of the blunder (in turn 3):
(12)
1. U2 (f ): Te entendemos igualmente… jjjjjjjjj
[We understand you as well… hahahaha]
2. U1 (m): ok, jajajajajjaajajajaja.
[ok, hahahahahahahaha]
3. U3 (m): Jijiii … cataremos algún cubata mientras cantamos,
[hehehe … we’ll taste some cocktails while singing]
Humor can also be occasionally used in more controversial cases, as a

way to tone down the disruption. As opposed to self-initiated humor, in
cases like this, the use of humor may be argued to be not as much a self-
repair strategy as an attempt to recover the group’s harmony by minimiz-
ing the initial disruption. In contrast to humorous streams like that of
extract (12), the use of humor as a “mitigator” of disruption tends to
occur in isolation. Extract (13) follows one of the user’s farewell to the
group. As in the previous example, each turn is followed by its transla-
tion. Given the length of the exchange (with 35 turns), it has been only
partially quoted. As can be observed, the most general tendency for the
interlocutors is to express their ignorance or to show their support for the
leaving member, as in the case of Users 1, 2 or 3. There are also some cases
of aggression (which act as a clear face-threat), as is the case of User 4.
However, users 5 and 6 employ humor to soften the disruption and as an

attempt to lighten up the atmosphere. Interestingly enough, user 6
explicitly refers to humor as a conflict mitigating strategy (he explicitly
writes that he is simply “trying to play things down”):
(13)
1. U1 (f ): Por que te vas [Name]?
[Why are you leaving, [Name]?]
2. U2 (f ): pero [Name] xq te vas??
[But [Name], why are you leaving??]
3. U3 (m): Pero que dices????
[What are you talking about????]
[…]
4. U4 (m): pues adios, a mi me parece que aqui hay mucha gente que
intenta darselas de victima para que le supliquen que no se vaya.
quien no quiera estar, pues ya sabe donde esta la puerta
[well, good-bye then. I think there are many people here who
play the victim to be begged not to go. The one who doesn’t want
to be here also knows where the exit is.]
5. U5 (m): Esto mas que un grupo parece el gobierno
[This looks like the government more than like a group]
[…]
6. U6 (m): Esto parece gran hermano Vip, Víctor Sandoval también se
quiere ir. Intentando quitar un poco de hierro al asunto.
[This seems VIP Big Brother, Víctor Sandoval5 also wants to
leave. Trying to play things down].
10.4.5 Aggression
Remarkably enough, aggression is employed in 12% of the cases. As a

self-repair strategy, aggression is used to reassert the offended user’s
strength and position within the group (Cupach & Metts, 1994). In
extract (14), the user employs aggression as a self-repair strategy by first
discrediting the member who previously offended her. She argues that he
cannot judge her or say she is playing the victim because he does not
know her at all. It could be argued that resorting to aggression may not
necessarily help her restore face in front of the group. However, she
might see her aggressive response as a reaffirmation of her personality and

self-value in front of the group, which implies face-restoration (at least in
her understanding). She ends up her update by disparaging the other
member as a nobody (“of course I don’t give a damn [about him]”;
“I didn’t even know you existed”) and by means of sarcasm (“I’ve never
had the PLEASURE to talk to him”; “well, I’m very sorry you have to
deal with me again, hahaha”).
(14)
Bien, vamos a ver, y sin que nadie se vaya a ofender, solamente la per-
sona a la que me voy a dirigir, que por supuesto me importa un bledo,
pues nunca he tenido el PLACER de hablar con el. Vamos a ver y aquí
se va a quedar todo, no entraré al trapo digas lo que digas, estas son mis
últimas palabras para ti. El susodicho es : [Name]. TU quien co … eres
tío?? Acaso me conoces de algo?? Una pena que no te vayas tú … de víc-
tima irás tú. Tu que sabes de mi?? Y te voy decir, suplicar no lo han echo,
pero llamadas y de todos contactos, todo lo que te puedas imaginar.Siento
mucho, bueno lo de siento es un decir, pues eso que siento mucho que me
tengas que aguantar otra vez jjjjj. Aunque contigo es la primera vez que
te menciono, ni siquiera sabía que existias, pues hala aquí estoy de
nuevo. Ya lo he dicho, no voy a comentar nada de nada, como si me
llamas lo que quieras. No ofende quien quiere sino quien puede. Y no es
tú caso. Guapo.
[Well, let’s see, and please nobody get offended, only the person
whom I am addressing, who of course I don’t give a damn about,
because I have never had the PLEASURE to talk to him. This per-
son is: [Name]. Who the f*** are YOU, man? Do you even know
me?? A real pity it is not you who have left … you may play the
victim, not me. What do you know about me?? And I’m telling you,
they haven’t begged me, but calls from all the contacts, everything
you can imagine. I’m very sorry, well, that’s a way of talking, well,
I’m very sorry you have to deal with me again, hahaha. Although it
is the first time I mention you, I didn’t even know you existed, well,
here I am again. I’ve said it before, I’m not responding to anything,
you can call me anything you like. Offence is not caused by those
who want to but by those who can. And you can’t. Handsome.]
Other cases of aggression are not intended as face-repair strategies

(although they may trigger face-repair). This is the case in extract (15),
where users do not try to avoid another user leaving the group but are
rather pleased she is doing so (cf. also extract U4’s comment in extract 13
above):
(15) Pues nada adiós. No seré yo quién te lo impida.

[Well, goodbye then. It won’t be me who stops you]
It can be argued that in cases like these, users contribute to other mem-
ber’s further face-loss by clearly opposing them. In other words, rather
than showing their support for the offended party, these other users pre-
fer to attack the former.
10.4.6 Remediation
According to Guerrero, Andersen, and Afifi (2014), remediation takes

place when there is a physical attempt to repair face-loss (e.g. by cleaning
up after spilling a glass of wine). In a disembodied context like CMC,
where no physical actions of that nature are possible, remediation takes
place linguistically, when users try to correct a previous face-threat, for
example, by promising future action in favor of the offended party. This
is illustrated in extract (16), where the user tries to remediate their part-
ner’s face-loss by promising to do their best in support of his idea (ini-
tially criticized by other members):
(16)
hatemos [sic] todo lo kr este en nuestras manos asi es todo por sentimiento
y por la muxica ke nos dejo [Cantante] [we’ll do our best, that is,
everything for this feeling and the music [Singer] left us]
However, remediation is less common in the data, with 13 cases (6%).

This might be due to the limited size of the dataset or to the disembodied
nature of CMC as opposed to face-to-face exchanges, where remediation
might happen more frequently by non-verbal means.
10.4.7 Appealing to Group’s Unity
Finally, the least frequent strategy found in the corpus is appealing to the
group’s common interests (5%). Individual users (whose face might have
been lost) are requested by other members to prioritize the group’s interest
and ignore their individual face-loss. It could be argued, hence, that more
than a face-repair strategy per se, appealing to unity is an attempt to
minimize the importance of individuals (“blurring” their individual face
and face-loss) in favor of communal harmony. A clear case is illustrated
by extract (17), where the user argues that anything “personal” should be
put aside (we should leave aside everything “personal”) and that the
group’s purpose is precisely to be united by a common interest (“Those in
the group are united by a common feeling of admiration and love for
[Singer]”). It is also remarkable how she abandons the singular personal
pronoun (“I don’t get it either, to be honest … but I respect people’s deci-
sions”) to move on to the plural personal person, an inclusive “we” stress-
ing the group’s unity:
(17)
Pues yo tampoco me entero, la verdad … pero respetolas decisiones Los
que formamos el grupo estamos unidos por un mismo sentimiento de
admiración y cariño por [Singer] y creo que deberíamos dejar a un lado
todo lo que fuera “personal”. Se trata de disfrutar con la página y con el
grupo … si alguien no está contento, pues es libre de hacer lo que quiera.
El protagonismo siempre debe ser de [CANTANTE].
[Well, I don’t get it either, to be honest … but I respect people’s
decisions. Those in the group are united by a common feeling of
admiration and love for [Singer] and we should leave aside every-
thing “personal”. This is about enjoying the page and the group …
if someone is not happy, they are free to do whatever they want.
The protagonist must always be [Singer]].
It could even be argued that, for most users, individual face (and face-
loss) should be obliterated in favor of the group and, especially, of the
Singer. If there is some face to save, it would be the Singer’s face, who
might get unhappy with these disruptions in the group (even if he passed
away years ago), as stated by the user in extract (18):
(18)
Bueno estas cosas …seguro … que nó [sic] … le gustan a [Cantante] …
besos para todos. [Well, I’m sure… [Singer] doesn’t like … these
things… Kisses for everybody.]
10.5 Discussion and Conclusions

Focusing on the corrective facework employed by a concrete Facebook
group, this chapter aimed to answer the following research questions,
repeated here for the sake of clarity:
1 . What face-repairing strategies are used by the participants?

2. Does self-repairing occur more frequently than face-repair by other
members?
With regard to the first question, inspection of the data reveals that
prior taxonomies of face-repairing strategies such as giving an account,
using humor or expressing ignorance (Guerrero, Andersen, & Afifi,
2014) account for most cases of face-repair. However, these strategies
focus on self-repair rather than other-repair. In the corpus under analysis,
it has been observed that other-repair tends to involve expressing support
for the others as well as for their decisions and actions. More specifically,
prior studies (Angouri & Sanderson, 2016; Maíz-Arévalo & Sánchez-
Moya, 2017) have shown that expressing support is a strategy commonly
used as other-repair. Furthermore, the ‘disembodied’ nature of the chan-
nel, where interaction takes place online, may explain why common
strategies in face-to-face exchanges, such as remediation, are hardly used
in this computer-mediated setting. To sum up, the most common strat-
egy employed is expressing support, followed by giving an account,
expressing ignorance of what is going on, using humor (typically
employed as a self-repair strategy), aggression (used either as a self-repair
strategy or not as face-repair but as a direct face-threat toward another
user), remediation and, finally, appealing to the unity of the group. The
latter strategy seems to be employed as an attempt to obliterate individu-
als’ face needs rather than to repair face-loss.
As for the second question, it was hypothesized that, in face-damaging

or face-threatening situations, users are commonly more concerned about
repairing the social relationship with those other users who are well
known to them rather than with those users who are either unknown or
whose relationship is weaker (Jeon & Mauney, 2014). In this case, and
given that the members of the group share a relatively weak relationship
based on their common interest in a singer, it was hypothesized that
corrective facework would be mostly self-repairing and other users were
expected to ignore or avoid conflict by remaining silent. However, results
show that other-repair is extremely frequent, and takes place in many dif-
ferent ways and, rather than remaining silent, users are very participative
when repairing the group’s harmony, which indicates they highly value
the group’s unity.
Finally, there are some limitations to the study which need acknowl-
edging. On the one hand, the size of the corpus is reduced and focused
upon a specific group, which makes it a case study. Further research
should focus on a larger corpus including different common interest
groups to find out possible generalizations. On the other, the limited size
of the corpus has not allowed to zero in on the possible effects of gender
in the choice of some face-repair strategies over others. In this respect,
past research has often claimed that “men repair their own faces more and
attend to others’ less than women do” (Hodgins, Liebeskind, & Schwartz,
1996, p. 301; see also Hamilton & Hagiwara, 1992; Petronio, 1984) and
that female interactants produce longer accounts than their male coun-
terparts (Hodgins, Liebeskind, & Schwartz, 1996). Further research is,
hence, intended to delve into the role played by gender in corrective
facework.
Notes
1. I asked for “friendship” to the community and was accepted in less than
ten minutes, which shows that the group is really open and anybody inter-
ested in the singer can join it.
2. In order to fully preserve the users’ anonymity, all the examples analyzed
have the names removed and users are merely referred to as U1, U2,
etc.—i.e. user 1, user 2; in the order in which they appear in the exchange
if more than one intervention is quoted. This is accompanied by their
users’ gender, when identified. When users address other users or the
singer within their post, these names have also been edited and substi-
tuted for [Name]. The translation into English is offered immediately
after each example, also in square brackets. No further edition (e.g. of
mistakes, misspellings, etc.) has been carried out so as to keep the data’s
integrity. Finally, all examples are in italics, since they are written in a dif-
ferent language from that of the paper.
3. As pointed out in their web page, Facebook only allows for real identities
to be used or pseudonyms reproducing the participants’ real-life name.
https://www.facebook.com/help/112146705538576
4. For the sake of clarity, turns in longer examples (consisting of more than
one turn) have been numbered. Likewise, translation has been offered
after each single turn to ease the reading.
5. Víctor Sandoval is popular on Spanish TV for presenting gossip programs
and taking part in reality shows, like VIP Big Brother.
References
Angouri, J., & Sanderson, T. (2016). ‘You’ll find lots of help here’ unpacking the
function of an online Rheumatoid Arthritis (RA) forum. Language and
Communication, 46, 1–13.
Arundale, R. B. (2006). Face as relational and interactional: A communication
framework for research on face, facework, and politeness. Journal of Politeness
Research. Language, Behaviour, Culture, 2(2), 193–216.
Barash, V., Ducheneaut, N., Isaacs, E., & Bellotti, V. (2010). Faceplant:
Impression (mis) management in Facebook status updates. Proceedings of the
Fourth International AAAI Conference on Weblogs and Social Media,
207–210.
Bronstein, J. (2013). Like me! Analyzing the 2012 presidential candidates’
Facebook pages. Online Information Review, 37(2), 173–192.
Brown, P., & Levinson, S. C. (1987). Politeness: Some universals in language usage
(Vol. 4). Cambridge: Cambridge University Press.
Carr, C. T., Schrok, D. B., & Dauterman, P. (2012). Speech acts within
Facebook status messages. Journal of Language and Social Psychology, 31(2),
176–196. https://doi.org/10.1177/0261927X12438535
Cupach, W. R., & Metts, S. (1994). Facework (Vol. 7). London: Sage.
Dalsgaard, S. (2008). Facework on Facebook: The presentation of self in virtual
life and its role in the US elections. Anthropology Today, 24(6), 8–12.
Davies, B. L., Merrison, A. J., & Goddard, A. (2007). Institutional apologies in
UK higher education: Getting back into the black before going into the red.
Journal of Politeness Research. Language, Behaviour, Culture, 3(1), 39–63.
Davies, J. (2012). Facework on Facebook as a new literacy practice. Computers
& Education, 59(1), 19–29.
Dyson, E. (2012). Face-to-Facebook: A blended approach to careers work.
Journal of the National Institute for Career Education and Counselling, 29(1),
27–32.
Fraser, B. (1981). On apologizing. In F. Coulmas (Ed.), Conversational routine:
Exploration in standardized communication situations and prepatterned speech
(pp. 259–271). The Hague: Mouton.
Georgakopoulou, A. (2006). Postscript: Computer-mediated communication
in sociolinguistics. Journal of Sociolinguistics, 10(4), 548–557.
Goffman, E. (1955). On face-work: An analysis of ritual elements in social
interaction. Psychiatry, 18(3), 213–231.
Goffman, E. (1967). Interaction ritual: Essays on face-to-face behavior. Allen
Lane: The Penguin Press.
Grainger, K., & Harris, S. (2007). Special issue: Apologies. Introduction. Journal
of Politeness Research, 3(1), 1–9.
Guerrero, L. K., Andersen, P. A., & Afifi, W. A. (2014). Close encounters:
Communication in relationships. London: Sage Publications.
Hamilton, V. L., & Hagiwara, S. (1992). Roles, responsibility, and accounts
across cultures. International Journal of Psychology, 27(2), 157–179.
Herring, S., Stein, D., & Virtanen, T. (Eds.). (2013). Pragmatics of computer-
mediated communication. Berlin: Mouton de Gruyter.
Hodgins, H. S., Liebeskind, E., & Schwartz, W. (1996). Getting out of hot
water: Facework in social predicaments. Journal of Personality and Social
Psychology, 71(2), 300–314.
Holmes, J. (1990). Apologies in New Zealand English. Language in Society,
19(2), 155–199.
Ilyas, S., & Khushi, Q. (2012). Facebook status update: A speech act analysis.
Academic Research International, 3(2), 500–507.
Jeon, L., & Mauney, S. (2014). “As much as I love you, I’ll never get you to under-
stand”: Political discourse and ‘Face’Work on Facebook. Proceedings of the 22nd
Annual Symposium about Language and Society-Austin, 67–75.
Keat, L. H. (2012). Marketing politicians on Facebook: An examination of the
Singapore general election 2011. Studies in Business and Economics, 7(1),
101–109.
Keehn, J. M. (2013). From face-work to Facebook: Revisiting self-presentation
in the age of connectivism. International Journal of Instructional Technology
and Distance Learning, 10(1), 3–12.
Kozinets, R. V. (2010). Netnography: Doing ethnographic research online. London:
Sage Publications.
Lim, S. S., Vadrevu, S., Chan, Y. H., & Basnyat, I. (2012). Facework on
Facebook: The online publicness of juvenile delinquents and youths-at-risk.
Journal of Broadcasting & Electronic Media, 56(3), 346–361.
Maíz-Arévalo, C., & Sánchez-Moya, A. (2017). “I know how you feel”:
Multifaceted insights into the expression of support strategies in computer-
mediated-communication. In C. Vargas Sierra (Ed.), AESLA2016 EPiC series
in language and linguistics (Vol. 2, pp. 214–223). Alicante: Universidad de
Alicante.
Mann, C., & Stewart, F. (2000). Internet communication and qualitative research.
London: Sage Publishers.
McKay, D. (2010). On the face of Facebook: Historical images and personhood
in Filipino social networking. History and Anthropology, 21(4), 479–498.
Olshtain, E. (1989). Apologies across languages. In S. Blum-Kulka, J. House, &
G. Kasper (Eds.), Cross-cultural pragmatics: Requests and apologies
(pp. 155–174). Norwood, NJ: Ablex.
Penman, R. (1990). Facework & politeness: Multiple goals in courtroom dis-
course. Journal of Language and Social Psychology, 9(1–2), 15–38.
Petronio, S. (1984). Communication strategies to reduce embarrassment differ-
ences between men and women. Western Journal of Communication, 48(1),
28–38.
Sauter, T. (2013). “What’s on your mind?” Writing on Facebook as a tool for
self-formation. New Media & Society, 16(5), 823–839.
Selwyn, N. (2009). Faceworking: Exploring students’ education-related use of
Facebook. Learning, Media and Technology, 34(2), 157–174.
Tannen, D., & Trester, A. M. (Eds.). (2013). Discourse 2.0: Language and new
media. Georgetown: Georgetown University Press.
Ting-Toomey, S. (1994). The challenge of facework: Cross-cultural and interper-
sonal issues. Albany: SUNY Press.
Ting-Toomey, S. (2005). The matrix of face: An updated face-negotiation the-

ory. In W. Gudykunst (Ed.), Theorizing about intercultural communication
(pp. 71–92). London: Sage.
Tracy, K. (1990). The many faces of facework. In H. Giles & W. P. Robinson
(Eds.), Handbook of language and social psychology (pp. 209–226). Oxford:
John Wiley & Sons.
Turnbull, W., & Saxton, K. L. (1997). Modal expressions as facework in refusals
to comply with requests: I think I should say ‘no’ right now. Journal of
Pragmatics, 27(2), 145–181.
West, L., & Trester, A. M. (2013). Facework on Facebook: Conversations on
social media. In D. Tannen & A. M. Trester (Eds.), Discourse 2.0.: Language
and new media (pp. 133–154). Georgetown: Georgetown University Press.
Wood, L. A., & Kroger, R. O. (1994). The analysis of facework in discourse
review and proposal. Journal of Language and Social Psychology, 13(3),
248–277.
Part V
Language and Media Ideologies
11
Sexting and Hegemonic Masculinity:
Interrogating Male Sexual Agency,
Empowerment and Dominant Gendered
Norms
Antonio García-Gómez
11.1 Introduction
As Martinez-Prather and Vandiver (2014) point out, digital technologies
mediate most aspects of young people’s social interactions and friend-
ships. Among the different tools available, this “born digital” generation
of youth seems to be certainly captivated and seduced by social network-
ing sites. As a result, researchers from various fields and disciplines, includ-
ing social psychology, critical media studies and sociology, have been
trying to ascertain the types of actions teenagers engage in online (boyd,
2008; Livingstone, Bober, & Helsper, 2005; Ringrose & Eriksson Barajas,
2011; among many others). An increasingly common behaviour that has
been of particular interest to researchers is the electronic swapping of sex-
ually provocative images and/or texts—commonly known as “sexting”
(Eleftheriou-Smith, 2015; Franks, 2015). This current sexual practice is
A. García-Gómez (*)
Departamento de Filología Moderna, University of Alcalá,
Alcalá de Henares, Spain
e-mail: antonio.garciag@uah.es

https://doi.org/10.1007/978-3-319-92663-6_11
314 A. García-Gómez
gaining popularity among adults (Lee, Crofts, Salter, Milivojevic, &

McGovern, 2013), young adolescents (Davidson, 2014; Schloms-
Madlener, 2013) and, in particular, among female adolescents (Dake,
Price, Maziarz, & Ward, 2012).
Owing to the pervasiveness of sexting (Livingstone, 2008), recent lead-
ing-edge studies have concentrated on (a) making sense of the discourse of
sexting (Thurlow, 2014); (b) determining its prevalence rates (Lounsbury,
Mitchell, & Finkelhor, 2011); (c) analysing its dangers and risks (Crofts
& Lee, 2013); and (d) regulating and criminalising the production, pos-
session and sharing of this material without consent (Albury & Crawford,
2012; Lee et al., 2013). These studies attempt to account for a set of inter-
related problems that may arise out of minors (a) becoming victims of
online predators: (b) having sexually explicit images of themselves dis-
seminated without their consent; (c) disseminating sexually explicit pic-
tures of themselves or of another minor; and (d) being in possession of
these pictures with or without having consented to receive them.
In addition to these studies, there is also a growing body of feminist
research that has mainly focused on Postfeminist (digital) cultures
(Dobson, 2015) and, more precisely, on a specific form of “empower-
ment”: women’s sexual agency (Ringrose, Gill, Livingstone, & Harvey,
2013). These studies have purported, on the one hand, to identify how
social media facilitate the sexual objectification of girls (Dobson, 2014;
Dobson & Ringrose, 2016; Ringrose et al., 2013) and, on the other, to
analyse how teen girls both perform and negotiate their sexual identity
(García-Gómez, 2017). Although sexting has been tackled from very dif-
ferent angles, scholars seem to coincide in reconsidering the way it is
currently approached (Dobson, 2015) and propose a re-thinking of the
way(s) in which social media facilitate the sexual objectification of girls
(Gill, 2007; Gonick et al., 2009; Hasinoff, 2014).
In spite of all these efforts, little attention has been paid to young men’s
understanding of sexting in general (Walker, Lena, & Temple Smith, 2013)
and, to my knowledge, there is no research on what the self-representations
of young men can tell us about the gendered discourses of youth sexuali-
ties. Using guided discussions and personal interviews with 27 British
young men (aged 18 to 21) as my data, the overall aim of this study is to
gain insights into the characteristics of sexualised youth c yberculture. In
Sexting and Hegemonic Masculinity: Interrogating Male… 315
particular, this study aims to explore the construction of multiple mascu-

line identities (Courtenay, 2000) within the context of sexting.
In what follows, these preliminary considerations will be given careful
thought. Section 11.2 presents an overview of the concepts and relevant
theoretical issues that aim to contextualise my research objectives. Section
11.3 deals with data collection and the coding process. Section 11.4
delves into the different sexual gendered discourses young men live out
when they narrate why they and their girlfriends sext. Finally, Sect. 11.5
ends by considering the main implications from the analysis.
11.2 T
he Construction of the Self
and Multiple (Masculine) Identities
In order to delve into the role that sexting plays in the construction of
young men’s masculine identities, the present study focuses on the role of
language in the construction of the self (Shotter & Gergen, 1989). By
exploring how gender is reproduced and performed in this type of cur-
rent sexual practice, this study emphasises not only the fact that language
does play a role in the construction of one’s own and others’ identity,1 but
also sheds light on how specific linguistic strategies impact young men’s
constructions of the self.2
Given that identity is constructed and negotiated in communication,
the construction of the self is therefore multi-faceted (Markus & Wurf,
1987); that is to say, both internal states and external circumstances
determine the specific information that becomes operational and domi-
nates a person’s sense of self (i.e. working self-concept) at a particular
point in time (Solomon & Theiss, 2012). In addition, the activation of a
specific set of features correlates with a particular self-attribution3 process
that allows people to “interpret people’s sense of what, who or where they
belong to” (Versluys, 2007, p. 90).
Although there are a number of studies that attempt to define the dif-
ferent forms of self (Markus & Kitayama, 1991), the analysis below of
young men’s self-representation is based on Hecht (1993) and Hecht,
Warren, Jung, and Krieger’s (2005) studies on the construction of self and
identity. The present study is therefore premised on the fact that there are
three facets or layers of the self: the personal layer of the self that is defined
by those perceptions we have about ourselves that we communicate to

others and make the self different from all others (e.g. this layer focuses
on the self-knowledge one person displays to others when they commu-
nicate); the relational layer of the self that is defined by the dyadic relation-
ship that assimilates the self to significant others (e.g. this layer includes
the characteristics of the self that are related to maintaining social rela-
tionships with other people); and the communal layer of the self that cap-
tures how one is embedded in their group characteristics which makes it
possible to differentiate “us” versus “them” (e.g. this layer captures those
characteristics of the self that allow people to categorise themselves and
others as belonging to a specific group).
Here I suggest that each layer or form of self is discursively constructed
by attaching a set of characteristics that makes it possible to differentiate
oneself from another (i.e. self-attribution process) and that helps people
make sense of their own experiences (Higgins, 1996) and find out who they
are (Hogg & Vaughan, 2002). As Hogg and Vaughan (2002, p. 122) rightly
note, this knowledge of identity “regulates and structures human interac-
tion; and in turn, interactive and societal structures provide identities for
us”. This ultimately suggests the dynamic nature of the construction of self.
In accordance with this, the multiplicity of self is inherent to the con-
struction of one’s identity (Davies & Harré, 1990; Giddens, 1991). As
most scholars agree, identity is not monolithic, but a continuum of “in-
groupness” and “out-groupness” in which one can take on different posi-
tions (Duszak, 2002). As Davies and Harré (1990, p. 49) argue, the
positions “created for oneself and the other are not part of a linear non-
contradictory autobiography” and, as such, they can be modified in the
processes of social interaction. Identity is, therefore, “more than just the
representation of some ‘real world’ group membership, it is its very consti-
tution” (Versluys, 2007, p. 94). In other words, identities are “established”
and “communicatively produced” (Kroskrity, 1999, p. 112).
Even though there is a wide range of psychological, sociological and criti-
cal media studies that focus on multiple identities, no work has been
grounded in first-hand accounts of how young men construct and negoti-
ate their sexual gendered identities within the context of sexting. The
absence of research seems to be the result of the legal and ethical concerns
about sexting and child pornography.4 Therefore, research has, on the one
hand, exclusively focused on the construction of young girls as experiencing
both a social crisis and an individual crisis-event (Peter & Valkenburg,

2009) and, on the other hand, placed an excessive burden on young women
and girls who are constantly asked to tackle the negative effects of early
sexualisation (Dake et al., 2012; Livingstone, 2008; Lounsbury et al., 2011;
Schloms-Madlener, 2013). By putting the blame on girls for their sexualisa-
tion, as Hasinoff (2015) suggests, the role of men has been obscured as
attention has been drawn away from the fact that men are not asked to
tackle the negative effects of early sexualisation (Dobson, 2015; Valkenburg
& Peter, 2011).
It can then be argued that there is a considerable need for research in
this field in order to make visible the role of young men and their percep-
tions of this behaviour (Baumgartner, Valkenburg, & Peter, 2010). In
this line of reasoning, the present paper focuses exclusively on hetero-
sexual young men sending sexts to their significant others as an attempt
to throw light on which linguistic strategies they use in order to con-
struct, reproduce and perform their (sexual) multiple masculine identi-
ties (Hearn & Connell, 2005).
11.3 Data and Methods

The data I analyse are drawn from a larger study I have conducted as part
of a conflict resolution programme in four different secondary schools in
northern England (García-Gómez, 2014). Since the study explores the
construction of multiple masculine identities within the context of sexting,
the main focus is on male teens who have consensually participated in
sexting. In order to provide a homogenous sample, the data selected come
from a specific subsection of the male population: British heterosexual
young men. All males range from 18 to 21 years of age and are predomi-
nantly white lower-middle class with low socio-economic status. Therefore,
four main dimensions of identity which affect the way that the gendered
selves are presented were considered: gender, age, race and social class.
In this paper, I used quantitative and qualitative methods, based on the
rigorous discourse analysis of guided discussions and personal interviews
(Steinar, 1996) with young men who, on the one hand, have willingly
consented to sexting with their significant others and, on the other hand,
comment on the social and personal consequences of sexting.
11.3.1 Data Collection
As part of a previous study, 68 young women—aged 15 to 17—were

asked to complete a questionnaire on their online habits, 45 of whom
admitted to sexting in the prior year. Once they were already taking part
in the conflict resolution programme and their parents had signed the
informed consent form for the research, they were encouraged to invite
their significant others to take part in one of the guided discussions.
Although some of them were initially reluctant, 31 young men attended
one guided discussion and they were subsequently invited to take part in
a parallel project in order to find out more about young men’s online
habits. In the end, 27 young men—aged 18 to 21—agreed to take part
in the project and signed the informed consent form.
The data consist of a 20-minute personal interview with each partici-
pant. After transcribing and carrying out a content-analysis of the inter-
views (Creswell, 2009), guided discussions involving four or five young
men were also held in order to debate about (the risks of ) sexting—each
lasting between 50 minutes and a hour. Initially, these guided discussions
were aimed at creating an appropriate forum for the discussion and
exchange of ideas, but they also made it possible to identify possible vari-
ations in the participants’ forms of self when being individually inter-
viewed and when taking part in a group discussion.
11.3.2 T
he Coding Process: Speech Act Variables
Considered
Personal interviews and guided discussions were recorded and tran-

scribed. The resulting data consisted of 12,317 utterances and a total of
16 hours and 43 minutes of recorded material. Firstly, a preliminary over-
view of all these utterances was done in order to segment them into topi-
cally defined passages: (a) description of their online habits; (b)
acknowledgment of having agreed to exchange sexually provocative pic-
tures and/or texts and how often they have done so and with whom; (c)
motivations behind the act of sexting; (d) analysis of the effects of sexting
on their lives and its implications.
These topically defined passages were understood as Macro Speech Acts

(Van Dijk, 1977) that allow the analyst to identify the specific set of charac-
teristics the participants attributed to them (i.e. self-attribution process) in
an attempt to unveil how the participants constructed the different forms of
self: the personal, the relational and the communal layers of the self. Furthermore,
the self-attribution process identified in the construction of each layer
reveals not only its discursive realisation, but also its purpose. Inspection of
these constructions made it possible to identify two main discursive func-
tions of utterances: eliciting and informative acts (Tsui, 1994).
The personal layer of the self, on the one hand, is mainly built up by
means of informative acts that allow the participants to explore who they
are, how they behave when sexting and why. On the other hand, the con-
struction of this form of self revolves around evaluating one’s own behav-
iour positively and evaluating others accordingly. The self-attribution
processes, in addition, identified in the construction of the relational and
communal layers of the self are mainly built up by means of eliciting and
informative acts that allow the participants to activate a specific set of
characteristics that regulate their relations not only to their significant
others, but also to other young men. In particular, the exploitation of
informative acts allows the participants to evaluate others’ behaviour in a
positive or negative manner according to whether or not they live up to
their own expectations, whereas eliciting acts are mainly used to confirm
their own beliefs, create a bond with other young men and develop a
sense of belonging. It is worth noting that a more detailed classification
with distinctive features will be offered as the analysis unfolds.
11.4 Constructing, Reproducing

and Performing Multiple Male Identities
When explaining why the participants in the study and their girlfriends
sext, they represent themselves by employing a distinctive self-attribution
process. In light of this, the exploration of these self-attribution processes
will not only illustrate how sexting has an impact on the way(s) the par-
ticipants perform their masculinities, but it will also make it possible to
delineate a possible interactive relation among the three layers of the self
(i.e. personal, relational and communal). In particular, I would argue that

one layer of the self predominates as the basis for their self-representation
and serves to construct, reproduce and perform a specific masculine iden-
tity within the context of sexting.
11.4.1 G
uided Discussions: Enacting Hegemonic
Masculinity
In general terms, the participants’ narratives in the guided discussions

revolve around two main issues: sexting is a common sexual practice for
them, on the one hand, and exposing themselves sexually is socially desir-
able, on the other. This shows how this digitally mediated sexual practice
needs to be understood as an attempt to inhabit “legitimate” subject posi-
tions (Butler, 1993). The participants in the study try to construct their
male identities by fulfilling norms of traditional gender recognition
(Skeggs & Wood, 2012). More specifically, the discourse analysis of their
narratives reveals that they construct their masculine identities by focus-
ing on the communal layer of the self. It is possible to argue that the activa-
tion of this particular layer is premised on the concept of hegemonic
masculinity (Connell, 2005) or the differences in gendered power (Coles,
2009); that is to say, on “the configuration of gender practice which […]
guarantees […] the dominant position of men and the subordination of
women” (Connell, 2005, p. 77).
As the analysis will show, sexting works in relation to the performance
of certain gendered masculine scripts (i.e. representation of idealised bod-
ies and ideal hegemonic masculinities) that allows the participants to dis-
cursively construct their identity, negotiate their identity, and therefore,
perform their masculinities. In particular, the construction of the com-
munal layer of the self allows the participants to protect the positivity of
their identity as well as support a favourable and assertive masculinity.
Here I argue that this layer of the self is based on the validation of hege-
monic masculinities and the creation of male-group distinctiveness.
Furthermore, the self-attribution process involved in the communal
layer of the self contributes to, first, creating a positive sense of in-group
membership (i.e. “we men do these things”) and, second, differentiating
their and other men’s behaviour from out-group members’—namely
Table 11.1 The communal layer of the self: enacting hegemonic masculinity
Frequency Ratio
Informative acts
(In-)direct positive self-evaluation of physical characteristics 1487 12%
(In-)direct positive appraisal of own sexual performance 1814 14.7%
Expressions of self-praise 753 6.1%
Exaggerations of male values 326 2.6%
(In-)direct positive appraisal of sexting 845 6.8%
(In-)direct negative/positive appraisal of women’s sexual 748 6%
behaviour
(In-)direct negative appraisal of adults’ attitude towards sex 311 2.5%
Total 6284
Eliciting acts
Search for agreement by confirming one’s own point of 434 3.5%
view
Evaluation of speaker’s and/or addressee’s point of view 989 8%
Commitment to a course of action for the speaker’s benefit 245 1.9%
Elicitation to validate one’s own point of view 387 3.1%
Total 2055
their significant others, women in general, and/or adults. Linguistically

speaking, the enactment of this hegemonic masculinity is done through
informative and eliciting acts. Table 11.1 shows the linguistic realisations
and frequency of the pragmatic meaning of each utterance:
As Table 11.1 shows, the construction of the communal layer of the self is
mainly built up by means of informative and eliciting acts that may be
classified into four main groups in terms of the function they fulfil:
1. Informative acts which self-evaluate the young man positively in a

direct or indirect way comprise utterances that validate their mascu-
linity (e.g. “I’m a real man”) by describing their sex life (e.g. “I drive
girls absolutely crazy. I know what women want in a man”). These acts
also include expressions of self-praise (e.g. “I get a hard-on easily. I’m
fucking amazing”) as well as exaggerations of male value (e.g. “Men
always know how to please a woman”);
2. Informative acts which are aimed at justifying socially unacceptable
behaviour (e.g. “Having good sex is not bad”). This category com-
prises utterances that generalise the act of sexting (e.g. “Everyone I
know has sent pics like this some time”) or criticise those who do not
do it (e.g. “Sex is a natural urge but adults are sexually repressed”).
3. Informative acts which are aimed at evaluating their girlfriends positively

in a direct or indirect way comprise utterances that validate their sexy
femininity in terms of their ability to show their submission and satisfy
their boyfriends’ sexual needs and desires (e.g. “She ain’t scared of getting
her knees dirty”, “She knows how to make me horny”). These positive
evaluations contrast with the ones of women who do not do so (e.g.
“These other girls are useless”). These acts also include negative appraisals
of adults’ attitude towards sex and their lack of understanding (e.g. “Adults
don’t know what sex is”, “My parents simply don’t understand”); and
4. Eliciting acts which invite the other interlocutors to agree with the
speaker and/or validate their own point of view (e.g. “We all do it.
Don’t you think so?”). These utterances attempt to play down their
sexual behaviour by normalising these practices.
Therefore, the self-attribution process behind the enactment of their

hegemonic masculinity identified in the construction of the communal
layer of the self not only allows them to create a sense of in-groupness, but
it also regulates the way they construct their own gendered sexual identity
(i.e. the personal layer of the self) and the way they relate to others (i.e. the
relational layer of the self). Consider the following excerpts from the data
where two young men explain what sexting means to them:
Excerpt 1 Participant 1: A 19 year old young man
Texting steamy messages to my girl is just, you know, a way of turning

myself on […] it’s some sort of sexual game, you’re in your room and
you’re really horny and you text your girlfriend saying how hot you are
and then she starts saying dirty things to you, things she knows that really
turn me on and there you go […] We can’t see each other as much as we
want and you know it’s like we boys masturbate a lot and we need some
sort of erm you know […]
Men my age don’t watch porn. I mean I don’t need to masturbate watch-
ing porn. It’s like watching another guy doing what you’d like to do, I
prefer doing my own stuff […] you know I’m always up for an adventure
((laughing)) and she knows I won’t take ‘no’ for an answer ((laughing))
you know erm I know how to turn a woman wild in bed […]
These two excerpts show how these participants’ discursive position-

ing has to do with a traditional patriarchal discourse that revolves around
the centrality of male sexuality to define men’s identity and enact their
hegemonic masculinity (Connell, 2005). Interestingly enough, most of
the participants agree on referring to sexting as a replacement for watch-
ing porn where they become agents and are able to experience their own
sexual fantasies. Furthermore, the participants’ conceptualisation of
sexual agency endorses the most common gender stereotypes and sexist
behaviours; that is to say, these young men portray themselves as having
a constant desire to have sex (e.g. “I’m always up for an adventure”) and
to be sexually irresistible (e.g. “she knows I won’t take ‘no’ for an answer”).
This, in turn, allows them to prove (and boast about) their virility (e.g.
“I know how to turn a woman wild in bed”). These examples reveal that
the sexual activity with young women and the hypersexualised perfor-
mance of masculinity allow them to construct and maintain high-status
masculinity.
It can be argued that the participants use discourse to demonstrate
masculinity and enact hegemonic masculinity to reaffirm their masculine
identity. Furthermore, I suggest that the predominance of the communal
layer of the self determines the personal layer insofar as they prove the valid-
ity of this layer by crediting themselves with being sexually active and
skilful, and characterising themselves as men who make their own choices
about sexual embodiment in order to please themselves (e.g. “Texting
steamy messages to my girl is just, you know, a way of turning myself on
[…]”). In other words, the perceptions they have about themselves seem
to be filtered through the qualities they ascribe to themselves by fitting
into the group (i.e. “what real men do”). Consider the following excerpts:
Excerpt 3 Participants 4 and 5: A 20 year old and a 19 year old man
P4: You know I’m in my room and I start playing with erm you know
and then I text her to see if she’s alone and
P5: Yeah I
P4: Then I ask her what she’s wearing and I start saying and start saying
‘I love you’, ‘I’m really lucky to have you’, ‘You make me so happy”
any of this rubbish ((laughing))
P5: Yeah I do that too ((laughing out loud))
P4: Then she melts and erm she can’t say no
P5: They all love that kind of romantic stuff
P4: Then I erm I bombard her with pics of mine and that’s basically it.
P5: Yeah they’re pathetic, aren’t they ((laughing))
Excerpt 4 Participants 7 and 26: two 20 year old men
P7: When I start dating a girl, I always have to teach how I like things
done […] My girlfriend knows I text her every time I’m horny
((laughing))
P26: Every day every minute ((both laugh))
P7: Every minute ((both laugh))
P26: My girlfriend says she feels embarrassed because she doesn’t like
her body and I always say to her, do not be embarrassed, every-
body has a body and there is no shame in showing it. I like my
body and I know how to have fun with it.
As Excerpts 3 and 4 show, the participants start speaking about their sex
life and what they think about sex and their interlocutors intensify sympa-
thy with them (e.g. “Yeah I do that too”). In doing so, it can be argued that
the construction of their male identity is premised on emphasising group
membership and solidarity. More specifically, they construct their mascu-
line identity in terms of their active sexuality (e.g. “I bombard her with
pics of mine”, “My girlfriend knows I text her every time I’m horny”).
This, in turn, stresses common membership in the group as they define
themselves as “sex addicts” who like their bodies and enjoy their sex life
(e.g. “do not be embarrassed, everybody has a body and there is no shame
in showing it. I like my body and I know how to have fun with it”).
Furthermore, participants 5 and 26 claim common ground with partici-
pants 4 and 7 by indicating they all do the same and share the same opinion
about women (e.g. “They all love that kind of romantic stuff, “Yeah they
are pathetic, aren’t they”). In discussing women this way, the participants
reinforce their in-group membership as an attempt to legitimise their supe-

rior masculine subject position. Therefore, it can be argued that these
online performative practices reveal the participants’ desire to fulfil norms
of idealised gendered recognition (e.g. men are superior to women).
In addition, inspection of the data shows that their self-presentation as
sexually agentive correlates with the treatment of their girlfriends as a
commodity or an object of male sexual desire. By demonstrating their
hegemonic masculinity and reaffirming their identity this way, I here
suggest that the predominance of the communal layer of the self also deter-
mines the relational layer of the self. As examples will illustrate, the partici-
pants’ narrations of their online sexual activities can not only be understood
as a digital performance of masculinity, but also it can be argued that the
practice of sexting includes the perception of themselves as well as the
way(s) they relate to their girlfriends, women in general, and adults.
Sexting, therefore, seems to signify high-status masculinity and serves as a
means of gaining recognition and value (Skeggs & Wood, 2012) insofar
as they meet the group membership requirements (i.e. “what real men
do”). Consider the following excerpts:
My girlfriend had never done it before, but there’s always a first time for
everything ((laughing)). She used to say that she didn’t like her body, but
we’re are in a relationship erm I started sending her a picture of my erm
you know and them some more and in the end she also started sending me
some […] She knows I’m erm sexually active and she sometimes complains
because my cock takes forever to come, but I genuinely enjoy myself […]
I’m pretty sure I’m a sex addict. I can have sex 5 or 6 times a day ((laugh-
ing)) but I sometimes cannot see her or we can’t find a place to have sex
and there you go erm I asked her to send me pics and erm you know […]
I know I put pressure on her but she knows I’m a man and I’ve got my
needs. She says I only want her for sex, but she’s wrong. I love her, but
we’re in a relationship so she must be prepared to give me what I want by
taking the title of ‘girlfriend’ […]
As these two excerpts show, participants 8 and 10 express a positive

attitude towards themselves by self-presenting as men in love (e.g. “I love
her”) who are in a committed relationship (e.g. “We are in a relation-
ship”). By the mere fact of being girlfriend and boyfriend (e.g. “we’re in a
relationship so she must be prepared to give me what I want by taking the
title of girlfriend”), their girlfriends seem to be forced to sext in order to
satisfy their sexual needs (e.g. “I asked her to send me pics and erm you
know […] I know I put pressure on her but she knows I’m a man”). In
this view, they seem to have accepted the idea of women subordinating
their own needs to men’s satisfaction as a natural component of their
relationship in spite of their girlfriends’ possible reluctance (e.g. “My girl-
friend had never done it before, but there’s always a first time for every-
thing”). In doing so, they devalue their partners by treating them as
sexual objects whose only aim is to keep them satisfied. Consider the
following excerpt:
Excerpt 7 Participants 6 and 22: A 20 year old and a 18 year old young man
P22: How many women have I slept with? Erm I don’t really know. I
started dating girls when I was 12 and I’ll turn 19 next month.
P6: I’ve never questioned my own number of sexual partners either.
We’re young. You met someone you like and there you go ((laugh-
ing)) […] I don’t talk about sex with my parents. I guess they
think I’m still a virgin ((laughing))
P22: Yeah they are old and they can’t understand. Many parents would
die if only they knew their daughters are sleeping around erm too
much, most of them are complete whores
P6: Now girls want to take charge in the bedroom you know and they
want you to get rough in bed. They think they know everything
because they get on their knees easily but then you ask them to
send you a pic and they panic ((laughing))
As Excerpt 7 shows, the enactment of hegemonic masculinity exacer-

bates gender inequality and manifests the sexual double standards behind
sexting since sexually active women in general are evaluated negatively (e.g.
“Many parents would die if only they knew their daughters are sleeping
around erm too much”). This time the fact that they are sexually active
women does not validate their sexy femininity as before, but they are deni-
grated and treated as sluts (e.g. “most of them are complete whores”).
As mentioned above, the predominance of the communal layer of the
self determines the relational layer and encompasses the way(s) they relate
to adults as well. In particular, the participants report that they sext in
order to express themselves sexually and experience the electronic swap-
ping of sexually provocative images and texts as the way young people live
their sex life in full—usually in contrast to adults. Consider the following
excerpts:
Yeah, you know, I guess this is the way we now flirt with girls and do our
things. My parents ((laughing)) don’t have a fucking clue what I’m doing.
They will not understand. They are erm you know, let’s say, conservative,
But things have changed much and now we are not afraid of doing these
things and erm […]
My girlfriend and I erm we send these kinda pictures every now and then.
We are bored at home and erm we we text each other. It’s kinda funny. I
say something dirty and she answers back. She knows I like it and I guess
she likes it too. We are showing our love I guess […] I have never told my
friends but I’m sure they do it. We all do it. We are young at the end of
the day […] parents have no opinion as they know nothing ((laughing))
my mum would have a heart attack if only she knew […]
As can be seen in Excerpts (8) and (9), the analysis of the self-attribution
process shows how these males and their girlfriends seem to have found
in this form of sexual behaviour their own code of communication (e.g.
“we send these kinda pictures every now and then”). Interestingly enough,
the fact of sending these types of pictures seems to define the group under
study as one who show their love by texting sexual images and provocative
messages as a natural part of their relationship (e.g. “We are showing our
love, I guess”); however, these young men do not talk openly about it
with either their parents, as they would never understand (e.g. “my mum
would have a heart attack if only she knew”), or with their own friends
(e.g. “I have never told my friends but I’m sure they do it”). In other
words, these young men claim common ground with other male friends
by indicating sexting is a natural/common thing, but, in spite of the
belief that it is common, they keep this practice to themselves.
11.4.2 P
ersonal Interviews: Attempts to Maintain
a Masculine Identity
When taking part in the personal interviews, the participants’ construc-

tion of their masculinity moves away from the communal layer of the self
to focusing on the personal layer of the self. The discourse analysis of these
narratives unveils, on the one hand, the existence of competing interwo-
ven narratives that range from agency, through forced agency to passivity
and, on the other hand, highlights that there is not one single pattern,
but appropriate, context-dependent strategies that are implemented to
perform and validate hegemonic masculinity (Noone & Stephens, 2008,
p. 713). This fact supports the view that sexting is connected with meet-
ing the group membership requirements (i.e. “what real men do”).
However, the participants’ attempts to maintain their high-status mascu-
linity and their search for recognition and value are not without contra-
diction. This claim makes it possible to suggest that this digital
hypersexualised performance of masculinity draws attention to a process
of mediating gendered norms of recognition and value. As examples will
illustrate, their narratives reveal the struggle between their frustration and
their lack of confidence in their ability to meet their girlfriends’ sexual
demands which seems to place them in “marginalised positions” (Connell,
2005) and their attempts to rely on aspects of dominant masculinity to
reaffirm their masculine identities and assume a dominant position
(Coles, 2009, p. 33).
In spite of young women having an active sexual agency—which is
interpreted as an essentially masculine territory—the fact that the personal
layer of the self predominates as the basis for the participants’ self-
presentation allows them to search for a dominant position and validate
themselves as men. Here, I argue that the self-attribution process involved

in this particular type of presentation of the self is an attempt to maintain
a masculine identity. This, in turn, suggests that these online sexual prac-
tices put additional demands on the participants that have to explore and
identify a set of personal characteristics that allow them to perform as
strong, virile and sexually active men. In order to maintain their high-
status masculinity and inhabit a “legitimate” subject position they attempt
to down play potential threats to their masculine positions and allow
them to incorporate values commonly related to hegemonic masculinity.
Linguistically speaking, the search for maintaining high-status masculin-
ity and the redefinition of hegemonic masculine identities are done
through informative acts. Table 11.2 shows the linguistic realisations and
frequency of the pragmatic meaning of each utterance:
As Table 11.2 shows, the construction of the personal layer of the self is
mainly built up by means of informative acts that may be classified into
three main groups in terms of the function they fulfil:
1. informatives which self-evaluate the participants positively in a direct

or indirect way. These comprise utterances that affirm their full sexual
potential and validate their masculinity (e.g. “Every morning when I
wake up, I’m very horny and ready to have sex”);
2. informatives which self-evaluate the young man negatively in a direct
or indirect way. These comprise utterances that express the young
man’s insecurities by the unsuccessful attempts to meet and fulfil his
girlfriend’s sexual desires (e.g. “I don’t like when she makes funny
comments about my, you know”);
3. informatives which are aimed at evaluating their girlfriends negatively
in a direct or indirect way. These comprise utterances that criticise
these girls’ sexual self-awareness and sexual liberation (e.g. “She’s not
normal, you know, she is erm she is sexually insatiable”).
Table 11.2 The personal layer of self: attempts to maintain a masculine identity
Informative acts Frequency Ratio
(In-)direct positive appraisal of own sexual performance 1294 10.5%
Expressions of self-praise 836 6.7%
Expressions of insecurity 861 6.9%
(In-)direct negative appraisal of women’s sexual behaviour 987 8%
Total 3978
The discourse analysis of the construction of the personal layer of the self
highlights the fact that young women’s active sexual agency seems to con-
tradict the participants’ traditional gender roles and gendered sexual
scripts. Here I suggest that the notion of sexual agency is gaining a new
dimension for both genders insofar as sexting is not exclusively a particular
sexual practice, but also it reveals that showing an increased sexual agency
becomes a mechanism for gaining recognition and value, and therefore,
inhabit a “legitimate” subject position. Consider the following excerpts
from the data where two men explain how they feel about their sexually
insatiable girlfriends who want them to sext for hours when they are alone
at home:
Yeah I am a man and I have my erm needs […] I’ve been dating about
twenty girls over the last two years or so and I’ve had sex with all them.
Having sex is very important in a relationship, well I make it important,
you know what I mean ((laughing)) […] My girl is really hot, she knows
how to turn you on in a sec, wow, she says and does things that erm well
you know what I mean […] she says things that erm make me think wow
this girl is a filthy bitch ((laughing)) you know it is usually really hard to
text one-handed, you know what I mean ((laughing))
She knows I’m the fucking man. She knows pretty well ((laughing)) I can
have sex 6 times in a row, but she sometimes goes over the top […] She
sometimes hits my phone up because she’s alone at home and she’s erm
you know she wants to play a little, but then there is no stopping, she
wants more and more and four hours later I’m exhausted and she still
wants more and complains and starts saying things about me not being a
real man. I know she’s joking but you know it hurts […] I sometimes
think she’s sick. I know this sounds wrong, but she is not normal. Not a
single girl I know behaves you know like erm she does. My friends say
she’s fucking insane or she’s just a whore.
These two excerpts illustrate how sexting underscores new processes of

mediating gendered recognition, which must be negotiated (Skeggs &
Wood, 2012). In particular, the narratives of participants 3 and 18 focus
on dominant aspects of masculinity (i.e. traditional gender scripts) which
they felt prescribed; that is, men should take the initiative as they are the
ones who always have an insatiable sexual desire. Interestingly, active
sexual agency is here understood as a practice of legitimisation around
the construction of identity (Butler, 1993). Therefore, the self-attribution
process identified in the construction of the personal layer of the self not
only allows them to construct a gendered sexual identity that validates
them as men—albeit within marginalised positions—but also regulates
the way they actually relate to their significant others (i.e. the relational
layer of the self) once the in-group pressure is relieved (Shariff, 2015).
Excerpts 10 and 11 illustrate how participants 3 and 18 find them-
selves in marginalised positions because their partners are the ones who
are always ready to have sex and they are the ones who give excuses since
they are unable to meet their sexual needs. That is to say, their partners’
active sexual agency challenges the participants’ attempt to maintain
high-status masculinity. This idea is reinforced by the fact that they admit
their frustration openly (e.g. “I am kinda stressed sometimes but […] I
know she left one of my friends because he couldn’t give her what she
wanted”). This clearly contrasts with their attempts to validate themselves
as men by self-praising (e.g. “Having sex is very important in a relation-
ship, well I make it important, you know what I mean ((laughing))”) and
magnifying their virility (e.g. “She knows I’m the fucking man. She
knows pretty well ((laughing)) I can have sex 6 times in a row”). This
attempt to draw on dominant masculinity is reinforced by discrediting
their girlfriends and using in-group membership to emphasise their lack
of “normality” (e.g. “she is not normal”, “My friends says she’s fucking
insane or she’s just a whore”).
All in all, inspection of these competing interwoven narratives reveal
that sexting can be understood as a social practice that plays a key role in
the construction of the participants’ masculinities. The struggle between
men’s taken-for-granted masculinities (Hearn & Connell, 2005) and the
presence of some less conventional discourses (Edwards, 2006) that
include features connected to men’s weaknesses and emotions states
makes it possible to argue that this social practice has an impact on the
construction of the participants’ masculinities and sense of value.
Consider the following excerpts:
You know it’s always been me the one telling girls do this or do that, but
Linda you know it’s funny ‘cos she was so naïve and then we started send-
ing pics and hot texts and all hell broke loose and now she wants it all the
time […] I’m at work and I can’t and then she blocks me for a while then
she asks for more. She’s so childish sometimes, but I love her and give her
what she wants […] I guess she forces me to send her pics and stroke my
you know when I’m not in the mood.
I’ve been going out with this girl, and well I knew she’d done most of my
friends and many other boys, but she is really good in bed […] she knows
what to say and how to do things that erm drive a man crazy. We started
texting each other and I was shocked the first time she sent me a hot mes-
sage, I must admit I was erm let’s say afraid she thought my erm you
know was not big enough and or she didn’t like something of me, but I
answered and then we text each other erm night and day […] It is like she
is horny all the time, always asking how hard I am even when I’m at
work. I am kinda stressed sometimes but […] I know she left one of my
friends because he couldn’t give her what she wanted.
These two excerpts illustrate how participants 6 and 7 found them-

selves in marginalised positions, having to admit their vulnerability (e.g.
“It is like she is horny all the time, always asking how hard I am even
when I’m at work. I am kinda stressed”), and their insecurities (e.g. “I
know she left one of my friends because he couldn’t give her what she
wanted.”). In spite of its popularity, sexting can be said to be a recent
phenomenon and (young) people are still defining the norms of behav-
iour. This fact may justify why the notion of sexual agency is gaining new
dimensions for both genders as men’s narratives show their confusion
about making their own choices about sexual embodiment and their
attempts to magnify aspects of dominant masculinity to protect the posi-

tivity of their masculine identity (e.g. “You know it’s always been me the
one telling girls do this or do that”) coexist with other marginalised posi-
tions which not only reflect pressure and insecurities about their bodies
(e.g. “I must admit I was erm let’s say afraid she thought my erm you
know was not big enough”), but also challenge those roles they felt pre-
scribed for men (e.g. “she forces me to send her pics and stroke my you
know when I’m not in the mood”).
11.5 Conclusions
This paper has attempted to contribute to the literature by including the
voice of young men in the study of sexting. Therefore, this study contrib-
utes to the understanding of gendered discourses of youth sexuality by
grounding in the first-hand accounts of how sexting plays a role in the
construction and negotiation of a group of young men’s sexual gendered
masculine identities. In particular, the analysis has given evidence of
competing interwoven narratives that range from hegemonic masculinity
to marginalised and subordinated forms. By exploring these multiple
masculinities, the paper has suggested that male sexters think they are
making their choices about sexual embodiment and attempt to cast this
sexual practice in a positive light. Furthermore, this study adds an impor-
tant dimension to the literature by focusing not only on the view point
of young men, but also on men’s experience as creators of sexts—rather
than solely as recipients.
However, the discursive construction of their masculine identities
reveals not only the benefits, but also the risks of conforming to and
abiding by the norms of these digital current sexual practices of teenagers.
Furthermore, the study of the ways they construct, reproduce and per-
form their multiple masculine identities sheds further light on Hasinoff’s
(2012, 2015) claims about the overly simplistic treatment of this social
phenomenon and contributes to this debate by showing that the notion
of sexual agency is gaining new dimensions for both genders. More spe-
cifically, the enactment of hegemonic masculinity and the strategies the
participants in the study exploit so as to draw on dominant masculinities
and validate themselves as men not only reveal men’s confusion about
sexual agency and choice, but also bring forward the effects of young
women’s sexually active agency and disinhibition.
Furthermore, the competing interwoven narratives found in the data
have shown the need to analyse the diverse modes of subjectivity; that is,
to analyse the ways in which young men engage with representational
mechanisms to perform the contradictory relations of gendered identi-
ties. In addition to this, it can be argued that sexting is not only a popular
sexual practice, but also an influential mechanism for claiming and gain-
ing social recognition and value that allows young people to inhabit a
“legitimate” subject position.
Natural directions for future research include a further interrogation of
both young men’s and women’s sexual choices by exploring how they
actually negotiate their sexual gendered identity if both partners are
interviewed together. It would be worth exploring how young men talk
about the consequences of sexting further as this will cast light on how
this social practice among teenagers creates the space for new forms of
knowledge and practice.
Acknowledgements The present study was financially supported by a grant

(ID No: FFI2013-47792-C2-2-P) from the Ministerio de Economía y
Competitividad. This article is part of the long-term research project “EMOtion
and language ‘at work’: The discursive emotive/evaluative FUNction in different
texts and context within corporate and institutional work: PROjectPERsuasion”
(EMO-FUNDETT: PROPER).
Notes
1. In accordance with social constructivism, language is here understood as
a discursive action (Harré & Moghaddam, 2003) that is key to under-
standing identity; that is to say, “identity is performatively constituted by
the very ‘expression’ that is said to be its results” (Butler, 1990, pp. 24–25).
2. In the literature it is commonly accepted, on the one hand, that identity
presumes the presence of multiple aspects of the self and, on the other,
that social context plays a key role in triggering different aspects of the self
that are presumed to constitute both our personal and social identity
(Hogg & Vaughan, 2002).
3. Baron and Byrne (1997, p. 152) defines self-attribution as “one’s self-
identity, a schema consisting of an organised collection of beliefs and feel-
ings about one self.”
4. The law has been unable to draw the line between sexting and child porn
insofar as “the laws surrounding the production of child pornography do
not seem to have any provisions for self-produced material or concern for
the age of the producer” (Lunceford, 2011, p. 105).
References
Albury, K., & Crawford, K. (2012). Sexting, consent and young peoples ethics:
Beyond Megan’s story. Continuum: Journal of Media and Cultural Studies,
26(3), 463–473.
Baron, R. A., & Byrne, D. (1997). Social psychology. Boston: Allyn and Bacon.
Baumgartner, S. E., Valkenburg, P. M., & Peter, J. (2010). Unwanted online
sexual solicitation and risky sexual online behavior across the lifespan. Journal
of Applied Developmental Psychology, 31, 439–447.
boyd, D. (2008). None of this is real. In J. Karaganis (Ed.), Structures of partici-
pation in digital culture (pp. 132–157). New York: Social Science Research
Council.
Butler, J. (1990). Gender trouble: Feminism and the subversion of identity. Thinking
gender. New York and London: Routledge.
Butler, J. (1993). Bodies that matter: On the discursive limits of ‘sex’. New York:
Routledge.
Coles, T. (2009). Negotiating the field of masculinity: The production and
reproduction of multiple dominant masculinities. Men and Masculinities,
12(1), 30–44.
Connell, R. W. (2005). Masculinities. Australia: Allen and Unwin.
Courtenay, W. H. (2000). Constructions of masculinity and their influence on
men’s well-being: A theory of gender and health. Social Science and Medicine,
50, 1385–1401.
Creswell, J. W. (2009). Research design: Qualitative, quantitative, and mixed
methods approaches. London: Sage Publications.
Crofts, T., & Lee, M. (2013). ‘Sexting,’children and child pornography. Sydney
Law Review, 35(1), 85–106.
Dake, J. A., Price, J. H., Maziarz, L., & Ward, B. (2012). Prevalence and cor-
relates of sexting behavior in adolescents. American Journal of Sexuality
Education, 7(1), 1–15.
Davidson, J. (2014). Sexting: Gender and teens. Rotterdam: Sense Publishers.

Dobson, A. (2014). Laddishness online: The possible significations and signifi-
cance of ‘performative shamelessness’ for young women in the post-feminist
context. Cultural Studies, 28(1), 142–164.
Dobson, A. (2015). Postfeminist digital cultures: Femininity, social media, and
self-representation. New York: Palgrave Macmillan.
Dobson, A., & Ringrose, J. (2016). Sext education: Pedagogies of sex, gender
and shame in the schoolyard of tagged and exposed. Sex Education: Sexuality,
Society and Learning, 16(1), 8–21.
Duszak, A. (2002). Us and others: An introduction. In A. Duszak (Ed.), Us and
others: Social identities across languages, discourses and cultures (pp. 1–28).
Amsterdam and Philadelphia: John Benjamins.
Edwards, T. (2006). Cultures of masculinity. London: Routledge.
Eleftheriou-Smith, L. (2015). ‘Revenge porn’ criminalised: What is it and what are
the consequences? Retrieved April 5, 2017, from http://www.independent.
co.uk/news/uk/crime/revenge-porn-criminalised-what-is-it-and-what-are-
the-consequences-10042291.html
Franks, M. A. (2015). Drafting an effective ‘Revenge porn’ law: A guide for
legislators. Social Science Research Network. Retrieved April 5, 2017, from
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2468823
García-Gómez, A. (2014). Deconstructing mean girls: Impolite verbal behav-
iours, on/offline self-representations and evaluative beliefs. In A. Sánchez
Macarro & A. B. Cabrejas Peñuelas (Eds.), New insights into gendered discur-
sive practices: Language, gender and identity construction, English in the World
Series (pp. 73–100). Valencia: Universitat de Valencia.
García-Gómez, A. (2017). Teen girls and sexual agency: Exploring the intraper-
sonal and intergroup dimensions of sexting. Media, Culture and Society,
39(3), 391–407.
Giddens, A. (1991). Modernity and self-identity: Self and society in the late modern
age. Stanford, CA: Stanford University Press.
Gill, R. (2007). Post-feminist media culture: Elements of a sensibility. European
Journal of Cultural Studies, 10(2), 147–166.
Gonick, M., Renold, E., Ringrose, J., & Weems, L. (Guest Eds.).(2009).
Rethinking agency and resistance: What comes after girl power? Girlhood
Studies, Special Issue: What comes after Girl power?, 2(2), 1–9.
Harré, R., & Moghaddam, F. (2003). The self and others: Positioning individuals
and groups in personal, political, and cultural contexts. Westport: Praeger.
Hasinoff, A. A. (2012). Sexting as media production: Rethinking social media

and sexuality. New Media & Society, 15(4), 449–465.
Hasinoff, A. A. (2014). Blaming sexualization for sexting. Girlhood Studies, 7(1),
102–120.
Hasinoff, A. A. (2015). Sexting panic: Rethinking criminalization, privacy, and
consent. Champaign: University of Illinois Press.
Hearn, J., & Connell, R. W. (Eds.). (2005). Handbook of studies on men and
masculinities. Thousand Oaks, CA: SAGE.
Hecht, M. L. (1993). A research odyssey: Toward the development of a com-
munication theory of identity. Communication Monographs, 60(1), 76–82.
Hecht, M. L., Warren, J. R., Jung, E., & Krieger, J. L. (2005). The communica-
tion theory of identity: Development, theoretical perspective, and future
directions. In W. B. Gudykunst (Ed.), Theorizing about intercultural commu-
nication (pp. 257–278). Thousand Oaks: Sage Publications.
Higgins, E. T. (1996). The “self digest”: Self-knowledge serving self-regulatory
functions. Journal of Personality and Social Psychology, 71(6), 1062–1083.
Hogg, M. A., & Vaughan, G. M. (2002). Social psychology. London: Prentice
Hall.
Kroskrity, P. V. (1999). Identity. Journal of Linguistic Anthropology, 9(1–2),
111–114.
Lee, M., Crofts, T., Salter, M., Milivojevic, S., & McGovern, A. (2013). ‘Let’s
get sexting’: Risk, power, sex and criminalisation in the moral domain.
International Journal for Crime and Justice, 2, 35–49.
Livingstone, S. (2008). Taking risky opportunities in youthful content creation:
Teenagers’ use of social networking sites for intimacy, privacy and self-
expression. New Media & Society, 10(3), 393–411.
Livingstone, S., Bober, M., & Helsper, E. (2005). Internet literacy among chil-
dren and young people: Findings from the UK children go online project [online].
London: LSE Research Online. Retrieved April 5, 2017, from http://eprints.
lse.ac.uk/archive/00000397
Lounsbury, K., Mitchell, K. J., & Finkelhor, D. (2011). The true prevalence of
sexting: Crimes against children. Research Center, University of New
Hampshire. Retrieved April 5, 2017, from http://www.unh.edu/ccrc/pdf/
Sexting%20Fact%20Sheet%204_29_11.pdf
Lunceford, B. (2011). The new pornographers: Legal and ethical considerations
of sexting. In B. E. Drushel & K. German (Eds.), The ethics of emerging media:
Information, social norms, and new media technology (pp. 99–118). New York:
Continuum.
Markus, H., & Wurf, E. (1987). The dynamic self-concept. Annual Review of
Psychology, 38, 299–337.
Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for
cognition, emotion and motivation. Psychological Review, 98, 224–253.
Martinez-Prather, K., & Vandiver, D. (2014). Sexting among teenagers in the
United States: A retrospective analysis of identifying motivating factors,
potential targets, and the role of a capable guardian. International Journal of
Cyber Criminology, 8(1), 21–35.
Noone, J. H., & Stephens, C. (2008). Men, masculine identities, and health
care utilisation. Sociology of Health & Illness, 30, 711–725.
Peter, J., & Valkenburg, P. M. (2009). Adolescents’ exposure to sexually explicit
Internet material and notions of women as sex objects: Assessing causality
and underlying processes. Journal of Communication, 59, 407–433.
Ringrose, J., & Eriksson Barajas, K. (2011). Gendered risks and opportunities?
Exploting teen girls’ digitized sexual identities in postfeminist media con-
texts. International Journal of Media and Cultural Politics, 7(2), 121–138.
Ringrose, J., Gill, R., Livingstone, S., & Harvey, L. (2013). Teen girls, sexual
double standards and ‘sexting’: Gendered value in digital image exchange.
Feminist Theory, 14(3), 305–323.
Schloms-Madlener, K. C. (2013). The prevalence and characteristics of sexting
behaviours among adolescents and adults in Cape Town, South Africa.
Unpublished Master’s Dissertation. Retrieved April 5, 2017, from http://
core.ac.uk/display/29057008
Shariff, S. (2015). Sexting and cyberbullying: Defining the line for digitally empow-
ered kids. New York: Cambridge University Press.
Shotter, J., & Gergen, K. (1989). Texts of identity. London: Sage Publications.
Skeggs, B., & Wood, H. (2012). Reacting to reality television: Performance, audi-
ence and value. London: Routledge.
Solomon, D., & Theiss, J. (2012). Interpersonal communication: Putting theory
into practice. New York: Routledge.
Steinar, K. (1996). Interviews an introduction to qualitative research interviewing.
London: Sage Publications.
Thurlow, C. (2014). Disciplining youth: Language ideologies and new tech-
nologies. In A. Jaworski & A. Coupland (Eds.), The discourse reader
Tsui, A. B. M. (1994). English conversation. Oxford: Oxford University Press.
Valkenburg, P. M., & Peter, J. (2011). Online communication among adoles-
cents: An integrated model of its attraction, opportunities, and risks. Journal
of Adolescent Health, 48, 121–127.
Van Dijk, T. (1977). Text and context: Explorations in the semantics and pragmat-
ics of discourse. London/New York: Longman.
Versluys, E. (2007). The notion of identity in discourse analysis: Some ‘dis-
course analytical’ remarks. RASK. International Journal of Language and
Communication, 26, 89–100. Retrieved April 5, 2017, from http://
static.sdu.dk/mediafiles//1/6/4/%7B164DE69B-9E30-499B-B5F9-
C6264E6293ED%7DEline%20Versluys,%2089-99.pdf
Walker, S., Lena, S., & TempleSmith, M. (2013). Sexting: Young women’s and
men’s views on its nature and origins. Journal of Adolescent Health, 52(6),
697–701.
12
Twitter, Politeness, Self-Presentation
Maria Sifianou and Spiridoula Bella
12.1 Introduction
Early discursive approaches to politeness research have highlighted, among
other things, the need for a distinction between first order, that is, lay con-
ceptualizations of politeness, and second order, that is, theoretical constru-
als of it (Eelen, 2001; Watts, 2003; Watts, Ide, & Ehlich, 1992). In this
connection, it has been argued that research should focus on first-order
understandings of im/politeness as they emerge in everyday interactions
(e.g., Locher & Watts, 2005; Mills, 2003; Watts, 2003). This emphasis has
been implemented primarily with analyses of the sequential development of
real-life interactions and from assessments made by participants (Ogiermann,
2009). However, this is a rather restricted approach because even though
practices are typically situation-specific, there may be pan-situational values
and assumptions (Spencer-Oatey, 2005) which may be missed.
M. Sifianou (*) • S. Bella

Department of English Language and Literature, School of Philosophy,
National and Kapodistrian University of Athens, Athens, Greece
e-mail: msifian@enl.uoa.gr; sbella@phil.uoa

https://doi.org/10.1007/978-3-319-92663-6_12
342 M. Sifianou and S. Bella
The main aim of this paper is to try to unravel some such more general
assumptions and practices and thus contribute to a deeper understanding
of how politeness is conceptualized in Greek, drawing evidence from
electronic corpora. The paper begins with a brief overview of relevant
theoretical issues and proceeds with a brief description followed by the
analysis of the data drawn from a corpus consisting of 19,950 brief mes-
sages posted to Twitter. For the purposes of this study 195 instances
which included the keyword phrases “politeness is/is not” were analyzed.
12.2 Theoretical Background

Despite the growing number of publications on politeness, there is no
agreement as to what politeness exactly is, how it should fruitfully be
investigated, or how it might best be defined. In early attempts to clear
this foggy landscape, researchers (e.g., Eelen, 2001; Watts, 2003; Watts
et al., 1992) highlighted the need for a distinction between lay under-
standings of politeness (first order or P1) and politeness as a theoretical
construct (second order or P2) and insisted on the significance of taking
native speakers’ perceptions seriously since these should constitute the
locus of attention in politeness research (e.g., Culpeper, 2011; Locher &
Watts, 2005; Watts, 2003). Even though subsequent research has shown
that it is difficult, if at all possible, to maintain a clear-cut distinction
between the two (e.g., Garcés-Conejos Blitvich, 2010a; Haugh, 2013;
Terkourafi, 2005, 2011), as one cannot but inform the other, it is more
or less agreed that lay perspectives can offer valuable insights and contrib-
ute to a fuller understanding of what politeness is. A similar view is
detailed by Kádár and Haugh (2013) who contend that there can be no
purely first-order and second-order approaches since anybody can have
manifold perspectives on politeness. They propose a refinement that goes
beyond a mere dichotomy and advocate that folk-theoretic understand-
ings constitute second-order conceptualizations since “they involve obser-
vation rather than participation in the social world” (Kádár & Haugh,
2013, p. 85, emphasis in the original).
Relatedly, Eelen (2001) identifies two possible aspects of politeness as a
lay concept: an action-related and a conceptual side, the former referring
Twitter, Politeness, Self-Presentation 343
to actual manifestations of politeness in interactions and the latter to

accounts by lay people of what the term ‘politeness’ means. Most relevant
research has concentrated on the former, that is, on how politeness arises
and is evaluated in ongoing interactions, whereas the latter, that is, lay
understandings and evaluations of politeness from a more detached per-
spective, have not received equal attention. However, as Ogiermann (2009)
observes, the conceptual side of politeness is particularly valuable especially
in cross-cultural studies since lay members’ views can provide insights into
the conceptualization of politeness in the specific socio-cultural milieu.
Since in recent discursive approaches to politeness research, politeness
is understood as situated in specific interactions, the relative merits of
naturally occurring data have been foregrounded (Haugh, 2010). To
investigate situated evaluations, many politeness researchers have recently
turned their attention to electronic corpora because they provide a rich,
readily available source of authentic data which can offer a wider scope
and richer insights into linguistic analyses. Even though corpus linguistics
and the pragmatic study of politeness took off at about the same time, it
is only relatively recently that they have begun to work together (Leech,
2014), probably because of misunderstandings about corpus-based stud-
ies, as Culpeper (2015) argues. Nowadays, a fruitful cooperation between
the two fields has started emerging in the service of various purposes (see,
e.g., Diani, 2015; Taylor, 2013). Broadly speaking, in relation to polite-
ness research, some researchers have constructed their own genre-specific
corpora (e.g., Clancy, 2013; Culpeper, 2011; Diani, 2015; Schnurr,
Marra, & Holmes, 2008), while others have either used available ones to
explore im/politeness devices/features in discourse (e.g., Aijmer, 2015;
Beeching, 2006; García Vizcaíno, 2006; Taylor, 2013) or a combination
of the two (e.g., McEnery, Baker, & Cheepen, 2002). Culpeper (2011,
p. 111) underscores the emerging importance of corpora in pragmatics
and im/politeness research and argues that “[m]etalanguage/metadis-
course can reveal people’s understandings of impoliteness”. However, few
researchers have been concerned with the conceptual side of im/polite-
ness as an everyday concept, through the exploration of metalinguistic
expressions and metapragmatic comments in corpora1 (e.g., Culpeper,
2009; Garcés-Conejos Blitvich, Lorenzo-Dus, & Bou-Franch, 2010).
The popularity of the web as a corpus is dynamically emerging in

computer-mediated communication (CMC)2 research since it offers a
rich source of naturally occurring, digital data (e.g., Androutsopoulos,
2006; Bolander & Locher, 2014; Bou-Franch & Garcés-Conejos Blitvich,
2014a; Garcés-Conejos Blitvich, 2010b; Locher, 2010). Various facets of
im/politeness were of the first topics to be explored in this new environ-
ment (Herring, Stein, & Virtanen, 2013, p. 3). Computer-mediated
sources have been used to explore aspects or specific features of politeness
in interaction along with gender differences whereas research on its con-
ceptualization has been rather overlooked. However, as Herring et al.
(2013, p. 8) contend “[t]he written, persistent nature of CMC makes
language more available for metalinguistic reflection than in the case of
speech”. Twitter, more specifically, has only recently started to attract
politeness scholars’ attention (see, e.g., Dayter, 2014; Page, 2012, 2014)
probably because it was initially perceived and used primarily as a “mes-
saging utility” with most tweets falling under unflattering categories such
as “pointless babble” and “daily chatter” (Baym, 2010, p. 30; Rogers,
2014).3 However, as Twitter has become a dynamically evolving medium
and a significant source of real-time information and opinions on more
or less everything, we thought of exploring the metalinguistics of polite-
ness as attested in a Greek dataset compiled from public Twitter mes-
sages.4 In other words, we are concerned with what lay people (tweeters
in this case) understand politeness to be. Twitter centers on text-focused
interaction (Squires, 2015) and since we are mostly interested in the con-
tent of tweets, we will concentrate primarily on their textual forms.
Moreover, as Herring et al. (2013, p. 8) point out, there is a significant
gap in social media on “language-focused research” which is “a rich field
of investigation for discourse analysts”, as she (Herring, 2013, p. 21)
notes elsewhere (see also Bou-Franch & Garcés-Conejos Blitvich, 2014b).
12.3 Twitter
Given the expansion and availability of the internet, microblogging has
become an increasingly popular means of interaction among its users. One
such free microblogging service is Twitter, which offers an easy platform for
registered users to post brief messages (maximum 140 characters, called
“tweets”) and to “follow” and/or “be followed by” other microbloggers

(e.g., boyd, Golder, & Lotan, 2010; Honeycutt & Herring, 2009). Through
these brief messages, tweeters interact in almost real time by posting mes-
sages on both private and public issues involving “debate, humour, updates,
news, products, gossip, and more besides” (Hardaker & McGlashan, 2016,
p. 81). It is thus “a valuable source of people’s opinions” (Pak & Paroubek,
2010, p. 1320) and personal thoughts (Papacharissi, 2012, p. 1993).
The posted tweets can be made either public or distributed within a
social network, but the dominant trend is for them to be open to the pub-
lic (Hargittai & Litt, 2011; Marwick & boyd, 2010; Page, 2012; Squires,
2015). Unless posters “expressly state otherwise, their updates, interests,
and ‘likes’ are instantly publicized to others” (Senft, 2012). The advantage
of this is the wide circulation of posters’ tweets since if a poster chooses to
restrict their tweets to approved followers, these will remain inaccessible to
the wider audience (Squires, 2015, p. 247).5 As a result of the mostly pub-
lic nature of interactions, visibility has emerged as a core property of the
system reinforcing posters’ concern with identity construction, “an ongo-
ing performance of identity” (Zappavigna, 2012, p. 38).
It should be noted here that content acquires special significance in
identity construction in this environment (Gilpin, 2011; Johnson,
Zhang, Bichard, & Seltzer, 2011). Posters’ potential anonymity and the
disembodied nature of identity (not connected to a physical appearance
or non-verbal behavior)6 along with the fact that Twitter user profiles are
very brief (maximum 140 characters) prompts followers to draw conclu-
sions primarily on the basis of the content of tweets.
Motivated by the informal and assumed spontaneous nature of com-
munication through social media channels and its potential value as a
source for studying lay people’s understandings of politeness, we have
been gathering public text data drawn from Twitter’s huge repository, via
its new advanced search engine.
12.4 Corpus Data for Greek

Twitter offers the advantage of masses of publicly available data which
can be used not only for understanding language use in CMC but human
language in general (Squires, 2015). Data collection methods can vary in
terms of technical sophistication, the simplest way being to target specific

keywords or hashtags. Thus, searching for instances of the keyword
ευγένεια (‘politeness’ in Greek) within text messages or Twitter tags
(hashtags #), we constructed the Twitter Corpus of Greek Politeness (TC-
GP) consisting of 345,000 words and 19,550 tweets released by Greek
tweeters from February 2009 to February 2015.7 Since Twitter Search
API displays only recent tweets, our data was collected manually through
Twitter Search. By performing a manual search of relevant tweets in
Twitter’s database, we have been able to monitor data material more suc-
cessfully, by excluding irrelevant content, duplicates, highly similar
tweets, and automated tweets sent by non-humans (e.g., organizations);
this has helped us reduce data noise in the corpus (Zappavigna, 2012).
The corpus has been developed as part of a more extended corpus-
based project and contains tweets that explicitly mention the word
ευγένεια ‘politeness’ along with other metadata, such as user ID, posting
time, date, and Twitter actions (e.g., reply (@), retweet (RT), favorite).
Thus, preceding tweets that did not contain the specific word and may
have triggered the specific comment and which, along with succeeding
ones, provide the context to these tweets, are not included in the corpus.
The topic of the post and the relationship among posters may also influ-
ence the number of responses a post receives. More specifically, as is well-
known, Twitter allows for asymmetrical relationships between users
(Hargittai & Litt, 2011; Marwick & boyd, 2010; Page, 2012; Zappavigna,
2012), in that there is no ‘conversational obligation’ to respond immedi-
ately or even respond at all, posts may but may not be preceded or fol-
lowed by something related. In addition, reaction to tweets may be related
to the assumed significance of or interest in the content of the post, that
is, breaking news, for instance, will obviously attract many readers’ com-
ments whereas topics like politeness may constrain the number of poten-
tial readers and respondents. Moreover, there are limitations in Twitter’s
design for threaded conversation which has been facilitated to a certain
extent by the use of the prefix @ which expresses addressivity (Halavais,
2014; Honeycutt & Herring, 2009; Squires, 2015). Thus, extended
sequences of interactions are difficult to recover, and since they require
different corpus collection strategies (Zappavigna, 2012, p. 23), they will
be considered at a later stage of this research.
The TC-GP corpus was loaded to Sketch Engine, a sophisticated tool

that offers corpus query functions (Kilgarriff et al., 2014). For the pur-
poses of this study, we isolated 195 tweets which included the keyword
phrases (word strings), “politeness is/is not” in an initial attempt to iden-
tify specific views on what politeness means to posters.
12.5 Analysis of the Data

This dataset is appealing, not least because, at first glance, it verifies earlier
findings that politeness in Greek is understood in very broad terms
(Sifianou, 1992; Sifianou & Tzanne, 2010). What is also notable is that
most posters appear to view politeness mainly as something abstract
rather than actual behavior per se. For instance, politeness is a treasure
(5),8 an (important) asset (87, 105, 118), a (great) virtue (13, 30, 95,
112, 133), an attitude of life (19, 119), a qualification (40), a panhuman
characteristic (148), among others. Some posters characterize politeness
in positive terms, whereas others see it as something negative (see Watts,
2003). For instance, for some, ευγένεια ‘politeness’ is an inherent virtue
coming from the heart/soul, a virtue which you either possess or not, and
is thus effortless. If effort is invested in being polite, this effort is evident
(49), thus expressing feigned politeness, which is worse than impoliteness
(91). For others, since it is a quality that can be cultivated (101) through
education (4, 195) it, therefore, does need some effort (184). Politeness
may be the greatest indicator of sincerity (65), but may also be the most
tolerated form of hypocrisy (15, 26, 37, 83), while sincerity without
politeness is little more than impoliteness (162). Perceiving politeness in
negative terms is attributed to the historical development of the term in
English (Watts, 1992) but it is also found in other communities, such as
Israel (Blum-Kulka, 1992) and Venezuela (Bolívar, 2008). So it can be
surmised that negative attributions to politeness may be a result of its
being associated with education and cultivated behavior and thus socio-
economic power, a relationship which excludes certain groups. This is an
interesting issue for theoretical accounts which pay no attention to this
aspect of politeness, viewing it mostly in positive terms.
Interestingly, some posters provide negative definitions, such as

“politeness is not the absence of insulting words” (172), “politeness is not
associated with theatrical movements” (192), and “politeness is not weak-
ness” (36). Some others use a combination of a negative definition fol-
lowed by a positive one, such as “politeness is not weakness … it is the
most refined form of love” (175) and “politeness is not a luxury but a
necessity” (176). Such negative definitions constitute a form of intertex-
tuality through negation, in that speakers position themselves outside
their words, which they attribute to others (Johnstone, 2008, p. 165). By
explicitly rejecting others’ views as to what politeness is, posters reinforce
their own views. It is noteworthy that the views expressed in the above
examples highlight an understanding of politeness that “goes well beyond
overt social etiquette” (Johnstone, 2008, p. 159). All these examples
bring to mind Bucholtz’s (1999) distinction between positive and nega-
tive identity practices, or between association and dissociation practices,
respectively (Kádár & Haugh, 2013), which either actively construct a
chosen identity or distance oneself from a rejected identity. In other
words, in expressing their views as to what politeness is or is not, our
posters present their beliefs and hence who they are or who they want to
show they are. In our case, by rejecting (through negation) what could be
viewed as superficial aspects of politeness, posters contribute (overtly or
covertly) to the construction of a self who endorses and presumably
adheres to deeper, more essential values.
In very few cases, swear words, indecent expressions and incivility were
incorporated in the definitions, as for example “politeness is everything,
fuck you” (98) and “formal politeness is what a friend of mine says about
orgasm: if you fake it you’ll make it” (161). Such tweets may be the result
of releasing negative emotions and may be interpreted as offensive and
inappropriate in some contexts but could also be seen as expressions of
solidarity and authenticity, because as Papacharissi (2012) points out,
Twitter offers a platform for “saying shocking or potentially provocative
things in public” and further concludes that “[p]rovocation is thus both
an outcome of authenticity and a possible strategy for authenticity”
(p. 1999). Interestingly, the poster of the second of the above tweets
assigns this to a friend in a possible attempt to distance him/herself from
the responsibility of a potentially offensive interpretation.
Some posters do not provide definitions, but specific examples of

actions deemed polite, such as politeness is expressed by the server in the
kiosk who gives you chewing gum for free because you trekked all the
way back to your car to bring him change (90). Some of these examples
are Twitter-specific such as “politeness is to write hehehehe followed by
numerous exclamation marks, even though you are expressionless and
have not even read the post” (54), reflecting the poster’s negative view of
politeness, as superficial and hypocritical.
The expected variation in perceptions of politeness noted above may
reflect “what our own personal folk notions of polite behaviour happen
to be” and, consequently, the “discursive struggle over the value of the
term”, that is, “the ways in which (im)polite behaviour is evaluated and
commented on by lay members” (Watts, 2003, p. 2, 8). They may also
reflect, in our case, the design features (e.g., the required brevity of mes-
sages) and technological affordances and constraints of the digital envi-
ronment (e.g., searchability and disembodiment), to which we will now
turn our attention.
12.5.1 Required Brevity of Messages
As is well known, what has been at the heart of Twitter’s uniqueness is the
limit of 140 characters per tweet.9 This undoubtedly demands that post-
ers be restrictive in what and how they express and invites them “to be
creative with their use of the textual space” (Halavais, 2014, p. 31). It
may be this creativity which turned into pleasurable an initially mocked
generic format (Squires, 2015).
Given this character limit, views on what politeness is are brief, such as
“politeness indicates civilization” (74) and “politeness is the organized
indifference” (42). Therefore, one may suggest that what is posted is the
most significant aspect of politeness for the specific user or the one that
comes to his/her mind first. This may be an inadequate explanation
because in some cases, posts may be hurried and embedded in other
ongoing activities, and in others, there may be conscious self-reflection of
what to share (Zappavigna, 2012). Such consciously strategic perfor-
mances, which appear to abound in our data, serve various purposes, and
self-presentation seems to be one of them.
Despite the character limit and the consequent brevity of the opinions
expressed or possibly because of it, many posters exploit the creative pos-
sibilities of the language. For instance, some posters use devices like met-
aphors such as “politeness is the master key to all doors” (85), and
“politeness is the lubricant of social contact” (61).10 Others use similes
such as “politeness is like a reefer. It should circulate” (127) and “polite-
ness is as rare as the truffle in the mountains” (72). As is well-known, the
function of metaphors is at least twofold (see, e.g., Gibbs, 1994; Kövecses,
2002). They typically enable the understanding of an abstract concept or
a complex issue and they often lead to the creation of vivid images.
Whereas metaphors do not signal their function by any conventional
linguistic means, similes explicitly mark how the audience should inter-
pret a particular expression by the insertion of linguistic material, such as
like and as. Thus, posts containing such devices become more accessible,
memorable, and potentially colorful. These are cases of what Carter
(2004) calls “pattern re-forming” forms, which speakers employ to make
their words acquire “a distinct and discernible impact” and “foreground
the ideas, feelings, attitudes and evaluations which are being expressed”
(Carter, 2004, p. 133).
Similar is the function of humor/joking, which is “a key creative source
for self-display, entertainment and communality” (Carter, 2004, p. 85;
drawing from Norrick, 1993). More specifically, as has been shown time
and again a major social function of humor/joking is the establishment
and/or maintenance of solidarity and in-groupness (see, e.g., Brown &
Levinson, 1987; Zappavigna, 2012). Humor research (see, e.g., Archakis
& Tsakona, 2005; Attardo, 2001) has shown that humor is based on some
kind of incongruity or unexpectedness. For instance, a poster says “polite-
ness is to give your chubby friend a smaller size jacket and tell them, it
would suit you if you lose ten kilos” (16) in a possible attempt at building
rapport or provoking hilarity among the readers, since his/her post contra-
dicts what is expected in the situation described. In addition, humorous
choices have been found to contribute to the construction of aspects of
social identity (see, e.g., Archakis & Tsakona, 2005).
Humor involves varying degrees of complexity and shared knowledge.
One poster crosses linguistic boundaries and uses another language (English
in this case) in his/her post “formal politeness is what a friend of mine says
about orgasm: if you fake it you’ll make it” (161). This is an example of what
Androutsopoulos (2013, p. 7) calls “networked multilingualism” which

demonstrates “how language resources can be appropriated, combined,
juxtaposed, and displayed to a networked audience ‘for fun’ and ‘for show’”.
Interesting humorous cases of intertextual reference include the examples
“politeness is the other half of nobility” (78) and “politeness is half of every-
thing, we are done with cleanliness” (135), also both being further exam-
ples of “pattern re-forming”.11 The former is based on a playful intertextual
reference to a well-known Greek proverb “cleanliness is half nobility”. The
latter seems to involve a doubly-reformed pattern which intertextually
refers to the Greek proverb “the beginning is half of everything” trans-
formed into “cleanliness is half of everything” and then creatively trans-
muted by the poster as “politeness is half of everything, we are done with
cleanliness”. In all the above cases, it is impossible to detect the posters’
knowledge and what they appropriate from the internet sources (see
Androutsopoulos, 2013). Both Papacharissi (2012) and Zappavigna (2012)
among others identify humor and playfulness as dominant performative
strategies in their data. Papacharissi (2012) further suggests that “individu-
als confronted with a restricted stage for self-presentation seek to overcome
expressive restrictions through imaginative strategies that include play”
(p. 1998). By posting attractive messages, one hopes to construct a likable
identity and attract followers whose numbers are displayed on one’s profile
page (boyd, 2010; Senft, 2012).
Thus, the system enforced message brevity does not mean that tweeters
do not invest thought and time in their posts or that they do not exploit
linguistic options that would make their posts smart and attractive. In
fact, “individuals approach these media with a measure of playfulness”
(Papacharissi, 2012, p. 2001). A further interesting illustration of the
preparation involved in the self-presentation through tweets is provided
in the following section and relates to the use of internet resources out-
side Twitter.
12.5.2 Access to Network Resources
Posters have various motivations to post messages and this entails

degrees of deliberation: in some cases, they may post something hur-
riedly and in others, they may take their time in constructing their
messages (Zappavigna, 2012). Several researchers have explored how

posters may bring other voices in their tweets by republishing tweets
and flagging topics. Through retweeting, tweets can travel within
Twitter but can also be transported outside the medium, via its new
embedding feature, practices which increase their visibility and, conse-
quently, that of their producers (Squires, 2015). However, what has
received less attention in the relevant literature is what posters bring
into their tweets from various network resources outside Twitter. This
transference of information from outside sources blurs the cultural shift
noted by Zappavigna (2011) “from searching purely for content, to
searching what other people are saying online” (789). Even though, as
the author suggests, once microblogging services like Twitter emerged,
search has acquired an affiliative function, search for information
appears to be part of it, at least in our data, in which these two forms of
the search facility frequently intermingle. In other words, some posters,
at least, not only retweet but also draw either on their background
knowledge or most probably from immediately available internet
sources. As Androutsopoulos (2013) contends, communication through
the media means that users have access to various resources which they
explore, appropriate, and recontextualize. This appears evident even
with a cursory look through our data. However, to verify this assump-
tion, we performed a Google search for the definitions “politeness indi-
cates civilization” and “politeness is the organized indifference”. This
test returned 61 results for the former and 573 results for the latter defi-
nition, which clearly testify appropriation and recontextualization.
Moreover, social media facilitate the circulation of various views, aph-
orisms, clichés and religious doctrines, which thus become widely and
easily accessible and perhaps also familiar and memorable. For instance,
for a couple of posters, “politeness has become so rare that people mistake
it for flirtation” (70, 132), a view recently discussed in various blogs. For
others “politeness is everything” (39, 51) for which a Google search
returned 426 results. For one poster “politeness is one of the most beauti-
ful blossoms of love” which was found on various religious sites.
Paraphrasing an aphorism on kindness, attributed to Mark Twain on
various sites, a couple of posters say that politeness is the language that
the deaf can hear and the blind can see (1, 46).
What is remarkable, in our view, is that some posters go as far as to cite

their sources, as for example “politeness is the neat dress of indifference”
(69), attributed to the well-known Greek poet, Katerina Angelaki-Rouke.
For others, “politeness is the most tolerated form of hypocrisy” (15, 26,
37, 83), attributed to various sources or to no source at all. Similarly,
“politeness is like pure gold which never rusts” (121) is attributed to
Mark Twain and “politeness is not just one act. Politeness is a habit” is
attributed to Aristotle (186). Even though Aristotle’s name appears in
capital letters in this post, we were unable to trace this specific view to any
internet source. It is more likely a paraphrase of his ideas on virtue rather
than politeness and seems more like a case of name-dropping in order to
impress.12 Another definition “politeness is a virtue of a kind character
which leads to good deeds and words” (28) is posted without a source
whereas it is attributed to Plato in a number of other sites. Such examples
question, at least to some extent, the originality of the views expressed by
posters and corroborate Androutsopoulos’ (2013) view of language bits
being appropriated by networked language users. They also confirm, in a
rather striking way, Johnstone’s (2008, p. 159) contention, drawing from
Bakhtin’s (1953/1981) work on heteroglossia and appropriation, that all
writing is “multi-voiced” since we always partly borrow and appropriate
views and ways of expressing them.
Given that “online information and communication technologies have
the potential to shape identity processes in meaningful ways” (Ellison,
2013, p. 2), we consider the practice of quoting or citing views of others
to have important identity implications.13 Irrespective of whether the
source is provided, posters appear to aim at the construction of a knowl-
edgeable identity by imparting a sophisticated, witty view possibly aim-
ing at a positive self-presentation. However, citing a source, especially
when reference is made to well-known figures, such as Mark Twain or
Aristotle, further enhances a positive self-presentation in that posters
appear not only knowledgeable but also educated individuals who know
that plagiarism should be avoided.
One could argue that, at least, some of these posts resemble self-
branding by aspiring micro-celebrities whose involvement is not neces-
sarily (but may be) intentional (Page, 2012). This is not surprising, “as we
now see ‘regular’ people adopting micro-celebrity tactics to gain status
online” (Marwick & boyd, 2011, p. 140). One of the celebrity practices
discussed by Marwick and boyd (2011; see also Senft, 2012) is the use of
quotes by famous people, a practice which marks shared values but also
evidences knowledgeability, one might add. All this reminds us of
Goffman’s (1959) presentation of self; that is, that people, like perform-
ers, use the available resources to create and sustain a favorable image of
themselves that tends to be idealized in several different ways. “Twitter
affords a platform for condensed yet potentially rich and variably public
or private performances of the self ” (Papacharissi, 2012, p. 1989).
Moreover, posting on an issue like politeness and showing deeper
knowledge on it may itself constitute an act of identity construction pro-
jecting expertise on and concern with a presumably significant yet con-
tested social issue. Citing what great thinkers have said is in a sense an act
of affiliating with them.
In discussing the drawbacks of elicited data for evaluations of polite-
ness, Kádár and Haugh (2013, p. 31) mention the possible influence of
‘social desirability effects’, that is, informants may wish to be seen as
saying the right thing or think of themselves as saying the right thing.
The authors (2013) elaborate on this further stressing that “a concern
with what others think of us” (p. 207) lies at the heart of politeness.
Therefore, the notion of ‘intersubjectivity’ that is, “how we interpret or
understand the perceptions, feelings, thoughts, beliefs, desires of others,
and in some cases reach agreement or a common understanding about
them” (Kádár & Haugh, 2013, p. 207) is considered central in the for-
mulation of such evaluations. In other words, politeness is not located
in the individual, but it is intersubjectively constructed. This may lead
to variability and contestedness or agreement as to what politeness is.
Even though our data is not elicited, it appears that the same principles
hold. By citing views of well-known and highly-esteemed individuals,
posters invest their own views with greater validity and reduce the pos-
sibility of them being challenged and their views contested. Hence, it
appears that posters attempt to achieve a positive self-representation and
simultaneously ensure the ‘common understanding’ discussed above.
Posters may also attempt to present themselves as proficient users of this
novel environment.
So “new and evolving social practices” emerge in the social media (Bou-
Franch & Garcés-Conejos Blitvich, 2014b, p. 1) and “social identities can
be indexed by styles of discourse” “sometimes self-consciously and strategi-
cally and sometimes as a matter of habit” (Johnstone, 2008, p. 151, 152).
Twitter users, whether individuals or collectives, have “aims that are strate-
gic, casual, or a dynamic combination of both” (Puschmann & Burgess,
2014, p. 47). However, as Bucholtz and Hall (2005) contend, even though
“individuals’ sense of self is certainly an important element of identity …
the only way that such self-conceptions enter the social world is via some
form of discourse” (p. 587). In other words, as has been shown time and
again, identity, much like politeness, is not a stable construct produced by
and located in the individual. Rather, it is a relational and socio-cultural
phenomenon that emerges in situated interactions (see, e.g., Archakis &
Tsakona, 2005; Bucholtz & Hall, 2005; Garcés-Conejos Blitvich &
Sifianou, 2017). However, as Twitter fosters asymmetrical connections,
the average Twitter poster does not know who or how many people will
read their tweets. Thus, the audience, actual or imagined, will be diverse
and self-presentation becomes rather demanding.14
12.5.3 Orientation to Networked Audiences
In offline contexts, individuals have different kinds of relationships with

different people in different contexts and this affects the way they use lan-
guage and the identities they present. Twitter, in particular, with followers
rather than friends, as on other social network sites such as Facebook, pro-
vides a diverse audience for any poster. This creates what has been called
“context collapse” (Marwick & boyd, 2010), that is, “the process through
which various connections representing different aspects of one’s identity
are flattened into a uniform, one-dimensional group such as ‘Friends’ or
‘contacts’” (Ellison, 2013, p. 8) which in many ways affects the style of
communication that occurs (Halavais, 2014, p. 31). Self-presentation to
diverse groups of people who have come together as a networked audience
can be challenging. When people interact face-to-face, “audience design” is
“sometimes fairly automatic” but when it comes to the readership of a
blog, for instance, “attempts to accommodate to their expectations may be
quite self-consciously strategic” (Johnstone, 2008, p. 159). Thus, posting

smart or humorous messages and drawing from internet sources and well-
known individuals may constitute safe ways to overcome the challenge,
attract followers, and thus enhance status.
Furthermore, quoting the views of others may be viewed as a ‘homog-
enization’ practice (see Bucholtz & Hall, 2003, 2005), that is, as an
attempt to impose resemblance, which is also related to issues of identity
construction and is achieved through processes of stereotyped generaliza-
tions and, more specifically, through downgrading (or even concealing)
the differences between the members of what can potentially be consid-
ered a group (e.g., Archakis & Tsakona, 2012).

This is a first tentative attempt to use a Twitter corpus to investigate con-
ceptualizations of politeness in Greek. What emerged from the analysis is
that politeness is viewed in both positive and negative, but always very
broad, terms and is, thus, not restricted to verbal or non-verbal behavior.
Posters express a wide variety of conflicting views as to what politeness
means to them, evidence of the discursive struggle over politeness dis-
cussed in recent politeness research (see, e.g., Watts, 2003). While pro-
cessing the data, it soon became evident that the design features and the
constraints and affordances of the specific medium exerted a strong influ-
ence on the content of these messages (Herring, 2007). The 140-character
limit on the content of tweets contributed to brief definitions, many of
which, however, involved various figures of speech such as metaphors and
similes and playful language. These may be a reflection of the ludic char-
acter of language use on the internet (cf. Herring et al., 2013, p. 7) but
in addition, contribute to specific self-representations.
However, it seems that searchability is the medium’s affordance which is
the most exploited since web search has become a routine activity among
users. Our posters appear to use this facility to appropriate and recontextual-
ize bits of language which involve established wisdom. This enables them to
project knowledgeability and thus claim authority before a widely varied
and unspecified audience. What is noticeable is that this knowledgeability is
not only content-related but also source-related, that is, showing that one
knows what well-known figures have said about the specific topic. In other
words, it appears that the focus is more on recycling and publicizing pre-
sumably established knowledge than on presenting personal understandings
and evaluations of what politeness means to them. This practice is directly
related to self-presentation.
Hirschon (2001) contends that “apparently unrelated phenomena can
be seen to make sense if interpreted in a holistic way within the overall
socio-cultural context”, further adding that linguistic conduct and com-
municative style can be interpreted with reference to key social values
(Hirschon, 2001, p. 17). Koutsantoni (2005) argues that knowledgeabil-
ity is one such value reflected in the Greek educational system which
until recently favored theoretical rather than practical studies. Thus, the
views presented may not only reflect the affordances of the specific
medium but also the facilitation it offers for the distribution of assumed
values in circulation. This facility may have wider repercussions as tweets
are frequently made doubly public by being retweeted or quoted in other
mass media, what Squires (2015, p. 247) calls the ‘appropriative’ use of
Twitter (see also Georgakopoulou, 2013).
One may legitimately question the representativeness of this sample
not only because Twitter users are a subset of the world’s population (e.g.,
Hargittai & Litt, 2011) but also because, in Greece, in particular, the
number of users is still quite limited (Tsaliki, 2010; Zafiropoulos,
Antoniadis, & Vrana, 2014). However, since Twitter users constitute a
highly heterogeneous group, diverse points of view are likely to surface,
some of which may even reflect current conceptualizations, such as the
relatively recent rather negative collocation αστική ευγένεια, ‘urban
politeness’, to refer to insincere formalities characteristic of cities.15 So
this new source of naturally occurring data can provide insights into the
perceptions of various groups of people who may not be accessed in other
ways. Consequently, such data analysis is not offered as a substitute but
as a complement for findings from other sources.
Acknowledgments This is part of a more extensive project on the concept of

im/politeness in Greek funded by the National and Kapodistrian University of
Athens (Special Research Account 70/4/11099). We are grateful to Harriet
Mitrakou, who compiled the corpus using her expertise in corpus linguistics.
Notes
1. See Culpeper (2011, p. 74) for the distinction between the two.
2. For a discussion on the appropriateness of this term among various
others, see Herring et al. (2013).
3. The signature bird logo of the medium supports the idea that tweets are
inconsequential chirpings (Rogers, 2014, p. xii).
4. As Puschmann, Bruns, Mahrt, Weller, and Burgess (2014) note “through
Twitter, researchers gain access to huge volumes of data, a treasure trove of
digital traces, waiting to be mined for precious insights into people’s
behaviours, their moods, their consumption patterns, their language, and
their voting behaviours” (p. 426). Twitter’s data set is to be archived by the
Library of Congress and accessible for research purposes (Rogers, 2014).
5. On ethical considerations on Twitter data, see discussion in Hardaker
and McGlashan (2016) and on ethical issues with social media data, see
Bolander and Locher (2014) and the forthcoming Special Issue of
Applied Linguistics Review.
6. It may be worth mentioning here that scholars, like Thurlow, Lengel,
and Tomic (2004) have criticized what they view as an overemphasis on
disembodiment in computer mediated communication since, as they
argue, all identity is multiple and dynamic given that interlocutors con-
tinually shift identities whether online or offline.
7. Mikros and Perifanos (2013) have compiled the first Greek Twitter
Corpus but it was unavailable to us due to the restrictions of redistribu-
tion imposed by the Twitter terms of use. As this is part of a more
extended project, this time span was chosen because we would like to
explore at a later stage whether the financial crisis has affected conceptu-
alizations of politeness.
8. The numbers in parentheses in this and the following sections refer to
the number of the tweet in the current corpus.
9. This restriction may be overcome by including hyperlinks and other
multimedia.
10. Interestingly, a Google search did not return any results for this specific
utterance but of the alternative “hypocrisy is the lubricant of social con-
tact” which is used as a headline by one blogger and attributed to him/
her in other blogs. However, the assumed relationship between the two
is evident.
11. In his discussion of YouTube videos, Androutsopoulos (2010) elaborates

on their intertextual relations and refers to users as ‘intertextual opera-
tors’ who modify existing multimodal texts. He stresses the significance
of intertextuality as a resource of meaning through appropriation and
recontextualization.
12. In this connection, we were amazed to discover a site entitled “wise peo-
ple’s sayings to pretend that you are smart on facebook and twitter” con-
taining almost 600 entries and a book (Metzger, 2016) entitled Highly
Tweetable: 5000+ Awesome Quotes to Use on Twitter and Social Media.
13. Interestingly, Twitter encourages the use of quoting through the retweet
(RT) facility which among other motivations can also be a selfish act of
drawing attention to oneself (boyd et al., 2010).
14. It may be worth noting here that scholars like Huberman, Romero, and
Wu (2008) argue that Twitter posters interact with few of those “listed”
in their network, that is, those that matter to them and that reciprocate
their attention. They further suggest that two different networks are in
fact created: a very dense network including followers and followees, and
a sparser and simpler one that includes actual friends.
15. It should be noted that this collocation is not listed in older dictionaries
and is found only in a recent one (Users’ Dictionary of Modern Greek,
2014, p. 632).
References
Aijmer, K. (2015). Pragmatic markers. In K. Aijmer & C. Rühlemann (Eds.),
Corpus pragmatics: A handbook (pp. 195–218). Cambridge: Cambridge
University Press.
Androutsopoulos, J. (2010). Localizing the global on the participatory web. In
N. Coupland (Ed.), The handbook of language and globalization (pp. 203–231).
Malden, MA: Wiley-Blackwell.
Androutsopoulos, J. (2013). Networked multilingualism: Some language prac-
tices on Facebook and their implications. International Journal of Bilingualism,
19(2), 185–205. https://doi.org/10.1177/1367006913489198
Archakis, A., & Tsakona, V. (2005). Analyzing conversational data in GTVH
terms: A new approach to the issue of identity construction via humor.
Humor: International Journal of Humor Research, 18, 41–68.
Archakis, A., & Tsakona, V. (2012). The narrative construction of identities in

critical education. Basingstoke: Palgrave Macmillan.
Attardo, S. (2001). Humorous texts: A semantic and pragmatic analysis. Berlin/
New York: Mouton de Gruyter.
Bakhtin, M. M. (1953/1981). The dialogic imagination: Four essays by M. M.
Bakhtin (M. Holquist, Ed., C. Emerson & M. Holquist, Trans.). Austin, TX:
University of Texas Press.
Baym, N. K. (2010). Personal connections in the digital age. Cambridge: Polity
Press.
Beeching, K. (2006). Politeness markers in French: Post-posed quoi in the tour-
ist office. Journal of Politeness Research, 2(1), 143–167.
Blum-Kulka, S. (1992). The metapragmatics of politeness in Israeli society. In
R. J. Watts, S. Ide, & K. Ehlich (Eds.), Politeness in language: Studies in its
history, theory and practice (pp. 255–280). Berlin: Mouton de Gruyter.
Bolander, B., & Locher, M. (2014). Doing sociolinguistic research on computer-
mediated data: A review of four methodological issues. Discourse, Context &
Media, 3, 14–26.
Bolívar, A. (2008). Perceptions of (im)politeness in Venezuelan Spanish: The
role of evaluation in interaction. Pragmatics, 18(4), 611–633.
Bou-Franch, P., & Garcés-Conejos Blitvich, P. (2014a). Conflict management
in massive polylogues: A case study from YouTube. Journal of Pragmatics, 73,
19–36.
Bou-Franch, P., & Garcés-Conejos Blitvich, P. (2014b). Editorial: The pragmat-
ics of textual participation in the social media. Journal of Pragmatics, 73, 1–3.
boyd, d. (2010). Social network sites as networked publics: Affordances, dynamics,
and implications. Retrieved January 20, 2015, from http://www.danah.org/
papers/2010/SNSasNetworkedPublics.pdf
boyd, d., Golder, S., & Lotan, G. (2010, January 6). Tweet, tweet, retweet:
Conversational aspects of retweeting on Twitter. HICSS-43. Kauai, HI: IEEE.
Brown, P., & Levinson, S. (1987). Politeness: Some universals in language usage.
Bucholtz, M. (1999). ‘Why be normal?’ Language and identity practices in a
community of nerd girls. Language in Society, 28, 203–223.
Bucholtz, M., & Hall, K. (2003). Language and identity. In A. Duranti (Ed.), A
companion to linguistic anthropology (pp. 368–394). Oxford: Blackwell.
guistic approach. Discourse Studies, 7, 584–614.
Carter, R. (2004). Language and creativity: The art of common talk. London/New
York: Routledge.
Clancy, B. (2013). Do you want to do it yourself like? Hedging in Irish traveller

and settled family discourse. In B. L. Davies, M. Haugh, & A. J. Merrison
(Eds.), Situated politeness (pp. 129–146). London/New York: Bloomsbury.
Culpeper, J. (2009). The metalanguage of impoliteness: Using sketch engine to
explore the Oxford English corpus. In P. Baker (Ed.), Contemporary corpus
linguistics (pp. 64–86). London: Continuum.
Culpeper, J. (2011). Impoliteness: Using language to cause offence. Cambridge:
Culpeper, J. (2015). Epilogue: The “how” and the “what” of (im)politeness. In
M. Terkourafi (Ed.), Interdisciplinary perspectives on im/politeness
(pp. 267–275). Amsterdam/Philadelphia: John Benjamins.
Dayter, D. (2014). Self-praise in microblogging. Journal of Pragmatics, 61,
91–102.
Diani, G. (2015). Politeness. In K. Aijmer & C. Rühlemann (Eds.), Corpus prag-
matics: A handbook (pp. 169–191). Cambridge: Cambridge University Press.
Eelen, G. (2001). A critique of politeness theories. Manchester: St. Jerome.
Ellison, N. (2013). Social media and identity. Retrieved from https://www.gov.
uk/government/uploads/system/uploads/attachment_data/file/275752/13-
505-social-media-and-identity.pdf
Garcés-Conejos Blitvich, P. (2010a). Introduction: The status-quo and quo vadis
of impoliteness research. Intercultural Pragmatics, 7(4), 535–559.
Garcés-Conejos Blitvich, P. (2010b). The YouTubification of politics, impolite-
ness and polarization. In R. Taiwo (Ed.), Handbook of research on discourse
behavior and digital communication: Language structure and social interaction
Garcés-Conejos Blitvich, P., Lorenzo-Dus, N., & Bou-Franch, P. (2010). A
genre approach to impoliteness in a Spanish television talk show: Evidence
from corpus-based analysis, questionnaires and focus groups. Intercultural
Pragmatics, 7(4), 689–723.
Garcés-Conejos Blitvich, P., & Sifianou, M. (2017). (Im)politeness and identity.
In J. Culpeper, M. Haugh, & D. Kádár (Eds.), The Palgrave handbook of lin-
guistic (im)politeness (pp. 227–256). Basingstoke: Palgrave Macmillan.
García Vizcaíno, M. J. (2006). Using oral corpora in contrastive studies of lin-
guistic politeness. In E. Fitzpatrick (Ed.), Corpus linguistics beyond the word:
Corpus research from phrase to discourse (pp. 117–142). Amsterdam/New
York: Rodopi.
Georgakopoulou, A. (2013). Storytelling on the go: Breaking news stories as a
travelling narrative genre. In M. Hatavara, L.-C. Hydén, & M. Hyvärinen
(Eds.), The travelling concepts of narrative (pp. 201–224). Amsterdam/
Gibbs, R. W., Jr. (1994). The poetics of mind: Figurative thought, language, and
understanding. Cambridge: Cambridge University Press.
Gilpin, D. R. (2011). Working the Twittersphere: Microblogging as professional
identity construction. In Z. Papacharissi (Ed.), A networked self: Identity, com-
munity, and culture on social network sites (pp. 232–250). New York, NY:
Routledge.
Goffman, E. (1959). The presentation of self in everyday life. New York: Doubleday.
Halavais, A. (2014). Structure of Twitter: Social and technical. In K. Weller,
A. Bruns, J. Burgess, M. Mahrt, & C. Puschmann (Eds.), Twitter and society
(pp. 29–41). New York: Peter Lang.
Hardaker, C., & McGlashan, M. (2016). “Real men don’t hate women”: Twitter
rape threats and group identity. Journal of Pragmatics, 91, 80–93.
Hargittai, E., & Litt, E. (2011). The tweet smell of celebrity success: Explaining
variation in Twitter adoption among a diverse group of young adults.
New Media & Society, 13(5), 824–842. https://doi.org/10.1177/
1461444811405805
Haugh, M. (2010). Intercultural (im)politeness and the micro-macro issue. In
A. Trosborg (Ed.), Pragmatics across languages and cultures (pp. 139–166).
Belin/New York: Walter de Gruyter.
Haugh, M. (2013). Im/politeness, social practice and the participation order.
Journal of Pragmatics, 58, 52–72.
discourse. Language@Internet, 4(1), 1–37. urn:nbn:de:0009-7-7611, SSN
1860-2029. Retrieved from http://www.languageatinternet.de
Herring, S. C. (2013). Discourse in Web 2.0: Familiar, reconfigured and emer-
gent. In D. Tannen & A. M. Trester (Eds.), Discourse Web 2.0: Language in
the media (pp. 1–26). Washington, DC: Georgetown University Press.
Herring, S. C., Stein, D., & Virtanen, T. (2013). Introduction to the pragmatics
of computer-mediated communication. In S. C. Herring, D. Stein, &
T. Virtanen (Eds.), Pragmatics of computer-mediated communication
(pp. 3–32). Berlin: Mouton de Gruyter.
Hirschon, R. (2001). Freedom, solidarity and obligation: The socio-cultural
context of Greek politeness. In A. Bayraktaroğlu & M. Sifianou (Eds.),
Linguistic politeness across boundaries: The case of Greek and Turkish (pp. 17–42).
Honeycutt, C., & Herring, S. C. (2009). Beyond microblogging: Conversation
and collaboration via Twitter. In Proceedings of the forty-second Hawai’i
IEEE Press.
Huberman, B. A., Romero, D. M., & Wu, F. (2008). Social networks that matter:
Twitter under the microscope. Retrieved April 2, 2017, from https://arxiv.org/
pdf/0812.1045.pdf
Johnson, T. J., Zhang, W., Bichard, S. L., & Seltzer, T. (2011). United we stand?
Online social network sites and civic engagement. In Z. Papacharissi (Ed.), A
networked self: Identity, community, and culture on social network sites
(pp. 185–207). New York, NY: Routledge.
Johnstone, B. (2008). Discourse analysis (2nd ed.). Oxford: Blackwell.
Kádár, D. Z., & Haugh, M. (2013). Understanding politeness. Cambridge:
Kilgarriff, A., Baisa, V., Bušta, J., Jakubíček, M., Kovář, V., Michelfeit, J., et al.
(2014). The sketch engine: Ten years on. Lexicography, 1(1), 7–36.
Koutsantoni, D. (2005). Greek cultural characteristics and academic writing.
Journal of Modern Greek Studies, 23, 97–138.
Kövecses, Z. (2002). Metaphor: A practical introduction. Oxford and New York:
Leech, G. (2014). The pragmatics of politeness. Oxford: Oxford University Press.
Locher, M. A. (2010). Introduction: Politeness and impoliteness in computer-
mediated communication. Journal of Politeness Research, 6, 1–5. https://doi.
org/10.1515/JPLR.2010.001
Locher, M. A., & Watts, R. J. (2005). Politeness theory and relational work.
Journal of Politeness Research, 1, 9–33.
Marwick, A. E., & boyd, d. (2010). I tweet honestly, I tweet passionately:
Twitter users, context collapse, and the imagined audience. New Media &
Society, 13(1), 114–133. https://doi.org/10.1177/1461444810365313
Marwick, A. E., & boyd, d. (2011). To see and be seen: Celebrity practice on
Twitter. Convergence: The International Journal of Research into New Media
Technologies, 17(2), 139–158.
McEnery, T., Baker, P., & Cheepen, C. (2002). Lexis, indirectness and politeness
in operator calls. In P. Peters, P. Collins, & A. Smith (Eds.), New frontiers in
corpus research (pp. 53–69). Amsterdam: Rodopi.
Metzger, D. (2016). Highly tweetable: 5000+ awesome quotes to use on Twitter
and social media. CreateSpace Independent Publishing Platform.
Mikros, G. K., & Perifanos, K. A. (2013). Authorship attribution in Greek
tweets using multilevel author’s n-gram profiles. In E. Hovy, V. Markman,
C. H. Martell, & E. Uthus (Eds.), Papers from the 2013 AAAI spring sympo-
sium analyzing microtext (pp. 25–27). Stanford, CA: AAAI Press.
Mills, S. (2003). Gender and politeness. Cambridge: Cambridge University Press.
Norrick, N. (1993). Conversational joking: Humor in everyday talk. Bloomington:

Indiana University Press.
Ogiermann, E. (2009). On apologising in negative and positive politeness cultures.
Page, R. (2012). The linguistics of self-branding and micro-celebrity in Twitter:
Page, R. (2014). Saying ‘sorry’: Corporate apologies posted on Twitter. Journal
of Pragmatics, 62, 30–45.
Pak, A., & Paroubek, P. (2010). Twitter as a corpus for sentiment analysis and
opinion mining. Retrieved January 20, 2015, from http://deepthoughtinc.
com/wp-content/uploads/2011/01/Twitter-as-a-corpus-for-sentiment-
analysis-and-opinion-mining.pdf
Papacharissi, Z. (2012). Without you, I’m nothing: Performances of the self on
Twitter. International Journal of Communication, 6, 1989–2006.
Puschmann, C., Bruns, A., Mahrt, M., Weller, K., & Burgess, J. (2014).
Epilogue: Why study Twitter? In K. Weller, A. Bruns, J. Burgess, M. Mahrt,
& C. Puschmann (Eds.), Twitter and society (pp. 425–432). New York: Peter
Lang.
Puschmann, C., & Burgess, J. (2014). The politics of Twitter data. In K. Weller,
A. Bruns, J. Burgess, M. Mahrt, & C. Puschmann (Eds.), Twitter and society
(pp. 43–54). New York: Peter Lang.
Rogers, R. (2014). Foreword: Debanalising Twitter: The transformation of an
object of study. In K. Weller, A. Bruns, J. Burgess, M. Mahrt, & C. Puschmann
(Eds.), Twitter and society (pp. ix–xxvi). New York: Peter Lang.
Schnurr, S., Marra, M., & Holmes, J. (2008). Impoliteness as a means of con-
testing power relations in the workplace. In D. Bousfield & M. A. Locher
(Eds.), Impoliteness in language: Studies on its interplay with power in theory
and practice (pp. 211–229). Berlin & New York: Mouton de Gruyter.
Senft, T. M. (2012). Microcelebrity and the branded self. Retrieved from https://
www.academia.edu/3775110/Micro-celebrity_and_the_Branded_Self
Sifianou, M. (1992). Politeness phenomena in England and Greece. Oxford:
Clarendon Press.
Sifianou, M., & Tzanne, A. (2010). Conceptualizations of politeness and impo-
liteness in Greek. Intercultural Pragmatics, 7(4), 661–687.
Spencer-Oatey, H. (2005). Rapport management theory and culture.
Intercultural Pragmatics, 2–3(3), 335–346.
Squires, L. (2015). Twitter: Design, discourse, and the implications of public
text. In A. Georgakopoulou & T. Spilioti (Eds.), The Routledge handbook of
language and digital communication (pp. 239–255). London and New York:
Routledge.
Taylor, C. (2013). Negative politeness forms and impoliteness functions in insti-

tutional discourse: A corpus-based approach. In B. L. Davies, M. Haugh, &
A. J. Merrison (Eds.), Situated politeness (pp. 209–231). London/New York:
Bloomsbury.
Terkourafi, M. (2005). Beyond the micro-level in politeness research. Journal of
Politeness Research, 1(2), 237–262.
Terkourafi, M. (2011). From politeness1 to politeness2: Tracking norms of im/
politeness across time and space. Journal of Politeness Research, 7(2), 159–185.
Thurlow, C., Lengel, L., & Tomic, A. (2004). Computer mediated communica-
tion: Social interaction and the internet. London, Thousand Oaks, New Delhi:
Sage Publications.
Tsaliki, L. (2010). Technologies of political mobilization and civil society in
Greece: The wildfires of summer 2007. Convergence: The International Journal
of Research into New Media Technologies, 16(2), 151–161. https://doi.
org/10.1177/1354856509357675
Watts, R. J. (1992). Linguistic politeness and politic behaviour: Reconsidering
claims for universality. In R. J. Watts, S. Ide, & K. Ehlich (Eds.), Politeness in
language: Studies in its history, theory and practice (pp. 43–69). Berlin/New
York: Mouton de Gruyter.
Watts, R. J. (2003). Politeness. Cambridge: Cambridge University Press.
Watts, R. J., Ide, S., & Ehlich, K. (1992). Introduction. In R. J. Watts, S. Ide,
& K. Ehlich (Eds.), Politeness in language: Studies in its history, theory and
practice (pp. 1–17). Berlin/New York: Mouton de Gruyter.
Zafiropoulos, K., Antoniadis, K., & Vrana, V. (2014). Sharing followers in
e-government Twitter accounts: The case of Greece. Future Internet, 6,
337–358. https://doi.org/10.3390/fi6020337
Zappavigna, M. (2011). Ambient affiliation: A linguistic perspective on Twitter.
New Media & Society, 13(5), 788–806.
Zappavigna, M. (2012). Discourse of Twitter and social media: How we use
language to create affiliation on the web. London: Continuum International.
13
Pedagogy, Audience, and Attitudes:
Influencing University Students’
Metalinguistic Awareness About
Texting Practices
Rebecca Roeder, Elizabeth Miller,
and Pilar Garcés-Conejos Blitvich
13.1 Introduction
A pedagogical focus in university linguistics classes on evolving norms
of practice for digitally mediated communication (DMC) is an emer-
gent phenomenon that has thus far received little attention in the
research community. This chapter reports on a pilot study examining
the effect of pedagogy on metalinguistic awareness of emergent norms
of communication in virtual world spaces among undergraduates at a
large state university in the American Southeast using both quantitative
and qualitative analyses. The specific foci of analysis are audience aware-
ness and attitudes about appropriate use of language and particular lan-
guage forms in text messages. The qualitative analysis also revealed
students’ awareness of how particular language forms in DMC affect
R. Roeder (*) • E. Miller • P. Garcés-Conejos Blitvich

Department of English, University of North Carolina at Charlotte,
Charlotte, NC, USA
e-mail: rroeder@uncc.edu; ermiller@uncc.edu; pgblitvi@uncc.edu

https://doi.org/10.1007/978-3-319-92663-6_13
368 R. Roeder et al.
their presentation of self. This work is patterned after survey studies that
have found that students’ attitudes toward non-standard language vari-
eties, such as African American Vernacular English, can change after
taking a single linguistics class (e.g., Bowie & Bond, 1994; Bündgens-
Kosten, 2009; Smitherman & Villanueva, 2000). Our findings support
this conclusion, in part, given that the target class (described below) did
appear to lead to students’ increased metapragmatic awareness of audi-
ence and the accompanying need to shift between formal/informal (i.e.,
“textese”) language forms in DMC. We also found that students in the
target class demonstrated greater acceptance for the appropriateness of
informal language forms (i.e., “textese”) in DMC although we also
observed that the classes had little effect on attitudes about prescriptive
norms in general.
Section 13.2 situates the current work in the context of previous
research on pragmatic awareness, language attitudes, and the effectiveness
of pedagogy on student outcomes, and ends by posing the research ques-
tions that guide the analysis. Section 13.3 describes the methodological
approach taken. Sections 13.4, 13.5, and 13.6 present and interpret both
quantitative and qualitative results in the areas of audience awareness
(Sect. 13.4), attitudes to prescriptive norms (Sect. 13.5), and appropri-
ateness of language use in texting (Sect. 13.6). Section 13.7 presents
qualitative findings with respect to students’ awareness of presentation of
self. Finally, Sect. 13.8 provides answers, albeit tentative, to the research
questions and provides suggestions for future research.
13.2 Background
Texting, as a genre of DMC, has received significant attention from a
multiplicity of fields, including computer science (Cook & Stevenson,
2009), education (van Dijk, van Witteloostuijn, Vasić, Avrutin, & Blom,
2016), rhetoric and composition (Aziz, Shamim, Aziz, & Avais, 2013),
psychology (Grace, 2013; Holtgraves & Paul, 2013), and linguistics
(Baron, Squires, Tench, & Thompson, 2005; Crystal, 2008; Hard af
Segerstad, 2005; Spilioti, 2009, 2011, 2014; Thurlow & Brown, 2003).
However, little attention has been given to attitudes related to language
Pedagogy, Audience, and Attitudes: Influencing University… 369
use in texting practices (but see Thurlow, 2006), and even less to how
pedagogical intervention may have an impact on changing those
attitudes. To our knowledge, the only partially related study to the one
we are presenting here is the one by Grace (2013). Grace investigated
factors that might be associated with variations in textism use between
individual phone users and concluded that the Australian students
included in her subject pool were able to discern situations in which
textism use is appropriate and that textisms were avoided in students’
formal writing.
Exploring young adults’ attitudes toward texting practices emerged as
a useful way to tap into their pragmatic competence regarding awareness
of audience and appropriateness and their attitudes toward notions of
correctness. Often, pragmatic choices are made subconsciously and atti-
tudes toward correctness are based on societal notions of prescriptivism
regarding language use. Pragmatic or communicative competence was
initially defined by Hymes (1972) as the ability to communicate in real-
life situations. Thus, utterances should be not only grammatically correct
but also contextually appropriate. For his part, Bachman (1990, p. 90)
understands the notion of pragmatic competence, as “…illocutionary
competence, or the knowledge of the pragmatic conventions for per-
forming acceptable language functions, and sociolinguistic competence,
or knowledge of the sociolinguistic conventions for performing language
functions appropriately in a given context”.
Thus, the notion of appropriateness emerges as key to understanding
pragmatic competence. Although from a pragmatic standpoint there is
no one correct way to use language, speakers usually have clear notions of
what constitutes appropriate use of language in different genre practices.
The reason for this, according to Celce-Murcia and Olshtain (2000,
p. 20), is that
pragmatic competence relies very heavily on conventional, culturally

appropriate, and socially acceptable ways of interacting. These rules of
appropriacy result in regular and expected behaviors in language use…
within a given social and cultural group, people usually know what is
expected and what is considered appropriate behavior, and this knowledge
enables them to interpret the language uses they encounter.
The need to communicate appropriately in context drives pragmatic

development from early on (Matthews, 2014). As Matthews (2014, p. 2)
argues, finessing communicative skills takes us into adolescence and
beyond. At the same time, and just as it has been observed in L2 situa-
tions, grammatical and pragmatic development may not evolve hand in
hand in L1 language acquisition (Kasper & Rose, 2002). Grammatically
competent speakers may need to learn and relearn new norms of appro-
priateness as they engage in novel practices. This is certainly the case for
those genre practices mediated by digital technology, such as texting.
In the study discussed here, we planned to test the Noticing Hypothesis
(Schmidt, 1993) that argues that awareness needs to be explicitly raised
regarding pragmatic functions and contextual factors since these are often
not salient and thus not noticed. Also, we followed Kasper and Rose
(2002) and Rose (2005) in implementing an explicit approach to raising
pragmatic awareness. All of the language courses included in our study
aimed at (1) debunking the notion that some uses and varieties of lan-
guage are inherently better than others; however, only two of the courses
focused explicitly on (2) raising students’ awareness of their situated uses
of language in DMC contexts. More specifically, the target class gives
particular emphasis to notions of appropriateness as they relate to lan-
guage forms associated with texting.
As mentioned above, students’ attitudes toward texting practices played
a central role in our study. The extensive body of scholarly work on lan-
guage attitudes spans research from a number of fields, including com-
munication, discourse analysis, the social psychology of language, sociology
of language, sociolinguistics, and anthropological linguistics. Relevant
sociolinguistic research indicates that attitudes about language use are
deeply entrenched in prescriptive concepts of correctness and pragmatic
concepts of friendliness (Preston, 1999), as well as macrolevel social cate-
gories such as gender (e.g., Eckert & McConnell-Ginet, 1992; Kiesling,
1998) and ethnicity (e.g., Garrett, Coupland, & Williams, 2003).
Assessments regarding notions of pragmatic appropriateness and lan-
guage attitudes come together as both are interpreted within the con-
text of a moral order (Kádár & Haugh, 2013). Pragmatic interpretations
of appropriateness are essentially a judgment of the individual as a
social being (McConachy & Liddicoat, 2016) and attitudes toward
non-standard forms of language are often associated with the Other,

that is, those who are regarded as not only different but socially inferior
and even morally defective (Lippi-Green, 1997).
It is against the theoretical background presented in Sect. 13.2 that our
study seeks to answer these research questions:
1. Can pedagogical intervention raise students’ metapragmatic awareness

of how informal language forms used in DMC relate to audience?
2. Can pedagogical intervention change students’ attitudes toward lan-
guage use in DMC from a primarily prescriptive to a more descriptive
approach?
3. How do students demonstrate their understanding of language and
DMC in qualitative responses?
In order to answer these research questions, the methodology detailed in

Sect. 13.3 was devised.
13.3 Method
The data for the present study were collected by means of a survey that
was administered to students in three undergraduate linguistics classes at
the University of North Carolina at Charlotte (UNCC) during the Spring
2015 and Spring 2016 semesters. The survey was comprised of multiple-
choice questions about audience awareness and judgments of appropriate
language use in text messages, and students were asked to fill out the
survey at both the beginning of the term (pre-test) and the end of the
term (post-test). The two versions of the survey were identical apart from
additional questions that were included at the beginning of the pre-test
survey requesting information on demographic factors (students’ age,
gender, previous linguistics classes), as well as several questions designed
to measure a participant’s general level of social media and texting usage.
Self-reporting as a data elicitation tool has well-known advantages and
disadvantages. The strengths of this method include the ability to effi-
ciently acquire explicit responses to targeted questions. The method is
easily replicable and affords data that can be readily quantified. Although
a self-report, multiple-choice survey allows for only fixed choice questions

and answers, that limitation was mitigated in the current study by provid-
ing students with open boxes allowing for free commentary. Furthermore,
although self-reporting reflects perception instead of actual usage, this
bias matches the goals of the current study, which was to gain insight into
students’ perceptions of emergent norms of texting practices.
Only data from students who completed a questionnaire at both the
beginning and the end of the term are included in the analysis, resulting
in 56 participants during the first year of the study and 52 during the
second year (108 total participants), distributed across classes as shown
below in Table 13.1.
The UNCC English department’s Language and Digital Technology
concentration includes three core linguistics courses. Students from two
of these core classes were included in the current project, as follows:
• ENGL 3162-Language and the Digital World (Target course, referred

to as Lang & DMC in Table 13.1)
• ENGL 4168-Multimodality and Text Description (referred to as
Multimodality in Table 13.1)
Both of these classes include discussion of language use in the virtual world.
The control group in our study included students in the following class:
• ENGL 3132 (Introduction to Contemporary American English,

referred to as Ling Intro Survey in Table 13.1)
The Ling Intro Survey class was chosen as the control because norms of
language use in digital spaces are not directly addressed as a topic and
most of the study participants from this class reported not having taken
Table 13.1 Distribution of study participants by class and year (N = 108)

Class Spring 2015 Spring 2016
Ling intro survey 24 19
Lang & DMC 19 16
Multimodality 13 17
Total 56 52
any prior linguistics classes. This course serves as a general introduction to

the field of linguistic study covering topics such as phonology, morphol-
ogy, and syntax, along with some attention to dialects of American English
and notions of prescriptive versus descriptive approaches to language. In
contrast, Lang & DMC (the target course) focuses specifically on lan-
guage use and practices in digitally mediated contexts. The course descrip-
tion provided in the syllabus indicates that the class will examine “pop
culture attitudes about changing linguistic norms, structural manipula-
tion of digitally-mediated text to fit diverse purposes, use of the web as a
language corpus for social science inquiries, and norms of politeness in
digitally-mediated communication”. The Multimodality class devotes
half of the semester to the analysis of multimodality in digital genres, with
special attention to the construction of meaning and audience awareness.
Two of the course goals identified in the syllabus for this class include: (1)
“Students will familiarize themselves with the role of multimodality in
digital discourse”, and (2) “Students will produce original empirical, mul-
timodal analysis of digital discourse and incorporate existing scholarly
research into their interpretation”. As such, it is clear that the target
course, Lang & DMC, focuses most explicitly on developing students’
metapragmatic awareness (i.e., Noticing) related to purpose or audience
as well as to exploring norms of politeness or language appropriateness.
13.4 R
esults and Discussion: Judgments
of Appropriate Language Use
As noted in previous work (e.g., Haas, Takayoshi, Carr, Hudson, &
Pollock, 2011), the affordances of texting—orthography, emoticons and
emojis—uniquely facilitate novel modes of stylistic variation. The sur-
veys in our study included nine questions that elicited students’ percep-
tions of the appropriateness of some of these texting practices as well as
their self-reported usage of specific features of texting practices. A sample
survey question as well as the directions for answering such a question
appear below.
Sample sentence from survey:
Excerpt 1 Directions: Please answer based on imagining INFORMAL

texts. In other words, the rest of the questions only apply to messages
where you are not worried about being formal or proper (e.g., with very
close friends).
Abbreviated slang phrases: omg, that is weird. lol.
(a) I sometimes text this way, but it’s not correct, even for informal tex-
ting purposes.
(b) I sometimes text this way, and it is appropriate and fine for informal
texting purposes.
(c) I don’t text this way, but it is appropriate and fine for informal texting
purposes.
(d) I don’t text this way, and it’s inappropriate, even for informal texting
purposes.
Among the many ways in which written language can be modified in

texting, the current study focuses on only a few features—ellipsis, letters
for words, emoji usage, omission of punctuation, and so on, as presented
in Table 13.2, which shows the conflated end-of-term results for the two
terms. The results are ordered according to the features students deemed
most appropriate in informal texting (top) to those regarded as least
appropriate (bottom).
There was little change from the beginning to the end of each term, or
between years, in students’ evaluations of appropriateness when asked
about these specific forms, suggesting that perceptions of appropriateness
are both deeply entrenched and were largely unaffected by classwork.
That said, differences in the evaluations of four or more students did
occur between 2015 and 2016 for question examples B8, B15, and B9,
as shown in Table 13.3.
Exit survey (B) sentence 14 See you tonight is accepted as appropriate
by all respondents in 2016. Although this sentence is informal in that it
includes dropping of the initial pronoun I and the auxiliary verb will, it
does not use the purely text-based modifications that the other three sen-
tences in Table 13.3 use (i.e., letters for words, numbers for words, and
Table 13.2 Students’ attitudes toward appropriateness of texting features, Spring

2015 and Spring 2016 combined results (N = 108)
Appropriate for Not appropriate,
informal even for informal
purposes (b, c) purposes (a, d)
Question examples Feature % n % n
B11: Thanks! Emoji 97.2 105 2.8 3
B13: I’m leaving Ellipsis w/dots 95.4 103 4.6 5
work now… see
you tonight
B14: See you Ellipsis (no dots) 94.4 102 5.6 6
tonight
B10: omg, that is Abbreviation 93.5 101 6.5 7
weird. lol
B7: im leaving No caps or 83.3 90 16.7 18
work now see punctuation
you tonight
B8: I’m leaving Letters for words 82.4 89 17.6 19
work now. c u
tonight
B12: HELLO!!!! Are Modification for 80.6 87 19.4 21
you there? emphasis
B15: I’ll see you # for 79.6 86 20.4 22
2nite sound + diff.
spelling
B9: im leaving Letters for 77.8 84 22.2 24
work now c u words + no
tonight punc.
lack of punctuation). These spelling shortcuts are entirely visual, and

were likely more prevalent when words had to be completely typed out
letter by letter, which involved laboriously scrolling through various
options on one button. Although usage-based evidence on the correla-
tion between technological developments in smart phones and changing
orthographic norms is beyond the scope of the current research, and the
sample is small, it is possible that the difference in attitudes about the
texting features included in questions B8, B15, and B9 at least in part
reflects the changing norms due to changing technology. With the devel-
opment of smartphone affordances such as predictive spelling, speech
recognition software, and better keyboards, orthographic modifications
such as letters or numbers for words are becoming less common (see
Drouin & Driver, 2014) and we found that there was a slight decrease
Table 13.3 Change in students’ attitudes about appropriateness of texting fea-

tures, from Spring 2015 to Spring 2016
Appropriate for Appropriate for
informal purposes informal purposes
2015 (N = 56) 2016 (N = 52)
Question Feature % n % n
B14: See you Ellipsis 89 50 100 52
tonight
B8: I’m leaving Letters for 89 50 76.5 39
work now. c u words + ellipsis
tonight
B15: I’ll see you # for sound + diff. 84 47 76.5 41
2nite spelling
B9: im leaving Letters for 82 46 74.5 38
work now c u words + no punc.
tonight
in the number of students who deemed these features as “Appropriate for

informal purposes” from 2015 to 2016. Further support for this conclu-
sion comes from the observation that the only difference between See you
tonight (B14) and the following row in Table 13.3, c u tonight (B8) is the
use of letters for words and the lack of capitalization of the first word.
However, unlike See you tonight (B14), the exit survey showed a drop in
the appropriateness rating for c u tonight (B8) from the beginning of the
semester to the end of the semester for both 2015 (7% drop) and 2016
(25.5% drop), which can only be interpreted as a decline in the perceived
appropriateness of the use of letters for words.
Several comments from students, shown below, directly addressed the
changing affordances of technology and how it affects their lack of use of
stereotypical texting features:
• With the emergence of smart phones with “autocorrect” capabilities,

one almost has to try to purposefully use texts such as “2nite” and “C
U l8er.” It’s easier now to use correct grammar and punctuation than
it used to be.1
• I would probably never say 2nite instead of tonight (1) because it is
basically the same amount of characters (2) it takes more work to get
to the numbers of my phone and (3) autocorrect helps where all I have
to type sometimes is “to” or “ton” and it will suggest tonight.
• It is harder to type things such as C U 2nite because of autocorrect. It’s

much easier to just type what I’m saying correctly.
Overall, the findings presented in Table 13.2 indicate a high tolerance for

non-standard forms in DMC, regardless of classroom exposure. For example,
the form that was ranked lowest in terms of appropriateness was still consid-
ered appropriate for informal purposes by 77.8% of students. It seems that
even with the advent of technological changes that appear to be leading to a
decrease in university students’ use of some texting features, such as letters for
words (“c u tonight”) or numbers for sounds (“2nite”), other “informal” lan-
guage forms will likely persist and be deemed appropriate for DMC practices
(see also Drouin, 2011; Drouin & Davis, 2009; Drouin & Driver, 2014).
13.5 R
esults and Discussion: Metapragmatic
Awareness and Audience
As illustrated in Table 13.2 above, students in all three classes demonstrated
significant pragmatic awareness of audience with respect to DMC already
at the beginning of each term. However, as is illustrated in Table 13.4
below, they did display greater metapragmatic sensitivity in their DMC
practices in terms of differentiating among audience types by the end of
each term, particularly those students in the target Lang & DMC class.
The following survey question specifically targeted students’ metaprag-
matic awareness about their language usage in DMC in relation to their
intended audience:
Excerpt 2 I would make a point of using proper grammar with…
Table 13.4 Change in metapragmatic awareness in relation to audience

Class 2015 pre → post 2015 % ∆ 2016 pre → post 2016 % ∆
Ling intro survey 37.5 → 41.7 4.2 31.6 → 47.4 15.8
Lang & DMC 31.6 → 57.9 26.3 25 → 62.5 37.5
Multimodality 53.8 → 76.9 23.1 41.2 → 52.9 11.7
Percentage of respondents, by class and year, who indicated awareness of
different texting methods in communications with intimates as compared to
superiors/strangers
(a) my parents
(b) my boss
(c) people older than me whom I don’t know very well
(d) people my own age whom I don’t know very well
(e) my close friends
(f ) my siblings
(g) everyone but a few people whom I know really well
(h) I use proper grammar with everyone I text
Table 13.4 presents the percent change from beginning to end of the
semester in students’ self-reports of “proper grammar” usage, based on their
intended audience. Responses (g) “everyone but a few people whom I know
really well” and (h) “I use proper grammar with everyone I text” are not
included in the calculations displayed in Table 13.4. An increase in percent-
age from the beginning of the term (left side of the arrow) to the end of the
term (right side of the arrow) therefore indicates students’ increased recog-
nition of pragmatic differentiation based on audience (e.g., parents, boss,
stranger, etc.). In other words, if the percentage is higher on the right side
of the arrow, which is true in every cell in Table 13.4, that means that more
respondents chose responses (a) through (f), instead of only (g) or (h), on
the post-test than on the pre-test, thus pointing to their heightened
metapragmatic awareness that they shift styles according to their audience.
The Ling Intro Survey class demonstrates some increased awareness,
especially in 2016 (15.8%). The Multimodality class shows the highest
rate of awareness at the beginning of each term and some change during
the term (23.1% and 11.7% respectively). However, the target class
(Lang & DMC), which focused specifically on DMC practices, demon-
strates the greatest change for both years (26.3% and 37.5%, respec-
tively). Due to the small sample size (N = 108), these numbers must be
taken as preliminary only. For example, the 23.1% change noted in 2015
for the Multimodality class reflects an increase in just 3 of 13 respon-
dents, and Chi square tests indicate that none of the differences listed in
Table 13.5 rise to the level of significance at p < 0.05. Nonetheless, these
findings do suggest that metalinguistic awareness can improve due to
targeted class work, and they warrant further investigation.
Student comments provide further evidence of their metapragmatic
awareness in relation to audience:
Table 13.5 Self-reporting of “I would make a point of using proper grammar

with everyone”
Spring 2015 Spring 2016
Class n Beg. End % change n Beg. End % change
Ling intro survey 24 9 8 −8.3 19 8 5 −15.8
Lang & DMC 19 7 3 −21 16 5 1 −25
Multimodality 13 4 1 −23.1 17 8 6 −11.8
• I would only change my texting if the context of the conversation was

in a more serious setting. If I am texting my Boss about a metting
tomorrow I will used proper spelling and grammar. If I’m texting my
roommate about going to the bar tomorrow then I won’t.
• When speaking to a boss or someone I don’t know well, I would use
correct grammar, and spelling, because I don’t want to appear stupid
or lazy. With friends and family, I use all of the text slang because it
doesn’t matter.
• I would only use emoticons with my significant other.
• Some of my close friends will text me using proper grammar, to which
I will respond with proper grammar. If someone texts me using infor-
mal language, I’m much more comfortable and likely to use informal
language back.
• I probably wouldn’t use emojis or “omg” if I was texting someone older
than me who I didn’t know very well unless they used them first.
Although to a lesser degree than accommodation for audience in terms

of their use of (in)formal language forms, some respondents also indi-
cated a conscious choice to use more formal texting language with par-
ents and peer strangers in order to be understood by their texting
interlocutors. Below are four comments from students about audience
accommodation for comprehension:
• I don’t use much abbreviation when texting, especially to my Dad who

might not understand the abbreviation.
• I wouldn’t use text speak with my parents because they wouldn’t under-
stand and then I’d have to explain it.
• Some of my friends or people I text aren’t text savy, so I’ll adjust how I
type with them. I wouldn’t say “c u tonight” to my dad b/c he might
not understand it. One of my friends would refuse to respond until I
wrote it out correctly.
• If I thought texting in specific ways would be confusing to the reader/
addressee I would adjust. I’d be more likely to use standard grammar.
The findings from our study lend support to research reporting on the
practice of shifting styles when interacting via social media with different
audiences (e.g., Pavalanathan & Eisenstein, 2015). Though we do not
have data related to students’ actual texting practices, it seems that the
students in the target class, who read and discussed multiple studies
focusing on DMC practices, developed greater awareness of how audi-
ence affects their choice of language forms. As such, our study indicates
that explicit attention to the pragmatic functions of language in DMC
can enhance students’ metapragmatic awareness, thus lending support
for the Noticing Hypothesis (Schmidt, 1993).
13.6 R
esults and Discussion: Attitudes
to Prescriptive Norms
Despite the evidence shown in Sect. 13.4 of students’ pragmatic compe-
tence in recognizing the appropriateness of texting features for informal
language use in DMC, along with indications of their developing
metapragmatic awareness in their choices to alter their DMC practices
according to audience (Sect. 13.5), we found that prescriptive expecta-
tions still often governed students’ metalinguistic evaluations of specific
examples, as illustrated in the following comments.
Comments from students about prescriptive expectations:
• Words should be spelled correctly; there is no reason to spell a word

incorrectly on purpose.
• Using poor grammar and/or poor spelling is never okay.
Comments from students about texting as bad grammar (see also

Thurlow, 2006):
• I work in the writing center on campus and I have seen people come
in with academic papers filled with text slang, thinking that it’s ok to
use in a formal paper.
• Although texting can help with writing and reading skills, texting can
lead to bad grammar as well. The younger generation that have had
technology like smart phones & autocorrect their entire lives will have
problems when they aren’t able to rely on their technology.
However, despite the persistence of prescriptive norms among students

across the three courses, we found that students from the target class
(Lang & DMC) were more likely than students from the other classes to
show evidence of a developing awareness that prescriptive norms need
not be adhered to in DMC language practices and that lack of adherence
should not be treated as grounds for negatively evaluating language users.
In fact, 56 percent of students in the target class (Lang & DMC) reported
that the course changed their perceptions regarding what is appropriate
use of language in texting but only 27 percent of students in the
Multimodality classes and only 19 percent in the Ling Intro Survey class
made the same claim. Furthermore, as shown in Table 13.5, by the end
of the 2015 term only 3 of 19 respondents in the target class still indi-
cated that they used “proper” grammar in all of their text messages (a
drop of 21%), and the 2016 number was even lower at 1 of 16 by the end
of the term (a drop of 25%). The multimodality class showed a similar
drop in 2015 (a drop of 23.1%), but the introductory linguistics class,
which provided the least amount of explicit discussion of DMC, showed
less than 20% change both terms.
The findings from the students in the target class also support the evi-
dence discussed in Sect. 13.6 which indicated that students in this class
displayed clearer differentiation by the end of the term regarding whom
they would make a point of using proper grammar with; that is, they
indicated that they were more likely to use “proper grammar” only with
a boss or an older stranger but not with friends and acquaintances in their
age cohorts.
Comments from Lang & DMC students about their changing atti-
tudes toward texting usage:
• I definitely don’t have as much of a prejudice regarding textese after

taking this class.
• I know now that using different spelling and grammar doesn’t define
one’s reading/writing ability. It’s okay to text however you want b/c
most of the time texts are informal.
• I believe at the beginning of the semester I was a bit more rigid in my
views. I have seen things a bit differently since the first survey and even
have relaxed my own texting habits…a bit (smiley face)
• I’ve learned that poor grammar & miss-spellings are not necessarily
bad & that texting is not ruining our youth. It is not making people
change the way they write or speak.
• I used to have a little bit more of a problem, I think, with people being
horrible with grammar through texting. It wasn’t a big issue for me,
but I think now it’s not an issue at all.
• I feel like my perspective of how texting affects people has changed.
Before I thought it made people do worse with english writing but
now I think it actually helps and lets people be creative. Also, not as
many people use drastic use of textese.
• My POV on the way people type online/in texts has changed. “2nite”
used to annoy me no end, but while I wouldn’t use it even now, I can
tolerate seeing it. Language is so versatile & varied, and there are so
many fun ways to play with it. Why limit yourself? (smile face)
The pedagogical goal, therefore, is to engage students in such a way

that the pragmatic competence with which they enter the classroom cor-
relates with their metapragmatic awareness about language usage and
prescriptive norms. Raising students’ awareness in relation to DMC is
not necessarily undertaken with any attempt to change their current
practices but rather to help them understand the role of audience and
purpose in influencing all of their writing choices and to enable them to
critique the role of prescriptive norms across all genres and contexts.
Though the results from our study show promise in relation to the role of
pedagogy in helping students to recognize and even resist powerful
prescriptive norms in evaluating language use in DMC contexts, the stu-

dents in general seemed to remain committed to socially validated beliefs
that standard, educated language forms are always better.
13.7 R
esults and Discussion: Presentation
of Self
The flip side of audience awareness, and a corollary of judgments of
appropriateness and norm adherence, is presentation of self, that is, desir-
ing to create a positive image of self through humor, through establishing
rapport, sometimes through accommodation to audience, sometimes
through usage of “proper grammar”. Although this topic was not explic-
itly addressed in the multiple-choice section of the survey, the open spaces
for comments provided students with opportunities to provide additional
explanatory information to support their responses, and in doing so, they
revealed their metalinguistic awareness of the potential for stylistic varia-
tion in DMC to serve as a resource for the positive presentation of self.
For example, students demonstrated an awareness that texting features
can be deployed to create a humorous or “cute” (see first comment below)
image of themselves.
Comments from students about humor in texting as presentation of self:
• I really only use textese if I’m intentionally trying to be cute or funny

and usually it is with people who understand I’m being silly right away.
I also use “wanna” and “gonna” if I’m trying to be cute & connect
w/certain friends.
• I generally always use (mostly) proper grammar but sometimes I use
incorrect grammar when being humorous.
• I will use improper grammar as a form of inside joke. Otherwise, I use
proper grammar exclusively.
• I don’t like when people put “ur” or “u” (Question #14) But, sometimes
I’ll spell things wrong in a text in terms of trying to be funny like “gurl”.
Scholars have long noted the important role of humor in DMC

(Baron, 2007; Baym, 1995; Crystal, 2008). In fact, Vandergriff (2010,
p. 237) views humor as a central feature of DMC, achieved to a large
extent through “form-based language play” in which orthographic, spell-

ing, and grammar features are manipulated for humorous effect. Such
uses of language play, can, as she puts it, “support the presentation of
positive face” (p. 243). They can also contribute to the “establishment of
personal rapport” (Vandergriff, 2010, p. 243), another important com-
ponent of a positive presentation of self. As shown below, students’ com-
ments also indicated that they view the use of texting features as a resource
for creating and maintaining friendly relationships.
Comments from students about rapport:
• The only time when I use textisms is when joking with a specific friend.
• I believe my approach is appropriate for me because when I text like
“sup brah” or anything like that it show me I am comfortable with the
person.
• Even though I do not naturally use “lol” or emojis, if the other person
does, I will, just to they do not think I am acting too superior.
At the same time, students demonstrated metapragmatic awareness of

how the injudicious use of some texting features might contribute to
creating negative images of self, leading them to avoid some informal
usages while incorporating others. Thus, students showed nuanced aware-
ness of the effects of language forms in DMC on self-presentation, such
as whether one should or should not use a period at the end of a text
message (see first comment below) to avoid “sound[ing] very angry”.
Comments from students that discuss avoidance of negative
self-presentation:
• I prefer to spell out words with the exception of a few colloquialisms,

such as “wtf ”, or droppin’ the “g” off of “ing” words. I also pay very
close attention to whether or not I should end the text with a period
vs. no punctuation. The period can easily make you sound very angry.
However, in some situations, NO period can make you sound angry.
• I have to be careful in my choice of words/lack of words, i wouldn’t
want to send the wrong message/tone of voice implied by the use of
caps. it may not have signficant consequences but the way i type/write
perform identity & I’m pretty aware of that.
• If I am trying to impress someone or seem older/more professional I

would not use those abbreviations.
• Some of the alterations like part of speech deletion can end up sound-
ing short or rude, which my parents would not take that way but my
friends would.
13.8 Conclusion
The main goal of our analysis was to respond to the research questions
that guided it:
1. Can pedagogical intervention raise students’ metapragmatic aware-

ness of how informal language forms used in DMC relate to audience
and/or purpose?
As was illustrated in Table 13.2, students in all three classes demon-

strated significant pragmatic awareness of audience with respect to DMC
already at the beginning of each term. However, as shown in Table 13.4,
they did display greater metapragmatic sensitivity in terms of differentiat-
ing among audience types in their DMC practices by the end of each
term, particularly those students in the target Lang & DMC class.
2. Can pedagogical intervention change students’ attitudes toward lan-

guage use in DMC from a primarily prescriptive to a more descriptive
approach?
Our observations support previous findings that explicit instruction

can lead to increased awareness of pragmatic norms (Kasper & Rose,
2002) and thus support the Noticing Hypothesis (Schmidt, 1993). We
tentatively conclude that meta-discussion, such as that which takes place
in a university classroom focused on DMC, can positively affect language
ideology and awareness in an abstract way, though we have no evidence
to show whether it has any effect on actual language usage.
3. How do students demonstrate their understanding of language and

DMC in qualitative responses?
While students’ voluntary comments indicated a persistent perception

that prescriptive norms of language should generally be adhered to, stu-
dents in the target class demonstrated that these powerful norms can be
rethought and challenged when exposed to studies and class discussions
that treat these practices as appropriate depending on one’s communica-
tive purpose and audience. These changing perspectives came through
most clearly in the comments produced by several students. They point
to an incipient change in students’ attitudes to “incorrect” DMC prac-
tices; that is, they revealed students’ awareness that such practices are not
necessarily “bad” or “ruining our youth”, as one student put it. In
addition, students’ qualitative comments in the surveys provide evidence
of their awareness of the pragmatic functions of DMC features such as
emojis and non-standard spelling in creating humor and maintaining
rapport, features which are important in their presentation of self.
It seems likely that ongoing technological changes will continue to
affect the affordances for language use in texting and other social media
and that students’ participation in a wide variety of DMC genres will only
increase (Drouin & Driver, 2014). As such, there is an ongoing need to
cultivate students’ metapragmatic awareness of their language use in DMC
according to audience and in relation to their attitudes toward prescriptive
norms. There is also a concomitant need to address students’ actual usage
and to incorporate their DMC productions into analytic projects in order
to enable them to recognize, analyze, and evaluate their own practices.
Note
1. Students’ comments are presented here exactly as they were written in the
questionnaires.
References
Aziz, S., Shamim, M., Aziz, M. F., & Avais, P. (2013). The impact of texting/
SMS language on academic writing of students: What do we need to panic
about? Elixir Linguistics and Translation, 55, 12884–12890.
Bachman, L. F. (1990). Fundamental considerations in language testing. Oxford:

Baron, N. S. (2007). My best day: Presentation of self and social manipulation
in Facebook and IM. Internet Research, 8, 17–20.
Baron, N. S., Squires, L., Tench, S., & Thompson, M. (2005). Tethered or
mobile? Use of away messages in instant messaging by American college stu-
dents. In R. Ling & P. E. Pedersen (Eds.), Mobile communications:
Re-negotiation of the social sphere (pp. 293–311). London: Springer.
Baym, N. K. (1995). The performance of humor in computer-mediated com-
munication. Journal of Computer-Mediated Communication, 1(2).
Bowie, R. L., & Bond, C. L. (1994). Influencing teachers’ attitudes toward
Black English: Arewe making a difference? Journal of Teacher Education, 45,
112–118.
Bündgens-Kosten, J. (2009). Teachers’ attitudes toward African American
Vernacular English:Influence of contact with linguistics on ambivalent attitudes.
Unpublished doctoral dissertation, RWTH Aachen University, Aachen,
Germany.
Celce-Murcia, M., & Olshtain, E. (2000). Discourse and context in language
teaching: A guide for language teachers. Stuttgart: Ernst Klett Sprachen.
Cook, P., & Stevenson, S. (2009). An unsupervised model for text message nor-
malization. In Proceedings of the Workshop on Computational Approaches to
Linguistic Creativity (pp. 71–78). Association for Computational Linguistics.
Crystal, D. (2008). Txtng: The Gr8 Db8. Oxford: Oxford University Press.
Drouin, M. A. (2011). College students’ text messaging, use of textese and lit-
eracy skills. Journal of Computer Assisted Learning, 27, 67–75.
Drouin, M. A., & Davis, C. (2009). R U texting? Is the use of text speak hurting
your literacy? Journal of Literacy Research, 41, 46–67.
Drouin, M. A., & Driver, B. (2014). Texting, textese and literacy abilities: A
naturalistic study. Journal of Research in Reading, 37, 250–267.
Eckert, P., & McConnell-Ginet, S. (1992). Think practically and look locally:
Language andgender as community-based practice. Annual Review of
Anthropology, 21, 461–490.
Garrett, P., Coupland, N., & Williams, A. (2003). Investigating language atti-
tudes, socialmeanings of dialect, ethnicity and performance. Cardiff: University
of Wales Press.
Grace, A. A. S. (2013). Mobile phone text messaging language: How and why
undergraduates use textisms. Unpublished doctoral dissertation, University of
Tasmania, Australia.
Haas, C., Takayoshi, P., Carr, B., Hudson, K., & Pollock, R. (2011). Young
people’s everyday literacies: The language features of instant messaging.
Research in the Teaching of English, 45, 378–404.
Hard af Segerstad, Y. (2005). Language in SMS: A socio-linguistic view. In
R. Harper, L. Palen, & A. Taylor (Eds.), The inside text: Social, cultural and
design perspectives on SMS (pp. 33–52). Norwell, MA: Springer.
Holtgraves, T., & Paul, K. (2013). Texting versus talking: An exploration in
telecommunication language. Telematics and Informatics, 30, 289–295.
Hymes, D. (1972). On communicative competence. In J. B. Pride & J. Holmes
(Eds.), Sociolinguistics (pp. 269–293). Halmondsworth: Penguin.
Kádár, D. Z., & Haugh, M. (2013). Understanding politeness. Cambridge:
Kasper, G., & Rose, K. R. (2002). Pragmatic development in a second language.
Oxford: Blackwell.
Kiesling, S. (1998). Men’s identities and sociolinguistic variation: The case of
fraternity men. Journal of Sociolinguistics, 2, 69–99.
Lippi-Green, R. (1997). English with an accent: Language, ideology, and discrimi-
nation in the United States. New York, NY: Routledge.
Matthews, D. (Ed.). (2014). Pragmatic development in first language acquisition
(Vol. 10). Amsterdam: John Benjamins Publishing Company.
McConachy, T., & Liddicoat, A. J. (2016). Meta-pragmatic awareness and
intercultural competence: The role of reflection and interpretation in inter-
cultural mediation. In F. Dervin & Z. Gross (Eds.), Intercultural competence
in education (pp. 13–30). London: Palgrave Macmillan.
Pavalanathan, U., & Eisenstein, J. (2015). Audience-modulated variation in
online socialmedia. American Speech, 90, 187–213.
Preston, D. (Ed.). (1999). Handbook of perceptual dialectology (Vol. 1).
Amsterdam: John Benjamins.
Rose, K. R. (2005). On the effects of instruction in second language pragmatics.
System, 33, 385–399.
Schmidt, R. (1993). Consciousness, learning and interlanguage pragmatics. In
G. Kasper (Ed.), Interlanguage pragmatics (pp. 21–42). Oxford: Oxford
University Press.
Smitherman, G., & Villanueva, V. (2000). Language knowledge and awareness
survey: Finalresearch report. Urbana, IL: Conference on College Composition
and Communication, National Council of Teachers of English.
Spilioti, T. (2009). Graphemic representation of text-messaging: Alphabet-
choice and code-switches in Greek SMS. Pragmatics, 19, 393–412.
Spilioti, T. (2011). Beyond genre: Closings and relational work in text-messaging.

In C. Thurlow & K. Mroczek (Eds.), Digital discourse: Language in the new
media (pp. 67–85). Oxford: Oxford University Press.
Spilioti, T. (2014). On the mobile again! Opening sequences in Greek text-
messaging. In M. Christodoulidou (Ed.), Analysing Greek talk-in-interaction
(pp. 240–261). Newcastle Upon Tyne: Cambridge Scholars Publishing.
Thurlow, C. (2006). From statistical panic to moral panic: The metadiscursive
construction ofpopular exaggeration of new media language in the print
media. Journal of Computer-Mediated Communication, 11, 667–701.
Thurlow, C., & Brown, A. (2003). Generation Txt? The sociolinguistics of
young people’s text-messaging. Discourse Analysis Online. Retrieved May 11,
2017, from http://extra.shu.ac.uk/daol/articles/v1/n1/a3/thurlow2002003-
paper.html
van Dijk, C. N., van Witteloostuijn, M., Vasić, N., Avrutin, S., & Blom, E.
(2016). The influence of texting language on grammar and executive func-
tions in primary school children. PLoS One, 11, e0152409.
Vandergriff, I. (2010). Humor and play in CMC. In R. Taiwo (Ed.), The hand-
book of research on discourse behavior and digital communication: Language
structures and social interaction (pp. 235–251). Hershey, NY: Information
Science Reference.
Index1
A Androutsopoulos, J., 4, 5, 11, 12,

Addressivity, 179, 346 14, 28, 29, 34, 37, 38, 40,
Advice column, 12, 219, 220, 49n7, 73, 74, 283, 344,
223–225, 227, 231, 245 350–353, 359n11
Affiliation, 14, 252, 263, 352, 354 Annotation, 73–77, 80, 88, 95–97
Affordances, 6, 9, 10, 13, 25, 42, 72, Anonymity, 34, 145, 164, 170,
74, 88, 90, 93, 107, 124, 134, 195, 196, 227, 269, 290,
136, 147, 164, 165, 181–183, 305n2, 345
195, 196, 213, 253, 257, 261, Archakis, A., 350, 355, 356
263–265, 269–272, 273n1, Asymmetry, 195, 221, 228
274n3, 349, 356, 357, 373, Audience, 11, 16, 17, 98n2, 182,
375, 376, 386 196, 197, 224, 235, 252, 256,
Agency, 13, 16, 106, 196, 207, 210, 258, 259, 263, 272, 345, 350,
211, 254, 255, 257, 259, 260, 351, 355–356, 367–386
262–264, 313–334 awareness, 17, 367–369, 371,
Aggression, 288, 299–302, 304 373, 377, 383, 385
Algorithm, 9, 134, 136, 200, 262, Audio, 7, 25, 33, 35, 36, 38, 40–42,
263, 270, 273 45–47, 48n1, 76, 106
Amazon, 12, 193–213 Avatars, 6, 7, 26, 42, 43, 45
Note: Page numbers followed by ‘n’ refer to notes.

1

https://doi.org/10.1007/978-3-319-92663-6
392 Index
B Computer-mediated discourse
Blogs, 36, 39, 40, 44, 72, 77, 78, (CMD), 4, 5, 26–30, 34–40,
137, 138, 168, 196, 199, 206, 42, 44, 46–48, 49n6, 49n7
214n2, 223, 252, 257, 352, Computer-mediated discourse
355, 358n10 analysis (CMDA),
Bou-Franch, P., 3, 9, 10, 12, 37, 74, 5, 25–48, 49n2
182, 343, 344, 355 Conflict, 27, 288, 289, 297, 300,
boyd, d., 11, 259, 262, 266, 313, 305, 317, 318, 356
345, 346, 351, 354, 355, Connectivity, 86, 262, 263, 265,
359n13 269, 270, 272, 274n3
Bucholtz, M., 10, 11, 33, 194, 199, Content analysis, 28, 34, 35, 318
224, 348, 355, 356 Conversation analysis, 27, 28, 34,
Butler, J., 37, 199, 213, 320, 331, 73, 83, 88, 136
334n1 Corpus linguistics, 27, 265, 343
Counseling, 219, 220, 224, 227,
228, 231, 235–237, 239–241,
C 244, 246
Channel, 25, 36, 38, 45, 47, 48n1, Credibility, 197, 220, 223–226, 228,
94, 147, 153, 157, 181, 183, 242, 245, 246, 247n2, 255,
264, 272, 304, 345 256, 264, 268
Chats, 8, 78, 164, 173, 177–181, Critical discourse analysis, 11, 27, 136
183, 185n5 Crystal, D., 73, 368, 383
Cognitive pragmatics, 8 Cyberpragmatics, 8
Comics, 7, 9, 46, 111, 112, 120,
124, 128
Communities, 4, 8, 11, 14, 15, 27, D
28, 34, 35, 39, 40, 47, 76, Datafication, 262, 264, 265, 269,
164, 165, 169–173, 179, 271, 272
181–183, 196, 220, 223, 228, Davies, B., 141, 143, 194, 213, 224,
236, 241–244, 252, 260, 263, 237, 284, 287, 316
274n3, 284, 289, 290, 305n1, De Fina, A., 10
347, 367 Digitally mediated communication
Computer-mediated communication (DMC), 367, 368, 370–373,
(CMC), 4, 5, 7, 13, 25–48, 72, 377, 378, 380–386
74, 75, 164–168, 172, 177, Discourse analysis, 4, 6, 9, 10, 12,
182, 183, 195, 198, 225, 258, 14, 15, 18, 25, 28, 29, 34, 35,
283, 287–290, 302, 344, 345 37, 46, 47, 73, 74, 135, 170,
modes of, 5, 9, 26, 30–34, 36, 47, 173, 179, 197, 199, 246, 273,
106, 165, 168 317, 320, 328, 330, 344, 370
Index
393
Discursive psychology, 74, 224, 237 G

Disembodiment, 72, 349, 358n6 Games, 6, 8, 42, 43, 50n10, 322
Garcés-Conejos Blitvich, P., 10, 12,
17, 37, 74, 342–344, 355, 367
E Gaze, 7, 28, 43, 73, 75–77, 79, 80,
Email, 13, 30, 31, 36, 39, 48, 89–95, 98n2
48n1, 168, 169, 195, 219, Gee, J. P., 4, 6, 7, 48, 50n10
220, 226, 227, 231, 235–240, Gender
253, 273n1 difference, 8, 165, 166, 168, 169,
Embeddedness, 3, 230–232, 245 173, 182, 183, 213, 320, 344
Emoji, 8, 26, 43, 44, 47, 48, identity, 12, 33, 208, 322, 331, 334
167–169, 172, 175, 179–181, ideology, 11, 12, 194, 199, 204,
183, 373, 374, 379, 384, 386 211–213
See also Emoticon; Graphicon inequality, 209, 212, 326
Emoticon, 8, 39, 43, 44, 47, 195, politics, 12, 206, 208, 210, 212
373, 379 role, 166, 169, 193–195, 197,
See also Emoji; Graphicon 207, 211, 213, 330
Ethics, 15, 36, 228, 288, 290–291 stereotype, 182, 194, 198, 204,
Ethnography, 7, 8, 27, 28, 46, 74, 283 205, 207, 210, 323
Expertise, 13, 219–221, 225, Genre, 5, 6, 9, 14, 18, 27, 71, 72,
227–232, 235–246, 247n1, 78, 133–155, 166, 197–199,
247n2, 354 213, 253, 274n4, 368–370,
See also Identity, expert 373, 382, 386
Georgakopoulou, A., 4–6, 8, 10, 11,
14, 15, 34, 73, 177, 283, 357
F GIFs, 8, 43, 44, 167
Face, 10–12, 14, 27, 44, 90, 93, Goffman, E., 14, 74, 199, 251, 255,
98n2, 114–116, 118, 119, 121, 257, 258, 285, 289
150, 168, 169, 175, 177–180, Google, 30, 71, 110, 254, 256, 352,
183, 185n5, 220, 224, 258, 358n10
261, 283–305, 382, 384 Graphicon, 8, 43, 45
Facebook, 11, 14, 40, 42, 45, 108, See also Emoji; Emoticon
134, 136, 164, 168, 183, Graphics, 33, 36, 38, 40–43, 45–47,
184n1, 206, 254, 257, 259, 48n1, 155, 179, 181, 182, 214n2
263, 283–305, 355, 359n12
Fairclough, N., 4, 136, 137, 252, 253
Forums, 13, 32–34, 36, 39, 46, 137, H
141, 164, 213, 219, 220, 223, Hall, K., 10, 32, 194, 199, 224,
226, 228, 231, 232, 236, 355, 356
240–244, 246, 272, 318 Halliday, M. A. K., 72, 75, 89
394 Index
Harré, R., 141, 143, 224, 237, 316 group, 175, 178, 180, 333
Haugh, M., 342, 343, 348, 354, 370 masculine, 315–317, 320, 323,
Health, 12, 13, 74, 219–246 324, 328–333
Herring, S. C., 4, 5, 7, 8, 10, 12, 14, placement, 259, 270
25–30, 32–40, 42–47, 48n1, professional, 13, 251–273
49n3, 49n4, 49n5, 49n7, relational, 11, 194, 200, 201,
72–74, 167, 168, 195, 222, 204–206, 212
225–227, 243, 253, 257, 259, sexual, 314, 316, 317, 322, 331
283, 344–346, 356, 358n2 social, 11, 12, 15, 193, 198, 199,
Holmes, J., 212, 287, 343 257, 334n2, 350, 355
Human–computer interaction, 7, 28, Ideologies
41, 46 gender, 11, 12, 194, 199, 204,
Human–robot interaction, 7, 46 211–213
Human-to-human communication, language, 15, 17, 40, 385
26, 42 Image
Humor, 37, 109, 114, 118, 119, macro, 8, 105–129 (see also
123, 126, 128, 183, 196, 198, Memes)
206–208, 219, 224, 235, moving, 35, 72, 75, 139
286–287, 298–300, 304, 350, still, 35, 149, 153
351, 356, 383, 384, 386 Impoliteness, see Politeness
Hyperlinks, 33, 34, 144–146, 231, Instant messaging (IM), 28, 33, 39,
358n9 40, 46, 48n1, 164, 166–169,
Hypertexts, 9, 134, 275n6 173, 181, 182
See also Texting
Interactional sociolinguistics,
I 27, 74, 173
Identity Interactivity, 27, 177, 225, 226, 228,
announcement, 259, 268, 270, 229, 236–241, 243–246, 254
272, 274n5 Interface, 15, 31, 111, 121, 253,
construction, 10, 12, 13, 15, 16, 262, 269
180, 205, 221–225, 232, Internet, 18, 30–34, 36, 38, 39,
235–237, 240, 244, 251, 254, 43, 49n4, 71, 88, 105, 106,
257–262, 273, 283, 315–317, 110, 164, 182, 195, 196,
319–333, 345, 348, 350, 351, 214n2, 221–223, 290, 344,
353, 354, 356 351–353, 356
expert, 12, 13, 219, 221, 225, Internet Relay Chat (IRC), 31, 32,
232–237, 240–242, 245, 246 34, 39, 195
gender, 12, 183, 194–197, 199, See also Chats
201, 204–206, 208, 213, 289, Intertextuality, 34, 39, 284, 348,
316, 322, 333, 334 359n11
Index
395
J Memes, 7–9, 26, 43, 105–129, 196

Jewitt, C., 6, 8, 9, 107, 121, 124 Metapragmatic awareness, 368, 371,
Johnstone, 348, 353, 355, 356 377–380, 382, 384–386
Jones, R. H., 4, 5, 13, 15, 37, 98n2, Mobile phones, 33, 164, 167, 177,
138, 155, 164, 181 181, 223
Mroczek, K., 4, 5, 14, 15, 37,
73, 257
K Multimedia, 42, 258, 358n9
Kádár, D. Z., 342, 348, 354, 370 Multimodality, 6, 7, 9, 10, 17, 26,
Kinesics, 7, 73, 75–77, 79, 89–95 29, 36, 38, 47, 77, 78,
Kozinets, R. V., 198, 283, 289–291 105–129, 133, 138, 139, 155,
Kress, G., 6, 48n1, 75, 110, 134, 139 372, 373, 378, 381
MySpace, 195, 259
L
Language N
ideologies, 15, 17, 40, 385 Narratives, 6, 18, 109, 137, 140,
non-standard, 17, 368 197, 198, 207, 208, 219, 244,
play, 194, 195, 213, 384 255, 269, 270, 272, 275n6,
standard, 146, 181 320, 328, 331–334
use, 220, 355, 369 Netnography, 283
LinkedIn, 13, 254–257, 259, News, 38, 42, 85, 109, 133–155, 157,
263–265, 268–273 158n1, 262, 273n1, 345, 346
Literacy, 9, 109, 180, 184, 193 Newsgroups, 31, 32, 34, 39,
Locher, M. A., 10, 12, 13, 38, 219, 168, 223
220, 222–225, 227, 234, 341, Newspapers, 134–136, 142, 157,
342, 344, 358n5 183, 199, 206
Logos, 222, 230, 231, 358n3 Norris, S., 5, 7, 73, 76, 89
Lorenzo-Dus, N., 10, 74, 182, 343
O
M O’Halloran, K., 6, 10
Maíz Arévalo, C., 12, 14, 168, 180, Orthography, 27, 31, 39, 177, 373
182, 291, 294, 304
Marwick, A. E., 11, 196, 257, 261,
345, 346, 354, 355 P
Masculinity, 15, 175, 201, 202, 208, Page, R., 10, 11, 194, 253, 257, 261,
212, 313–334 344–346, 353
hegemonic, 15, 313–334 Papacharissi, Z., 10, 254, 259, 345,
See also Identity, masculine 348, 351, 354
396 Index
Participation, 9, 37, 49n3, 84, 109, Relevance theory, 8, 113, 117, 124
134, 137, 155, 165, 172, 226, Remediation, 9, 154, 273n1, 287,
260, 263, 284, 342, 386 302, 304
Petroni, S., 12, 13, 78, 97, Representation, 15, 16, 45, 73, 76,
252, 253 97, 105, 141, 169, 182, 253,
Politeness, 16, 37, 285, 341–357, 289, 299, 315, 316, 320
358n7, 373 Reputation, 231, 254, 256–262,
Popularity, 157, 206, 261–265, 264, 265, 268, 270, 271
269–272, 314, 332, 344 Review
Positioning, 13, 92, 93, 138, 143, Amazon, 12, 193
149, 151, 152, 154, 224, 237, consumer, 197–199
238, 241, 246, 323 opinion, 9, 133–155
Pragmatic competence, 94, 369, parody, 194, 197–200, 206–213,
380, 382 214n1
Pragmatics, 8, 16, 17, 27, 28, press, 134, 135, 138, 142, 154
36, 40, 48, 73, 94, 96, Robot, 7, 26, 42, 45–46
106, 107, 125, 128, 139–141,
172, 201, 284, 321, 329, 343,
368–370, 377, 378, 380, 382, S
385, 386 Self
Programmability, 262–265, 269, -attribution, 315, 316, 319, 320,
270, 272 322, 327, 329, 331, 335n3
Proxemics, 7, 73, 75–77, 79, 89–95 -branding, 13, 14, 251–273, 353
Punctuation, 86, 88, 177, -expression, 196, 259, 269, 271
374–376, 384 -image, 258, 261, 285, 384
-presentation, 10, 11, 13, 15,
16, 47, 251, 257–261, 264,
Q 266–269, 283, 314, 320,
Quotations, 9, 133–155, 158n2 325, 328, 341–357, 368,
383–386
-profiling, 253, 256, 273
R -promotion, 164, 253, 259,
Rapport, 179, 180, 298, 299, 350, 261, 271
383, 384, 386 -repair, 285, 288, 291, 296, 297,
Recontextualization, 9, 134, 299, 300, 304, 305
138, 140, 141, 154, 352, Semiotics, 4, 6, 7, 15, 28, 42, 43, 46,
359n11 72, 73, 75, 77, 78, 89, 91, 92,
Index
397
94, 95, 97, 109, 111, 116, Spilioti, T., 4–6, 8, 14, 15, 166, 170,
120, 121, 128, 134, 140 171, 178, 368
semiotic modes, 5, 7, 9, 10, 18, Squires, L., 46, 181, 344, 345, 349,
26, 41–43, 46, 47, 48n1, 106, 352, 357
120, 134, 137, 139 Stance, 106, 137, 141, 143, 147,
Sexting, 15, 16, 313–334, 335n4 149–152, 172, 202, 224,
Shareability, 251, 252 237, 243
Sindoni, M. G., 7, 28, 43, 46, Stereotypes, see under Gender
72–74, 77, 79, 88, 90, 92, Stickers, 8, 43, 45, 167, 168, 181
94–96
Skype, 7, 42, 71, 73, 76–79,
88, 89 T
Smartphone, 25, 71, 89–93, 165, Tannen, D., 98, 179, 198, 212, 283
166, 169, 181, 375 Textese, 177, 368, 382, 383
SMS, 33, 39, 44, 166, 167, Texting, 17, 165–167, 172, 179,
170, 172 181, 322, 323, 327, 332,
Social distance, 45, 47, 79, 89, 367–386
91, 92 See also Instant messaging (IM)
Social media, 10, 11, 14, 15, 40, Thurlow, C., 4, 5, 11, 14–17,
42, 43, 46, 133–137, 147, 37, 73, 165–167, 177,
154, 167, 173, 195, 210, 179, 180, 257, 314,
251–273, 314, 344, 345, 358n6, 368, 381
352, 355, 358n5, 371, Transcription, 7, 9, 71–97, 290
380, 386 Tsakona, V., 108, 111, 350,
Social networking sites (SNSs), 355, 356
11, 12, 14, 164, 195, 196, Tumblr, 195, 196
284, 313 Turn-taking, 32, 34, 39, 45, 77,
Social practice, 4, 195, 251, 256, 93, 228
331, 332, 334, 355 Twitter, 16, 40, 136, 144, 145, 147,
Social semiotics, 28, 75, 111, 116, 195, 257, 263, 341–357,
120, 121 358n4, 358n5, 358n7,
Sociolinguistics, 4, 10, 16, 27, 29, 359n12, 359n13, 359n14
34, 40, 73–75, 168, 169, 173,
198, 369, 370
Sociopragmatics, 9, 135, 136 V
Solidarity, 180, 285, 324, 348, 350 van Dijk, T., 11, 135, 319, 368
Spelling, 31, 85, 166, 375, 379, 380, van Leeuwen, T., 6, 17, 48n1, 75,
382, 384, 386 106, 110, 111, 120, 121, 137
398 Index
Video, 7–9, 25, 28, 33, 35, 36, Websites, 12, 13, 36, 39, 77, 89,
38–47, 48n1, 50n10, 71–97, 193, 197, 220, 222, 226, 227,
106, 134–139, 141, 142, 145, 230–232, 235, 237, 240, 245,
147–149, 154, 155, 157, 253, 246, 256, 273n1
258, 264, 289, 359n11 WhatsApp, 8, 71, 108, 164–184,
Virtual worlds, 7, 26, 33, 35, 39, 43, 184n1
45, 48n1, 367, 372
Y
W YouTube, 40, 42, 44, 257, 359n11
Web Yus, F., 7–9, 36, 37, 73, 107, 114,
1.0, 5, 26, 29, 32–35, 118, 124, 168, 172, 173,
39, 257 177–180
2.0, 5, 25, 26, 29, 36–38, 49n6,
72, 251, 257, 261
3.0, 49n6 Z
pre-, 5, 26, 29–32, 39 Zappavigna, M., 14, 73, 252, 259,
Webcam, 79, 92, 93 261, 263, 345, 346, 349–352

Patricia Bou - Analyzing Digital Discourse PDF

Uploaded by

Copyright:

Available Formats

Patricia Bou - Analyzing Digital Discourse PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Patricia Bou - Analyzing Digital Discourse PDF

Uploaded by

Copyright:

Available Formats

What topics does the book cover?

What topics does the book cover?

What perspectives does it take on digital discourse?

What perspectives does it take on digital discourse?

Edited by Patricia Bou-Franch and Pilar Garcés-Conejos Blitvich

Analyzing Digital Discourse

New Insights and Future Directions

“With contributions by established as well as new scholars, this volume pulls

ISBN 978-3-319-92662-9 ISBN 978-3-319-92663-6 (eBook)

Library of Congress Control Number: 2018948842

© The Editor(s) (if applicable) and The Author(s) 2019

Cover design by Tom Howey

Part I Introduction 1

1 Introduction to Analyzing Digital Discourse: New

Part II Past, Present and Future 23

2 The Coevolution of Computer-Mediated Communication

Part III Multimodality 69

3 “Of course I’m married!” Communicative Strategies and

4 Multimodality in Memes: A Cyberpragmatic Approach 105

5 Digital and Written Quotations in a News Text:

6 Emoticons in Relational Writing Practices on WhatsApp:

Part IV Face and Identity 191

7 From “My Manly Husband…” to “… Sitting Down

8 Linguistic Expert Creation in Online Health Practices 219

9 How Social Media Shape Identities and Discourses

10 Losing Face on Facebook: Linguistic Strategies to Repair

Part V Language and Media Ideologies 311

11 Sexting and Hegemonic Masculinity: Interrogating Male

12 Twitter, Politeness, Self-Presentation 341

13 Pedagogy, Audience, and Attitudes: Influencing University

Spiridoula Bella is Professor of Linguistics at the National and Kapodistrian

Antonio García-Gómez is Associate Professor of Linguistics at the University

Marie-Thérèse Rudolf von Rohr completed her Ph.D. in English linguistics at

Camilla Vásquez is Professor of Applied Linguistics in the Department of

Fig. 2.1 CMC reconceptualized as inherently multimodal 41

Table 2.1 The computer-mediated discourse analysis (CMDA) ‘toolkit’ 27

Table 8.1 Overview of our four data sets, ordered according to

© The Author(s) 2019 3

and identifying the ideologies that underpin the construction of digital

c­ ommunication therein. Thus, we should place “the concepts of multi-

(by Sindoni, Perez-Sabater, Johansson, and Yus) and a significant part of

for the analyst to relate microinteractions with societal ideologies at the

­implicatures. Consequently, the interpreting of a meme entails the pro-

Blitvich, 2018). Therefore, we see a unique opportunity here through

In their analysis of Facebook, for instance, Tagg and Seargeant (2014,

technologies have brought about a “demotic turn” (Turner, 2010) through

taining exchanges of letters between a professional persona and lay

architecture of the technology itself. In this way, the author addresses

identifying ways in which social processes underlying the formation and

semiotic modes and genres. Another interesting contribution to research

© The Author(s) 2019 25

discourse. It also tends to assume that online communication takes place

2.2 Computer-Mediated Discourse Analysis

Table 2.1 The computer-mediated discourse analysis (CMDA) ‘toolkit’

social behavior, reflecting a progressive broadening of focus from the micro

Over the years, modifications have been proposed to specific CMDA

My historical overview highlights connections between CMDA research

developments, choosing to study newer modes more often than older

2.3.2 Phase I: Pre-Web (1983–1993)

Part I Introduction 1

1 Introduction to Analyzing Digital Discourse: New

Part II Past, Present and Future 23

2 The Coevolution of Computer-Mediated Communication

Part III Multimodality 69

3 “Of course I’m married!” Communicative Strategies and

4 Multimodality in Memes: A Cyberpragmatic Approach 105

5 Digital and Written Quotations in a News Text:

6 Emoticons in Relational Writing Practices on WhatsApp:

Part IV Face and Identity 191

7 From “My Manly Husband…” to “… Sitting Down

8 Linguistic Expert Creation in Online Health Practices 219

9 How Social Media Shape Identities and Discourses

10 Losing Face on Facebook: Linguistic Strategies to Repair

Part V Language and Media Ideologies 311

11 Sexting and Hegemonic Masculinity: Interrogating Male

12 Twitter, Politeness, Self-Presentation 341

13 Pedagogy, Audience, and Attitudes: Influencing University

Fig. 2.1 CMC reconceptualized as inherently multimodal 41

Table 2.1 The computer-mediated discourse analysis (CMDA) ‘toolkit’ 27

c ommunication therein. Thus, we should place “the concepts of multi-

implicatures. Consequently, the interpreting of a meme entails the pro-

2.2 Computer-Mediated Discourse Analysis

2.3.2 Phase I: Pre-Web (1983–1993)

2.3.3 Phase II: Web 1.0 (1994–2004)

2.3.4 Phase III: Web 2.0 (2004–2017)6

and graphics • Deception, hoaxes, and • Critiques by

2.4.1 Interactive Multimodal Platforms

2.4.2 Graphical Communication

2.4.3 Telepresence Robot-Mediated Communication

2.5 Implications for CMDA

• What is students’ perception and understanding of both video-